From 2f19c81be4b54c72ebb6b67c47663e012f4a9c12 Mon Sep 17 00:00:00 2001
From: Klaus Post <klauspost@gmail.com>
Date: Thu, 24 Mar 2022 05:25:40 -0700
Subject: [PATCH] Reduce generated code (#185)

* Reduce generated code

Use a define (with hacks)
---
 .github/workflows/go.yml |     1 +
 _gen/cleanup.go          |    14 +-
 _gen/gen.go              |    20 +-
 galois_gen_amd64.go      |     2 +
 galois_gen_amd64.s       | 65114 ++++---------------------------------
 5 files changed, 6743 insertions(+), 58408 deletions(-)

diff --git a/.github/workflows/go.yml b/.github/workflows/go.yml
index 5451542..7a95a6b 100644
--- a/.github/workflows/go.yml
+++ b/.github/workflows/go.yml
@@ -87,4 +87,5 @@ jobs:
       run: go test -no-avx512 -no-avx2 -no-ssse3 -short -race .
 
     - name: Test Microarch v4
+      shell: bash {0}
       run: go run testlevel.go 4;if [ $? -eq 0 ]; then GOAMD64=v4 go test -no-avx512 ./...; else true; fi
diff --git a/_gen/cleanup.go b/_gen/cleanup.go
index bcd7964..6dd2a30 100644
--- a/_gen/cleanup.go
+++ b/_gen/cleanup.go
@@ -21,7 +21,19 @@ func main() {
 		if err != nil {
 			log.Fatalln(err)
 		}
-		data = bytes.Replace(data, []byte("\t// #"), []byte("#"), -1)
+		data = bytes.ReplaceAll(data, []byte("\t// #"), []byte("#"))
+		data = bytes.ReplaceAll(data, []byte("\t// @"), []byte(""))
+		data = bytes.ReplaceAll(data, []byte("VPTERNLOGQ"), []byte("XOR3WAY("))
+		split := bytes.Split(data, []byte("\n"))
+		// Add closing ')'
+		want := []byte("\tXOR3WAY(")
+		for i, b := range split {
+			if bytes.Contains(b, want) {
+				b = []byte(string(b) + ")")
+				split[i] = b
+			}
+		}
+		data = bytes.Join(split, []byte("\n"))
 		data, err = asmfmt.Format(bytes.NewBuffer(data))
 		if err != nil {
 			log.Fatalln(err)
diff --git a/_gen/gen.go b/_gen/gen.go
index 40f1227..50b8043 100644
--- a/_gen/gen.go
+++ b/_gen/gen.go
@@ -38,6 +38,17 @@ func main() {
 	Constraint(buildtags.Not("nogen").ToConstraint())
 	Constraint(buildtags.Term("gc").ToConstraint())
 
+	TEXT("_dummy_", 0, "func()")
+	Comment("#ifdef GOAMD64_v4")
+	Comment("#define XOR3WAY(ignore, a, b, dst)\\")
+	Comment("@\tVPTERNLOGD $0x96, a, b, dst")
+	Comment("#else")
+	Comment("#define XOR3WAY(ignore, a, b, dst)\\")
+	Comment("@\tVPXOR a, dst, dst\\")
+	Comment("@\tVPXOR b, dst, dst")
+	Comment("#endif")
+	RET()
+
 	const perLoopBits = 6
 	const perLoop = 1 << perLoopBits
 
@@ -123,13 +134,8 @@ func galMulSlicesAvx2Xor(matrix []byte, in, out [][]byte, start, stop int) int {
 
 // VPXOR3way will 3-way xor a and b and dst.
 func VPXOR3way(a, b, dst reg.VecVirtual) {
-	Comment("#ifdef GOAMD64_v4")
-	// AVX512F and AVX512VL required
-	VPTERNLOGD(U8(0x96), a, b, dst)
-	Comment("#else")
-	VPXOR(a, dst, dst) // dst = a^dst
-	VPXOR(b, dst, dst) // dst = (a^dst)^b
-	Comment("#endif")
+	// VPTERNLOGQ is replaced by XOR3WAY - we just use an equivalent operation
+	VPTERNLOGQ(U8(0), a, b, dst)
 }
 
 func genMulAvx2(name string, inputs int, outputs int, xor bool) {
diff --git a/galois_gen_amd64.go b/galois_gen_amd64.go
index 817c7ea..72c4ca4 100644
--- a/galois_gen_amd64.go
+++ b/galois_gen_amd64.go
@@ -5,6 +5,8 @@
 
 package reedsolomon
 
+func _dummy_()
+
 // mulAvxTwo_1x1 takes 1 inputs and produces 1 outputs.
 // The output is initialized to 0.
 //go:noescape
diff --git a/galois_gen_amd64.s b/galois_gen_amd64.s
index 5d24082..97ead9c 100644
--- a/galois_gen_amd64.s
+++ b/galois_gen_amd64.s
@@ -5,6 +5,20 @@
 
 #include "textflag.h"
 
+// func _dummy_()
+TEXT ·_dummy_(SB), $0
+#ifdef GOAMD64_v4
+#define XOR3WAY(ignore, a, b, dst) \
+	VPTERNLOGD $0x96, a, b, dst
+
+#else
+#define XOR3WAY(ignore, a, b, dst) \
+	VPXOR a, dst, dst \
+	VPXOR b, dst, dst
+
+#endif
+	RET
+
 // func mulAvxTwo_1x1(matrix []byte, in [][]byte, out [][]byte, start int, n int)
 // Requires: AVX, AVX2, SSE2
 TEXT ·mulAvxTwo_1x1(SB), NOSPLIT, $0-88
@@ -154,15 +168,8 @@ mulAvxTwo_1x1Xor_loop:
 	VMOVDQU (DX), Y2
 	VPSHUFB Y4, Y0, Y4
 	VPSHUFB Y5, Y1, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y2, (DX)
 	ADDQ    $0x20, DX
@@ -223,23 +230,9 @@ mulAvxTwo_1x1_64Xor_loop:
 	VPSHUFB Y7, Y0, Y7
 	VPSHUFB Y6, Y1, Y6
 	VPSHUFB Y8, Y1, Y8
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y2, (DX)
 	VMOVDQU Y3, 32(DX)
@@ -428,27 +421,12 @@ mulAvxTwo_1x2Xor_loop:
 	VMOVDQU (BX), Y4
 	VPSHUFB Y9, Y0, Y7
 	VPSHUFB Y10, Y1, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (DX), Y5
 	VPSHUFB Y9, Y2, Y7
 	VPSHUFB Y10, Y3, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y4, (BX)
 	ADDQ    $0x20, BX
@@ -515,46 +493,17 @@ mulAvxTwo_1x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (SI)
 	VMOVDQU Y1, 32(SI)
@@ -772,39 +721,16 @@ mulAvxTwo_1x3Xor_loop:
 	VMOVDQU (BX), Y6
 	VPSHUFB Y12, Y0, Y10
 	VPSHUFB Y13, Y1, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU (SI), Y7
 	VPSHUFB Y12, Y2, Y10
 	VPSHUFB Y13, Y3, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU (DX), Y8
 	VPSHUFB Y12, Y4, Y10
 	VPSHUFB Y13, Y5, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y6, (BX)
 	ADDQ    $0x20, BX
@@ -877,69 +803,25 @@ mulAvxTwo_1x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (SI)
 	VMOVDQU Y1, 32(SI)
@@ -1081,57 +963,26 @@ mulAvxTwo_1x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (BX), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -1283,71 +1134,32 @@ mulAvxTwo_1x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (BX), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -1512,85 +1324,38 @@ mulAvxTwo_1x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (BX), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -1768,99 +1533,44 @@ mulAvxTwo_1x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R11), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (BX), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -2051,113 +1761,50 @@ mulAvxTwo_1x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU (R11), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU (R12), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU (BX), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -2361,127 +2008,56 @@ mulAvxTwo_1x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU (R11), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU (R12), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU (R13), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU (BX), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -2698,141 +2274,62 @@ mulAvxTwo_1x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU (R11), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU (R12), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU (R13), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU (R14), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU (BX), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	VMOVDQU Y0, (SI)
 	ADDQ    $0x20, SI
@@ -2914,15 +2411,8 @@ mulAvxTwo_2x1_loop:
 	VPAND   Y5, Y7, Y7
 	VPSHUFB Y6, Y2, Y6
 	VPSHUFB Y7, Y3, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y4, (BX)
 	ADDQ    $0x20, BX
@@ -3000,23 +2490,9 @@ mulAvxTwo_2x1_64_loop:
 	VPSHUFB Y9, Y2, Y9
 	VPSHUFB Y8, Y3, Y8
 	VPSHUFB Y10, Y3, Y10
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y4, (BX)
 	VMOVDQU Y5, 32(BX)
@@ -3072,15 +2548,8 @@ mulAvxTwo_2x1Xor_loop:
 	VMOVDQU (BX), Y4
 	VPSHUFB Y6, Y0, Y6
 	VPSHUFB Y7, Y1, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (CX), Y6
 	ADDQ    $0x20, CX
@@ -3089,15 +2558,8 @@ mulAvxTwo_2x1Xor_loop:
 	VPAND   Y5, Y7, Y7
 	VPSHUFB Y6, Y2, Y6
 	VPSHUFB Y7, Y3, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y4, (BX)
 	ADDQ    $0x20, BX
@@ -3162,23 +2624,9 @@ mulAvxTwo_2x1_64Xor_loop:
 	VPSHUFB Y9, Y0, Y9
 	VPSHUFB Y8, Y1, Y8
 	VPSHUFB Y10, Y1, Y10
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (CX), Y7
 	VMOVDQU 32(CX), Y9
@@ -3193,23 +2641,9 @@ mulAvxTwo_2x1_64Xor_loop:
 	VPSHUFB Y9, Y2, Y9
 	VPSHUFB Y8, Y3, Y8
 	VPSHUFB Y10, Y3, Y10
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y4, (BX)
 	VMOVDQU Y5, 32(BX)
@@ -3283,26 +2717,11 @@ mulAvxTwo_2x2_loop:
 	VPAND   Y10, Y14, Y14
 	VPSHUFB Y13, Y4, Y11
 	VPSHUFB Y14, Y5, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VPSHUFB Y13, Y6, Y11
 	VPSHUFB Y14, Y7, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y8, (SI)
 	ADDQ    $0x20, SI
@@ -3392,46 +2811,17 @@ mulAvxTwo_2x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -3496,27 +2886,12 @@ mulAvxTwo_2x2Xor_loop:
 	VMOVDQU (SI), Y8
 	VPSHUFB Y13, Y0, Y11
 	VPSHUFB Y14, Y1, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU (BX), Y9
 	VPSHUFB Y13, Y2, Y11
 	VPSHUFB Y14, Y3, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (CX), Y13
 	ADDQ    $0x20, CX
@@ -3525,26 +2900,11 @@ mulAvxTwo_2x2Xor_loop:
 	VPAND   Y10, Y14, Y14
 	VPSHUFB Y13, Y4, Y11
 	VPSHUFB Y14, Y5, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VPSHUFB Y13, Y6, Y11
 	VPSHUFB Y14, Y7, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y8, (SI)
 	ADDQ    $0x20, SI
@@ -3613,46 +2973,17 @@ mulAvxTwo_2x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -3669,46 +3000,17 @@ mulAvxTwo_2x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -3790,41 +3092,18 @@ mulAvxTwo_2x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -3926,69 +3205,25 @@ mulAvxTwo_2x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -4052,43 +3287,20 @@ mulAvxTwo_2x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (SI), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -4099,41 +3311,18 @@ mulAvxTwo_2x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -4208,69 +3397,25 @@ mulAvxTwo_2x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -4287,69 +3432,25 @@ mulAvxTwo_2x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -4441,54 +3542,23 @@ mulAvxTwo_2x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -4553,57 +3623,26 @@ mulAvxTwo_2x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (SI), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -4614,54 +3653,23 @@ mulAvxTwo_2x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -4759,67 +3767,28 @@ mulAvxTwo_2x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -4888,71 +3857,32 @@ mulAvxTwo_2x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (SI), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -4963,67 +3893,28 @@ mulAvxTwo_2x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -5130,80 +4021,33 @@ mulAvxTwo_2x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -5276,85 +4120,38 @@ mulAvxTwo_2x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (SI), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -5365,80 +4162,33 @@ mulAvxTwo_2x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -5554,93 +4304,38 @@ mulAvxTwo_2x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -5717,99 +4412,44 @@ mulAvxTwo_2x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R12), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (SI), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -5820,93 +4460,38 @@ mulAvxTwo_2x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -6031,106 +4616,43 @@ mulAvxTwo_2x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -6211,113 +4733,50 @@ mulAvxTwo_2x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU (R12), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU (R13), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU (SI), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -6328,106 +4787,43 @@ mulAvxTwo_2x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -6561,119 +4957,48 @@ mulAvxTwo_2x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -6758,127 +5083,56 @@ mulAvxTwo_2x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU (R12), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU (R13), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU (R14), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU (SI), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -6889,119 +5143,48 @@ mulAvxTwo_2x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -7144,132 +5327,53 @@ mulAvxTwo_2x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -7358,141 +5462,62 @@ mulAvxTwo_2x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU (R12), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU (R13), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU (R14), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU (R15), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU (SI), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -7503,132 +5528,53 @@ mulAvxTwo_2x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -7714,15 +5660,8 @@ mulAvxTwo_3x1_loop:
 	VPAND   Y7, Y9, Y9
 	VPSHUFB Y8, Y2, Y8
 	VPSHUFB Y9, Y3, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (CX), Y8
 	ADDQ    $0x20, CX
@@ -7731,15 +5670,8 @@ mulAvxTwo_3x1_loop:
 	VPAND   Y7, Y9, Y9
 	VPSHUFB Y8, Y4, Y8
 	VPSHUFB Y9, Y5, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y6, (SI)
 	ADDQ    $0x20, SI
@@ -7819,23 +5751,9 @@ mulAvxTwo_3x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -7852,23 +5770,9 @@ mulAvxTwo_3x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -7928,15 +5832,8 @@ mulAvxTwo_3x1Xor_loop:
 	VMOVDQU (SI), Y6
 	VPSHUFB Y8, Y0, Y8
 	VPSHUFB Y9, Y1, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (BX), Y8
 	ADDQ    $0x20, BX
@@ -7945,15 +5842,8 @@ mulAvxTwo_3x1Xor_loop:
 	VPAND   Y7, Y9, Y9
 	VPSHUFB Y8, Y2, Y8
 	VPSHUFB Y9, Y3, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (CX), Y8
 	ADDQ    $0x20, CX
@@ -7962,15 +5852,8 @@ mulAvxTwo_3x1Xor_loop:
 	VPAND   Y7, Y9, Y9
 	VPSHUFB Y8, Y4, Y8
 	VPSHUFB Y9, Y5, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y6, (SI)
 	ADDQ    $0x20, SI
@@ -8035,23 +5918,9 @@ mulAvxTwo_3x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -8068,23 +5937,9 @@ mulAvxTwo_3x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -8101,23 +5956,9 @@ mulAvxTwo_3x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (DI)
 	VMOVDQU Y1, 32(DI)
@@ -8191,28 +6032,13 @@ mulAvxTwo_3x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -8223,28 +6049,13 @@ mulAvxTwo_3x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -8336,46 +6147,17 @@ mulAvxTwo_3x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -8392,46 +6174,17 @@ mulAvxTwo_3x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -8492,29 +6245,14 @@ mulAvxTwo_3x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (DI), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -8525,28 +6263,13 @@ mulAvxTwo_3x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -8557,28 +6280,13 @@ mulAvxTwo_3x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -8649,46 +6357,17 @@ mulAvxTwo_3x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -8705,46 +6384,17 @@ mulAvxTwo_3x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -8761,46 +6411,17 @@ mulAvxTwo_3x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -8884,41 +6505,18 @@ mulAvxTwo_3x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -8929,41 +6527,18 @@ mulAvxTwo_3x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -9067,69 +6642,25 @@ mulAvxTwo_3x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -9146,69 +6677,25 @@ mulAvxTwo_3x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -9274,43 +6761,20 @@ mulAvxTwo_3x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (DI), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -9321,41 +6785,18 @@ mulAvxTwo_3x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -9366,41 +6807,18 @@ mulAvxTwo_3x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -9477,69 +6895,25 @@ mulAvxTwo_3x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -9556,69 +6930,25 @@ mulAvxTwo_3x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -9635,69 +6965,25 @@ mulAvxTwo_3x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -9791,54 +7077,23 @@ mulAvxTwo_3x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -9849,54 +7104,23 @@ mulAvxTwo_3x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -9963,57 +7187,26 @@ mulAvxTwo_3x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (DI), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -10024,54 +7217,23 @@ mulAvxTwo_3x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -10082,54 +7244,23 @@ mulAvxTwo_3x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -10229,67 +7360,28 @@ mulAvxTwo_3x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -10300,67 +7392,28 @@ mulAvxTwo_3x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -10431,71 +7484,32 @@ mulAvxTwo_3x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (DI), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -10506,67 +7520,28 @@ mulAvxTwo_3x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -10577,67 +7552,28 @@ mulAvxTwo_3x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -10746,80 +7682,33 @@ mulAvxTwo_3x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -10830,80 +7719,33 @@ mulAvxTwo_3x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -10978,85 +7820,38 @@ mulAvxTwo_3x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R12), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (DI), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -11067,80 +7862,33 @@ mulAvxTwo_3x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -11151,80 +7899,33 @@ mulAvxTwo_3x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -11342,93 +8043,38 @@ mulAvxTwo_3x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -11439,93 +8085,38 @@ mulAvxTwo_3x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -11604,99 +8195,44 @@ mulAvxTwo_3x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R12), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R13), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (DI), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -11707,93 +8243,38 @@ mulAvxTwo_3x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -11804,93 +8285,38 @@ mulAvxTwo_3x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -12017,106 +8443,43 @@ mulAvxTwo_3x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -12127,106 +8490,43 @@ mulAvxTwo_3x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -12309,113 +8609,50 @@ mulAvxTwo_3x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU (R12), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU (R13), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU (R14), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU (DI), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -12426,106 +8663,43 @@ mulAvxTwo_3x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -12536,106 +8710,43 @@ mulAvxTwo_3x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -12771,119 +8882,48 @@ mulAvxTwo_3x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -12894,119 +8934,48 @@ mulAvxTwo_3x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -13093,127 +9062,56 @@ mulAvxTwo_3x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU (R12), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU (R13), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU (R14), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU (R15), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU (DI), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -13224,119 +9122,48 @@ mulAvxTwo_3x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -13347,119 +9174,48 @@ mulAvxTwo_3x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -13606,132 +9362,53 @@ mulAvxTwo_3x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (AX), Y13
 	ADDQ    $0x20, AX
@@ -13742,132 +9419,53 @@ mulAvxTwo_3x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -13960,141 +9558,62 @@ mulAvxTwo_3x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU (R12), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU (R13), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU (R14), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU (R15), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU (SI), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (BX), Y13
 	ADDQ    $0x20, BX
@@ -14105,132 +9624,53 @@ mulAvxTwo_3x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (AX), Y13
 	ADDQ    $0x20, AX
@@ -14241,132 +9681,53 @@ mulAvxTwo_3x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	VMOVDQU Y0, (DI)
 	ADDQ    $0x20, DI
@@ -14456,15 +9817,8 @@ mulAvxTwo_4x1_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y2, Y10
 	VPSHUFB Y11, Y3, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -14473,15 +9827,8 @@ mulAvxTwo_4x1_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y4, Y10
 	VPSHUFB Y11, Y5, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (CX), Y10
 	ADDQ    $0x20, CX
@@ -14490,15 +9837,8 @@ mulAvxTwo_4x1_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y6, Y10
 	VPSHUFB Y11, Y7, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y8, (DI)
 	ADDQ    $0x20, DI
@@ -14580,23 +9920,9 @@ mulAvxTwo_4x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -14613,23 +9939,9 @@ mulAvxTwo_4x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -14646,23 +9958,9 @@ mulAvxTwo_4x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -14726,15 +10024,8 @@ mulAvxTwo_4x1Xor_loop:
 	VMOVDQU (DI), Y8
 	VPSHUFB Y10, Y0, Y10
 	VPSHUFB Y11, Y1, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (BX), Y10
 	ADDQ    $0x20, BX
@@ -14743,15 +10034,8 @@ mulAvxTwo_4x1Xor_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y2, Y10
 	VPSHUFB Y11, Y3, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -14760,15 +10044,8 @@ mulAvxTwo_4x1Xor_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y4, Y10
 	VPSHUFB Y11, Y5, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (CX), Y10
 	ADDQ    $0x20, CX
@@ -14777,15 +10054,8 @@ mulAvxTwo_4x1Xor_loop:
 	VPAND   Y9, Y11, Y11
 	VPSHUFB Y10, Y6, Y10
 	VPSHUFB Y11, Y7, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y8, (DI)
 	ADDQ    $0x20, DI
@@ -14852,23 +10122,9 @@ mulAvxTwo_4x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -14885,23 +10141,9 @@ mulAvxTwo_4x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -14918,23 +10160,9 @@ mulAvxTwo_4x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -14951,23 +10179,9 @@ mulAvxTwo_4x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R8)
 	VMOVDQU Y1, 32(R8)
@@ -15043,28 +10257,13 @@ mulAvxTwo_4x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -15075,28 +10274,13 @@ mulAvxTwo_4x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -15107,28 +10291,13 @@ mulAvxTwo_4x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -15222,46 +10391,17 @@ mulAvxTwo_4x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -15278,46 +10418,17 @@ mulAvxTwo_4x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -15334,46 +10445,17 @@ mulAvxTwo_4x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -15436,29 +10518,14 @@ mulAvxTwo_4x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R8), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -15469,28 +10536,13 @@ mulAvxTwo_4x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -15501,28 +10553,13 @@ mulAvxTwo_4x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -15533,28 +10570,13 @@ mulAvxTwo_4x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -15627,46 +10649,17 @@ mulAvxTwo_4x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -15683,46 +10676,17 @@ mulAvxTwo_4x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -15739,46 +10703,17 @@ mulAvxTwo_4x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -15795,46 +10730,17 @@ mulAvxTwo_4x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -15920,41 +10826,18 @@ mulAvxTwo_4x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -15965,41 +10848,18 @@ mulAvxTwo_4x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -16010,41 +10870,18 @@ mulAvxTwo_4x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -16150,69 +10987,25 @@ mulAvxTwo_4x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -16229,69 +11022,25 @@ mulAvxTwo_4x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -16308,69 +11057,25 @@ mulAvxTwo_4x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -16438,43 +11143,20 @@ mulAvxTwo_4x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R8), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -16485,41 +11167,18 @@ mulAvxTwo_4x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -16530,41 +11189,18 @@ mulAvxTwo_4x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -16575,41 +11211,18 @@ mulAvxTwo_4x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -16688,69 +11301,25 @@ mulAvxTwo_4x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -16767,69 +11336,25 @@ mulAvxTwo_4x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -16846,69 +11371,25 @@ mulAvxTwo_4x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -16925,69 +11406,25 @@ mulAvxTwo_4x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -17083,54 +11520,23 @@ mulAvxTwo_4x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -17141,54 +11547,23 @@ mulAvxTwo_4x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -17199,54 +11574,23 @@ mulAvxTwo_4x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -17315,57 +11659,26 @@ mulAvxTwo_4x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R8), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -17376,54 +11689,23 @@ mulAvxTwo_4x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -17434,54 +11716,23 @@ mulAvxTwo_4x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -17492,54 +11743,23 @@ mulAvxTwo_4x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -17641,67 +11861,28 @@ mulAvxTwo_4x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -17712,67 +11893,28 @@ mulAvxTwo_4x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -17783,67 +11925,28 @@ mulAvxTwo_4x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -17916,71 +12019,32 @@ mulAvxTwo_4x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (R8), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -17991,67 +12055,28 @@ mulAvxTwo_4x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -18062,67 +12087,28 @@ mulAvxTwo_4x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -18133,67 +12119,28 @@ mulAvxTwo_4x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -18304,80 +12251,33 @@ mulAvxTwo_4x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -18388,80 +12288,33 @@ mulAvxTwo_4x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -18472,80 +12325,33 @@ mulAvxTwo_4x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -18622,85 +12428,38 @@ mulAvxTwo_4x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R13), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (R8), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -18711,80 +12470,33 @@ mulAvxTwo_4x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -18795,80 +12507,33 @@ mulAvxTwo_4x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -18879,80 +12544,33 @@ mulAvxTwo_4x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -19072,93 +12690,38 @@ mulAvxTwo_4x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -19169,93 +12732,38 @@ mulAvxTwo_4x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -19266,93 +12774,38 @@ mulAvxTwo_4x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -19433,99 +12886,44 @@ mulAvxTwo_4x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R13), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R14), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (R8), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -19536,93 +12934,38 @@ mulAvxTwo_4x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -19633,93 +12976,38 @@ mulAvxTwo_4x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -19730,93 +13018,38 @@ mulAvxTwo_4x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -19945,106 +13178,43 @@ mulAvxTwo_4x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -20055,106 +13225,43 @@ mulAvxTwo_4x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -20165,106 +13272,43 @@ mulAvxTwo_4x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -20349,113 +13393,50 @@ mulAvxTwo_4x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU (R13), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU (R14), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU (R15), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU (R8), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -20466,106 +13447,43 @@ mulAvxTwo_4x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -20576,106 +13494,43 @@ mulAvxTwo_4x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -20686,106 +13541,43 @@ mulAvxTwo_4x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -20925,119 +13717,48 @@ mulAvxTwo_4x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -21048,119 +13769,48 @@ mulAvxTwo_4x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (AX), Y12
 	ADDQ    $0x20, AX
@@ -21171,119 +13821,48 @@ mulAvxTwo_4x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -21374,127 +13953,56 @@ mulAvxTwo_4x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU (R12), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU (R13), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU (R14), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU (R15), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU (DI), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (BX), Y12
 	ADDQ    $0x20, BX
@@ -21505,119 +14013,48 @@ mulAvxTwo_4x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -21628,119 +14065,48 @@ mulAvxTwo_4x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (AX), Y12
 	ADDQ    $0x20, AX
@@ -21751,119 +14117,48 @@ mulAvxTwo_4x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	VMOVDQU Y0, (R8)
 	ADDQ    $0x20, R8
@@ -21988,132 +14283,53 @@ mulAvxTwo_4x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -22124,132 +14340,53 @@ mulAvxTwo_4x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -22260,132 +14397,53 @@ mulAvxTwo_4x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R8), R10
 	VMOVDQU Y0, (R10)(R9*1)
@@ -22458,150 +14516,71 @@ mulAvxTwo_4x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R8), R10
 	VMOVDQU (R10)(R9*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R8), R10
 	VMOVDQU (R10)(R9*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R8), R10
 	VMOVDQU (R10)(R9*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R8), R10
 	VMOVDQU (R10)(R9*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R8), R10
 	VMOVDQU (R10)(R9*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R8), R10
 	VMOVDQU (R10)(R9*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R8), R10
 	VMOVDQU (R10)(R9*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R8), R10
 	VMOVDQU (R10)(R9*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R8), R10
 	VMOVDQU (R10)(R9*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -22612,132 +14591,53 @@ mulAvxTwo_4x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -22748,132 +14648,53 @@ mulAvxTwo_4x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -22884,132 +14705,53 @@ mulAvxTwo_4x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R8), R10
 	VMOVDQU Y0, (R10)(R9*1)
@@ -23104,15 +14846,8 @@ mulAvxTwo_5x1_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y2, Y12
 	VPSHUFB Y13, Y3, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -23121,15 +14856,8 @@ mulAvxTwo_5x1_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y4, Y12
 	VPSHUFB Y13, Y5, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -23138,15 +14866,8 @@ mulAvxTwo_5x1_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y6, Y12
 	VPSHUFB Y13, Y7, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (CX), Y12
 	ADDQ    $0x20, CX
@@ -23155,15 +14876,8 @@ mulAvxTwo_5x1_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y8, Y12
 	VPSHUFB Y13, Y9, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y10, (R8)
 	ADDQ    $0x20, R8
@@ -23247,23 +14961,9 @@ mulAvxTwo_5x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -23280,23 +14980,9 @@ mulAvxTwo_5x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -23313,23 +14999,9 @@ mulAvxTwo_5x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -23346,23 +15018,9 @@ mulAvxTwo_5x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -23430,15 +15088,8 @@ mulAvxTwo_5x1Xor_loop:
 	VMOVDQU (R8), Y10
 	VPSHUFB Y12, Y0, Y12
 	VPSHUFB Y13, Y1, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (BX), Y12
 	ADDQ    $0x20, BX
@@ -23447,15 +15098,8 @@ mulAvxTwo_5x1Xor_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y2, Y12
 	VPSHUFB Y13, Y3, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -23464,15 +15108,8 @@ mulAvxTwo_5x1Xor_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y4, Y12
 	VPSHUFB Y13, Y5, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -23481,15 +15118,8 @@ mulAvxTwo_5x1Xor_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y6, Y12
 	VPSHUFB Y13, Y7, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (CX), Y12
 	ADDQ    $0x20, CX
@@ -23498,15 +15128,8 @@ mulAvxTwo_5x1Xor_loop:
 	VPAND   Y11, Y13, Y13
 	VPSHUFB Y12, Y8, Y12
 	VPSHUFB Y13, Y9, Y13
+	XOR3WAY( $0x00, Y12, Y13, Y10)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y12, Y13, Y10
-
-#else
-	VPXOR Y12, Y10, Y10
-	VPXOR Y13, Y10, Y10
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y10, (R8)
 	ADDQ    $0x20, R8
@@ -23575,23 +15198,9 @@ mulAvxTwo_5x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -23608,23 +15217,9 @@ mulAvxTwo_5x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -23641,23 +15236,9 @@ mulAvxTwo_5x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -23674,23 +15255,9 @@ mulAvxTwo_5x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -23707,23 +15274,9 @@ mulAvxTwo_5x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R9)
 	VMOVDQU Y1, 32(R9)
@@ -23801,28 +15354,13 @@ mulAvxTwo_5x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -23833,28 +15371,13 @@ mulAvxTwo_5x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -23865,28 +15388,13 @@ mulAvxTwo_5x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -23897,28 +15405,13 @@ mulAvxTwo_5x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -24014,46 +15507,17 @@ mulAvxTwo_5x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -24070,46 +15534,17 @@ mulAvxTwo_5x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -24126,46 +15561,17 @@ mulAvxTwo_5x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -24182,46 +15588,17 @@ mulAvxTwo_5x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -24286,29 +15663,14 @@ mulAvxTwo_5x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R9), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -24319,28 +15681,13 @@ mulAvxTwo_5x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -24351,28 +15698,13 @@ mulAvxTwo_5x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -24383,28 +15715,13 @@ mulAvxTwo_5x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -24415,28 +15732,13 @@ mulAvxTwo_5x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -24511,46 +15813,17 @@ mulAvxTwo_5x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -24567,46 +15840,17 @@ mulAvxTwo_5x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -24623,46 +15867,17 @@ mulAvxTwo_5x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -24679,46 +15894,17 @@ mulAvxTwo_5x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -24735,46 +15921,17 @@ mulAvxTwo_5x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -24862,41 +16019,18 @@ mulAvxTwo_5x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -24907,41 +16041,18 @@ mulAvxTwo_5x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -24952,41 +16063,18 @@ mulAvxTwo_5x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -24997,41 +16085,18 @@ mulAvxTwo_5x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -25139,69 +16204,25 @@ mulAvxTwo_5x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -25218,69 +16239,25 @@ mulAvxTwo_5x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -25297,69 +16274,25 @@ mulAvxTwo_5x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -25376,69 +16309,25 @@ mulAvxTwo_5x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -25508,43 +16397,20 @@ mulAvxTwo_5x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R9), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -25555,41 +16421,18 @@ mulAvxTwo_5x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -25600,41 +16443,18 @@ mulAvxTwo_5x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -25645,41 +16465,18 @@ mulAvxTwo_5x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -25690,41 +16487,18 @@ mulAvxTwo_5x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -25805,69 +16579,25 @@ mulAvxTwo_5x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -25884,69 +16614,25 @@ mulAvxTwo_5x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -25963,69 +16649,25 @@ mulAvxTwo_5x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -26042,69 +16684,25 @@ mulAvxTwo_5x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -26121,69 +16719,25 @@ mulAvxTwo_5x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -26281,54 +16835,23 @@ mulAvxTwo_5x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -26339,54 +16862,23 @@ mulAvxTwo_5x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -26397,54 +16889,23 @@ mulAvxTwo_5x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -26455,54 +16916,23 @@ mulAvxTwo_5x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -26573,57 +17003,26 @@ mulAvxTwo_5x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R9), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -26634,54 +17033,23 @@ mulAvxTwo_5x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -26692,54 +17060,23 @@ mulAvxTwo_5x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -26750,54 +17087,23 @@ mulAvxTwo_5x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -26808,54 +17114,23 @@ mulAvxTwo_5x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -26959,67 +17234,28 @@ mulAvxTwo_5x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -27030,67 +17266,28 @@ mulAvxTwo_5x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -27101,67 +17298,28 @@ mulAvxTwo_5x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -27172,67 +17330,28 @@ mulAvxTwo_5x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -27307,71 +17426,32 @@ mulAvxTwo_5x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R13), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (R9), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -27382,67 +17462,28 @@ mulAvxTwo_5x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -27453,67 +17494,28 @@ mulAvxTwo_5x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -27524,67 +17526,28 @@ mulAvxTwo_5x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -27595,67 +17558,28 @@ mulAvxTwo_5x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -27768,80 +17692,33 @@ mulAvxTwo_5x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -27852,80 +17729,33 @@ mulAvxTwo_5x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -27936,80 +17766,33 @@ mulAvxTwo_5x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -28020,80 +17803,33 @@ mulAvxTwo_5x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -28172,85 +17908,38 @@ mulAvxTwo_5x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R13), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R14), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (R9), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -28261,80 +17950,33 @@ mulAvxTwo_5x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -28345,80 +17987,33 @@ mulAvxTwo_5x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -28429,80 +18024,33 @@ mulAvxTwo_5x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -28513,80 +18061,33 @@ mulAvxTwo_5x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -28708,93 +18209,38 @@ mulAvxTwo_5x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -28805,93 +18251,38 @@ mulAvxTwo_5x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -28902,93 +18293,38 @@ mulAvxTwo_5x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -28999,93 +18335,38 @@ mulAvxTwo_5x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -29168,99 +18449,44 @@ mulAvxTwo_5x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R13), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R14), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R15), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (R9), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -29271,93 +18497,38 @@ mulAvxTwo_5x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -29368,93 +18539,38 @@ mulAvxTwo_5x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -29465,93 +18581,38 @@ mulAvxTwo_5x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -29562,93 +18623,38 @@ mulAvxTwo_5x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -29781,106 +18787,43 @@ mulAvxTwo_5x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -29891,106 +18834,43 @@ mulAvxTwo_5x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -30001,106 +18881,43 @@ mulAvxTwo_5x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (AX), Y11
 	ADDQ    $0x20, AX
@@ -30111,106 +18928,43 @@ mulAvxTwo_5x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -30299,113 +19053,50 @@ mulAvxTwo_5x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU (R13), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU (R14), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU (R15), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU (R8), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (BX), Y11
 	ADDQ    $0x20, BX
@@ -30416,106 +19107,43 @@ mulAvxTwo_5x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -30526,106 +19154,43 @@ mulAvxTwo_5x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -30636,106 +19201,43 @@ mulAvxTwo_5x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (AX), Y11
 	ADDQ    $0x20, AX
@@ -30746,106 +19248,43 @@ mulAvxTwo_5x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	VMOVDQU Y0, (R9)
 	ADDQ    $0x20, R9
@@ -30965,119 +19404,48 @@ mulAvxTwo_5x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -31088,119 +19456,48 @@ mulAvxTwo_5x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -31211,119 +19508,48 @@ mulAvxTwo_5x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -31334,119 +19560,48 @@ mulAvxTwo_5x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R9), R11
 	VMOVDQU Y0, (R11)(R10*1)
@@ -31519,135 +19674,64 @@ mulAvxTwo_5x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R9), R11
 	VMOVDQU (R11)(R10*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R9), R11
 	VMOVDQU (R11)(R10*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R9), R11
 	VMOVDQU (R11)(R10*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R9), R11
 	VMOVDQU (R11)(R10*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R9), R11
 	VMOVDQU (R11)(R10*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R9), R11
 	VMOVDQU (R11)(R10*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R9), R11
 	VMOVDQU (R11)(R10*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R9), R11
 	VMOVDQU (R11)(R10*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -31658,119 +19742,48 @@ mulAvxTwo_5x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -31781,119 +19794,48 @@ mulAvxTwo_5x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -31904,119 +19846,48 @@ mulAvxTwo_5x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -32027,119 +19898,48 @@ mulAvxTwo_5x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R9), R11
 	VMOVDQU Y0, (R11)(R10*1)
@@ -32267,132 +20067,53 @@ mulAvxTwo_5x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -32403,132 +20124,53 @@ mulAvxTwo_5x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -32539,132 +20181,53 @@ mulAvxTwo_5x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -32675,132 +20238,53 @@ mulAvxTwo_5x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R9), R11
 	VMOVDQU Y0, (R11)(R10*1)
@@ -32875,150 +20359,71 @@ mulAvxTwo_5x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R9), R11
 	VMOVDQU (R11)(R10*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R9), R11
 	VMOVDQU (R11)(R10*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R9), R11
 	VMOVDQU (R11)(R10*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R9), R11
 	VMOVDQU (R11)(R10*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R9), R11
 	VMOVDQU (R11)(R10*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R9), R11
 	VMOVDQU (R11)(R10*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R9), R11
 	VMOVDQU (R11)(R10*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R9), R11
 	VMOVDQU (R11)(R10*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R9), R11
 	VMOVDQU (R11)(R10*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -33029,132 +20434,53 @@ mulAvxTwo_5x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -33165,132 +20491,53 @@ mulAvxTwo_5x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -33301,132 +20548,53 @@ mulAvxTwo_5x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -33437,132 +20605,53 @@ mulAvxTwo_5x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R9), R11
 	VMOVDQU Y0, (R11)(R10*1)
@@ -33661,15 +20750,8 @@ mulAvxTwo_6x1_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y2, Y14
 	VPSHUFB Y15, Y3, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y14
 	ADDQ    $0x20, SI
@@ -33678,15 +20760,8 @@ mulAvxTwo_6x1_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y4, Y14
 	VPSHUFB Y15, Y5, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (DI), Y14
 	ADDQ    $0x20, DI
@@ -33695,15 +20770,8 @@ mulAvxTwo_6x1_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y6, Y14
 	VPSHUFB Y15, Y7, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R8), Y14
 	ADDQ    $0x20, R8
@@ -33712,15 +20780,8 @@ mulAvxTwo_6x1_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y8, Y14
 	VPSHUFB Y15, Y9, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (CX), Y14
 	ADDQ    $0x20, CX
@@ -33729,15 +20790,8 @@ mulAvxTwo_6x1_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y10, Y14
 	VPSHUFB Y15, Y11, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y12, (R9)
 	ADDQ    $0x20, R9
@@ -33823,23 +20877,9 @@ mulAvxTwo_6x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -33856,23 +20896,9 @@ mulAvxTwo_6x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -33889,23 +20915,9 @@ mulAvxTwo_6x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -33922,23 +20934,9 @@ mulAvxTwo_6x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -33955,23 +20953,9 @@ mulAvxTwo_6x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -34043,15 +21027,8 @@ mulAvxTwo_6x1Xor_loop:
 	VMOVDQU (R9), Y12
 	VPSHUFB Y14, Y0, Y14
 	VPSHUFB Y15, Y1, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (BX), Y14
 	ADDQ    $0x20, BX
@@ -34060,15 +21037,8 @@ mulAvxTwo_6x1Xor_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y2, Y14
 	VPSHUFB Y15, Y3, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (SI), Y14
 	ADDQ    $0x20, SI
@@ -34077,15 +21047,8 @@ mulAvxTwo_6x1Xor_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y4, Y14
 	VPSHUFB Y15, Y5, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (DI), Y14
 	ADDQ    $0x20, DI
@@ -34094,15 +21057,8 @@ mulAvxTwo_6x1Xor_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y6, Y14
 	VPSHUFB Y15, Y7, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R8), Y14
 	ADDQ    $0x20, R8
@@ -34111,15 +21067,8 @@ mulAvxTwo_6x1Xor_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y8, Y14
 	VPSHUFB Y15, Y9, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (CX), Y14
 	ADDQ    $0x20, CX
@@ -34128,15 +21077,8 @@ mulAvxTwo_6x1Xor_loop:
 	VPAND   Y13, Y15, Y15
 	VPSHUFB Y14, Y10, Y14
 	VPSHUFB Y15, Y11, Y15
+	XOR3WAY( $0x00, Y14, Y15, Y12)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y14, Y15, Y12
-
-#else
-	VPXOR Y14, Y12, Y12
-	VPXOR Y15, Y12, Y12
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y12, (R9)
 	ADDQ    $0x20, R9
@@ -34207,23 +21149,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -34240,23 +21168,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -34273,23 +21187,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -34306,23 +21206,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -34339,23 +21225,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -34372,23 +21244,9 @@ mulAvxTwo_6x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R10)
 	VMOVDQU Y1, 32(R10)
@@ -34468,28 +21326,13 @@ mulAvxTwo_6x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -34500,28 +21343,13 @@ mulAvxTwo_6x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -34532,28 +21360,13 @@ mulAvxTwo_6x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -34564,28 +21377,13 @@ mulAvxTwo_6x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -34596,28 +21394,13 @@ mulAvxTwo_6x2_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -34715,46 +21498,17 @@ mulAvxTwo_6x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -34771,46 +21525,17 @@ mulAvxTwo_6x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -34827,46 +21552,17 @@ mulAvxTwo_6x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -34883,46 +21579,17 @@ mulAvxTwo_6x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -34939,46 +21606,17 @@ mulAvxTwo_6x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -35045,29 +21683,14 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R10), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -35078,28 +21701,13 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -35110,28 +21718,13 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -35142,28 +21735,13 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -35174,28 +21752,13 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -35206,28 +21769,13 @@ mulAvxTwo_6x2Xor_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -35304,46 +21852,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -35360,46 +21879,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -35416,46 +21906,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -35472,46 +21933,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -35528,46 +21960,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -35584,46 +21987,17 @@ mulAvxTwo_6x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -35713,41 +22087,18 @@ mulAvxTwo_6x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -35758,41 +22109,18 @@ mulAvxTwo_6x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -35803,41 +22131,18 @@ mulAvxTwo_6x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -35848,41 +22153,18 @@ mulAvxTwo_6x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -35893,41 +22175,18 @@ mulAvxTwo_6x3_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -36037,69 +22296,25 @@ mulAvxTwo_6x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -36116,69 +22331,25 @@ mulAvxTwo_6x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -36195,69 +22366,25 @@ mulAvxTwo_6x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -36274,69 +22401,25 @@ mulAvxTwo_6x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -36353,69 +22436,25 @@ mulAvxTwo_6x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -36487,43 +22526,20 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R10), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -36534,41 +22550,18 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -36579,41 +22572,18 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -36624,41 +22594,18 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -36669,41 +22616,18 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -36714,41 +22638,18 @@ mulAvxTwo_6x3Xor_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -36831,69 +22732,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -36910,69 +22767,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -36989,69 +22802,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -37068,69 +22837,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -37147,69 +22872,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -37226,69 +22907,25 @@ mulAvxTwo_6x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -37388,54 +23025,23 @@ mulAvxTwo_6x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -37446,54 +23052,23 @@ mulAvxTwo_6x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -37504,54 +23079,23 @@ mulAvxTwo_6x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -37562,54 +23106,23 @@ mulAvxTwo_6x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -37620,54 +23133,23 @@ mulAvxTwo_6x4_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -37740,57 +23222,26 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R10), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -37801,54 +23252,23 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -37859,54 +23279,23 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -37917,54 +23306,23 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -37975,54 +23333,23 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -38033,54 +23360,23 @@ mulAvxTwo_6x4Xor_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -38186,67 +23482,28 @@ mulAvxTwo_6x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -38257,67 +23514,28 @@ mulAvxTwo_6x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -38328,67 +23546,28 @@ mulAvxTwo_6x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -38399,67 +23578,28 @@ mulAvxTwo_6x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -38470,67 +23610,28 @@ mulAvxTwo_6x5_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -38607,71 +23708,32 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R14), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (R10), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -38682,67 +23744,28 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -38753,67 +23776,28 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -38824,67 +23808,28 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -38895,67 +23840,28 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -38966,67 +23872,28 @@ mulAvxTwo_6x5Xor_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -39141,80 +24008,33 @@ mulAvxTwo_6x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -39225,80 +24045,33 @@ mulAvxTwo_6x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -39309,80 +24082,33 @@ mulAvxTwo_6x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -39393,80 +24119,33 @@ mulAvxTwo_6x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -39477,80 +24156,33 @@ mulAvxTwo_6x6_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -39631,85 +24263,38 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R14), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R15), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (R10), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -39720,80 +24305,33 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -39804,80 +24342,33 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -39888,80 +24379,33 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -39972,80 +24416,33 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -40056,80 +24453,33 @@ mulAvxTwo_6x6Xor_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -40255,93 +24605,38 @@ mulAvxTwo_6x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -40352,93 +24647,38 @@ mulAvxTwo_6x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -40449,93 +24689,38 @@ mulAvxTwo_6x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -40546,93 +24731,38 @@ mulAvxTwo_6x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (AX), Y10
 	ADDQ    $0x20, AX
@@ -40643,93 +24773,38 @@ mulAvxTwo_6x7_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -40816,99 +24891,44 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU (R13), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU (R14), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU (R15), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU (R9), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (BX), Y10
 	ADDQ    $0x20, BX
@@ -40919,93 +24939,38 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -41016,93 +24981,38 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -41113,93 +25023,38 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -41210,93 +25065,38 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (AX), Y10
 	ADDQ    $0x20, AX
@@ -41307,93 +25107,38 @@ mulAvxTwo_6x7Xor_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	VMOVDQU Y0, (R10)
 	ADDQ    $0x20, R10
@@ -41508,106 +25253,43 @@ mulAvxTwo_6x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -41618,106 +25300,43 @@ mulAvxTwo_6x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -41728,106 +25347,43 @@ mulAvxTwo_6x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -41838,106 +25394,43 @@ mulAvxTwo_6x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -41948,106 +25441,43 @@ mulAvxTwo_6x8_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -42120,120 +25550,57 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	MOVQ    24(R10), R12
 	VMOVDQU (R12)(R11*1), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	MOVQ    48(R10), R12
 	VMOVDQU (R12)(R11*1), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	MOVQ    72(R10), R12
 	VMOVDQU (R12)(R11*1), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	MOVQ    96(R10), R12
 	VMOVDQU (R12)(R11*1), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	MOVQ    120(R10), R12
 	VMOVDQU (R12)(R11*1), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	MOVQ    144(R10), R12
 	VMOVDQU (R12)(R11*1), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	MOVQ    168(R10), R12
 	VMOVDQU (R12)(R11*1), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -42244,106 +25611,43 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -42354,106 +25658,43 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -42464,106 +25705,43 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -42574,106 +25752,43 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -42684,106 +25799,43 @@ mulAvxTwo_6x8Xor_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -42906,119 +25958,48 @@ mulAvxTwo_6x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -43029,119 +26010,48 @@ mulAvxTwo_6x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -43152,119 +26062,48 @@ mulAvxTwo_6x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -43275,119 +26114,48 @@ mulAvxTwo_6x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -43398,119 +26166,48 @@ mulAvxTwo_6x9_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -43585,135 +26282,64 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R10), R12
 	VMOVDQU (R12)(R11*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R10), R12
 	VMOVDQU (R12)(R11*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R10), R12
 	VMOVDQU (R12)(R11*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R10), R12
 	VMOVDQU (R12)(R11*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R10), R12
 	VMOVDQU (R12)(R11*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R10), R12
 	VMOVDQU (R12)(R11*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R10), R12
 	VMOVDQU (R12)(R11*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R10), R12
 	VMOVDQU (R12)(R11*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -43724,119 +26350,48 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -43847,119 +26402,48 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -43970,119 +26454,48 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -44093,119 +26506,48 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -44216,119 +26558,48 @@ mulAvxTwo_6x9Xor_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -44458,132 +26729,53 @@ mulAvxTwo_6x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -44594,132 +26786,53 @@ mulAvxTwo_6x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -44730,132 +26843,53 @@ mulAvxTwo_6x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -44866,132 +26900,53 @@ mulAvxTwo_6x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -45002,132 +26957,53 @@ mulAvxTwo_6x10_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -45204,150 +27080,71 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R10), R12
 	VMOVDQU (R12)(R11*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R10), R12
 	VMOVDQU (R12)(R11*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R10), R12
 	VMOVDQU (R12)(R11*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R10), R12
 	VMOVDQU (R12)(R11*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R10), R12
 	VMOVDQU (R12)(R11*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R10), R12
 	VMOVDQU (R12)(R11*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R10), R12
 	VMOVDQU (R12)(R11*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R10), R12
 	VMOVDQU (R12)(R11*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R10), R12
 	VMOVDQU (R12)(R11*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -45358,132 +27155,53 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -45494,132 +27212,53 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -45630,132 +27269,53 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -45766,132 +27326,53 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -45902,132 +27383,53 @@ mulAvxTwo_6x10Xor_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R10), R12
 	VMOVDQU Y0, (R12)(R11*1)
@@ -46120,15 +27522,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -46139,15 +27534,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -46158,15 +27546,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -46177,15 +27558,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -46196,15 +27570,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -46215,15 +27582,8 @@ mulAvxTwo_7x1_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -46311,23 +27671,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -46344,23 +27690,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -46377,23 +27709,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -46410,23 +27728,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -46443,23 +27747,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -46476,23 +27766,9 @@ mulAvxTwo_7x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -46556,15 +27832,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 32(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y4
 	ADDQ    $0x20, SI
@@ -46575,15 +27844,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -46594,15 +27856,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -46613,15 +27868,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -46632,15 +27880,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -46651,15 +27892,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -46670,15 +27904,8 @@ mulAvxTwo_7x1Xor_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -46751,23 +27978,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -46784,23 +27997,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -46817,23 +28016,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -46850,23 +28035,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -46883,23 +28054,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -46916,23 +28073,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -46949,23 +28092,9 @@ mulAvxTwo_7x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R11)
 	VMOVDQU Y1, 32(R11)
@@ -47047,28 +28176,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -47079,28 +28193,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -47111,28 +28210,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -47143,28 +28227,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -47175,28 +28244,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -47207,28 +28261,13 @@ mulAvxTwo_7x2_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -47328,46 +28367,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -47384,46 +28394,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -47440,46 +28421,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -47496,46 +28448,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -47552,46 +28475,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -47608,46 +28502,17 @@ mulAvxTwo_7x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -47716,29 +28581,14 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R11), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -47749,28 +28599,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -47781,28 +28616,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -47813,28 +28633,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -47845,28 +28650,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -47877,28 +28667,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -47909,28 +28684,13 @@ mulAvxTwo_7x2Xor_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -48009,46 +28769,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -48065,46 +28796,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -48121,46 +28823,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -48177,46 +28850,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -48233,46 +28877,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -48289,46 +28904,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -48345,46 +28931,17 @@ mulAvxTwo_7x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -48476,41 +29033,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -48521,41 +29055,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -48566,41 +29077,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -48611,41 +29099,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -48656,41 +29121,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -48701,41 +29143,18 @@ mulAvxTwo_7x3_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -48847,69 +29266,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -48926,69 +29301,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -49005,69 +29336,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -49084,69 +29371,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -49163,69 +29406,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -49242,69 +29441,25 @@ mulAvxTwo_7x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -49378,43 +29533,20 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R13), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R11), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -49425,41 +29557,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -49470,41 +29579,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -49515,41 +29601,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -49560,41 +29623,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -49605,41 +29645,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -49650,41 +29667,18 @@ mulAvxTwo_7x3Xor_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -49769,69 +29763,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -49848,69 +29798,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -49927,69 +29833,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -50006,69 +29868,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -50085,69 +29903,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -50164,69 +29938,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -50243,69 +29973,25 @@ mulAvxTwo_7x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -50407,54 +30093,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -50465,54 +30120,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -50523,54 +30147,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -50581,54 +30174,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -50639,54 +30201,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -50697,54 +30228,23 @@ mulAvxTwo_7x4_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -50819,57 +30319,26 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R13), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R14), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R11), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -50880,54 +30349,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -50938,54 +30376,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -50996,54 +30403,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -51054,54 +30430,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -51112,54 +30457,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -51170,54 +30484,23 @@ mulAvxTwo_7x4Xor_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -51325,67 +30608,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -51396,67 +30640,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -51467,67 +30672,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -51538,67 +30704,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -51609,67 +30736,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -51680,67 +30768,28 @@ mulAvxTwo_7x5_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -51819,71 +30868,32 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R13), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R14), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R15), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -51894,67 +30904,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -51965,67 +30936,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -52036,67 +30968,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -52107,67 +31000,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -52178,67 +31032,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -52249,67 +31064,28 @@ mulAvxTwo_7x5Xor_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -52428,80 +31204,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -52512,80 +31241,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -52596,80 +31278,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -52680,80 +31315,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -52764,80 +31352,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (AX), Y9
 	ADDQ    $0x20, AX
@@ -52848,80 +31389,33 @@ mulAvxTwo_7x6_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -53006,85 +31500,38 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU (R14), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU (R15), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU (R10), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (BX), Y9
 	ADDQ    $0x20, BX
@@ -53095,80 +31542,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -53179,80 +31579,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -53263,80 +31616,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -53347,80 +31653,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -53431,80 +31690,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (AX), Y9
 	ADDQ    $0x20, AX
@@ -53515,80 +31727,33 @@ mulAvxTwo_7x6Xor_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	VMOVDQU Y0, (R11)
 	ADDQ    $0x20, R11
@@ -53698,93 +31863,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -53795,93 +31905,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -53892,93 +31947,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -53989,93 +31989,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -54086,93 +32031,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -54183,93 +32073,38 @@ mulAvxTwo_7x7_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -54342,105 +32177,50 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	MOVQ    24(R11), R13
 	VMOVDQU (R13)(R12*1), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	MOVQ    48(R11), R13
 	VMOVDQU (R13)(R12*1), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	MOVQ    72(R11), R13
 	VMOVDQU (R13)(R12*1), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	MOVQ    96(R11), R13
 	VMOVDQU (R13)(R12*1), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	MOVQ    120(R11), R13
 	VMOVDQU (R13)(R12*1), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	MOVQ    144(R11), R13
 	VMOVDQU (R13)(R12*1), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -54451,93 +32231,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -54548,93 +32273,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -54645,93 +32315,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -54742,93 +32357,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -54839,93 +32399,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -54936,93 +32441,38 @@ mulAvxTwo_7x7Xor_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -55140,106 +32590,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -55250,106 +32637,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -55360,106 +32684,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -55470,106 +32731,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -55580,106 +32778,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -55690,106 +32825,43 @@ mulAvxTwo_7x8_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -55864,120 +32936,57 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	MOVQ    24(R11), R13
 	VMOVDQU (R13)(R12*1), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	MOVQ    48(R11), R13
 	VMOVDQU (R13)(R12*1), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	MOVQ    72(R11), R13
 	VMOVDQU (R13)(R12*1), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	MOVQ    96(R11), R13
 	VMOVDQU (R13)(R12*1), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	MOVQ    120(R11), R13
 	VMOVDQU (R13)(R12*1), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	MOVQ    144(R11), R13
 	VMOVDQU (R13)(R12*1), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	MOVQ    168(R11), R13
 	VMOVDQU (R13)(R12*1), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -55988,106 +32997,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -56098,106 +33044,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -56208,106 +33091,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -56318,106 +33138,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -56428,106 +33185,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -56538,106 +33232,43 @@ mulAvxTwo_7x8Xor_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -56762,119 +33393,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -56885,119 +33445,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -57008,119 +33497,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -57131,119 +33549,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -57254,119 +33601,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -57377,119 +33653,48 @@ mulAvxTwo_7x9_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -57566,135 +33771,64 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R11), R13
 	VMOVDQU (R13)(R12*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R11), R13
 	VMOVDQU (R13)(R12*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R11), R13
 	VMOVDQU (R13)(R12*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R11), R13
 	VMOVDQU (R13)(R12*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R11), R13
 	VMOVDQU (R13)(R12*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R11), R13
 	VMOVDQU (R13)(R12*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R11), R13
 	VMOVDQU (R13)(R12*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R11), R13
 	VMOVDQU (R13)(R12*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -57705,119 +33839,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -57828,119 +33891,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -57951,119 +33943,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -58074,119 +33995,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -58197,119 +34047,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -58320,119 +34099,48 @@ mulAvxTwo_7x9Xor_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -58564,132 +34272,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -58700,132 +34329,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -58836,132 +34386,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -58972,132 +34443,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -59108,132 +34500,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -59244,132 +34557,53 @@ mulAvxTwo_7x10_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -59448,150 +34682,71 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R11), R13
 	VMOVDQU (R13)(R12*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R11), R13
 	VMOVDQU (R13)(R12*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R11), R13
 	VMOVDQU (R13)(R12*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R11), R13
 	VMOVDQU (R13)(R12*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R11), R13
 	VMOVDQU (R13)(R12*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R11), R13
 	VMOVDQU (R13)(R12*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R11), R13
 	VMOVDQU (R13)(R12*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R11), R13
 	VMOVDQU (R13)(R12*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R11), R13
 	VMOVDQU (R13)(R12*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -59602,132 +34757,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -59738,132 +34814,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -59874,132 +34871,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -60010,132 +34928,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -60146,132 +34985,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -60282,132 +35042,53 @@ mulAvxTwo_7x10Xor_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R11), R13
 	VMOVDQU Y0, (R13)(R12*1)
@@ -60502,15 +35183,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -60521,15 +35195,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -60540,15 +35207,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -60559,15 +35219,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -60578,15 +35231,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -60597,15 +35243,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -60616,15 +35255,8 @@ mulAvxTwo_8x1_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -60714,23 +35346,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -60747,23 +35365,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -60780,23 +35384,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -60813,23 +35403,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -60846,23 +35422,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -60879,23 +35441,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -60912,23 +35460,9 @@ mulAvxTwo_8x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -60994,15 +35528,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 32(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y4
 	ADDQ    $0x20, SI
@@ -61013,15 +35540,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -61032,15 +35552,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -61051,15 +35564,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -61070,15 +35576,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -61089,15 +35588,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -61108,15 +35600,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -61127,15 +35612,8 @@ mulAvxTwo_8x1Xor_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -61210,23 +35688,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -61243,23 +35707,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -61276,23 +35726,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -61309,23 +35745,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -61342,23 +35764,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -61375,23 +35783,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -61408,23 +35802,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -61441,23 +35821,9 @@ mulAvxTwo_8x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R12)
 	VMOVDQU Y1, 32(R12)
@@ -61541,28 +35907,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -61573,28 +35924,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -61605,28 +35941,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -61637,28 +35958,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -61669,28 +35975,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -61701,28 +35992,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -61733,28 +36009,13 @@ mulAvxTwo_8x2_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -61856,46 +36117,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -61912,46 +36144,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -61968,46 +36171,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -62024,46 +36198,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -62080,46 +36225,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -62136,46 +36252,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -62192,46 +36279,17 @@ mulAvxTwo_8x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -62302,29 +36360,14 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R12), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -62335,28 +36378,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -62367,28 +36395,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -62399,28 +36412,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -62431,28 +36429,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -62463,28 +36446,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -62495,28 +36463,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -62527,28 +36480,13 @@ mulAvxTwo_8x2Xor_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -62629,46 +36567,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -62685,46 +36594,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -62741,46 +36621,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -62797,46 +36648,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -62853,46 +36675,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -62909,46 +36702,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -62965,46 +36729,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -63021,46 +36756,17 @@ mulAvxTwo_8x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -63154,41 +36860,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -63199,41 +36882,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -63244,41 +36904,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -63289,41 +36926,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -63334,41 +36948,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -63379,41 +36970,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -63424,41 +36992,18 @@ mulAvxTwo_8x3_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -63572,69 +37117,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -63651,69 +37152,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -63730,69 +37187,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -63809,69 +37222,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -63888,69 +37257,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -63967,69 +37292,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -64046,69 +37327,25 @@ mulAvxTwo_8x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -64184,43 +37421,20 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R14), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R12), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -64231,41 +37445,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -64276,41 +37467,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -64321,41 +37489,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -64366,41 +37511,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -64411,41 +37533,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -64456,41 +37555,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -64501,41 +37577,18 @@ mulAvxTwo_8x3Xor_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -64622,69 +37675,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -64701,69 +37710,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -64780,69 +37745,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -64859,69 +37780,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -64938,69 +37815,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -65017,69 +37850,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -65096,69 +37885,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -65175,69 +37920,25 @@ mulAvxTwo_8x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -65341,54 +38042,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -65399,54 +38069,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -65457,54 +38096,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -65515,54 +38123,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -65573,54 +38150,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -65631,54 +38177,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -65689,54 +38204,23 @@ mulAvxTwo_8x4_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -65813,57 +38297,26 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R14), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R15), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -65874,54 +38327,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -65932,54 +38354,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -65990,54 +38381,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -66048,54 +38408,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -66106,54 +38435,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -66164,54 +38462,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -66222,54 +38489,23 @@ mulAvxTwo_8x4Xor_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -66381,67 +38617,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -66452,67 +38649,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -66523,67 +38681,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -66594,67 +38713,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -66665,67 +38745,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -66736,67 +38777,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (AX), Y8
 	ADDQ    $0x20, AX
@@ -66807,67 +38809,28 @@ mulAvxTwo_8x5_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -66950,71 +38913,32 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU (R13), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU (R14), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU (R15), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU (R11), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (BX), Y8
 	ADDQ    $0x20, BX
@@ -67025,67 +38949,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -67096,67 +38981,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -67167,67 +39013,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -67238,67 +39045,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -67309,67 +39077,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -67380,67 +39109,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (AX), Y8
 	ADDQ    $0x20, AX
@@ -67451,67 +39141,28 @@ mulAvxTwo_8x5Xor_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	VMOVDQU Y0, (R12)
 	ADDQ    $0x20, R12
@@ -67616,80 +39267,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -67700,80 +39304,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -67784,80 +39341,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -67868,80 +39378,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -67952,80 +39415,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -68036,80 +39452,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -68120,80 +39489,33 @@ mulAvxTwo_8x6_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -68266,90 +39588,43 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	MOVQ    24(R12), R14
 	VMOVDQU (R14)(R13*1), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	MOVQ    48(R12), R14
 	VMOVDQU (R14)(R13*1), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	MOVQ    72(R12), R14
 	VMOVDQU (R14)(R13*1), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	MOVQ    96(R12), R14
 	VMOVDQU (R14)(R13*1), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	MOVQ    120(R12), R14
 	VMOVDQU (R14)(R13*1), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -68360,80 +39635,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -68444,80 +39672,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -68528,80 +39709,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -68612,80 +39746,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -68696,80 +39783,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -68780,80 +39820,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -68864,80 +39857,33 @@ mulAvxTwo_8x6Xor_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -69050,93 +39996,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -69147,93 +40038,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -69244,93 +40080,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -69341,93 +40122,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -69438,93 +40164,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -69535,93 +40206,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -69632,93 +40248,38 @@ mulAvxTwo_8x7_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -69793,105 +40354,50 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	MOVQ    24(R12), R14
 	VMOVDQU (R14)(R13*1), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	MOVQ    48(R12), R14
 	VMOVDQU (R14)(R13*1), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	MOVQ    72(R12), R14
 	VMOVDQU (R14)(R13*1), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	MOVQ    96(R12), R14
 	VMOVDQU (R14)(R13*1), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	MOVQ    120(R12), R14
 	VMOVDQU (R14)(R13*1), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	MOVQ    144(R12), R14
 	VMOVDQU (R14)(R13*1), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -69902,93 +40408,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -69999,93 +40450,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -70096,93 +40492,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -70193,93 +40534,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -70290,93 +40576,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -70387,93 +40618,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -70484,93 +40660,38 @@ mulAvxTwo_8x7Xor_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -70690,106 +40811,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -70800,106 +40858,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -70910,106 +40905,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -71020,106 +40952,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -71130,106 +40999,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -71240,106 +41046,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -71350,106 +41093,43 @@ mulAvxTwo_8x8_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -71526,120 +41206,57 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	MOVQ    24(R12), R14
 	VMOVDQU (R14)(R13*1), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	MOVQ    48(R12), R14
 	VMOVDQU (R14)(R13*1), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	MOVQ    72(R12), R14
 	VMOVDQU (R14)(R13*1), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	MOVQ    96(R12), R14
 	VMOVDQU (R14)(R13*1), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	MOVQ    120(R12), R14
 	VMOVDQU (R14)(R13*1), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	MOVQ    144(R12), R14
 	VMOVDQU (R14)(R13*1), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	MOVQ    168(R12), R14
 	VMOVDQU (R14)(R13*1), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -71650,106 +41267,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -71760,106 +41314,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -71870,106 +41361,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -71980,106 +41408,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -72090,106 +41455,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -72200,106 +41502,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -72310,106 +41549,43 @@ mulAvxTwo_8x8Xor_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -72536,119 +41712,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -72659,119 +41764,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -72782,119 +41816,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -72905,119 +41868,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -73028,119 +41920,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -73151,119 +41972,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -73274,119 +42024,48 @@ mulAvxTwo_8x9_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -73465,135 +42144,64 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R12), R14
 	VMOVDQU (R14)(R13*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R12), R14
 	VMOVDQU (R14)(R13*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R12), R14
 	VMOVDQU (R14)(R13*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R12), R14
 	VMOVDQU (R14)(R13*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R12), R14
 	VMOVDQU (R14)(R13*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R12), R14
 	VMOVDQU (R14)(R13*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R12), R14
 	VMOVDQU (R14)(R13*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R12), R14
 	VMOVDQU (R14)(R13*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -73604,119 +42212,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -73727,119 +42264,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -73850,119 +42316,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -73973,119 +42368,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -74096,119 +42420,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -74219,119 +42472,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -74342,119 +42524,48 @@ mulAvxTwo_8x9Xor_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -74588,132 +42699,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -74724,132 +42756,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -74860,132 +42813,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -74996,132 +42870,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -75132,132 +42927,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -75268,132 +42984,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -75404,132 +43041,53 @@ mulAvxTwo_8x10_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -75610,150 +43168,71 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R12), R14
 	VMOVDQU (R14)(R13*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R12), R14
 	VMOVDQU (R14)(R13*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R12), R14
 	VMOVDQU (R14)(R13*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R12), R14
 	VMOVDQU (R14)(R13*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R12), R14
 	VMOVDQU (R14)(R13*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R12), R14
 	VMOVDQU (R14)(R13*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R12), R14
 	VMOVDQU (R14)(R13*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R12), R14
 	VMOVDQU (R14)(R13*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R12), R14
 	VMOVDQU (R14)(R13*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -75764,132 +43243,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -75900,132 +43300,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -76036,132 +43357,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -76172,132 +43414,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -76308,132 +43471,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -76444,132 +43528,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -76580,132 +43585,53 @@ mulAvxTwo_8x10Xor_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R12), R14
 	VMOVDQU Y0, (R14)(R13*1)
@@ -76802,15 +43728,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -76821,15 +43740,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -76840,15 +43752,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -76859,15 +43764,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -76878,15 +43776,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -76897,15 +43788,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y4
 	ADDQ    $0x20, R12
@@ -76916,15 +43800,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 8 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -76935,15 +43812,8 @@ mulAvxTwo_9x1_loop:
 	VMOVDQU 544(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -77035,23 +43905,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -77068,23 +43924,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -77101,23 +43943,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -77134,23 +43962,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -77167,23 +43981,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -77200,23 +44000,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y6
 	VMOVDQU 32(R12), Y5
@@ -77233,23 +44019,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 8 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -77266,23 +44038,9 @@ mulAvxTwo_9x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -77350,15 +44108,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 32(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y4
 	ADDQ    $0x20, SI
@@ -77369,15 +44120,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -77388,15 +44132,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -77407,15 +44144,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -77426,15 +44156,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -77445,15 +44168,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -77464,15 +44180,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y4
 	ADDQ    $0x20, R12
@@ -77483,15 +44192,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 8 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -77502,15 +44204,8 @@ mulAvxTwo_9x1Xor_loop:
 	VMOVDQU 544(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -77587,23 +44282,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -77620,23 +44301,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -77653,23 +44320,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -77686,23 +44339,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -77719,23 +44358,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -77752,23 +44377,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -77785,23 +44396,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y6
 	VMOVDQU 32(R12), Y5
@@ -77818,23 +44415,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 8 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -77851,23 +44434,9 @@ mulAvxTwo_9x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R13)
 	VMOVDQU Y1, 32(R13)
@@ -77953,28 +44522,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -77985,28 +44539,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -78017,28 +44556,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -78049,28 +44573,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -78081,28 +44590,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -78113,28 +44607,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y5
 	ADDQ    $0x20, R12
@@ -78145,28 +44624,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 8 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -78177,28 +44641,13 @@ mulAvxTwo_9x2_loop:
 	VMOVDQU 1056(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1088(CX), Y3
 	VMOVDQU 1120(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -78302,46 +44751,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -78358,46 +44778,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -78414,46 +44805,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -78470,46 +44832,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -78526,46 +44859,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -78582,46 +44886,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y9
 	VMOVDQU 32(R12), Y11
@@ -78638,46 +44913,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 8 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -78694,46 +44940,17 @@ mulAvxTwo_9x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -78806,29 +45023,14 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R13), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -78839,28 +45041,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -78871,28 +45058,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -78903,28 +45075,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -78935,28 +45092,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -78967,28 +45109,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -78999,28 +45126,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y5
 	ADDQ    $0x20, R12
@@ -79031,28 +45143,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 8 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -79063,28 +45160,13 @@ mulAvxTwo_9x2Xor_loop:
 	VMOVDQU 1056(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1088(CX), Y3
 	VMOVDQU 1120(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -79167,46 +45249,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -79223,46 +45276,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -79279,46 +45303,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -79335,46 +45330,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -79391,46 +45357,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -79447,46 +45384,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -79503,46 +45411,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y9
 	VMOVDQU 32(R12), Y11
@@ -79559,46 +45438,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 8 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -79615,46 +45465,17 @@ mulAvxTwo_9x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -79750,41 +45571,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -79795,41 +45593,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -79840,41 +45615,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -79885,41 +45637,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -79930,41 +45659,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -79975,41 +45681,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (R12), Y6
 	ADDQ    $0x20, R12
@@ -80020,41 +45703,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 8 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -80065,41 +45725,18 @@ mulAvxTwo_9x3_loop:
 	VMOVDQU 1568(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1600(CX), Y4
 	VMOVDQU 1632(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1664(CX), Y4
 	VMOVDQU 1696(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -80215,69 +45852,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -80294,69 +45887,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -80373,69 +45922,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -80452,69 +45957,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -80531,69 +45992,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -80610,69 +46027,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (R12), Y11
 	VMOVDQU 32(R12), Y13
@@ -80689,69 +46062,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 8 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -80768,69 +46097,25 @@ mulAvxTwo_9x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -80908,43 +46193,20 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R15), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -80955,41 +46217,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -81000,41 +46239,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -81045,41 +46261,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -81090,41 +46283,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -81135,41 +46305,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -81180,41 +46327,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (R12), Y6
 	ADDQ    $0x20, R12
@@ -81225,41 +46349,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 8 to 3 outputs
 	VMOVDQU (DX), Y6
 	ADDQ    $0x20, DX
@@ -81270,41 +46371,18 @@ mulAvxTwo_9x3Xor_loop:
 	VMOVDQU 1568(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1600(CX), Y4
 	VMOVDQU 1632(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1664(CX), Y4
 	VMOVDQU 1696(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -81393,69 +46471,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -81472,69 +46506,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -81551,69 +46541,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -81630,69 +46576,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -81709,69 +46611,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -81788,69 +46646,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -81867,69 +46681,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (R12), Y11
 	VMOVDQU 32(R12), Y13
@@ -81946,69 +46716,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 8 to 3 outputs
 	VMOVDQU (DX), Y11
 	VMOVDQU 32(DX), Y13
@@ -82025,69 +46751,25 @@ mulAvxTwo_9x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -82195,54 +46877,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -82253,54 +46904,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -82311,54 +46931,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -82369,54 +46958,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -82427,54 +46985,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -82485,54 +47012,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -82543,54 +47039,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 8 to 4 outputs
 	VMOVDQU (AX), Y7
 	ADDQ    $0x20, AX
@@ -82601,54 +47066,23 @@ mulAvxTwo_9x4_loop:
 	VMOVDQU 2080(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2112(CX), Y5
 	VMOVDQU 2144(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2176(CX), Y5
 	VMOVDQU 2208(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2240(CX), Y5
 	VMOVDQU 2272(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -82729,57 +47163,26 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU (R14), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU (R15), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU (R12), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (BX), Y7
 	ADDQ    $0x20, BX
@@ -82790,54 +47193,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -82848,54 +47220,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -82906,54 +47247,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -82964,54 +47274,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -83022,54 +47301,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -83080,54 +47328,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -83138,54 +47355,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 8 to 4 outputs
 	VMOVDQU (AX), Y7
 	ADDQ    $0x20, AX
@@ -83196,54 +47382,23 @@ mulAvxTwo_9x4Xor_loop:
 	VMOVDQU 2080(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2112(CX), Y5
 	VMOVDQU 2144(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2176(CX), Y5
 	VMOVDQU 2208(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2240(CX), Y5
 	VMOVDQU 2272(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	VMOVDQU Y0, (R13)
 	ADDQ    $0x20, R13
@@ -83343,67 +47498,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -83414,67 +47530,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -83485,67 +47562,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -83556,67 +47594,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -83627,67 +47626,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R11), Y8
 	ADDQ    $0x20, R11
@@ -83698,67 +47658,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (R12), Y8
 	ADDQ    $0x20, R12
@@ -83769,67 +47690,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 8 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -83840,67 +47722,28 @@ mulAvxTwo_9x5_loop:
 	VMOVDQU 2592(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2624(CX), Y6
 	VMOVDQU 2656(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2688(CX), Y6
 	VMOVDQU 2720(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2752(CX), Y6
 	VMOVDQU 2784(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2816(CX), Y6
 	VMOVDQU 2848(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -83973,75 +47816,36 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -84052,67 +47856,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -84123,67 +47888,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -84194,67 +47920,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -84265,67 +47952,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -84336,67 +47984,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R11), Y8
 	ADDQ    $0x20, R11
@@ -84407,67 +48016,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (R12), Y8
 	ADDQ    $0x20, R12
@@ -84478,67 +48048,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 8 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -84549,67 +48080,28 @@ mulAvxTwo_9x5Xor_loop:
 	VMOVDQU 2592(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2624(CX), Y6
 	VMOVDQU 2656(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2688(CX), Y6
 	VMOVDQU 2720(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2752(CX), Y6
 	VMOVDQU 2784(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2816(CX), Y6
 	VMOVDQU 2848(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -84717,80 +48209,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -84801,80 +48246,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -84885,80 +48283,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -84969,80 +48320,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -85053,80 +48357,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -85137,80 +48394,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (R12), Y9
 	ADDQ    $0x20, R12
@@ -85221,80 +48431,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 8 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -85305,80 +48468,33 @@ mulAvxTwo_9x6_loop:
 	VMOVDQU 3104(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3136(CX), Y7
 	VMOVDQU 3168(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3200(CX), Y7
 	VMOVDQU 3232(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3264(CX), Y7
 	VMOVDQU 3296(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3328(CX), Y7
 	VMOVDQU 3360(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3392(CX), Y7
 	VMOVDQU 3424(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -85453,90 +48569,43 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	MOVQ    120(R13), R15
 	VMOVDQU (R15)(R14*1), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -85547,80 +48616,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -85631,80 +48653,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -85715,80 +48690,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -85799,80 +48727,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -85883,80 +48764,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -85967,80 +48801,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (R12), Y9
 	ADDQ    $0x20, R12
@@ -86051,80 +48838,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 8 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -86135,80 +48875,33 @@ mulAvxTwo_9x6Xor_loop:
 	VMOVDQU 3104(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3136(CX), Y7
 	VMOVDQU 3168(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3200(CX), Y7
 	VMOVDQU 3232(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3264(CX), Y7
 	VMOVDQU 3296(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3328(CX), Y7
 	VMOVDQU 3360(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3392(CX), Y7
 	VMOVDQU 3424(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -86323,93 +49016,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -86420,93 +49058,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -86517,93 +49100,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -86614,93 +49142,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -86711,93 +49184,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -86808,93 +49226,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (R12), Y10
 	ADDQ    $0x20, R12
@@ -86905,93 +49268,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 8 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -87002,93 +49310,38 @@ mulAvxTwo_9x7_loop:
 	VMOVDQU 3616(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3648(CX), Y8
 	VMOVDQU 3680(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3712(CX), Y8
 	VMOVDQU 3744(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3776(CX), Y8
 	VMOVDQU 3808(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3840(CX), Y8
 	VMOVDQU 3872(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3904(CX), Y8
 	VMOVDQU 3936(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3968(CX), Y8
 	VMOVDQU 4000(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -87165,105 +49418,50 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	MOVQ    120(R13), R15
 	VMOVDQU (R15)(R14*1), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	MOVQ    144(R13), R15
 	VMOVDQU (R15)(R14*1), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -87274,93 +49472,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -87371,93 +49514,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -87468,93 +49556,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -87565,93 +49598,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -87662,93 +49640,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -87759,93 +49682,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (R12), Y10
 	ADDQ    $0x20, R12
@@ -87856,93 +49724,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 8 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -87953,93 +49766,38 @@ mulAvxTwo_9x7Xor_loop:
 	VMOVDQU 3616(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3648(CX), Y8
 	VMOVDQU 3680(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3712(CX), Y8
 	VMOVDQU 3744(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3776(CX), Y8
 	VMOVDQU 3808(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3840(CX), Y8
 	VMOVDQU 3872(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3904(CX), Y8
 	VMOVDQU 3936(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3968(CX), Y8
 	VMOVDQU 4000(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -88161,106 +49919,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -88271,106 +49966,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -88381,106 +50013,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -88491,106 +50060,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -88601,106 +50107,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -88711,106 +50154,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (R12), Y11
 	ADDQ    $0x20, R12
@@ -88821,106 +50201,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 8 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -88931,106 +50248,43 @@ mulAvxTwo_9x8_loop:
 	VMOVDQU 4128(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4160(CX), Y9
 	VMOVDQU 4192(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4224(CX), Y9
 	VMOVDQU 4256(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4288(CX), Y9
 	VMOVDQU 4320(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4352(CX), Y9
 	VMOVDQU 4384(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4416(CX), Y9
 	VMOVDQU 4448(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4480(CX), Y9
 	VMOVDQU 4512(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4544(CX), Y9
 	VMOVDQU 4576(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -89109,120 +50363,57 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	MOVQ    120(R13), R15
 	VMOVDQU (R15)(R14*1), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	MOVQ    144(R13), R15
 	VMOVDQU (R15)(R14*1), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	MOVQ    168(R13), R15
 	VMOVDQU (R15)(R14*1), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -89233,106 +50424,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -89343,106 +50471,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -89453,106 +50518,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -89563,106 +50565,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -89673,106 +50612,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -89783,106 +50659,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (R12), Y11
 	ADDQ    $0x20, R12
@@ -89893,106 +50706,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 8 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -90003,106 +50753,43 @@ mulAvxTwo_9x8Xor_loop:
 	VMOVDQU 4128(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4160(CX), Y9
 	VMOVDQU 4192(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4224(CX), Y9
 	VMOVDQU 4256(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4288(CX), Y9
 	VMOVDQU 4320(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4352(CX), Y9
 	VMOVDQU 4384(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4416(CX), Y9
 	VMOVDQU 4448(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4480(CX), Y9
 	VMOVDQU 4512(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4544(CX), Y9
 	VMOVDQU 4576(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -90231,119 +50918,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -90354,119 +50970,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -90477,119 +51022,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -90600,119 +51074,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -90723,119 +51126,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -90846,119 +51178,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (R12), Y12
 	ADDQ    $0x20, R12
@@ -90969,119 +51230,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 8 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -91092,119 +51282,48 @@ mulAvxTwo_9x9_loop:
 	VMOVDQU 4640(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4672(CX), Y10
 	VMOVDQU 4704(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4736(CX), Y10
 	VMOVDQU 4768(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4800(CX), Y10
 	VMOVDQU 4832(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4864(CX), Y10
 	VMOVDQU 4896(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4928(CX), Y10
 	VMOVDQU 4960(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4992(CX), Y10
 	VMOVDQU 5024(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5056(CX), Y10
 	VMOVDQU 5088(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5120(CX), Y10
 	VMOVDQU 5152(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -91285,135 +51404,64 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R13), R15
 	VMOVDQU (R15)(R14*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R13), R15
 	VMOVDQU (R15)(R14*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R13), R15
 	VMOVDQU (R15)(R14*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R13), R15
 	VMOVDQU (R15)(R14*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -91424,119 +51472,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -91547,119 +51524,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -91670,119 +51576,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -91793,119 +51628,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -91916,119 +51680,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -92039,119 +51732,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (R12), Y12
 	ADDQ    $0x20, R12
@@ -92162,119 +51784,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 8 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -92285,119 +51836,48 @@ mulAvxTwo_9x9Xor_loop:
 	VMOVDQU 4640(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4672(CX), Y10
 	VMOVDQU 4704(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4736(CX), Y10
 	VMOVDQU 4768(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4800(CX), Y10
 	VMOVDQU 4832(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4864(CX), Y10
 	VMOVDQU 4896(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4928(CX), Y10
 	VMOVDQU 4960(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4992(CX), Y10
 	VMOVDQU 5024(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5056(CX), Y10
 	VMOVDQU 5088(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5120(CX), Y10
 	VMOVDQU 5152(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -92533,132 +52013,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -92669,132 +52070,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -92805,132 +52127,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -92941,132 +52184,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -93077,132 +52241,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -93213,132 +52298,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (R12), Y13
 	ADDQ    $0x20, R12
@@ -93349,132 +52355,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 8 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -93485,132 +52412,53 @@ mulAvxTwo_9x10_loop:
 	VMOVDQU 5152(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5184(CX), Y11
 	VMOVDQU 5216(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5248(CX), Y11
 	VMOVDQU 5280(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5312(CX), Y11
 	VMOVDQU 5344(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 5376(CX), Y11
 	VMOVDQU 5408(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 5440(CX), Y11
 	VMOVDQU 5472(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 5504(CX), Y11
 	VMOVDQU 5536(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 5568(CX), Y11
 	VMOVDQU 5600(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 5632(CX), Y11
 	VMOVDQU 5664(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5696(CX), Y11
 	VMOVDQU 5728(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -93693,150 +52541,71 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R13), R15
 	VMOVDQU (R15)(R14*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R13), R15
 	VMOVDQU (R15)(R14*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R13), R15
 	VMOVDQU (R15)(R14*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R13), R15
 	VMOVDQU (R15)(R14*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R13), R15
 	VMOVDQU (R15)(R14*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R13), R15
 	VMOVDQU (R15)(R14*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R13), R15
 	VMOVDQU (R15)(R14*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R13), R15
 	VMOVDQU (R15)(R14*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R13), R15
 	VMOVDQU (R15)(R14*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -93847,132 +52616,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -93983,132 +52673,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -94119,132 +52730,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -94255,132 +52787,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -94391,132 +52844,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -94527,132 +52901,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (R12), Y13
 	ADDQ    $0x20, R12
@@ -94663,132 +52958,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 8 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -94799,132 +53015,53 @@ mulAvxTwo_9x10Xor_loop:
 	VMOVDQU 5152(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5184(CX), Y11
 	VMOVDQU 5216(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5248(CX), Y11
 	VMOVDQU 5280(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5312(CX), Y11
 	VMOVDQU 5344(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 5376(CX), Y11
 	VMOVDQU 5408(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 5440(CX), Y11
 	VMOVDQU 5472(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 5504(CX), Y11
 	VMOVDQU 5536(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 5568(CX), Y11
 	VMOVDQU 5600(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 5632(CX), Y11
 	VMOVDQU 5664(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5696(CX), Y11
 	VMOVDQU 5728(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R13), R15
 	VMOVDQU Y0, (R15)(R14*1)
@@ -95023,15 +53160,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -95042,15 +53172,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -95061,15 +53184,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -95080,15 +53196,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -95099,15 +53208,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -95118,15 +53220,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y4
 	ADDQ    $0x20, R12
@@ -95137,15 +53232,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 8 to 1 outputs
 	VMOVDQU (R13), Y4
 	ADDQ    $0x20, R13
@@ -95156,15 +53244,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 544(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 9 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -95175,15 +53256,8 @@ mulAvxTwo_10x1_loop:
 	VMOVDQU 608(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -95277,23 +53351,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -95310,23 +53370,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -95343,23 +53389,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -95376,23 +53408,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -95409,23 +53427,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -95442,23 +53446,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y6
 	VMOVDQU 32(R12), Y5
@@ -95475,23 +53465,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 8 to 1 outputs
 	VMOVDQU (R13), Y6
 	VMOVDQU 32(R13), Y5
@@ -95508,23 +53484,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 9 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -95541,23 +53503,9 @@ mulAvxTwo_10x1_64_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -95627,15 +53575,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 32(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y4
 	ADDQ    $0x20, SI
@@ -95646,15 +53587,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 96(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y4
 	ADDQ    $0x20, DI
@@ -95665,15 +53599,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 160(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y4
 	ADDQ    $0x20, R8
@@ -95684,15 +53611,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 224(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y4
 	ADDQ    $0x20, R9
@@ -95703,15 +53623,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 288(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y4
 	ADDQ    $0x20, R10
@@ -95722,15 +53635,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 352(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y4
 	ADDQ    $0x20, R11
@@ -95741,15 +53647,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 416(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y4
 	ADDQ    $0x20, R12
@@ -95760,15 +53659,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 480(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 8 to 1 outputs
 	VMOVDQU (R13), Y4
 	ADDQ    $0x20, R13
@@ -95779,15 +53671,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 544(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Load and process 32 bytes from input 9 to 1 outputs
 	VMOVDQU (DX), Y4
 	ADDQ    $0x20, DX
@@ -95798,15 +53683,8 @@ mulAvxTwo_10x1Xor_loop:
 	VMOVDQU 608(CX), Y3
 	VPSHUFB Y4, Y2, Y2
 	VPSHUFB Y5, Y3, Y3
+	XOR3WAY( $0x00, Y2, Y3, Y0)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y2, Y3, Y0
-
-#else
-	VPXOR Y2, Y0, Y0
-	VPXOR Y3, Y0, Y0
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -95885,23 +53763,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 1 to 1 outputs
 	VMOVDQU (SI), Y6
 	VMOVDQU 32(SI), Y5
@@ -95918,23 +53782,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 2 to 1 outputs
 	VMOVDQU (DI), Y6
 	VMOVDQU 32(DI), Y5
@@ -95951,23 +53801,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 3 to 1 outputs
 	VMOVDQU (R8), Y6
 	VMOVDQU 32(R8), Y5
@@ -95984,23 +53820,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 4 to 1 outputs
 	VMOVDQU (R9), Y6
 	VMOVDQU 32(R9), Y5
@@ -96017,23 +53839,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 5 to 1 outputs
 	VMOVDQU (R10), Y6
 	VMOVDQU 32(R10), Y5
@@ -96050,23 +53858,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 6 to 1 outputs
 	VMOVDQU (R11), Y6
 	VMOVDQU 32(R11), Y5
@@ -96083,23 +53877,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 7 to 1 outputs
 	VMOVDQU (R12), Y6
 	VMOVDQU 32(R12), Y5
@@ -96116,23 +53896,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 8 to 1 outputs
 	VMOVDQU (R13), Y6
 	VMOVDQU 32(R13), Y5
@@ -96149,23 +53915,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Load and process 64 bytes from input 9 to 1 outputs
 	VMOVDQU (DX), Y6
 	VMOVDQU 32(DX), Y5
@@ -96182,23 +53934,9 @@ mulAvxTwo_10x1_64Xor_loop:
 	VPSHUFB Y6, Y3, Y3
 	VPSHUFB Y8, Y4, Y6
 	VPSHUFB Y7, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y0)
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
 	// Store 1 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -96286,28 +54024,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -96318,28 +54041,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -96350,28 +54058,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -96382,28 +54075,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -96414,28 +54092,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -96446,28 +54109,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y5
 	ADDQ    $0x20, R12
@@ -96478,28 +54126,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 8 to 2 outputs
 	VMOVDQU (R13), Y5
 	ADDQ    $0x20, R13
@@ -96510,28 +54143,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 1056(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1088(CX), Y3
 	VMOVDQU 1120(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 9 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -96542,28 +54160,13 @@ mulAvxTwo_10x2_loop:
 	VMOVDQU 1184(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1216(CX), Y3
 	VMOVDQU 1248(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R15)
 	ADDQ    $0x20, R15
@@ -96669,46 +54272,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -96725,46 +54299,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -96781,46 +54326,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -96837,46 +54353,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -96893,46 +54380,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -96949,46 +54407,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y9
 	VMOVDQU 32(R12), Y11
@@ -97005,46 +54434,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 8 to 2 outputs
 	VMOVDQU (R13), Y9
 	VMOVDQU 32(R13), Y11
@@ -97061,46 +54461,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 9 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -97117,46 +54488,17 @@ mulAvxTwo_10x2_64_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R15)
 	VMOVDQU Y1, 32(R15)
@@ -97231,29 +54573,14 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 32(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU (R14), Y1
 	VMOVDQU 64(CX), Y3
 	VMOVDQU 96(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y5
 	ADDQ    $0x20, SI
@@ -97264,28 +54591,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 160(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 192(CX), Y3
 	VMOVDQU 224(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y5
 	ADDQ    $0x20, DI
@@ -97296,28 +54608,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 288(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 320(CX), Y3
 	VMOVDQU 352(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y5
 	ADDQ    $0x20, R8
@@ -97328,28 +54625,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 416(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 448(CX), Y3
 	VMOVDQU 480(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y5
 	ADDQ    $0x20, R9
@@ -97360,28 +54642,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 544(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 576(CX), Y3
 	VMOVDQU 608(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y5
 	ADDQ    $0x20, R10
@@ -97392,28 +54659,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 672(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 704(CX), Y3
 	VMOVDQU 736(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y5
 	ADDQ    $0x20, R11
@@ -97424,28 +54676,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 800(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 832(CX), Y3
 	VMOVDQU 864(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y5
 	ADDQ    $0x20, R12
@@ -97456,28 +54693,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 928(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 960(CX), Y3
 	VMOVDQU 992(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 8 to 2 outputs
 	VMOVDQU (R13), Y5
 	ADDQ    $0x20, R13
@@ -97488,28 +54710,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 1056(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1088(CX), Y3
 	VMOVDQU 1120(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Load and process 32 bytes from input 9 to 2 outputs
 	VMOVDQU (DX), Y5
 	ADDQ    $0x20, DX
@@ -97520,28 +54727,13 @@ mulAvxTwo_10x2Xor_loop:
 	VMOVDQU 1184(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y0
-
-#else
-	VPXOR Y3, Y0, Y0
-	VPXOR Y4, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y3, Y4, Y0)
 	VMOVDQU 1216(CX), Y3
 	VMOVDQU 1248(CX), Y4
 	VPSHUFB Y5, Y3, Y3
 	VPSHUFB Y6, Y4, Y4
+	XOR3WAY( $0x00, Y3, Y4, Y1)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y3, Y4, Y1
-
-#else
-	VPXOR Y3, Y1, Y1
-	VPXOR Y4, Y1, Y1
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R15)
 	ADDQ    $0x20, R15
@@ -97626,46 +54818,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 1 to 2 outputs
 	VMOVDQU (SI), Y9
 	VMOVDQU 32(SI), Y11
@@ -97682,46 +54845,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 2 to 2 outputs
 	VMOVDQU (DI), Y9
 	VMOVDQU 32(DI), Y11
@@ -97738,46 +54872,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 3 to 2 outputs
 	VMOVDQU (R8), Y9
 	VMOVDQU 32(R8), Y11
@@ -97794,46 +54899,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 4 to 2 outputs
 	VMOVDQU (R9), Y9
 	VMOVDQU 32(R9), Y11
@@ -97850,46 +54926,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 5 to 2 outputs
 	VMOVDQU (R10), Y9
 	VMOVDQU 32(R10), Y11
@@ -97906,46 +54953,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 6 to 2 outputs
 	VMOVDQU (R11), Y9
 	VMOVDQU 32(R11), Y11
@@ -97962,46 +54980,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 7 to 2 outputs
 	VMOVDQU (R12), Y9
 	VMOVDQU 32(R12), Y11
@@ -98018,46 +55007,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 8 to 2 outputs
 	VMOVDQU (R13), Y9
 	VMOVDQU 32(R13), Y11
@@ -98074,46 +55034,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Load and process 64 bytes from input 9 to 2 outputs
 	VMOVDQU (DX), Y9
 	VMOVDQU 32(DX), Y11
@@ -98130,46 +55061,17 @@ mulAvxTwo_10x2_64Xor_loop:
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y11, Y5, Y7
 	VPSHUFB Y9, Y5, Y5
 	VPSHUFB Y12, Y6, Y8
 	VPSHUFB Y10, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y2)
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
 	// Store 2 outputs
 	VMOVDQU Y0, (R15)
 	VMOVDQU Y1, 32(R15)
@@ -98269,41 +55171,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -98314,41 +55193,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -98359,41 +55215,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -98404,41 +55237,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -98449,41 +55259,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -98494,41 +55281,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -98539,41 +55303,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 8 to 3 outputs
 	VMOVDQU (R12), Y6
 	ADDQ    $0x20, R12
@@ -98584,41 +55325,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 1568(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1600(CX), Y4
 	VMOVDQU 1632(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1664(CX), Y4
 	VMOVDQU 1696(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 9 to 3 outputs
 	VMOVDQU (AX), Y6
 	ADDQ    $0x20, AX
@@ -98629,41 +55347,18 @@ mulAvxTwo_10x3_loop:
 	VMOVDQU 1760(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1792(CX), Y4
 	VMOVDQU 1824(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1856(CX), Y4
 	VMOVDQU 1888(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -98785,69 +55480,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -98864,69 +55515,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -98943,69 +55550,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -99022,69 +55585,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -99101,69 +55620,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -99180,69 +55655,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -99259,69 +55690,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 8 to 3 outputs
 	VMOVDQU (R12), Y11
 	VMOVDQU 32(R12), Y13
@@ -99338,69 +55725,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 9 to 3 outputs
 	VMOVDQU (AX), Y11
 	VMOVDQU 32(AX), Y13
@@ -99417,69 +55760,25 @@ mulAvxTwo_10x3_64_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -99561,43 +55860,20 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 32(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU (R15), Y1
 	VMOVDQU 64(CX), Y4
 	VMOVDQU 96(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU (R13), Y2
 	VMOVDQU 128(CX), Y4
 	VMOVDQU 160(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 1 to 3 outputs
 	VMOVDQU (BX), Y6
 	ADDQ    $0x20, BX
@@ -99608,41 +55884,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 224(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 256(CX), Y4
 	VMOVDQU 288(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 320(CX), Y4
 	VMOVDQU 352(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 2 to 3 outputs
 	VMOVDQU (SI), Y6
 	ADDQ    $0x20, SI
@@ -99653,41 +55906,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 416(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 448(CX), Y4
 	VMOVDQU 480(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 512(CX), Y4
 	VMOVDQU 544(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 3 to 3 outputs
 	VMOVDQU (DI), Y6
 	ADDQ    $0x20, DI
@@ -99698,41 +55928,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 608(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 640(CX), Y4
 	VMOVDQU 672(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 704(CX), Y4
 	VMOVDQU 736(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 4 to 3 outputs
 	VMOVDQU (R8), Y6
 	ADDQ    $0x20, R8
@@ -99743,41 +55950,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 800(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 832(CX), Y4
 	VMOVDQU 864(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 896(CX), Y4
 	VMOVDQU 928(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 5 to 3 outputs
 	VMOVDQU (R9), Y6
 	ADDQ    $0x20, R9
@@ -99788,41 +55972,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 992(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1024(CX), Y4
 	VMOVDQU 1056(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1088(CX), Y4
 	VMOVDQU 1120(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 6 to 3 outputs
 	VMOVDQU (R10), Y6
 	ADDQ    $0x20, R10
@@ -99833,41 +55994,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 1184(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1216(CX), Y4
 	VMOVDQU 1248(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1280(CX), Y4
 	VMOVDQU 1312(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 7 to 3 outputs
 	VMOVDQU (R11), Y6
 	ADDQ    $0x20, R11
@@ -99878,41 +56016,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 1376(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1408(CX), Y4
 	VMOVDQU 1440(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1472(CX), Y4
 	VMOVDQU 1504(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 8 to 3 outputs
 	VMOVDQU (R12), Y6
 	ADDQ    $0x20, R12
@@ -99923,41 +56038,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 1568(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1600(CX), Y4
 	VMOVDQU 1632(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1664(CX), Y4
 	VMOVDQU 1696(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Load and process 32 bytes from input 9 to 3 outputs
 	VMOVDQU (AX), Y6
 	ADDQ    $0x20, AX
@@ -99968,41 +56060,18 @@ mulAvxTwo_10x3Xor_loop:
 	VMOVDQU 1760(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y0
-
-#else
-	VPXOR Y4, Y0, Y0
-	VPXOR Y5, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y0)
 	VMOVDQU 1792(CX), Y4
 	VMOVDQU 1824(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y1
-
-#else
-	VPXOR Y4, Y1, Y1
-	VPXOR Y5, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y4, Y5, Y1)
 	VMOVDQU 1856(CX), Y4
 	VMOVDQU 1888(CX), Y5
 	VPSHUFB Y6, Y4, Y4
 	VPSHUFB Y7, Y5, Y5
+	XOR3WAY( $0x00, Y4, Y5, Y2)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y4, Y5, Y2
-
-#else
-	VPXOR Y4, Y2, Y2
-	VPXOR Y5, Y2, Y2
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	ADDQ    $0x20, R14
@@ -100097,69 +56166,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 1 to 3 outputs
 	VMOVDQU (BX), Y11
 	VMOVDQU 32(BX), Y13
@@ -100176,69 +56201,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 2 to 3 outputs
 	VMOVDQU (SI), Y11
 	VMOVDQU 32(SI), Y13
@@ -100255,69 +56236,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 3 to 3 outputs
 	VMOVDQU (DI), Y11
 	VMOVDQU 32(DI), Y13
@@ -100334,69 +56271,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 4 to 3 outputs
 	VMOVDQU (R8), Y11
 	VMOVDQU 32(R8), Y13
@@ -100413,69 +56306,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 5 to 3 outputs
 	VMOVDQU (R9), Y11
 	VMOVDQU 32(R9), Y13
@@ -100492,69 +56341,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 6 to 3 outputs
 	VMOVDQU (R10), Y11
 	VMOVDQU 32(R10), Y13
@@ -100571,69 +56376,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 7 to 3 outputs
 	VMOVDQU (R11), Y11
 	VMOVDQU 32(R11), Y13
@@ -100650,69 +56411,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 8 to 3 outputs
 	VMOVDQU (R12), Y11
 	VMOVDQU 32(R12), Y13
@@ -100729,69 +56446,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Load and process 64 bytes from input 9 to 3 outputs
 	VMOVDQU (AX), Y11
 	VMOVDQU 32(AX), Y13
@@ -100808,69 +56481,25 @@ mulAvxTwo_10x3_64Xor_loop:
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y13, Y7, Y9
 	VPSHUFB Y11, Y7, Y7
 	VPSHUFB Y14, Y8, Y10
 	VPSHUFB Y12, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y4)
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
 	// Store 3 outputs
 	VMOVDQU Y0, (R14)
 	VMOVDQU Y1, 32(R14)
@@ -100968,54 +56597,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -101026,54 +56624,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -101084,54 +56651,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -101142,54 +56678,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -101200,54 +56705,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -101258,54 +56732,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (R12), Y7
 	ADDQ    $0x20, R12
@@ -101316,54 +56759,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 8 to 4 outputs
 	VMOVDQU (R13), Y7
 	ADDQ    $0x20, R13
@@ -101374,54 +56786,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 2080(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2112(CX), Y5
 	VMOVDQU 2144(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2176(CX), Y5
 	VMOVDQU 2208(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2240(CX), Y5
 	VMOVDQU 2272(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 9 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -101432,54 +56813,23 @@ mulAvxTwo_10x4_loop:
 	VMOVDQU 2336(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2368(CX), Y5
 	VMOVDQU 2400(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2432(CX), Y5
 	VMOVDQU 2464(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2496(CX), Y5
 	VMOVDQU 2528(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -101552,60 +56902,29 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 32(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y5
 	VMOVDQU 96(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y5
 	VMOVDQU 160(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y5
 	VMOVDQU 224(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 1 to 4 outputs
 	VMOVDQU (SI), Y7
 	ADDQ    $0x20, SI
@@ -101616,54 +56935,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 288(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 320(CX), Y5
 	VMOVDQU 352(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 384(CX), Y5
 	VMOVDQU 416(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 448(CX), Y5
 	VMOVDQU 480(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 2 to 4 outputs
 	VMOVDQU (DI), Y7
 	ADDQ    $0x20, DI
@@ -101674,54 +56962,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 544(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 576(CX), Y5
 	VMOVDQU 608(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 640(CX), Y5
 	VMOVDQU 672(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 704(CX), Y5
 	VMOVDQU 736(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 3 to 4 outputs
 	VMOVDQU (R8), Y7
 	ADDQ    $0x20, R8
@@ -101732,54 +56989,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 800(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 832(CX), Y5
 	VMOVDQU 864(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 896(CX), Y5
 	VMOVDQU 928(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 960(CX), Y5
 	VMOVDQU 992(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 4 to 4 outputs
 	VMOVDQU (R9), Y7
 	ADDQ    $0x20, R9
@@ -101790,54 +57016,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 1056(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1088(CX), Y5
 	VMOVDQU 1120(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1152(CX), Y5
 	VMOVDQU 1184(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1216(CX), Y5
 	VMOVDQU 1248(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 5 to 4 outputs
 	VMOVDQU (R10), Y7
 	ADDQ    $0x20, R10
@@ -101848,54 +57043,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 1312(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1344(CX), Y5
 	VMOVDQU 1376(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1408(CX), Y5
 	VMOVDQU 1440(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1472(CX), Y5
 	VMOVDQU 1504(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 6 to 4 outputs
 	VMOVDQU (R11), Y7
 	ADDQ    $0x20, R11
@@ -101906,54 +57070,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 1568(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1600(CX), Y5
 	VMOVDQU 1632(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1664(CX), Y5
 	VMOVDQU 1696(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1728(CX), Y5
 	VMOVDQU 1760(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 7 to 4 outputs
 	VMOVDQU (R12), Y7
 	ADDQ    $0x20, R12
@@ -101964,54 +57097,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 1824(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 1856(CX), Y5
 	VMOVDQU 1888(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 1920(CX), Y5
 	VMOVDQU 1952(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 1984(CX), Y5
 	VMOVDQU 2016(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 8 to 4 outputs
 	VMOVDQU (R13), Y7
 	ADDQ    $0x20, R13
@@ -102022,54 +57124,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 2080(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2112(CX), Y5
 	VMOVDQU 2144(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2176(CX), Y5
 	VMOVDQU 2208(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2240(CX), Y5
 	VMOVDQU 2272(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Load and process 32 bytes from input 9 to 4 outputs
 	VMOVDQU (DX), Y7
 	ADDQ    $0x20, DX
@@ -102080,54 +57151,23 @@ mulAvxTwo_10x4Xor_loop:
 	VMOVDQU 2336(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y0
-
-#else
-	VPXOR Y5, Y0, Y0
-	VPXOR Y6, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y0)
 	VMOVDQU 2368(CX), Y5
 	VMOVDQU 2400(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y1
-
-#else
-	VPXOR Y5, Y1, Y1
-	VPXOR Y6, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y1)
 	VMOVDQU 2432(CX), Y5
 	VMOVDQU 2464(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y2
-
-#else
-	VPXOR Y5, Y2, Y2
-	VPXOR Y6, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y5, Y6, Y2)
 	VMOVDQU 2496(CX), Y5
 	VMOVDQU 2528(CX), Y6
 	VPSHUFB Y7, Y5, Y5
 	VPSHUFB Y8, Y6, Y6
+	XOR3WAY( $0x00, Y5, Y6, Y3)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y5, Y6, Y3
-
-#else
-	VPXOR Y5, Y3, Y3
-	VPXOR Y6, Y3, Y3
-
-#endif
 	// Store 4 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -102230,67 +57270,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -102301,67 +57302,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -102372,67 +57334,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -102443,67 +57366,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -102514,67 +57398,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R11), Y8
 	ADDQ    $0x20, R11
@@ -102585,67 +57430,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (R12), Y8
 	ADDQ    $0x20, R12
@@ -102656,67 +57462,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 8 to 5 outputs
 	VMOVDQU (R13), Y8
 	ADDQ    $0x20, R13
@@ -102727,67 +57494,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 2592(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2624(CX), Y6
 	VMOVDQU 2656(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2688(CX), Y6
 	VMOVDQU 2720(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2752(CX), Y6
 	VMOVDQU 2784(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2816(CX), Y6
 	VMOVDQU 2848(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 9 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -102798,67 +57526,28 @@ mulAvxTwo_10x5_loop:
 	VMOVDQU 2912(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2944(CX), Y6
 	VMOVDQU 2976(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 3008(CX), Y6
 	VMOVDQU 3040(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 3072(CX), Y6
 	VMOVDQU 3104(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 3136(CX), Y6
 	VMOVDQU 3168(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -102933,75 +57622,36 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 32(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y6
 	VMOVDQU 96(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y6
 	VMOVDQU 160(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y6
 	VMOVDQU 224(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y6
 	VMOVDQU 288(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 1 to 5 outputs
 	VMOVDQU (SI), Y8
 	ADDQ    $0x20, SI
@@ -103012,67 +57662,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 352(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 384(CX), Y6
 	VMOVDQU 416(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 448(CX), Y6
 	VMOVDQU 480(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 512(CX), Y6
 	VMOVDQU 544(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 576(CX), Y6
 	VMOVDQU 608(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 2 to 5 outputs
 	VMOVDQU (DI), Y8
 	ADDQ    $0x20, DI
@@ -103083,67 +57694,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 672(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 704(CX), Y6
 	VMOVDQU 736(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 768(CX), Y6
 	VMOVDQU 800(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 832(CX), Y6
 	VMOVDQU 864(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 896(CX), Y6
 	VMOVDQU 928(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 3 to 5 outputs
 	VMOVDQU (R8), Y8
 	ADDQ    $0x20, R8
@@ -103154,67 +57726,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 992(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1024(CX), Y6
 	VMOVDQU 1056(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1088(CX), Y6
 	VMOVDQU 1120(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1152(CX), Y6
 	VMOVDQU 1184(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1216(CX), Y6
 	VMOVDQU 1248(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 4 to 5 outputs
 	VMOVDQU (R9), Y8
 	ADDQ    $0x20, R9
@@ -103225,67 +57758,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 1312(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1344(CX), Y6
 	VMOVDQU 1376(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1408(CX), Y6
 	VMOVDQU 1440(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1472(CX), Y6
 	VMOVDQU 1504(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1536(CX), Y6
 	VMOVDQU 1568(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 5 to 5 outputs
 	VMOVDQU (R10), Y8
 	ADDQ    $0x20, R10
@@ -103296,67 +57790,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 1632(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1664(CX), Y6
 	VMOVDQU 1696(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 1728(CX), Y6
 	VMOVDQU 1760(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 1792(CX), Y6
 	VMOVDQU 1824(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 1856(CX), Y6
 	VMOVDQU 1888(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 6 to 5 outputs
 	VMOVDQU (R11), Y8
 	ADDQ    $0x20, R11
@@ -103367,67 +57822,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 1952(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 1984(CX), Y6
 	VMOVDQU 2016(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2048(CX), Y6
 	VMOVDQU 2080(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2112(CX), Y6
 	VMOVDQU 2144(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2176(CX), Y6
 	VMOVDQU 2208(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 7 to 5 outputs
 	VMOVDQU (R12), Y8
 	ADDQ    $0x20, R12
@@ -103438,67 +57854,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 2272(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2304(CX), Y6
 	VMOVDQU 2336(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2368(CX), Y6
 	VMOVDQU 2400(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2432(CX), Y6
 	VMOVDQU 2464(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2496(CX), Y6
 	VMOVDQU 2528(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 8 to 5 outputs
 	VMOVDQU (R13), Y8
 	ADDQ    $0x20, R13
@@ -103509,67 +57886,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 2592(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2624(CX), Y6
 	VMOVDQU 2656(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 2688(CX), Y6
 	VMOVDQU 2720(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 2752(CX), Y6
 	VMOVDQU 2784(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 2816(CX), Y6
 	VMOVDQU 2848(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Load and process 32 bytes from input 9 to 5 outputs
 	VMOVDQU (DX), Y8
 	ADDQ    $0x20, DX
@@ -103580,67 +57918,28 @@ mulAvxTwo_10x5Xor_loop:
 	VMOVDQU 2912(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y0
-
-#else
-	VPXOR Y6, Y0, Y0
-	VPXOR Y7, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y0)
 	VMOVDQU 2944(CX), Y6
 	VMOVDQU 2976(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y1
-
-#else
-	VPXOR Y6, Y1, Y1
-	VPXOR Y7, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y1)
 	VMOVDQU 3008(CX), Y6
 	VMOVDQU 3040(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y2
-
-#else
-	VPXOR Y6, Y2, Y2
-	VPXOR Y7, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y2)
 	VMOVDQU 3072(CX), Y6
 	VMOVDQU 3104(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y3
-
-#else
-	VPXOR Y6, Y3, Y3
-	VPXOR Y7, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y6, Y7, Y3)
 	VMOVDQU 3136(CX), Y6
 	VMOVDQU 3168(CX), Y7
 	VPSHUFB Y8, Y6, Y6
 	VPSHUFB Y9, Y7, Y7
+	XOR3WAY( $0x00, Y6, Y7, Y4)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y6, Y7, Y4
-
-#else
-	VPXOR Y6, Y4, Y4
-	VPXOR Y7, Y4, Y4
-
-#endif
 	// Store 5 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -103750,80 +58049,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -103834,80 +58086,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -103918,80 +58123,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -104002,80 +58160,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -104086,80 +58197,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -104170,80 +58234,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (R12), Y9
 	ADDQ    $0x20, R12
@@ -104254,80 +58271,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 8 to 6 outputs
 	VMOVDQU (R13), Y9
 	ADDQ    $0x20, R13
@@ -104338,80 +58308,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 3104(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3136(CX), Y7
 	VMOVDQU 3168(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3200(CX), Y7
 	VMOVDQU 3232(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3264(CX), Y7
 	VMOVDQU 3296(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3328(CX), Y7
 	VMOVDQU 3360(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3392(CX), Y7
 	VMOVDQU 3424(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 9 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -104422,80 +58345,33 @@ mulAvxTwo_10x6_loop:
 	VMOVDQU 3488(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3520(CX), Y7
 	VMOVDQU 3552(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3584(CX), Y7
 	VMOVDQU 3616(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3648(CX), Y7
 	VMOVDQU 3680(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3712(CX), Y7
 	VMOVDQU 3744(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3776(CX), Y7
 	VMOVDQU 3808(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -104572,90 +58448,43 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 32(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y7
 	VMOVDQU 96(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y7
 	VMOVDQU 160(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y7
 	VMOVDQU 224(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y7
 	VMOVDQU 288(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	MOVQ    120(R14), BP
 	VMOVDQU (BP)(R15*1), Y5
 	VMOVDQU 320(CX), Y7
 	VMOVDQU 352(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 1 to 6 outputs
 	VMOVDQU (SI), Y9
 	ADDQ    $0x20, SI
@@ -104666,80 +58495,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 416(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 448(CX), Y7
 	VMOVDQU 480(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 512(CX), Y7
 	VMOVDQU 544(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 576(CX), Y7
 	VMOVDQU 608(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 640(CX), Y7
 	VMOVDQU 672(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 704(CX), Y7
 	VMOVDQU 736(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 2 to 6 outputs
 	VMOVDQU (DI), Y9
 	ADDQ    $0x20, DI
@@ -104750,80 +58532,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 800(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 832(CX), Y7
 	VMOVDQU 864(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 896(CX), Y7
 	VMOVDQU 928(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 960(CX), Y7
 	VMOVDQU 992(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1024(CX), Y7
 	VMOVDQU 1056(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1088(CX), Y7
 	VMOVDQU 1120(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 3 to 6 outputs
 	VMOVDQU (R8), Y9
 	ADDQ    $0x20, R8
@@ -104834,80 +58569,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 1184(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1216(CX), Y7
 	VMOVDQU 1248(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1280(CX), Y7
 	VMOVDQU 1312(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1344(CX), Y7
 	VMOVDQU 1376(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1408(CX), Y7
 	VMOVDQU 1440(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1472(CX), Y7
 	VMOVDQU 1504(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 4 to 6 outputs
 	VMOVDQU (R9), Y9
 	ADDQ    $0x20, R9
@@ -104918,80 +58606,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 1568(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1600(CX), Y7
 	VMOVDQU 1632(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 1664(CX), Y7
 	VMOVDQU 1696(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 1728(CX), Y7
 	VMOVDQU 1760(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 1792(CX), Y7
 	VMOVDQU 1824(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 1856(CX), Y7
 	VMOVDQU 1888(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 5 to 6 outputs
 	VMOVDQU (R10), Y9
 	ADDQ    $0x20, R10
@@ -105002,80 +58643,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 1952(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 1984(CX), Y7
 	VMOVDQU 2016(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2048(CX), Y7
 	VMOVDQU 2080(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2112(CX), Y7
 	VMOVDQU 2144(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2176(CX), Y7
 	VMOVDQU 2208(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2240(CX), Y7
 	VMOVDQU 2272(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 6 to 6 outputs
 	VMOVDQU (R11), Y9
 	ADDQ    $0x20, R11
@@ -105086,80 +58680,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 2336(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2368(CX), Y7
 	VMOVDQU 2400(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2432(CX), Y7
 	VMOVDQU 2464(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2496(CX), Y7
 	VMOVDQU 2528(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2560(CX), Y7
 	VMOVDQU 2592(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 2624(CX), Y7
 	VMOVDQU 2656(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 7 to 6 outputs
 	VMOVDQU (R12), Y9
 	ADDQ    $0x20, R12
@@ -105170,80 +58717,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 2720(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 2752(CX), Y7
 	VMOVDQU 2784(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 2816(CX), Y7
 	VMOVDQU 2848(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 2880(CX), Y7
 	VMOVDQU 2912(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 2944(CX), Y7
 	VMOVDQU 2976(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3008(CX), Y7
 	VMOVDQU 3040(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 8 to 6 outputs
 	VMOVDQU (R13), Y9
 	ADDQ    $0x20, R13
@@ -105254,80 +58754,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 3104(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3136(CX), Y7
 	VMOVDQU 3168(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3200(CX), Y7
 	VMOVDQU 3232(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3264(CX), Y7
 	VMOVDQU 3296(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3328(CX), Y7
 	VMOVDQU 3360(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3392(CX), Y7
 	VMOVDQU 3424(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Load and process 32 bytes from input 9 to 6 outputs
 	VMOVDQU (DX), Y9
 	ADDQ    $0x20, DX
@@ -105338,80 +58791,33 @@ mulAvxTwo_10x6Xor_loop:
 	VMOVDQU 3488(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y0
-
-#else
-	VPXOR Y7, Y0, Y0
-	VPXOR Y8, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y0)
 	VMOVDQU 3520(CX), Y7
 	VMOVDQU 3552(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y1
-
-#else
-	VPXOR Y7, Y1, Y1
-	VPXOR Y8, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y1)
 	VMOVDQU 3584(CX), Y7
 	VMOVDQU 3616(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y2
-
-#else
-	VPXOR Y7, Y2, Y2
-	VPXOR Y8, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y2)
 	VMOVDQU 3648(CX), Y7
 	VMOVDQU 3680(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y3
-
-#else
-	VPXOR Y7, Y3, Y3
-	VPXOR Y8, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y3)
 	VMOVDQU 3712(CX), Y7
 	VMOVDQU 3744(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y4
-
-#else
-	VPXOR Y7, Y4, Y4
-	VPXOR Y8, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y7, Y8, Y4)
 	VMOVDQU 3776(CX), Y7
 	VMOVDQU 3808(CX), Y8
 	VPSHUFB Y9, Y7, Y7
 	VPSHUFB Y10, Y8, Y8
+	XOR3WAY( $0x00, Y7, Y8, Y5)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y7, Y8, Y5
-
-#else
-	VPXOR Y7, Y5, Y5
-	VPXOR Y8, Y5, Y5
-
-#endif
 	// Store 6 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -105528,93 +58934,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -105625,93 +58976,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -105722,93 +59018,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -105819,93 +59060,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -105916,93 +59102,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -106013,93 +59144,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (R12), Y10
 	ADDQ    $0x20, R12
@@ -106110,93 +59186,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 8 to 7 outputs
 	VMOVDQU (R13), Y10
 	ADDQ    $0x20, R13
@@ -106207,93 +59228,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 3616(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3648(CX), Y8
 	VMOVDQU 3680(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3712(CX), Y8
 	VMOVDQU 3744(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3776(CX), Y8
 	VMOVDQU 3808(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3840(CX), Y8
 	VMOVDQU 3872(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3904(CX), Y8
 	VMOVDQU 3936(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3968(CX), Y8
 	VMOVDQU 4000(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 9 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -106304,93 +59270,38 @@ mulAvxTwo_10x7_loop:
 	VMOVDQU 4064(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 4096(CX), Y8
 	VMOVDQU 4128(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 4160(CX), Y8
 	VMOVDQU 4192(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 4224(CX), Y8
 	VMOVDQU 4256(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 4288(CX), Y8
 	VMOVDQU 4320(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 4352(CX), Y8
 	VMOVDQU 4384(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 4416(CX), Y8
 	VMOVDQU 4448(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -106469,105 +59380,50 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 32(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y8
 	VMOVDQU 96(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y8
 	VMOVDQU 160(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y8
 	VMOVDQU 224(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y8
 	VMOVDQU 288(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	MOVQ    120(R14), BP
 	VMOVDQU (BP)(R15*1), Y5
 	VMOVDQU 320(CX), Y8
 	VMOVDQU 352(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	MOVQ    144(R14), BP
 	VMOVDQU (BP)(R15*1), Y6
 	VMOVDQU 384(CX), Y8
 	VMOVDQU 416(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 1 to 7 outputs
 	VMOVDQU (SI), Y10
 	ADDQ    $0x20, SI
@@ -106578,93 +59434,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 480(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 512(CX), Y8
 	VMOVDQU 544(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 576(CX), Y8
 	VMOVDQU 608(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 640(CX), Y8
 	VMOVDQU 672(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 704(CX), Y8
 	VMOVDQU 736(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 768(CX), Y8
 	VMOVDQU 800(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 832(CX), Y8
 	VMOVDQU 864(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 2 to 7 outputs
 	VMOVDQU (DI), Y10
 	ADDQ    $0x20, DI
@@ -106675,93 +59476,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 928(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 960(CX), Y8
 	VMOVDQU 992(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1024(CX), Y8
 	VMOVDQU 1056(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1088(CX), Y8
 	VMOVDQU 1120(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1152(CX), Y8
 	VMOVDQU 1184(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1216(CX), Y8
 	VMOVDQU 1248(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1280(CX), Y8
 	VMOVDQU 1312(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 3 to 7 outputs
 	VMOVDQU (R8), Y10
 	ADDQ    $0x20, R8
@@ -106772,93 +59518,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 1376(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1408(CX), Y8
 	VMOVDQU 1440(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1472(CX), Y8
 	VMOVDQU 1504(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1536(CX), Y8
 	VMOVDQU 1568(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 1600(CX), Y8
 	VMOVDQU 1632(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 1664(CX), Y8
 	VMOVDQU 1696(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 1728(CX), Y8
 	VMOVDQU 1760(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 4 to 7 outputs
 	VMOVDQU (R9), Y10
 	ADDQ    $0x20, R9
@@ -106869,93 +59560,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 1824(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 1856(CX), Y8
 	VMOVDQU 1888(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 1920(CX), Y8
 	VMOVDQU 1952(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 1984(CX), Y8
 	VMOVDQU 2016(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2048(CX), Y8
 	VMOVDQU 2080(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2112(CX), Y8
 	VMOVDQU 2144(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2176(CX), Y8
 	VMOVDQU 2208(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 5 to 7 outputs
 	VMOVDQU (R10), Y10
 	ADDQ    $0x20, R10
@@ -106966,93 +59602,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 2272(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2304(CX), Y8
 	VMOVDQU 2336(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2368(CX), Y8
 	VMOVDQU 2400(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2432(CX), Y8
 	VMOVDQU 2464(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2496(CX), Y8
 	VMOVDQU 2528(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 2560(CX), Y8
 	VMOVDQU 2592(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 2624(CX), Y8
 	VMOVDQU 2656(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 6 to 7 outputs
 	VMOVDQU (R11), Y10
 	ADDQ    $0x20, R11
@@ -107063,93 +59644,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 2720(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 2752(CX), Y8
 	VMOVDQU 2784(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 2816(CX), Y8
 	VMOVDQU 2848(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 2880(CX), Y8
 	VMOVDQU 2912(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 2944(CX), Y8
 	VMOVDQU 2976(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3008(CX), Y8
 	VMOVDQU 3040(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3072(CX), Y8
 	VMOVDQU 3104(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 7 to 7 outputs
 	VMOVDQU (R12), Y10
 	ADDQ    $0x20, R12
@@ -107160,93 +59686,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 3168(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3200(CX), Y8
 	VMOVDQU 3232(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3264(CX), Y8
 	VMOVDQU 3296(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3328(CX), Y8
 	VMOVDQU 3360(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3392(CX), Y8
 	VMOVDQU 3424(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3456(CX), Y8
 	VMOVDQU 3488(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3520(CX), Y8
 	VMOVDQU 3552(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 8 to 7 outputs
 	VMOVDQU (R13), Y10
 	ADDQ    $0x20, R13
@@ -107257,93 +59728,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 3616(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 3648(CX), Y8
 	VMOVDQU 3680(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 3712(CX), Y8
 	VMOVDQU 3744(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 3776(CX), Y8
 	VMOVDQU 3808(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 3840(CX), Y8
 	VMOVDQU 3872(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 3904(CX), Y8
 	VMOVDQU 3936(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 3968(CX), Y8
 	VMOVDQU 4000(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Load and process 32 bytes from input 9 to 7 outputs
 	VMOVDQU (DX), Y10
 	ADDQ    $0x20, DX
@@ -107354,93 +59770,38 @@ mulAvxTwo_10x7Xor_loop:
 	VMOVDQU 4064(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y0
-
-#else
-	VPXOR Y8, Y0, Y0
-	VPXOR Y9, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y0)
 	VMOVDQU 4096(CX), Y8
 	VMOVDQU 4128(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y1
-
-#else
-	VPXOR Y8, Y1, Y1
-	VPXOR Y9, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y1)
 	VMOVDQU 4160(CX), Y8
 	VMOVDQU 4192(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y2
-
-#else
-	VPXOR Y8, Y2, Y2
-	VPXOR Y9, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y2)
 	VMOVDQU 4224(CX), Y8
 	VMOVDQU 4256(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y3
-
-#else
-	VPXOR Y8, Y3, Y3
-	VPXOR Y9, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y3)
 	VMOVDQU 4288(CX), Y8
 	VMOVDQU 4320(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y4
-
-#else
-	VPXOR Y8, Y4, Y4
-	VPXOR Y9, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y4)
 	VMOVDQU 4352(CX), Y8
 	VMOVDQU 4384(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y5
-
-#else
-	VPXOR Y8, Y5, Y5
-	VPXOR Y9, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y8, Y9, Y5)
 	VMOVDQU 4416(CX), Y8
 	VMOVDQU 4448(CX), Y9
 	VPSHUFB Y10, Y8, Y8
 	VPSHUFB Y11, Y9, Y9
+	XOR3WAY( $0x00, Y8, Y9, Y6)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y8, Y9, Y6
-
-#else
-	VPXOR Y8, Y6, Y6
-	VPXOR Y9, Y6, Y6
-
-#endif
 	// Store 7 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -107564,106 +59925,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -107674,106 +59972,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -107784,106 +60019,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -107894,106 +60066,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -108004,106 +60113,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -108114,106 +60160,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (R12), Y11
 	ADDQ    $0x20, R12
@@ -108224,106 +60207,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 8 to 8 outputs
 	VMOVDQU (R13), Y11
 	ADDQ    $0x20, R13
@@ -108334,106 +60254,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 4128(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4160(CX), Y9
 	VMOVDQU 4192(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4224(CX), Y9
 	VMOVDQU 4256(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4288(CX), Y9
 	VMOVDQU 4320(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4352(CX), Y9
 	VMOVDQU 4384(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4416(CX), Y9
 	VMOVDQU 4448(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4480(CX), Y9
 	VMOVDQU 4512(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4544(CX), Y9
 	VMOVDQU 4576(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 9 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -108444,106 +60301,43 @@ mulAvxTwo_10x8_loop:
 	VMOVDQU 4640(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4672(CX), Y9
 	VMOVDQU 4704(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4736(CX), Y9
 	VMOVDQU 4768(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4800(CX), Y9
 	VMOVDQU 4832(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4864(CX), Y9
 	VMOVDQU 4896(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4928(CX), Y9
 	VMOVDQU 4960(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4992(CX), Y9
 	VMOVDQU 5024(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 5056(CX), Y9
 	VMOVDQU 5088(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -108624,120 +60418,57 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 32(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y9
 	VMOVDQU 96(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y9
 	VMOVDQU 160(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y9
 	VMOVDQU 224(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y9
 	VMOVDQU 288(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	MOVQ    120(R14), BP
 	VMOVDQU (BP)(R15*1), Y5
 	VMOVDQU 320(CX), Y9
 	VMOVDQU 352(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	MOVQ    144(R14), BP
 	VMOVDQU (BP)(R15*1), Y6
 	VMOVDQU 384(CX), Y9
 	VMOVDQU 416(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	MOVQ    168(R14), BP
 	VMOVDQU (BP)(R15*1), Y7
 	VMOVDQU 448(CX), Y9
 	VMOVDQU 480(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 1 to 8 outputs
 	VMOVDQU (SI), Y11
 	ADDQ    $0x20, SI
@@ -108748,106 +60479,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 544(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 576(CX), Y9
 	VMOVDQU 608(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 640(CX), Y9
 	VMOVDQU 672(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 704(CX), Y9
 	VMOVDQU 736(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 768(CX), Y9
 	VMOVDQU 800(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 832(CX), Y9
 	VMOVDQU 864(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 896(CX), Y9
 	VMOVDQU 928(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 960(CX), Y9
 	VMOVDQU 992(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 2 to 8 outputs
 	VMOVDQU (DI), Y11
 	ADDQ    $0x20, DI
@@ -108858,106 +60526,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 1056(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1088(CX), Y9
 	VMOVDQU 1120(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1152(CX), Y9
 	VMOVDQU 1184(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1216(CX), Y9
 	VMOVDQU 1248(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1280(CX), Y9
 	VMOVDQU 1312(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1344(CX), Y9
 	VMOVDQU 1376(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1408(CX), Y9
 	VMOVDQU 1440(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1472(CX), Y9
 	VMOVDQU 1504(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 3 to 8 outputs
 	VMOVDQU (R8), Y11
 	ADDQ    $0x20, R8
@@ -108968,106 +60573,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 1568(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 1600(CX), Y9
 	VMOVDQU 1632(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 1664(CX), Y9
 	VMOVDQU 1696(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 1728(CX), Y9
 	VMOVDQU 1760(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 1792(CX), Y9
 	VMOVDQU 1824(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 1856(CX), Y9
 	VMOVDQU 1888(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 1920(CX), Y9
 	VMOVDQU 1952(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 1984(CX), Y9
 	VMOVDQU 2016(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 4 to 8 outputs
 	VMOVDQU (R9), Y11
 	ADDQ    $0x20, R9
@@ -109078,106 +60620,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 2080(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2112(CX), Y9
 	VMOVDQU 2144(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2176(CX), Y9
 	VMOVDQU 2208(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2240(CX), Y9
 	VMOVDQU 2272(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2304(CX), Y9
 	VMOVDQU 2336(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2368(CX), Y9
 	VMOVDQU 2400(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2432(CX), Y9
 	VMOVDQU 2464(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 2496(CX), Y9
 	VMOVDQU 2528(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 5 to 8 outputs
 	VMOVDQU (R10), Y11
 	ADDQ    $0x20, R10
@@ -109188,106 +60667,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 2592(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 2624(CX), Y9
 	VMOVDQU 2656(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 2688(CX), Y9
 	VMOVDQU 2720(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 2752(CX), Y9
 	VMOVDQU 2784(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 2816(CX), Y9
 	VMOVDQU 2848(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 2880(CX), Y9
 	VMOVDQU 2912(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 2944(CX), Y9
 	VMOVDQU 2976(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3008(CX), Y9
 	VMOVDQU 3040(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 6 to 8 outputs
 	VMOVDQU (R11), Y11
 	ADDQ    $0x20, R11
@@ -109298,106 +60714,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 3104(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3136(CX), Y9
 	VMOVDQU 3168(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3200(CX), Y9
 	VMOVDQU 3232(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3264(CX), Y9
 	VMOVDQU 3296(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3328(CX), Y9
 	VMOVDQU 3360(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3392(CX), Y9
 	VMOVDQU 3424(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3456(CX), Y9
 	VMOVDQU 3488(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 3520(CX), Y9
 	VMOVDQU 3552(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 7 to 8 outputs
 	VMOVDQU (R12), Y11
 	ADDQ    $0x20, R12
@@ -109408,106 +60761,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 3616(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 3648(CX), Y9
 	VMOVDQU 3680(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 3712(CX), Y9
 	VMOVDQU 3744(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 3776(CX), Y9
 	VMOVDQU 3808(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 3840(CX), Y9
 	VMOVDQU 3872(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 3904(CX), Y9
 	VMOVDQU 3936(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 3968(CX), Y9
 	VMOVDQU 4000(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4032(CX), Y9
 	VMOVDQU 4064(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 8 to 8 outputs
 	VMOVDQU (R13), Y11
 	ADDQ    $0x20, R13
@@ -109518,106 +60808,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 4128(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4160(CX), Y9
 	VMOVDQU 4192(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4224(CX), Y9
 	VMOVDQU 4256(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4288(CX), Y9
 	VMOVDQU 4320(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4352(CX), Y9
 	VMOVDQU 4384(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4416(CX), Y9
 	VMOVDQU 4448(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4480(CX), Y9
 	VMOVDQU 4512(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 4544(CX), Y9
 	VMOVDQU 4576(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Load and process 32 bytes from input 9 to 8 outputs
 	VMOVDQU (DX), Y11
 	ADDQ    $0x20, DX
@@ -109628,106 +60855,43 @@ mulAvxTwo_10x8Xor_loop:
 	VMOVDQU 4640(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y0
-
-#else
-	VPXOR Y9, Y0, Y0
-	VPXOR Y10, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y0)
 	VMOVDQU 4672(CX), Y9
 	VMOVDQU 4704(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y1
-
-#else
-	VPXOR Y9, Y1, Y1
-	VPXOR Y10, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y1)
 	VMOVDQU 4736(CX), Y9
 	VMOVDQU 4768(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y2
-
-#else
-	VPXOR Y9, Y2, Y2
-	VPXOR Y10, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y2)
 	VMOVDQU 4800(CX), Y9
 	VMOVDQU 4832(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y3
-
-#else
-	VPXOR Y9, Y3, Y3
-	VPXOR Y10, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y3)
 	VMOVDQU 4864(CX), Y9
 	VMOVDQU 4896(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y4
-
-#else
-	VPXOR Y9, Y4, Y4
-	VPXOR Y10, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y4)
 	VMOVDQU 4928(CX), Y9
 	VMOVDQU 4960(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y5
-
-#else
-	VPXOR Y9, Y5, Y5
-	VPXOR Y10, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y5)
 	VMOVDQU 4992(CX), Y9
 	VMOVDQU 5024(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y6
-
-#else
-	VPXOR Y9, Y6, Y6
-	VPXOR Y10, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y9, Y10, Y6)
 	VMOVDQU 5056(CX), Y9
 	VMOVDQU 5088(CX), Y10
 	VPSHUFB Y11, Y9, Y9
 	VPSHUFB Y12, Y10, Y10
+	XOR3WAY( $0x00, Y9, Y10, Y7)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y9, Y10, Y7
-
-#else
-	VPXOR Y9, Y7, Y7
-	VPXOR Y10, Y7, Y7
-
-#endif
 	// Store 8 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -109858,119 +61022,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -109981,119 +61074,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -110104,119 +61126,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -110227,119 +61178,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -110350,119 +61230,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -110473,119 +61282,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (R12), Y12
 	ADDQ    $0x20, R12
@@ -110596,119 +61334,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 8 to 9 outputs
 	VMOVDQU (R13), Y12
 	ADDQ    $0x20, R13
@@ -110719,119 +61386,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 4640(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4672(CX), Y10
 	VMOVDQU 4704(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4736(CX), Y10
 	VMOVDQU 4768(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4800(CX), Y10
 	VMOVDQU 4832(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4864(CX), Y10
 	VMOVDQU 4896(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4928(CX), Y10
 	VMOVDQU 4960(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4992(CX), Y10
 	VMOVDQU 5024(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5056(CX), Y10
 	VMOVDQU 5088(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5120(CX), Y10
 	VMOVDQU 5152(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 9 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -110842,119 +61438,48 @@ mulAvxTwo_10x9_loop:
 	VMOVDQU 5216(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 5248(CX), Y10
 	VMOVDQU 5280(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 5312(CX), Y10
 	VMOVDQU 5344(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 5376(CX), Y10
 	VMOVDQU 5408(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 5440(CX), Y10
 	VMOVDQU 5472(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 5504(CX), Y10
 	VMOVDQU 5536(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 5568(CX), Y10
 	VMOVDQU 5600(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5632(CX), Y10
 	VMOVDQU 5664(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5696(CX), Y10
 	VMOVDQU 5728(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -111037,135 +61562,64 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 32(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y10
 	VMOVDQU 96(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y10
 	VMOVDQU 160(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y10
 	VMOVDQU 224(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y10
 	VMOVDQU 288(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	MOVQ    120(R14), BP
 	VMOVDQU (BP)(R15*1), Y5
 	VMOVDQU 320(CX), Y10
 	VMOVDQU 352(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	MOVQ    144(R14), BP
 	VMOVDQU (BP)(R15*1), Y6
 	VMOVDQU 384(CX), Y10
 	VMOVDQU 416(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	MOVQ    168(R14), BP
 	VMOVDQU (BP)(R15*1), Y7
 	VMOVDQU 448(CX), Y10
 	VMOVDQU 480(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	MOVQ    192(R14), BP
 	VMOVDQU (BP)(R15*1), Y8
 	VMOVDQU 512(CX), Y10
 	VMOVDQU 544(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 1 to 9 outputs
 	VMOVDQU (SI), Y12
 	ADDQ    $0x20, SI
@@ -111176,119 +61630,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 608(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 640(CX), Y10
 	VMOVDQU 672(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 704(CX), Y10
 	VMOVDQU 736(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 768(CX), Y10
 	VMOVDQU 800(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 832(CX), Y10
 	VMOVDQU 864(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 896(CX), Y10
 	VMOVDQU 928(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 960(CX), Y10
 	VMOVDQU 992(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1024(CX), Y10
 	VMOVDQU 1056(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1088(CX), Y10
 	VMOVDQU 1120(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 2 to 9 outputs
 	VMOVDQU (DI), Y12
 	ADDQ    $0x20, DI
@@ -111299,119 +61682,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 1184(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1216(CX), Y10
 	VMOVDQU 1248(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1280(CX), Y10
 	VMOVDQU 1312(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1344(CX), Y10
 	VMOVDQU 1376(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1408(CX), Y10
 	VMOVDQU 1440(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 1472(CX), Y10
 	VMOVDQU 1504(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 1536(CX), Y10
 	VMOVDQU 1568(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 1600(CX), Y10
 	VMOVDQU 1632(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 1664(CX), Y10
 	VMOVDQU 1696(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 3 to 9 outputs
 	VMOVDQU (R8), Y12
 	ADDQ    $0x20, R8
@@ -111422,119 +61734,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 1760(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 1792(CX), Y10
 	VMOVDQU 1824(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 1856(CX), Y10
 	VMOVDQU 1888(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 1920(CX), Y10
 	VMOVDQU 1952(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 1984(CX), Y10
 	VMOVDQU 2016(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2048(CX), Y10
 	VMOVDQU 2080(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2112(CX), Y10
 	VMOVDQU 2144(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2176(CX), Y10
 	VMOVDQU 2208(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2240(CX), Y10
 	VMOVDQU 2272(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 4 to 9 outputs
 	VMOVDQU (R9), Y12
 	ADDQ    $0x20, R9
@@ -111545,119 +61786,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 2336(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2368(CX), Y10
 	VMOVDQU 2400(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 2432(CX), Y10
 	VMOVDQU 2464(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 2496(CX), Y10
 	VMOVDQU 2528(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 2560(CX), Y10
 	VMOVDQU 2592(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 2624(CX), Y10
 	VMOVDQU 2656(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 2688(CX), Y10
 	VMOVDQU 2720(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 2752(CX), Y10
 	VMOVDQU 2784(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 2816(CX), Y10
 	VMOVDQU 2848(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 5 to 9 outputs
 	VMOVDQU (R10), Y12
 	ADDQ    $0x20, R10
@@ -111668,119 +61838,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 2912(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 2944(CX), Y10
 	VMOVDQU 2976(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3008(CX), Y10
 	VMOVDQU 3040(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3072(CX), Y10
 	VMOVDQU 3104(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3136(CX), Y10
 	VMOVDQU 3168(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3200(CX), Y10
 	VMOVDQU 3232(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3264(CX), Y10
 	VMOVDQU 3296(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3328(CX), Y10
 	VMOVDQU 3360(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3392(CX), Y10
 	VMOVDQU 3424(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 6 to 9 outputs
 	VMOVDQU (R11), Y12
 	ADDQ    $0x20, R11
@@ -111791,119 +61890,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 3488(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 3520(CX), Y10
 	VMOVDQU 3552(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 3584(CX), Y10
 	VMOVDQU 3616(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 3648(CX), Y10
 	VMOVDQU 3680(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 3712(CX), Y10
 	VMOVDQU 3744(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 3776(CX), Y10
 	VMOVDQU 3808(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 3840(CX), Y10
 	VMOVDQU 3872(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 3904(CX), Y10
 	VMOVDQU 3936(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 3968(CX), Y10
 	VMOVDQU 4000(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 7 to 9 outputs
 	VMOVDQU (R12), Y12
 	ADDQ    $0x20, R12
@@ -111914,119 +61942,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 4064(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4096(CX), Y10
 	VMOVDQU 4128(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4160(CX), Y10
 	VMOVDQU 4192(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4224(CX), Y10
 	VMOVDQU 4256(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4288(CX), Y10
 	VMOVDQU 4320(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4352(CX), Y10
 	VMOVDQU 4384(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4416(CX), Y10
 	VMOVDQU 4448(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 4480(CX), Y10
 	VMOVDQU 4512(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 4544(CX), Y10
 	VMOVDQU 4576(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 8 to 9 outputs
 	VMOVDQU (R13), Y12
 	ADDQ    $0x20, R13
@@ -112037,119 +61994,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 4640(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 4672(CX), Y10
 	VMOVDQU 4704(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 4736(CX), Y10
 	VMOVDQU 4768(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 4800(CX), Y10
 	VMOVDQU 4832(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 4864(CX), Y10
 	VMOVDQU 4896(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 4928(CX), Y10
 	VMOVDQU 4960(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 4992(CX), Y10
 	VMOVDQU 5024(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5056(CX), Y10
 	VMOVDQU 5088(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5120(CX), Y10
 	VMOVDQU 5152(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Load and process 32 bytes from input 9 to 9 outputs
 	VMOVDQU (DX), Y12
 	ADDQ    $0x20, DX
@@ -112160,119 +62046,48 @@ mulAvxTwo_10x9Xor_loop:
 	VMOVDQU 5216(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y0
-
-#else
-	VPXOR Y10, Y0, Y0
-	VPXOR Y11, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y0)
 	VMOVDQU 5248(CX), Y10
 	VMOVDQU 5280(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y1
-
-#else
-	VPXOR Y10, Y1, Y1
-	VPXOR Y11, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y1)
 	VMOVDQU 5312(CX), Y10
 	VMOVDQU 5344(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y2
-
-#else
-	VPXOR Y10, Y2, Y2
-	VPXOR Y11, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y2)
 	VMOVDQU 5376(CX), Y10
 	VMOVDQU 5408(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y3
-
-#else
-	VPXOR Y10, Y3, Y3
-	VPXOR Y11, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y3)
 	VMOVDQU 5440(CX), Y10
 	VMOVDQU 5472(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y4
-
-#else
-	VPXOR Y10, Y4, Y4
-	VPXOR Y11, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y4)
 	VMOVDQU 5504(CX), Y10
 	VMOVDQU 5536(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y5
-
-#else
-	VPXOR Y10, Y5, Y5
-	VPXOR Y11, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y5)
 	VMOVDQU 5568(CX), Y10
 	VMOVDQU 5600(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y6
-
-#else
-	VPXOR Y10, Y6, Y6
-	VPXOR Y11, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y6)
 	VMOVDQU 5632(CX), Y10
 	VMOVDQU 5664(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y7
-
-#else
-	VPXOR Y10, Y7, Y7
-	VPXOR Y11, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y10, Y11, Y7)
 	VMOVDQU 5696(CX), Y10
 	VMOVDQU 5728(CX), Y11
 	VPSHUFB Y12, Y10, Y10
 	VPSHUFB Y13, Y11, Y11
+	XOR3WAY( $0x00, Y10, Y11, Y8)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y10, Y11, Y8
-
-#else
-	VPXOR Y10, Y8, Y8
-	VPXOR Y11, Y8, Y8
-
-#endif
 	// Store 9 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -112410,132 +62225,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -112546,132 +62282,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -112682,132 +62339,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -112818,132 +62396,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -112954,132 +62453,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -113090,132 +62510,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (R12), Y13
 	ADDQ    $0x20, R12
@@ -113226,132 +62567,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 8 to 10 outputs
 	VMOVDQU (R13), Y13
 	ADDQ    $0x20, R13
@@ -113362,132 +62624,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 5152(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5184(CX), Y11
 	VMOVDQU 5216(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5248(CX), Y11
 	VMOVDQU 5280(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5312(CX), Y11
 	VMOVDQU 5344(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 5376(CX), Y11
 	VMOVDQU 5408(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 5440(CX), Y11
 	VMOVDQU 5472(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 5504(CX), Y11
 	VMOVDQU 5536(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 5568(CX), Y11
 	VMOVDQU 5600(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 5632(CX), Y11
 	VMOVDQU 5664(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5696(CX), Y11
 	VMOVDQU 5728(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 9 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -113498,132 +62681,53 @@ mulAvxTwo_10x10_loop:
 	VMOVDQU 5792(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5824(CX), Y11
 	VMOVDQU 5856(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5888(CX), Y11
 	VMOVDQU 5920(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5952(CX), Y11
 	VMOVDQU 5984(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 6016(CX), Y11
 	VMOVDQU 6048(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 6080(CX), Y11
 	VMOVDQU 6112(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 6144(CX), Y11
 	VMOVDQU 6176(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 6208(CX), Y11
 	VMOVDQU 6240(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 6272(CX), Y11
 	VMOVDQU 6304(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 6336(CX), Y11
 	VMOVDQU 6368(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)
@@ -113708,150 +62812,71 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 32(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	MOVQ    24(R14), BP
 	VMOVDQU (BP)(R15*1), Y1
 	VMOVDQU 64(CX), Y11
 	VMOVDQU 96(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	MOVQ    48(R14), BP
 	VMOVDQU (BP)(R15*1), Y2
 	VMOVDQU 128(CX), Y11
 	VMOVDQU 160(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	MOVQ    72(R14), BP
 	VMOVDQU (BP)(R15*1), Y3
 	VMOVDQU 192(CX), Y11
 	VMOVDQU 224(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	MOVQ    96(R14), BP
 	VMOVDQU (BP)(R15*1), Y4
 	VMOVDQU 256(CX), Y11
 	VMOVDQU 288(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	MOVQ    120(R14), BP
 	VMOVDQU (BP)(R15*1), Y5
 	VMOVDQU 320(CX), Y11
 	VMOVDQU 352(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	MOVQ    144(R14), BP
 	VMOVDQU (BP)(R15*1), Y6
 	VMOVDQU 384(CX), Y11
 	VMOVDQU 416(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	MOVQ    168(R14), BP
 	VMOVDQU (BP)(R15*1), Y7
 	VMOVDQU 448(CX), Y11
 	VMOVDQU 480(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	MOVQ    192(R14), BP
 	VMOVDQU (BP)(R15*1), Y8
 	VMOVDQU 512(CX), Y11
 	VMOVDQU 544(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	MOVQ    216(R14), BP
 	VMOVDQU (BP)(R15*1), Y9
 	VMOVDQU 576(CX), Y11
 	VMOVDQU 608(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 1 to 10 outputs
 	VMOVDQU (SI), Y13
 	ADDQ    $0x20, SI
@@ -113862,132 +62887,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 672(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 704(CX), Y11
 	VMOVDQU 736(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 768(CX), Y11
 	VMOVDQU 800(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 832(CX), Y11
 	VMOVDQU 864(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 896(CX), Y11
 	VMOVDQU 928(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 960(CX), Y11
 	VMOVDQU 992(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1024(CX), Y11
 	VMOVDQU 1056(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1088(CX), Y11
 	VMOVDQU 1120(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1152(CX), Y11
 	VMOVDQU 1184(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1216(CX), Y11
 	VMOVDQU 1248(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 2 to 10 outputs
 	VMOVDQU (DI), Y13
 	ADDQ    $0x20, DI
@@ -113998,132 +62944,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 1312(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1344(CX), Y11
 	VMOVDQU 1376(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 1408(CX), Y11
 	VMOVDQU 1440(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 1472(CX), Y11
 	VMOVDQU 1504(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 1536(CX), Y11
 	VMOVDQU 1568(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 1600(CX), Y11
 	VMOVDQU 1632(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 1664(CX), Y11
 	VMOVDQU 1696(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 1728(CX), Y11
 	VMOVDQU 1760(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 1792(CX), Y11
 	VMOVDQU 1824(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 1856(CX), Y11
 	VMOVDQU 1888(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 3 to 10 outputs
 	VMOVDQU (R8), Y13
 	ADDQ    $0x20, R8
@@ -114134,132 +63001,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 1952(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 1984(CX), Y11
 	VMOVDQU 2016(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2048(CX), Y11
 	VMOVDQU 2080(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2112(CX), Y11
 	VMOVDQU 2144(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2176(CX), Y11
 	VMOVDQU 2208(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2240(CX), Y11
 	VMOVDQU 2272(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2304(CX), Y11
 	VMOVDQU 2336(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 2368(CX), Y11
 	VMOVDQU 2400(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 2432(CX), Y11
 	VMOVDQU 2464(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 2496(CX), Y11
 	VMOVDQU 2528(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 4 to 10 outputs
 	VMOVDQU (R9), Y13
 	ADDQ    $0x20, R9
@@ -114270,132 +63058,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 2592(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 2624(CX), Y11
 	VMOVDQU 2656(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 2688(CX), Y11
 	VMOVDQU 2720(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 2752(CX), Y11
 	VMOVDQU 2784(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 2816(CX), Y11
 	VMOVDQU 2848(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 2880(CX), Y11
 	VMOVDQU 2912(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 2944(CX), Y11
 	VMOVDQU 2976(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3008(CX), Y11
 	VMOVDQU 3040(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3072(CX), Y11
 	VMOVDQU 3104(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3136(CX), Y11
 	VMOVDQU 3168(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 5 to 10 outputs
 	VMOVDQU (R10), Y13
 	ADDQ    $0x20, R10
@@ -114406,132 +63115,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 3232(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3264(CX), Y11
 	VMOVDQU 3296(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3328(CX), Y11
 	VMOVDQU 3360(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 3392(CX), Y11
 	VMOVDQU 3424(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 3456(CX), Y11
 	VMOVDQU 3488(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 3520(CX), Y11
 	VMOVDQU 3552(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 3584(CX), Y11
 	VMOVDQU 3616(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 3648(CX), Y11
 	VMOVDQU 3680(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 3712(CX), Y11
 	VMOVDQU 3744(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 3776(CX), Y11
 	VMOVDQU 3808(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 6 to 10 outputs
 	VMOVDQU (R11), Y13
 	ADDQ    $0x20, R11
@@ -114542,132 +63172,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 3872(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 3904(CX), Y11
 	VMOVDQU 3936(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 3968(CX), Y11
 	VMOVDQU 4000(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4032(CX), Y11
 	VMOVDQU 4064(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4096(CX), Y11
 	VMOVDQU 4128(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4160(CX), Y11
 	VMOVDQU 4192(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4224(CX), Y11
 	VMOVDQU 4256(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4288(CX), Y11
 	VMOVDQU 4320(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4352(CX), Y11
 	VMOVDQU 4384(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 4416(CX), Y11
 	VMOVDQU 4448(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 7 to 10 outputs
 	VMOVDQU (R12), Y13
 	ADDQ    $0x20, R12
@@ -114678,132 +63229,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 4512(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 4544(CX), Y11
 	VMOVDQU 4576(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 4608(CX), Y11
 	VMOVDQU 4640(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 4672(CX), Y11
 	VMOVDQU 4704(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 4736(CX), Y11
 	VMOVDQU 4768(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 4800(CX), Y11
 	VMOVDQU 4832(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 4864(CX), Y11
 	VMOVDQU 4896(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 4928(CX), Y11
 	VMOVDQU 4960(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 4992(CX), Y11
 	VMOVDQU 5024(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5056(CX), Y11
 	VMOVDQU 5088(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 8 to 10 outputs
 	VMOVDQU (R13), Y13
 	ADDQ    $0x20, R13
@@ -114814,132 +63286,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 5152(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5184(CX), Y11
 	VMOVDQU 5216(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5248(CX), Y11
 	VMOVDQU 5280(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5312(CX), Y11
 	VMOVDQU 5344(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 5376(CX), Y11
 	VMOVDQU 5408(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 5440(CX), Y11
 	VMOVDQU 5472(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 5504(CX), Y11
 	VMOVDQU 5536(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 5568(CX), Y11
 	VMOVDQU 5600(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 5632(CX), Y11
 	VMOVDQU 5664(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 5696(CX), Y11
 	VMOVDQU 5728(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Load and process 32 bytes from input 9 to 10 outputs
 	VMOVDQU (DX), Y13
 	ADDQ    $0x20, DX
@@ -114950,132 +63343,53 @@ mulAvxTwo_10x10Xor_loop:
 	VMOVDQU 5792(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y0
-
-#else
-	VPXOR Y11, Y0, Y0
-	VPXOR Y12, Y0, Y0
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y0)
 	VMOVDQU 5824(CX), Y11
 	VMOVDQU 5856(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y1
-
-#else
-	VPXOR Y11, Y1, Y1
-	VPXOR Y12, Y1, Y1
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y1)
 	VMOVDQU 5888(CX), Y11
 	VMOVDQU 5920(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y2
-
-#else
-	VPXOR Y11, Y2, Y2
-	VPXOR Y12, Y2, Y2
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y2)
 	VMOVDQU 5952(CX), Y11
 	VMOVDQU 5984(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y3
-
-#else
-	VPXOR Y11, Y3, Y3
-	VPXOR Y12, Y3, Y3
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y3)
 	VMOVDQU 6016(CX), Y11
 	VMOVDQU 6048(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y4
-
-#else
-	VPXOR Y11, Y4, Y4
-	VPXOR Y12, Y4, Y4
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y4)
 	VMOVDQU 6080(CX), Y11
 	VMOVDQU 6112(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y5
-
-#else
-	VPXOR Y11, Y5, Y5
-	VPXOR Y12, Y5, Y5
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y5)
 	VMOVDQU 6144(CX), Y11
 	VMOVDQU 6176(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y6
-
-#else
-	VPXOR Y11, Y6, Y6
-	VPXOR Y12, Y6, Y6
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y6)
 	VMOVDQU 6208(CX), Y11
 	VMOVDQU 6240(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y7
-
-#else
-	VPXOR Y11, Y7, Y7
-	VPXOR Y12, Y7, Y7
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y7)
 	VMOVDQU 6272(CX), Y11
 	VMOVDQU 6304(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
-
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y8
-
-#else
-	VPXOR Y11, Y8, Y8
-	VPXOR Y12, Y8, Y8
-
-#endif
+	XOR3WAY( $0x00, Y11, Y12, Y8)
 	VMOVDQU 6336(CX), Y11
 	VMOVDQU 6368(CX), Y12
 	VPSHUFB Y13, Y11, Y11
 	VPSHUFB Y14, Y12, Y12
+	XOR3WAY( $0x00, Y11, Y12, Y9)
 
-#ifdef GOAMD64_v4
-	VPTERNLOGD $0x96, Y11, Y12, Y9
-
-#else
-	VPXOR Y11, Y9, Y9
-	VPXOR Y12, Y9, Y9
-
-#endif
 	// Store 10 outputs
 	MOVQ    (R14), BP
 	VMOVDQU Y0, (BP)(R15*1)