diff --git a/helper-i386.c b/helper-i386.c
index 293d46af80..1182be8ce7 100644
--- a/helper-i386.c
+++ b/helper-i386.c
@@ -19,6 +19,57 @@
  */
 #include "exec-i386.h"
 
+const uint8_t parity_table[256] = {
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
+    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
+};
+
+/* modulo 17 table */
+const uint8_t rclw_table[32] = {
+    0, 1, 2, 3, 4, 5, 6, 7, 
+    8, 9,10,11,12,13,14,15,
+   16, 0, 1, 2, 3, 4, 5, 6,
+    7, 8, 9,10,11,12,13,14,
+};
+
+/* modulo 9 table */
+const uint8_t rclb_table[32] = {
+    0, 1, 2, 3, 4, 5, 6, 7, 
+    8, 0, 1, 2, 3, 4, 5, 6,
+    7, 8, 0, 1, 2, 3, 4, 5, 
+    6, 7, 8, 0, 1, 2, 3, 4,
+};
+
 const CPU86_LDouble f15rk[7] =
 {
     0.00000000000000000000L,
@@ -693,7 +744,29 @@ void helper_fsincos(void)
 
 void helper_frndint(void)
 {
-    ST0 = rint(ST0);
+    CPU86_LDouble a;
+
+    a = ST0;
+#ifdef __arm__
+    switch(env->fpuc & RC_MASK) {
+    default:
+    case RC_NEAR:
+        asm("rndd %0, %1" : "=f" (a) : "f"(a));
+        break;
+    case RC_DOWN:
+        asm("rnddm %0, %1" : "=f" (a) : "f"(a));
+        break;
+    case RC_UP:
+        asm("rnddp %0, %1" : "=f" (a) : "f"(a));
+        break;
+    case RC_CHOP:
+        asm("rnddz %0, %1" : "=f" (a) : "f"(a));
+        break;
+    }
+#else
+    a = rint(a);
+#endif
+    ST0 = a;
 }
 
 void helper_fscale(void)
diff --git a/op-i386.c b/op-i386.c
index e3f26b4e02..c2f92638bd 100644
--- a/op-i386.c
+++ b/op-i386.c
@@ -19,57 +19,6 @@
  */
 #include "exec-i386.h"
 
-uint8_t parity_table[256] = {
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    CC_P, 0, 0, CC_P, 0, CC_P, CC_P, 0,
-    0, CC_P, CC_P, 0, CC_P, 0, 0, CC_P,
-};
-
-/* modulo 17 table */
-const uint8_t rclw_table[32] = {
-    0, 1, 2, 3, 4, 5, 6, 7, 
-    8, 9,10,11,12,13,14,15,
-   16, 0, 1, 2, 3, 4, 5, 6,
-    7, 8, 9,10,11,12,13,14,
-};
-
-/* modulo 9 table */
-const uint8_t rclb_table[32] = {
-    0, 1, 2, 3, 4, 5, 6, 7, 
-    8, 0, 1, 2, 3, 4, 5, 6,
-    7, 8, 0, 1, 2, 3, 4, 5, 
-    6, 7, 8, 0, 1, 2, 3, 4,
-};
-
 /* n must be a constant to be efficient */
 static inline int lshift(int x, int n)
 {
@@ -624,33 +573,6 @@ void OPPROTO op_cmpxchg8b(void)
     helper_cmpxchg8b();
 }
 
-#if defined(__powerpc__)
-
-/* on PowerPC we patch the jump instruction directly */
-#define JUMP_TB(tbparam, n, eip)\
-do {\
-    static void __attribute__((unused)) *__op_label ## n = &&label ## n;\
-    asm volatile ("b %0" : : "i" (&__op_jmp ## n));\
-label ## n:\
-    T0 = (long)(tbparam) + (n);\
-    EIP = eip;\
-} while (0)
-
-#else
-
-/* jump to next block operations (more portable code, does not need
-   cache flushing, but slower because of indirect jump) */
-#define JUMP_TB(tbparam, n, eip)\
-do {\
-    static void __attribute__((unused)) *__op_label ## n = &&label ## n;\
-    goto *(void *)(((TranslationBlock *)tbparam)->tb_next[n]);\
-label ## n:\
-    T0 = (long)(tbparam) + (n);\
-    EIP = eip;\
-} while (0)
-
-#endif
-
 void OPPROTO op_jmp_tb_next(void)
 {
     JUMP_TB(PARAM1, 0, PARAM2);
@@ -1561,6 +1483,8 @@ void OPPROTO op_fist_ST0_A0(void)
 
     d = ST0;
     val = lrint(d);
+    if (val != (int16_t)val)
+        val = -32768;
     stw((void *)A0, val);
 }