Compare commits

...

25 Commits

Author SHA1 Message Date
e651c93a90 Release 0.9.6
All checks were successful
Test / test_interrupted_rebalance (push) Successful in 1m59s
Test / test_interrupted_rebalance_imm (push) Successful in 3m41s
Test / test_interrupted_rebalance_ec (push) Successful in 1m53s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m29s
Test / test_failure_domain (push) Successful in 50s
Test / test_snapshot (push) Successful in 45s
Test / test_snapshot_ec (push) Successful in 23s
Test / test_minsize_1 (push) Successful in 15s
Test / test_move_reappear (push) Successful in 18s
Test / test_rm (push) Successful in 15s
Test / test_snapshot_chain (push) Successful in 2m23s
Test / test_snapshot_chain_ec (push) Successful in 3m1s
Test / test_snapshot_down (push) Successful in 31s
Test / test_snapshot_down_ec (push) Successful in 32s
Test / test_splitbrain (push) Successful in 19s
Test / test_rebalance_verify (push) Successful in 3m34s
Test / test_rebalance_verify_imm (push) Successful in 3m31s
Test / test_rebalance_verify_ec (push) Successful in 5m14s
Test / test_rebalance_verify_ec_imm (push) Successful in 5m14s
Test / test_write (push) Successful in 44s
Test / test_write_xor (push) Successful in 54s
Test / test_write_no_same (push) Successful in 15s
Test / test_heal_pg_size_2 (push) Successful in 4m38s
Test / test_heal_ec (push) Successful in 3m56s
Test / test_scrub (push) Successful in 36s
Test / test_scrub_zero_osd_2 (push) Successful in 35s
Test / test_scrub_xor (push) Successful in 31s
Test / test_scrub_pg_size_3 (push) Successful in 46s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 25s
Test / test_scrub_ec (push) Successful in 24s
- Fix vitastor-disk partition zeroing (sometimes it was writing garbage instead of zeroes)
- Fix incorrect EC space statistics in `vitastor-cli status`
- Several bug fixes for NFS:
  - Add . and .. in NFS directory listings
  - Return FILE_SYNC from NFS writes if immediate_commit is enabled
  - Return the same "verifier" in NFS COMMIT as in NFS WRITE
  - Make parallel NFS extending writes work correctly, without conflicts
  - Handle parallel NFS extending writes without imposing extra load on etcd
- Support UTF-8 in vitastor-cli table output
- Also allow "0" and "no" as false for inmemory_metadata and inmemory_journal
- Use HDD defaults for HDD-only in automatic `vitastor-disk prepare` mode
2023-07-29 10:54:00 +03:00
988e90be69 Fix vitastor-disk partition zeroing (it was writing random garbage instead of zeroes :D)
All checks were successful
Test / test_interrupted_rebalance (push) Successful in 1m59s
Test / test_interrupted_rebalance_imm (push) Successful in 1m38s
Test / test_interrupted_rebalance_ec (push) Successful in 1m12s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m18s
Test / test_failure_domain (push) Successful in 8s
Test / test_snapshot (push) Successful in 19s
Test / test_snapshot_ec (push) Successful in 20s
Test / test_minsize_1 (push) Successful in 12s
Test / test_move_reappear (push) Successful in 18s
Test / test_rm (push) Successful in 12s
Test / test_snapshot_chain (push) Successful in 1m3s
Test / test_snapshot_chain_ec (push) Successful in 1m25s
Test / test_snapshot_down (push) Successful in 20s
Test / test_snapshot_down_ec (push) Successful in 20s
Test / test_splitbrain (push) Successful in 12s
Test / test_rebalance_verify (push) Successful in 3m16s
Test / test_rebalance_verify_imm (push) Successful in 3m40s
Test / test_rebalance_verify_ec (push) Successful in 4m17s
Test / test_rebalance_verify_ec_imm (push) Successful in 4m20s
Test / test_write (push) Successful in 40s
Test / test_write_xor (push) Successful in 33s
Test / test_write_no_same (push) Successful in 13s
Test / test_heal_pg_size_2 (push) Successful in 3m54s
Test / test_heal_ec (push) Successful in 4m7s
Test / test_scrub (push) Successful in 58s
Test / test_scrub_zero_osd_2 (push) Successful in 53s
Test / test_scrub_xor (push) Successful in 40s
Test / test_scrub_pg_size_3 (push) Successful in 1m6s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 37s
Test / test_scrub_ec (push) Successful in 39s
2023-07-28 12:29:07 +03:00
272a45ad63 Fix modprobe command in docs 2023-07-27 23:57:02 +03:00
25a15d24cf Fix incorrect EC space statistics in vitastor-cli status
All checks were successful
Test / test_etcd_fail (push) Successful in 1m21s
Test / test_interrupted_rebalance_imm (push) Successful in 2m9s
Test / test_interrupted_rebalance_ec (push) Successful in 1m52s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m25s
Test / test_failure_domain (push) Successful in 10s
Test / test_snapshot (push) Successful in 28s
Test / test_snapshot_ec (push) Successful in 30s
Test / test_minsize_1 (push) Successful in 15s
Test / test_move_reappear (push) Successful in 17s
Test / test_rm (push) Successful in 11s
Test / test_snapshot_chain (push) Successful in 2m1s
Test / test_snapshot_chain_ec (push) Successful in 2m41s
Test / test_snapshot_down (push) Successful in 23s
Test / test_snapshot_down_ec (push) Successful in 24s
Test / test_splitbrain (push) Successful in 17s
Test / test_rebalance_verify (push) Successful in 3m9s
Test / test_rebalance_verify_imm (push) Successful in 3m9s
Test / test_rebalance_verify_ec (push) Successful in 3m23s
Test / test_rebalance_verify_ec_imm (push) Successful in 5m38s
Test / test_write (push) Successful in 33s
Test / test_write_xor (push) Successful in 43s
Test / test_write_no_same (push) Successful in 14s
Test / test_heal_pg_size_2 (push) Successful in 4m16s
Test / test_heal_ec (push) Successful in 5m0s
Test / test_scrub (push) Successful in 56s
Test / test_scrub_zero_osd_2 (push) Successful in 41s
Test / test_scrub_xor (push) Successful in 32s
Test / test_scrub_pg_size_3 (push) Successful in 53s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 45s
Test / test_scrub_ec (push) Successful in 40s
2023-07-27 02:26:17 +00:00
700e0e9bff Handle parallel NFS extending writes without imposing extra load on etcd 2023-07-27 02:26:17 +00:00
ab0ca7c00f Return FILE_SYNC from NFS writes if immediate_commit is enabled 2023-07-26 02:09:47 +03:00
f153bc950b Return the same "verifier" in NFS COMMIT as in NFS WRITE
This fixes buffered (not O_DIRECT) NFS writes in Linux - previously they were
hanging in an infinite loop because COMMIT didn't return the same verifier as
previous WRITEs, and NFS kernel client was infinitely retrying the same writes.

Also this probably allows for correct NFS failover, at least for the same
buffered writes, because NFS clients repeat all write requests until a COMMIT
confirms them.
2023-07-26 02:09:47 +03:00
425ff8818d Add . and .. in NFS directory listings
MC, for example, hangs with infinite listing retries without them
2023-07-26 02:09:47 +03:00
9e287a7778 Handle extending writes correctly in NFS proxy
Previously, multiple parallel writes extending file size through NFS were
racing with each other and triggering deletions of part of the written data

I.e. if you mounted vitastor-nfs and just copied a file into it in MC then
you could end up with only a part of the file actually written
2023-07-26 02:09:43 +03:00
f52f58b9e9 Support UTF-8 in vitastor-cli table output 2023-07-25 01:48:57 +00:00
1fe6b0c0e2 Also allow "0" and "no" as false for inmemory_metadata and inmemory_journal 2023-07-25 01:48:57 +00:00
e4237e9ed8 Enable HDD defaults for HDD-only in automatic vitastor-disk prepare mode
Some checks failed
Test / test_interrupted_rebalance (push) Successful in 2m20s
Test / test_interrupted_rebalance_imm (push) Failing after 10m5s
Test / test_interrupted_rebalance_ec (push) Successful in 2m4s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m19s
Test / test_failure_domain (push) Successful in 39s
Test / test_snapshot (push) Successful in 34s
Test / test_snapshot_ec (push) Successful in 23s
Test / test_minsize_1 (push) Successful in 12s
Test / test_move_reappear (push) Successful in 19s
Test / test_rm (push) Successful in 12s
Test / test_snapshot_chain (push) Successful in 1m59s
Test / test_snapshot_chain_ec (push) Successful in 2m40s
Test / test_snapshot_down (push) Successful in 22s
Test / test_snapshot_down_ec (push) Successful in 24s
Test / test_splitbrain (push) Successful in 15s
Test / test_rebalance_verify (push) Successful in 3m10s
Test / test_rebalance_verify_imm (push) Successful in 3m11s
Test / test_rebalance_verify_ec (push) Successful in 3m29s
Test / test_rebalance_verify_ec_imm (push) Successful in 5m54s
Test / test_write (push) Successful in 32s
Test / test_write_xor (push) Successful in 35s
Test / test_write_no_same (push) Successful in 18s
Test / test_heal_pg_size_2 (push) Successful in 3m59s
Test / test_heal_ec (push) Successful in 5m12s
Test / test_scrub (push) Successful in 59s
Test / test_scrub_zero_osd_2 (push) Successful in 47s
Test / test_scrub_xor (push) Successful in 30s
Test / test_scrub_pg_size_3 (push) Successful in 53s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 38s
Test / test_scrub_ec (push) Successful in 22s
2023-07-23 02:33:22 +03:00
10a5fd6abb Release 0.9.5
All checks were successful
Test / test_etcd_fail (push) Successful in 50s
Test / test_interrupted_rebalance (push) Successful in 2m27s
Test / test_interrupted_rebalance_imm (push) Successful in 1m39s
Test / test_interrupted_rebalance_ec (push) Successful in 1m49s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m29s
Test / test_failure_domain (push) Successful in 19s
Test / test_snapshot (push) Successful in 31s
Test / test_snapshot_ec (push) Successful in 26s
Test / test_minsize_1 (push) Successful in 15s
Test / test_rm (push) Successful in 18s
Test / test_snapshot_chain (push) Successful in 1m49s
Test / test_snapshot_chain_ec (push) Successful in 2m51s
Test / test_snapshot_down (push) Successful in 26s
Test / test_snapshot_down_ec (push) Successful in 24s
Test / test_splitbrain (push) Successful in 18s
Test / test_rebalance_verify (push) Successful in 3m8s
Test / test_rebalance_verify_imm (push) Successful in 3m13s
Test / test_rebalance_verify_ec (push) Successful in 3m36s
Test / test_rebalance_verify_ec_imm (push) Successful in 5m59s
Test / test_write (push) Successful in 48s
Test / test_write_xor (push) Successful in 37s
Test / test_write_no_same (push) Successful in 14s
Test / test_heal_pg_size_2 (push) Successful in 3m43s
Test / test_heal_ec (push) Successful in 4m6s
Test / test_scrub (push) Successful in 35s
Test / test_scrub_zero_osd_2 (push) Successful in 34s
Test / test_scrub_xor (push) Successful in 42s
Test / test_scrub_pg_size_3 (push) Successful in 52s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 33s
Test / test_scrub_ec (push) Successful in 28s
A hotfix to 0.9.4 containing only one bugfix: 100% CPU usage in the new QEMU
driver caused by the lack of eventfd reset on io_uring event handling :)
2023-07-21 00:04:41 +03:00
1c316ef350 Reset eventfd on every ringloop::loop() 2023-07-21 00:04:41 +03:00
0b2d12eef1 Remove has_work, it was unnecessary 2023-07-21 00:04:37 +03:00
1c10430ae1 Release 0.9.4
All checks were successful
Test / test_interrupted_rebalance (push) Successful in 1m54s
Test / test_interrupted_rebalance_imm (push) Successful in 2m4s
Test / test_interrupted_rebalance_ec (push) Successful in 1m40s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m25s
Test / test_failure_domain (push) Successful in 15s
Test / test_snapshot (push) Successful in 25s
Test / test_snapshot_ec (push) Successful in 20s
Test / test_minsize_1 (push) Successful in 13s
Test / test_move_reappear (push) Successful in 16s
Test / test_rm (push) Successful in 13s
Test / test_snapshot_chain (push) Successful in 1m56s
Test / test_snapshot_chain_ec (push) Successful in 2m33s
Test / test_snapshot_down (push) Successful in 23s
Test / test_snapshot_down_ec (push) Successful in 22s
Test / test_splitbrain (push) Successful in 16s
Test / test_rebalance_verify (push) Successful in 3m3s
Test / test_rebalance_verify_imm (push) Successful in 3m2s
Test / test_rebalance_verify_ec (push) Successful in 3m13s
Test / test_rebalance_verify_ec_imm (push) Successful in 8m35s
Test / test_write (push) Successful in 33s
Test / test_write_xor (push) Successful in 40s
Test / test_write_no_same (push) Successful in 15s
Test / test_heal_pg_size_2 (push) Successful in 4m25s
Test / test_heal_ec (push) Successful in 3m9s
Test / test_scrub (push) Successful in 1m0s
Test / test_scrub_zero_osd_2 (push) Successful in 46s
Test / test_scrub_xor (push) Successful in 1m1s
Test / test_scrub_pg_size_3 (push) Successful in 1m55s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 1m25s
Test / test_scrub_ec (push) Successful in 52s
- Improve QEMU driver performance by integrating io_uring in it (up to 1.5x total iops improvement)
- Fix QEMU driver deadlocks which started to reproduce in qemu-img after iothread fixes
- Fix `vitastor-cli status` reporting more etcds than actually exists (fix etcd address duplication in config on reload)
- Fix `vitastor-cli ls` crashing on inodes in non-existing pools
- Delete old garbage /pool/stats/ keys for non-existing (deleted) pools
- Reduce memory usage of etcds initialized by make-etcd script
- Fix OSDs almost always crashing on etcd restart due to "revisions were compacted" (support reloading state from etcd)
- Fix a crash and a stall possible mostly in HDD setups with small journal and big (512k, 900k) random writes
- Add notes about HDDs to documentation. You are officially allowed to use HDD-only Vitastor with HGST/Toshiba/EXOS :)
2023-07-19 02:50:30 +03:00
dfce91d168 Change git url in docs, correct block/vitastor.c path 2023-07-19 01:02:12 +03:00
332a13ba30 Build patched QEMU against local packages 2023-07-19 00:05:02 +03:00
d0e257ee81 Fix non-existing pool handling in vitastor-cli ls
All checks were successful
Test / test_interrupted_rebalance (push) Successful in 2m6s
Test / test_interrupted_rebalance_imm (push) Successful in 3m11s
Test / test_interrupted_rebalance_ec (push) Successful in 2m6s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 2m20s
Test / test_failure_domain (push) Successful in 22s
Test / test_snapshot (push) Successful in 50s
Test / test_snapshot_ec (push) Successful in 33s
Test / test_minsize_1 (push) Successful in 13s
Test / test_move_reappear (push) Successful in 1m23s
Test / test_rm (push) Successful in 13s
Test / test_snapshot_chain (push) Successful in 2m22s
Test / test_snapshot_chain_ec (push) Successful in 3m6s
Test / test_snapshot_down (push) Successful in 24s
Test / test_snapshot_down_ec (push) Successful in 22s
Test / test_splitbrain (push) Successful in 19s
Test / test_rebalance_verify (push) Successful in 3m28s
Test / test_rebalance_verify_imm (push) Successful in 3m27s
Test / test_rebalance_verify_ec (push) Successful in 9m10s
Test / test_rebalance_verify_ec_imm (push) Successful in 9m29s
Test / test_write (push) Successful in 1m36s
Test / test_write_xor (push) Successful in 2m17s
Test / test_write_no_same (push) Successful in 36s
Test / test_heal_pg_size_2 (push) Successful in 6m27s
Test / test_heal_ec (push) Successful in 5m53s
Test / test_scrub (push) Successful in 44s
Test / test_scrub_zero_osd_2 (push) Successful in 35s
Test / test_scrub_xor (push) Successful in 36s
Test / test_scrub_pg_size_3 (push) Successful in 1m1s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 46s
Test / test_scrub_ec (push) Successful in 36s
2023-07-18 23:52:02 +03:00
004912aac0 Add RPM spec patches for 6.2-el8 and 7.2-el9
Some checks failed
Test / test_interrupted_rebalance (push) Successful in 1m57s
Test / test_interrupted_rebalance_imm (push) Successful in 3m16s
Test / test_interrupted_rebalance_ec (push) Successful in 1m52s
Test / test_interrupted_rebalance_ec_imm (push) Successful in 1m25s
Test / test_failure_domain (push) Failing after 47s
Test / test_snapshot (push) Successful in 40s
Test / test_snapshot_ec (push) Successful in 24s
Test / test_minsize_1 (push) Successful in 16s
Test / test_move_reappear (push) Failing after 52s
Test / test_rm (push) Successful in 19s
Test / test_snapshot_chain (push) Successful in 2m27s
Test / test_snapshot_chain_ec (push) Failing after 3m9s
Test / test_snapshot_down (push) Successful in 22s
Test / test_snapshot_down_ec (push) Successful in 21s
Test / test_splitbrain (push) Successful in 21s
Test / test_rebalance_verify (push) Successful in 3m34s
Test / test_rebalance_verify_imm (push) Successful in 3m32s
Test / test_rebalance_verify_ec (push) Successful in 5m14s
Test / test_rebalance_verify_ec_imm (push) Successful in 5m18s
Test / test_write (push) Successful in 49s
Test / test_write_xor (push) Successful in 58s
Test / test_write_no_same (push) Successful in 13s
Test / test_heal_pg_size_2 (push) Successful in 3m55s
Test / test_heal_ec (push) Failing after 10m39s
Test / test_scrub (push) Successful in 33s
Test / test_scrub_zero_osd_2 (push) Successful in 29s
Test / test_scrub_xor (push) Successful in 30s
Test / test_scrub_pg_size_3 (push) Successful in 1m1s
Test / test_scrub_pg_size_6_pg_minsize_4_osd_count_6_ec (push) Successful in 44s
Test / test_scrub_ec (push) Successful in 25s
2023-07-18 23:38:14 +03:00
c18e92273e Copy qemu 5.1 -> 5.2 patch for convenience 2023-07-18 23:37:53 +03:00
9815d70ffc It is impossible to use io_uring with older vitastor-client because it does not have vitastor_c_uring_has_work() 2023-07-18 23:37:53 +03:00
4a4627dcab Do not use bool in C library 2023-07-18 23:37:53 +03:00
b963f2fd93 Add QEMU 2.12 patch (basically the same as 3.1) 2023-07-18 23:37:06 +03:00
ba7427020e Fix deadlocks possible in qemu-img after fixing iothread
Deadlock was caused by switching QEMU coroutines directly inside
vitastor_co_read_bitmap_cb() callback. The correct way is to schedule a BH
/BH is a QEMU term for setImmediate() :)/, same as in read and write callbacks.
2023-07-18 23:32:16 +03:00
42 changed files with 867 additions and 141 deletions

View File

@@ -2,6 +2,6 @@ cmake_minimum_required(VERSION 2.8.12)
project(vitastor)
set(VERSION "0.9.3")
set(VERSION "0.9.6")
add_subdirectory(src)

View File

@@ -1,4 +1,4 @@
VERSION ?= v0.9.3
VERSION ?= v0.9.6
all: build push

View File

@@ -49,7 +49,7 @@ spec:
capabilities:
add: ["SYS_ADMIN"]
allowPrivilegeEscalation: true
image: vitalif/vitastor-csi:v0.9.3
image: vitalif/vitastor-csi:v0.9.6
args:
- "--node=$(NODE_ID)"
- "--endpoint=$(CSI_ENDPOINT)"

View File

@@ -116,7 +116,7 @@ spec:
privileged: true
capabilities:
add: ["SYS_ADMIN"]
image: vitalif/vitastor-csi:v0.9.3
image: vitalif/vitastor-csi:v0.9.6
args:
- "--node=$(NODE_ID)"
- "--endpoint=$(CSI_ENDPOINT)"

View File

@@ -5,7 +5,7 @@ package vitastor
const (
vitastorCSIDriverName = "csi.vitastor.io"
vitastorCSIDriverVersion = "0.9.3"
vitastorCSIDriverVersion = "0.9.6"
)
// Config struct fills the parameters of request or user input

4
debian/changelog vendored
View File

@@ -1,10 +1,10 @@
vitastor (0.9.3-1) unstable; urgency=medium
vitastor (0.9.6-1) unstable; urgency=medium
* Bugfixes
-- Vitaliy Filippov <vitalif@yourcmc.ru> Fri, 03 Jun 2022 02:09:44 +0300
vitastor (0.9.3-1) unstable; urgency=medium
vitastor (0.9.6-1) unstable; urgency=medium
* Implement NFS proxy
* Add documentation

View File

@@ -28,13 +28,19 @@ RUN apt-get --download-only source qemu
ADD patches /root/vitastor/patches
ADD src/qemu_driver.c /root/vitastor/src/qemu_driver.c
#RUN set -e; \
# apt-get install -y wget; \
# wget -q -O /etc/apt/trusted.gpg.d/vitastor.gpg https://vitastor.io/debian/pubkey.gpg; \
# (echo deb http://vitastor.io/debian $REL main > /etc/apt/sources.list.d/vitastor.list); \
# (echo "APT::Install-Recommends false;" > /etc/apt/apt.conf) && \
# apt-get update; \
# apt-get install -y vitastor-client vitastor-client-dev quilt
RUN set -e; \
apt-get install -y wget; \
wget -q -O /etc/apt/trusted.gpg.d/vitastor.gpg https://vitastor.io/debian/pubkey.gpg; \
(echo deb http://vitastor.io/debian $REL main > /etc/apt/sources.list.d/vitastor.list); \
(echo "APT::Install-Recommends false;" > /etc/apt/apt.conf) && \
dpkg -i /root/packages/vitastor-$REL/vitastor-client_*.deb /root/packages/vitastor-$REL/vitastor-client-dev_*.deb; \
apt-get update; \
apt-get install -y vitastor-client vitastor-client-dev quilt; \
apt-get install -y quilt; \
mkdir -p /root/packages/qemu-$REL; \
rm -rf /root/packages/qemu-$REL/*; \
cd /root/packages/qemu-$REL; \

View File

@@ -35,8 +35,8 @@ RUN set -e -x; \
mkdir -p /root/packages/vitastor-$REL; \
rm -rf /root/packages/vitastor-$REL/*; \
cd /root/packages/vitastor-$REL; \
cp -r /root/vitastor vitastor-0.9.3; \
cd vitastor-0.9.3; \
cp -r /root/vitastor vitastor-0.9.6; \
cd vitastor-0.9.6; \
ln -s /root/fio-build/fio-*/ ./fio; \
FIO=$(head -n1 fio/debian/changelog | perl -pe 's/^.*\((.*?)\).*$/$1/'); \
ls /usr/include/linux/raw.h || cp ./debian/raw.h /usr/include/linux/raw.h; \
@@ -49,8 +49,8 @@ RUN set -e -x; \
rm -rf a b; \
echo "dep:fio=$FIO" > debian/fio_version; \
cd /root/packages/vitastor-$REL; \
tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.9.3.orig.tar.xz vitastor-0.9.3; \
cd vitastor-0.9.3; \
tar --sort=name --mtime='2020-01-01' --owner=0 --group=0 --exclude=debian -cJf vitastor_0.9.6.orig.tar.xz vitastor-0.9.6; \
cd vitastor-0.9.6; \
V=$(head -n1 debian/changelog | perl -pe 's/^.*\((.*?)\).*$/$1/'); \
DEBFULLNAME="Vitaliy Filippov <vitalif@yourcmc.ru>" dch -D $REL -v "$V""$REL" "Rebuild for $REL"; \
DEB_BUILD_OPTIONS=nocheck dpkg-buildpackage --jobs=auto -sa; \

View File

@@ -21,7 +21,7 @@
## Basic instructions
Download source, for example using git: `git clone --recurse-submodules https://yourcmc.ru/git/vitalif/vitastor/`
Download source, for example using git: `git clone --recurse-submodules https://git.yourcmc.ru/vitalif/vitastor/`
Get `fio` source and symlink it into `<vitastor>/fio`. If you don't want to build fio engine,
you can disable it by passing `-DWITH_FIO=no` to cmake.
@@ -41,7 +41,7 @@ It's recommended to build the QEMU driver (qemu_driver.c) in-tree, as a part of
QEMU build process. To do that:
- Install vitastor client library headers (from source or from vitastor-client-dev package)
- Take a corresponding patch from `patches/qemu-*-vitastor.patch` and apply it to QEMU source
- Copy `src/qemu_driver.c` to QEMU source directory as `block/block-vitastor.c`
- Copy `src/qemu_driver.c` to QEMU source directory as `block/vitastor.c`
- Build QEMU as usual
But it is also possible to build it out-of-tree. To do that:

View File

@@ -21,7 +21,7 @@
## Базовая инструкция
Скачайте исходные коды, например, из git: `git clone --recurse-submodules https://yourcmc.ru/git/vitalif/vitastor/`
Скачайте исходные коды, например, из git: `git clone --recurse-submodules https://git.yourcmc.ru/vitalif/vitastor/`
Скачайте исходные коды пакета `fio`, распакуйте их и создайте символическую ссылку на них
в директории исходников Vitastor: `<vitastor>/fio`. Либо, если вы не хотите собирать плагин fio,
@@ -41,7 +41,7 @@ cmake .. && make -j8 install
Драйвер QEMU (qemu_driver.c) рекомендуется собирать вместе с самим QEMU. Для этого:
- Установите заголовки клиентской библиотеки Vitastor (из исходников или из пакета vitastor-client-dev)
- Возьмите соответствующий патч из `patches/qemu-*-vitastor.patch` и примените его к исходникам QEMU
- Скопируйте [src/qemu_driver.c](../../src/qemu_driver.c) в директорию исходников QEMU как `block/block-vitastor.c`
- Скопируйте [src/qemu_driver.c](../../src/qemu_driver.c) в директорию исходников QEMU как `block/vitastor.c`
- Соберите QEMU как обычно
Однако в целях отладки драйвер также можно собирать отдельно от QEMU. Для этого:
@@ -60,7 +60,7 @@ cmake .. && make -j8 install
* Для QEMU 2.0+: `<qemu>/qapi-types.h` &rarr; `<vitastor>/qemu/b/qemu/qapi-types.h`
- `config-host.h` и `qapi` нужны, т.к. в них содержатся автогенерируемые заголовки
- Сконфигурируйте cmake Vitastor с `WITH_QEMU=yes` (`cmake .. -DWITH_QEMU=yes`) и, если вы
используете RHEL-подобый дистрибутив, также с `QEMU_PLUGINDIR=qemu-kvm`.
используете RHEL-подобный дистрибутив, также с `QEMU_PLUGINDIR=qemu-kvm`.
- После этого в процессе сборки Vitastor также будет собираться подходящий для вашей
версии QEMU `block-vitastor.so`.
- Таким образом можно использовать драйвер даже с немодифицированным QEMU, но в этом случае

View File

@@ -107,7 +107,8 @@ disabled by now, so if you want to try it on Debian, use a kernel from Ubuntu
Commands to attach Vitastor image as a VDUSE device:
```
modprobe vduse virtio-vdpa
modprobe vduse
modprobe virtio-vdpa
qemu-storage-daemon --daemonize --blockdev '{"node-name":"test1","driver":"vitastor",\
"etcd-host":"192.168.7.2:2379/v3","image":"testosd1","cache":{"direct":true,"no-flush":false},"discard":"unmap"}' \
--export vduse-blk,id=test1,node-name=test1,name=test1,num-queues=16,queue-size=128,writable=true

View File

@@ -111,7 +111,8 @@ VDUSE (CONFIG_VIRTIO_VDPA=m и CONFIG_VDPA_USER=m). В ядрах в Debian Linu
Команды для подключения виртуального диска через VDUSE:
```
modprobe vduse virtio-vdpa
modprobe vduse
modprobe virtio-vdpa
qemu-storage-daemon --daemonize --blockdev '{"node-name":"test1","driver":"vitastor",\
"etcd-host":"192.168.7.2:2379/v3","image":"testosd1","cache":{"direct":true,"no-flush":false},"discard":"unmap"}' \
--export vduse-blk,id=test1,node-name=test1,name=test1,num-queues=16,queue-size=128,writable=true

View File

@@ -1497,10 +1497,14 @@ class Mon
break;
}
}
const pool_cfg = (this.state.config.pools[pool_id]||{});
if (!object_size)
{
object_size = (this.state.config.pools[pool_id]||{}).block_size ||
this.config.block_size || 131072;
object_size = pool_cfg.block_size || this.config.block_size || 131072;
}
if (pool_cfg.scheme !== 'replicated')
{
object_size *= ((pool_cfg.pg_size||0) - (pool_cfg.parity_chunks||0));
}
object_size = BigInt(object_size);
for (const pg_num in this.state.pg.stats[pool_id])

View File

@@ -50,7 +50,7 @@ from cinder.volume import configuration
from cinder.volume import driver
from cinder.volume import volume_utils
VERSION = '0.9.3'
VERSION = '0.9.6'
LOG = logging.getLogger(__name__)

View File

@@ -0,0 +1,176 @@
diff --git a/block/Makefile.objs b/block/Makefile.objs
index d644bac60a..e404236291 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -19,6 +19,7 @@ block-obj-$(if $(CONFIG_LIBISCSI),y,n) += iscsi-opts.o
block-obj-$(CONFIG_LIBNFS) += nfs.o
block-obj-$(CONFIG_CURL) += curl.o
block-obj-$(CONFIG_RBD) += rbd.o
+block-obj-$(CONFIG_VITASTOR) += vitastor.o
block-obj-$(CONFIG_GLUSTERFS) += gluster.o
block-obj-$(CONFIG_VXHS) += vxhs.o
block-obj-$(CONFIG_LIBSSH2) += ssh.o
@@ -39,6 +40,8 @@ curl.o-cflags := $(CURL_CFLAGS)
curl.o-libs := $(CURL_LIBS)
rbd.o-cflags := $(RBD_CFLAGS)
rbd.o-libs := $(RBD_LIBS)
+vitastor.o-cflags := $(VITASTOR_CFLAGS)
+vitastor.o-libs := $(VITASTOR_LIBS)
gluster.o-cflags := $(GLUSTERFS_CFLAGS)
gluster.o-libs := $(GLUSTERFS_LIBS)
vxhs.o-libs := $(VXHS_LIBS)
diff --git a/configure b/configure
index 0a19b033bc..58b7fbf24c 100755
--- a/configure
+++ b/configure
@@ -398,6 +398,7 @@ trace_backends="log"
trace_file="trace"
spice=""
rbd=""
+vitastor=""
smartcard=""
libusb=""
usb_redir=""
@@ -1213,6 +1214,10 @@ for opt do
;;
--enable-rbd) rbd="yes"
;;
+ --disable-vitastor) vitastor="no"
+ ;;
+ --enable-vitastor) vitastor="yes"
+ ;;
--disable-xfsctl) xfs="no"
;;
--enable-xfsctl) xfs="yes"
@@ -1601,6 +1606,7 @@ disabled with --disable-FEATURE, default is enabled if available:
vhost-crypto vhost-crypto acceleration support
spice spice
rbd rados block device (rbd)
+ vitastor vitastor block device
libiscsi iscsi support
libnfs nfs support
smartcard smartcard support (libcacard)
@@ -3594,6 +3600,27 @@ EOF
fi
fi
+##########################################
+# vitastor probe
+if test "$vitastor" != "no" ; then
+ cat > $TMPC <<EOF
+#include <vitastor_c.h>
+int main(void) {
+ vitastor_c_create_qemu(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+ return 0;
+}
+EOF
+ vitastor_libs="-lvitastor_client"
+ if compile_prog "" "$vitastor_libs" ; then
+ vitastor=yes
+ else
+ if test "$vitastor" = "yes" ; then
+ feature_not_found "vitastor block device" "Install vitastor-client-dev"
+ fi
+ vitastor=no
+ fi
+fi
+
##########################################
# libssh2 probe
min_libssh2_version=1.2.8
@@ -5837,6 +5864,7 @@ echo "Trace output file $trace_file-<pid>"
fi
echo "spice support $spice $(echo_version $spice $spice_protocol_version/$spice_server_version)"
echo "rbd support $rbd"
+echo "vitastor support $vitastor"
echo "xfsctl support $xfs"
echo "smartcard support $smartcard"
echo "libusb $libusb"
@@ -6416,6 +6444,11 @@ if test "$rbd" = "yes" ; then
echo "RBD_CFLAGS=$rbd_cflags" >> $config_host_mak
echo "RBD_LIBS=$rbd_libs" >> $config_host_mak
fi
+if test "$vitastor" = "yes" ; then
+ echo "CONFIG_VITASTOR=m" >> $config_host_mak
+ echo "VITASTOR_CFLAGS=$vitastor_cflags" >> $config_host_mak
+ echo "VITASTOR_LIBS=$vitastor_libs" >> $config_host_mak
+fi
echo "CONFIG_COROUTINE_BACKEND=$coroutine" >> $config_host_mak
if test "$coroutine_pool" = "yes" ; then
diff --git a/qapi/block-core.json b/qapi/block-core.json
index c50517bff3..c780bb2c1c 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -2514,7 +2514,7 @@
'dmg', 'file', 'ftp', 'ftps', 'gluster', 'host_cdrom',
'host_device', 'http', 'https', 'iscsi', 'luks', 'nbd', 'nfs',
'null-aio', 'null-co', 'nvme', 'parallels', 'qcow', 'qcow2', 'qed',
- 'quorum', 'raw', 'rbd', 'replication', 'sheepdog', 'ssh',
+ 'quorum', 'raw', 'rbd', 'vitastor', 'replication', 'sheepdog', 'ssh',
'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat', 'vxhs' ] }
##
@@ -3217,6 +3217,28 @@
'*snap-id': 'uint32',
'*tag': 'str' } }
+##
+# @BlockdevOptionsVitastor:
+#
+# Driver specific block device options for vitastor
+#
+# @image: Image name
+# @inode: Inode number
+# @pool: Pool ID
+# @size: Desired image size in bytes
+# @config-path: Path to Vitastor configuration
+# @etcd-host: etcd connection address(es)
+# @etcd-prefix: etcd key/value prefix
+##
+{ 'struct': 'BlockdevOptionsVitastor',
+ 'data': { '*inode': 'uint64',
+ '*pool': 'uint64',
+ '*size': 'uint64',
+ '*image': 'str',
+ '*config-path': 'str',
+ '*etcd-host': 'str',
+ '*etcd-prefix': 'str' } }
+
##
# @ReplicationMode:
#
@@ -3547,6 +3569,7 @@
'rbd': 'BlockdevOptionsRbd',
'replication':'BlockdevOptionsReplication',
'sheepdog': 'BlockdevOptionsSheepdog',
+ 'vitastor': 'BlockdevOptionsVitastor',
'ssh': 'BlockdevOptionsSsh',
'throttle': 'BlockdevOptionsThrottle',
'vdi': 'BlockdevOptionsGenericFormat',
@@ -3991,6 +4014,17 @@
'*subformat': 'BlockdevVhdxSubformat',
'*block-state-zero': 'bool' } }
+##
+# @BlockdevCreateOptionsVitastor:
+#
+# Driver specific image creation options for Vitastor.
+#
+# @size: Size of the virtual disk in bytes
+##
+{ 'struct': 'BlockdevCreateOptionsVitastor',
+ 'data': { 'location': 'BlockdevOptionsVitastor',
+ 'size': 'size' } }
+
##
# @BlockdevVpcSubformat:
#
@@ -4074,6 +4108,7 @@
'rbd': 'BlockdevCreateOptionsRbd',
'replication': 'BlockdevCreateNotSupported',
'sheepdog': 'BlockdevCreateOptionsSheepdog',
+ 'vitastor': 'BlockdevCreateOptionsVitastor',
'ssh': 'BlockdevCreateOptionsSsh',
'throttle': 'BlockdevCreateNotSupported',
'vdi': 'BlockdevCreateOptionsVdi',

View File

@@ -0,0 +1,181 @@
Index: qemu-5.2+dfsg/qapi/block-core.json
===================================================================
--- qemu-5.2+dfsg.orig/qapi/block-core.json
+++ qemu-5.2+dfsg/qapi/block-core.json
@@ -2831,7 +2831,7 @@
'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
{ 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
- 'sheepdog',
+ 'sheepdog', 'vitastor',
'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
##
@@ -3668,6 +3668,28 @@
'*tag': 'str' } }
##
+# @BlockdevOptionsVitastor:
+#
+# Driver specific block device options for vitastor
+#
+# @image: Image name
+# @inode: Inode number
+# @pool: Pool ID
+# @size: Desired image size in bytes
+# @config-path: Path to Vitastor configuration
+# @etcd-host: etcd connection address(es)
+# @etcd-prefix: etcd key/value prefix
+##
+{ 'struct': 'BlockdevOptionsVitastor',
+ 'data': { '*inode': 'uint64',
+ '*pool': 'uint64',
+ '*size': 'uint64',
+ '*image': 'str',
+ '*config-path': 'str',
+ '*etcd-host': 'str',
+ '*etcd-prefix': 'str' } }
+
+##
# @ReplicationMode:
#
# An enumeration of replication modes.
@@ -4015,6 +4037,7 @@
'replication': { 'type': 'BlockdevOptionsReplication',
'if': 'defined(CONFIG_REPLICATION)' },
'sheepdog': 'BlockdevOptionsSheepdog',
+ 'vitastor': 'BlockdevOptionsVitastor',
'ssh': 'BlockdevOptionsSsh',
'throttle': 'BlockdevOptionsThrottle',
'vdi': 'BlockdevOptionsGenericFormat',
@@ -4404,6 +4427,17 @@
'*cluster-size' : 'size' } }
##
+# @BlockdevCreateOptionsVitastor:
+#
+# Driver specific image creation options for Vitastor.
+#
+# @size: Size of the virtual disk in bytes
+##
+{ 'struct': 'BlockdevCreateOptionsVitastor',
+ 'data': { 'location': 'BlockdevOptionsVitastor',
+ 'size': 'size' } }
+
+##
# @BlockdevVmdkSubformat:
#
# Subformat options for VMDK images
@@ -4665,6 +4699,7 @@
'qed': 'BlockdevCreateOptionsQed',
'rbd': 'BlockdevCreateOptionsRbd',
'sheepdog': 'BlockdevCreateOptionsSheepdog',
+ 'vitastor': 'BlockdevCreateOptionsVitastor',
'ssh': 'BlockdevCreateOptionsSsh',
'vdi': 'BlockdevCreateOptionsVdi',
'vhdx': 'BlockdevCreateOptionsVhdx',
Index: qemu-5.2+dfsg/block/meson.build
===================================================================
--- qemu-5.2+dfsg.orig/block/meson.build
+++ qemu-5.2+dfsg/block/meson.build
@@ -76,6 +76,7 @@ foreach m : [
['CONFIG_LIBNFS', 'nfs', libnfs, 'nfs.c'],
['CONFIG_LIBSSH', 'ssh', libssh, 'ssh.c'],
['CONFIG_RBD', 'rbd', rbd, 'rbd.c'],
+ ['CONFIG_VITASTOR', 'vitastor', vitastor, 'vitastor.c'],
]
if config_host.has_key(m[0])
if enable_modules
Index: qemu-5.2+dfsg/configure
===================================================================
--- qemu-5.2+dfsg.orig/configure
+++ qemu-5.2+dfsg/configure
@@ -372,6 +372,7 @@ trace_backends="log"
trace_file="trace"
spice=""
rbd=""
+vitastor=""
smartcard=""
u2f="auto"
libusb=""
@@ -1263,6 +1264,10 @@ for opt do
;;
--enable-rbd) rbd="yes"
;;
+ --disable-vitastor) vitastor="no"
+ ;;
+ --enable-vitastor) vitastor="yes"
+ ;;
--disable-xfsctl) xfs="no"
;;
--enable-xfsctl) xfs="yes"
@@ -1827,6 +1832,7 @@ disabled with --disable-FEATURE, default
vhost-vdpa vhost-vdpa kernel backend support
spice spice
rbd rados block device (rbd)
+ vitastor vitastor block device
libiscsi iscsi support
libnfs nfs support
smartcard smartcard support (libcacard)
@@ -3719,6 +3725,27 @@ EOF
fi
##########################################
+# vitastor probe
+if test "$vitastor" != "no" ; then
+ cat > $TMPC <<EOF
+#include <vitastor_c.h>
+int main(void) {
+ vitastor_c_create_qemu(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+ return 0;
+}
+EOF
+ vitastor_libs="-lvitastor_client"
+ if compile_prog "" "$vitastor_libs" ; then
+ vitastor=yes
+ else
+ if test "$vitastor" = "yes" ; then
+ feature_not_found "vitastor block device" "Install vitastor-client-dev"
+ fi
+ vitastor=no
+ fi
+fi
+
+##########################################
# libssh probe
if test "$libssh" != "no" ; then
if $pkg_config --exists libssh; then
@@ -6456,6 +6483,10 @@ if test "$rbd" = "yes" ; then
echo "CONFIG_RBD=y" >> $config_host_mak
echo "RBD_LIBS=$rbd_libs" >> $config_host_mak
fi
+if test "$vitastor" = "yes" ; then
+ echo "CONFIG_VITASTOR=y" >> $config_host_mak
+ echo "VITASTOR_LIBS=$vitastor_libs" >> $config_host_mak
+fi
echo "CONFIG_COROUTINE_BACKEND=$coroutine" >> $config_host_mak
if test "$coroutine_pool" = "yes" ; then
Index: qemu-5.2+dfsg/meson.build
===================================================================
--- qemu-5.2+dfsg.orig/meson.build
+++ qemu-5.2+dfsg/meson.build
@@ -596,6 +596,10 @@ rbd = not_found
if 'CONFIG_RBD' in config_host
rbd = declare_dependency(link_args: config_host['RBD_LIBS'].split())
endif
+vitastor = not_found
+if 'CONFIG_VITASTOR' in config_host
+ vitastor = declare_dependency(link_args: config_host['VITASTOR_LIBS'].split())
+endif
glusterfs = not_found
if 'CONFIG_GLUSTERFS' in config_host
glusterfs = declare_dependency(compile_args: config_host['GLUSTERFS_CFLAGS'].split(),
@@ -2145,6 +2149,7 @@ endif
# TODO: add back protocol and server version
summary_info += {'spice support': config_host.has_key('CONFIG_SPICE')}
summary_info += {'rbd support': config_host.has_key('CONFIG_RBD')}
+summary_info += {'vitastor support': config_host.has_key('CONFIG_VITASTOR')}
summary_info += {'xfsctl support': config_host.has_key('CONFIG_XFS')}
summary_info += {'smartcard support': config_host.has_key('CONFIG_SMARTCARD')}
summary_info += {'U2F support': u2f.found()}

View File

@@ -24,4 +24,4 @@ rm fio
mv fio-copy fio
FIO=`rpm -qi fio | perl -e 'while(<>) { /^Epoch[\s:]+(\S+)/ && print "$1:"; /^Version[\s:]+(\S+)/ && print $1; /^Release[\s:]+(\S+)/ && print "-$1"; }'`
perl -i -pe 's/(Requires:\s*fio)([^\n]+)?/$1 = '$FIO'/' $VITASTOR/rpm/vitastor-el$EL.spec
tar --transform 's#^#vitastor-0.9.3/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.9.3$(rpm --eval '%dist').tar.gz *
tar --transform 's#^#vitastor-0.9.6/#' --exclude 'rpm/*.rpm' -czf $VITASTOR/../vitastor-0.9.6$(rpm --eval '%dist').tar.gz *

View File

@@ -22,7 +22,7 @@
Name: qemu-kvm
Version: 4.2.0
-Release: 29.vitastor%{?dist}.6
+Release: 32.vitastor%{?dist}.6
+Release: 34.vitastor%{?dist}.6
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
Epoch: 15
License: GPLv2 and GPLv2+ and CC-BY

View File

@@ -13,7 +13,7 @@
Name: qemu-kvm
Version: 4.2.0
-Release: 29%{?dist}.6
+Release: 32.vitastor%{?dist}.6
+Release: 33.vitastor%{?dist}.6
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
Epoch: 15
License: GPLv2 and GPLv2+ and CC-BY

View File

@@ -0,0 +1,103 @@
--- qemu-kvm-6.2.spec.orig 2023-07-18 13:52:57.636625440 +0000
+++ qemu-kvm-6.2.spec 2023-07-18 13:52:19.011683886 +0000
@@ -73,6 +73,7 @@ Requires: %{name}-hw-usbredir = %{epoch}
%endif \
Requires: %{name}-block-iscsi = %{epoch}:%{version}-%{release} \
Requires: %{name}-block-rbd = %{epoch}:%{version}-%{release} \
+Requires: %{name}-block-vitastor = %{epoch}:%{version}-%{release}\
Requires: %{name}-block-ssh = %{epoch}:%{version}-%{release}
# Macro to properly setup RHEL/RHEV conflict handling
@@ -83,7 +84,7 @@ Obsoletes: %1-rhev <= %{epoch}:%{version
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 6.2.0
-Release: 32%{?rcrel}%{?dist}
+Release: 32.vitastor%{?rcrel}%{?dist}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
Epoch: 15
License: GPLv2 and GPLv2+ and CC-BY
@@ -122,6 +123,7 @@ Source37: tests_data_acpi_pc_SSDT.dimmpx
Source38: tests_data_acpi_q35_FACP.slic
Source39: tests_data_acpi_q35_SSDT.dimmpxm
Source40: tests_data_acpi_virt_SSDT.memhp
+Source41: qemu-vitastor.c
Patch0001: 0001-redhat-Adding-slirp-to-the-exploded-tree.patch
Patch0005: 0005-Initial-redhat-build.patch
@@ -652,6 +654,7 @@ Patch255: kvm-scsi-protect-req-aiocb-wit
Patch256: kvm-dma-helpers-prevent-dma_blk_cb-vs-dma_aio_cancel-rac.patch
# For bz#2090990 - qemu crash with error scsi_req_unref(SCSIRequest *): Assertion `req->refcount > 0' failed or scsi_dma_complete(void *, int): Assertion `r->req.aiocb != NULL' failed [8.7.0]
Patch257: kvm-virtio-scsi-reset-SCSI-devices-from-main-loop-thread.patch
+Patch258: qemu-6.2-vitastor.patch
BuildRequires: wget
BuildRequires: rpm-build
@@ -689,6 +692,7 @@ BuildRequires: libcurl-devel
BuildRequires: libssh-devel
BuildRequires: librados-devel
BuildRequires: librbd-devel
+BuildRequires: vitastor-client-devel
%if %{have_gluster}
# For gluster block driver
BuildRequires: glusterfs-api-devel
@@ -926,6 +930,14 @@ Install this package if you want to acce
using the rbd protocol.
+%package block-vitastor
+Summary: QEMU Vitastor block driver
+Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
+
+%description block-vitastor
+This package provides the additional Vitastor block driver for QEMU.
+
+
%package block-ssh
Summary: QEMU SSH block driver
Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
@@ -979,6 +991,7 @@ This package provides usbredir support.
rm -fr slirp
mkdir slirp
%autopatch -p1
+cp %{SOURCE41} ./block/vitastor.c
%global qemu_kvm_build qemu_kvm_build
mkdir -p %{qemu_kvm_build}
@@ -994,7 +1007,7 @@ cp -f %{SOURCE40} tests/data/acpi/virt/S
# --build-id option is used for giving info to the debug packages.
buildldflags="VL_LDFLAGS=-Wl,--build-id"
-%global block_drivers_list qcow2,raw,file,host_device,nbd,iscsi,rbd,blkdebug,luks,null-co,nvme,copy-on-read,throttle
+%global block_drivers_list qcow2,raw,file,host_device,nbd,iscsi,rbd,vitastor,blkdebug,luks,null-co,nvme,copy-on-read,throttle
%if 0%{have_gluster}
%global block_drivers_list %{block_drivers_list},gluster
@@ -1149,9 +1162,7 @@ pushd %{qemu_kvm_build}
--firmwarepath=%{_prefix}/share/qemu-firmware \
--meson="git" \
--target-list="%{buildarch}" \
- --block-drv-rw-whitelist=%{block_drivers_list} \
--audio-drv-list= \
- --block-drv-ro-whitelist=vmdk,vhdx,vpc,https,ssh \
--with-coroutine=ucontext \
--with-git=git \
--tls-priority=@QEMU,SYSTEM \
@@ -1197,6 +1208,7 @@ pushd %{qemu_kvm_build}
%endif
--enable-pie \
--enable-rbd \
+ --enable-vitastor \
%if 0%{have_librdma}
--enable-rdma \
%endif
@@ -1794,6 +1806,9 @@ sh %{_sysconfdir}/sysconfig/modules/kvm.
%files block-rbd
%{_libdir}/qemu-kvm/block-rbd.so
+%files block-vitastor
+%{_libdir}/qemu-kvm/block-vitastor.so
+
%files block-ssh
%{_libdir}/qemu-kvm/block-ssh.so

View File

@@ -0,0 +1,93 @@
--- qemu-kvm-7.2.spec.orig 2023-06-22 13:56:19.000000000 +0000
+++ qemu-kvm-7.2.spec 2023-07-18 07:55:22.347090196 +0000
@@ -100,8 +100,6 @@
%endif
%global target_list %{kvm_target}-softmmu
-%global block_drivers_rw_list qcow2,raw,file,host_device,nbd,iscsi,rbd,blkdebug,luks,null-co,nvme,copy-on-read,throttle,compress
-%global block_drivers_ro_list vdi,vmdk,vhdx,vpc,https
%define qemudocdir %{_docdir}/%{name}
%global firmwaredirs "%{_datadir}/qemu-firmware:%{_datadir}/ipxe/qemu:%{_datadir}/seavgabios:%{_datadir}/seabios"
@@ -126,6 +124,7 @@ Requires: %{name}-device-usb-host = %{ep
Requires: %{name}-device-usb-redirect = %{epoch}:%{version}-%{release} \
%endif \
Requires: %{name}-block-rbd = %{epoch}:%{version}-%{release} \
+Requires: %{name}-block-vitastor = %{epoch}:%{version}-%{release}\
Requires: %{name}-audio-pa = %{epoch}:%{version}-%{release}
# Since SPICE is removed from RHEL-9, the following Obsoletes:
@@ -148,7 +147,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}
Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm
Version: 7.2.0
-Release: 14%{?rcrel}%{?dist}%{?cc_suffix}.1
+Release: 14.vitastor%{?rcrel}%{?dist}%{?cc_suffix}.1
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
# Epoch 15 used for RHEL 8
# Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@@ -171,6 +170,7 @@ Source28: 95-kvm-memlock.conf
Source30: kvm-s390x.conf
Source31: kvm-x86.conf
Source36: README.tests
+Source37: qemu-vitastor.c
Patch0004: 0004-Initial-redhat-build.patch
@@ -418,6 +418,7 @@ Patch134: kvm-target-i386-Fix-BZHI-instr
Patch135: kvm-intel-iommu-fail-DEVIOTLB_UNMAP-without-dt-mode.patch
# For bz#2203745 - Disk detach is unsuccessful while the guest is still booting [rhel-9.2.0.z]
Patch136: kvm-acpi-pcihp-allow-repeating-hot-unplug-requests.patch
+Patch137: qemu-7.2-vitastor.patch
%if %{have_clang}
BuildRequires: clang
@@ -449,6 +450,7 @@ BuildRequires: libcurl-devel
%if %{have_block_rbd}
BuildRequires: librbd-devel
%endif
+BuildRequires: vitastor-client-devel
# We need both because the 'stap' binary is probed for by configure
BuildRequires: systemtap
BuildRequires: systemtap-sdt-devel
@@ -642,6 +644,14 @@ using the rbd protocol.
%endif
+%package block-vitastor
+Summary: QEMU Vitastor block driver
+Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
+
+%description block-vitastor
+This package provides the additional Vitastor block driver for QEMU.
+
+
%package audio-pa
Summary: QEMU PulseAudio audio driver
Requires: %{name}-common%{?_isa} = %{epoch}:%{version}-%{release}
@@ -719,6 +729,7 @@ This package provides usbredir support.
%prep
%setup -q -n qemu-%{version}%{?rcstr}
%autopatch -p1
+cp %{SOURCE37} ./block/vitastor.c
%global qemu_kvm_build qemu_kvm_build
mkdir -p %{qemu_kvm_build}
@@ -946,6 +957,7 @@ run_configure \
%if %{have_block_rbd}
--enable-rbd \
%endif
+ --enable-vitastor \
%if %{have_librdma}
--enable-rdma \
%endif
@@ -1426,6 +1438,9 @@ useradd -r -u 107 -g qemu -G kvm -d / -s
%files block-rbd
%{_libdir}/%{name}/block-rbd.so
%endif
+%files block-vitastor
+%{_libdir}/%{name}/block-vitastor.so
+
%files audio-pa
%{_libdir}/%{name}/audio-pa.so

View File

@@ -35,7 +35,7 @@ ADD . /root/vitastor
RUN set -e; \
cd /root/vitastor/rpm; \
sh build-tarball.sh; \
cp /root/vitastor-0.9.3.el7.tar.gz ~/rpmbuild/SOURCES; \
cp /root/vitastor-0.9.6.el7.tar.gz ~/rpmbuild/SOURCES; \
cp vitastor-el7.spec ~/rpmbuild/SPECS/vitastor.spec; \
cd ~/rpmbuild/SPECS/; \
rpmbuild -ba vitastor.spec; \

View File

@@ -1,11 +1,11 @@
Name: vitastor
Version: 0.9.3
Version: 0.9.6
Release: 1%{?dist}
Summary: Vitastor, a fast software-defined clustered block storage
License: Vitastor Network Public License 1.1
URL: https://vitastor.io/
Source0: vitastor-0.9.3.el7.tar.gz
Source0: vitastor-0.9.6.el7.tar.gz
BuildRequires: liburing-devel >= 0.6
BuildRequires: gperftools-devel

View File

@@ -35,7 +35,7 @@ ADD . /root/vitastor
RUN set -e; \
cd /root/vitastor/rpm; \
sh build-tarball.sh; \
cp /root/vitastor-0.9.3.el8.tar.gz ~/rpmbuild/SOURCES; \
cp /root/vitastor-0.9.6.el8.tar.gz ~/rpmbuild/SOURCES; \
cp vitastor-el8.spec ~/rpmbuild/SPECS/vitastor.spec; \
cd ~/rpmbuild/SPECS/; \
rpmbuild -ba vitastor.spec; \

View File

@@ -1,11 +1,11 @@
Name: vitastor
Version: 0.9.3
Version: 0.9.6
Release: 1%{?dist}
Summary: Vitastor, a fast software-defined clustered block storage
License: Vitastor Network Public License 1.1
URL: https://vitastor.io/
Source0: vitastor-0.9.3.el8.tar.gz
Source0: vitastor-0.9.6.el8.tar.gz
BuildRequires: liburing-devel >= 0.6
BuildRequires: gperftools-devel

View File

@@ -18,7 +18,7 @@ ADD . /root/vitastor
RUN set -e; \
cd /root/vitastor/rpm; \
sh build-tarball.sh; \
cp /root/vitastor-0.9.3.el9.tar.gz ~/rpmbuild/SOURCES; \
cp /root/vitastor-0.9.6.el9.tar.gz ~/rpmbuild/SOURCES; \
cp vitastor-el9.spec ~/rpmbuild/SPECS/vitastor.spec; \
cd ~/rpmbuild/SPECS/; \
rpmbuild -ba vitastor.spec; \

View File

@@ -1,11 +1,11 @@
Name: vitastor
Version: 0.9.3
Version: 0.9.6
Release: 1%{?dist}
Summary: Vitastor, a fast software-defined clustered block storage
License: Vitastor Network Public License 1.1
URL: https://vitastor.io/
Source0: vitastor-0.9.3.el9.tar.gz
Source0: vitastor-0.9.6.el9.tar.gz
BuildRequires: liburing-devel >= 0.6
BuildRequires: gperftools-devel

View File

@@ -16,7 +16,7 @@ if("${CMAKE_INSTALL_PREFIX}" MATCHES "^/usr/local/?$")
set(CMAKE_INSTALL_RPATH "${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_LIBDIR}")
endif()
add_definitions(-DVERSION="0.9.3")
add_definitions(-DVERSION="0.9.6")
add_definitions(-Wall -Wno-sign-compare -Wno-comment -Wno-parentheses -Wno-pointer-arith -fdiagnostics-color=always -I ${CMAKE_SOURCE_DIR}/src)
if (${WITH_ASAN})
add_definitions(-fsanitize=address -fno-omit-frame-pointer)

View File

@@ -85,11 +85,13 @@ void blockstore_impl_t::parse_config(blockstore_config_t & config, bool init)
immediate_commit = IMMEDIATE_SMALL;
}
metadata_buf_size = strtoull(config["meta_buf_size"].c_str(), NULL, 10);
inmemory_meta = config["inmemory_metadata"] != "false";
inmemory_meta = config["inmemory_metadata"] != "false" && config["inmemory_metadata"] != "0" &&
config["inmemory_metadata"] != "no";
journal.sector_count = strtoull(config["journal_sector_buffer_count"].c_str(), NULL, 10);
journal.no_same_sector_overwrites = config["journal_no_same_sector_overwrites"] == "true" ||
config["journal_no_same_sector_overwrites"] == "1" || config["journal_no_same_sector_overwrites"] == "yes";
journal.inmemory = config["inmemory_journal"] != "false";
journal.inmemory = config["inmemory_journal"] != "false" && config["inmemory_journal"] != "0" &&
config["inmemory_journal"] != "no";
// Validate
if (journal.sector_count < 2)
{

View File

@@ -56,14 +56,15 @@ struct image_lister_t
{
continue;
}
auto & pool_cfg = parent->cli->st_cli.pool_config.at(INODE_POOL(ic.second.num));
auto pool_it = parent->cli->st_cli.pool_config.find(INODE_POOL(ic.second.num));
bool good_pool = pool_it != parent->cli->st_cli.pool_config.end();
auto item = json11::Json::object {
{ "name", ic.second.name },
{ "size", ic.second.size },
{ "used_size", 0 },
{ "readonly", ic.second.readonly },
{ "pool_id", (uint64_t)INODE_POOL(ic.second.num) },
{ "pool_name", pool_cfg.name },
{ "pool_name", good_pool ? pool_it->second.name : "? (ID:"+std::to_string(INODE_POOL(ic.second.num))+")" },
{ "inode_num", INODE_NO_POOL(ic.second.num) },
{ "inode_id", ic.second.num },
};
@@ -247,6 +248,8 @@ resume_1:
if (state == 1)
goto resume_1;
get_list();
if (state == 100)
return;
if (show_stats)
{
resume_1:
@@ -269,7 +272,7 @@ resume_1:
{ "key", "name" },
{ "title", "NAME" },
});
if (!list_pool_id)
if (list_pool_name == "")
{
cols.push_back(json11::Json::object{
{ "key", "pool_name" },
@@ -376,16 +379,18 @@ resume_1:
std::string print_table(json11::Json items, json11::Json header, bool use_esc)
{
int header_sizes[header.array_items().size()];
std::vector<int> sizes;
for (int i = 0; i < header.array_items().size(); i++)
{
sizes.push_back(header[i]["title"].string_value().length());
header_sizes[i] = utf8_length(header[i]["title"].string_value());
sizes.push_back(header_sizes[i]);
}
for (auto & item: items.array_items())
{
for (int i = 0; i < header.array_items().size(); i++)
{
int l = item[header[i]["key"].string_value()].as_string().length();
int l = utf8_length(item[header[i]["key"].string_value()].as_string());
sizes[i] = sizes[i] < l ? l : sizes[i];
}
}
@@ -397,7 +402,7 @@ std::string print_table(json11::Json items, json11::Json header, bool use_esc)
// Separator
str += " ";
}
int pad = sizes[i]-header[i]["title"].string_value().length();
int pad = sizes[i]-header_sizes[i];
if (header[i]["right"].bool_value())
{
// Align right
@@ -425,7 +430,7 @@ std::string print_table(json11::Json items, json11::Json header, bool use_esc)
// Separator
str += " ";
}
int pad = sizes[i] - item[header[i]["key"].string_value()].as_string().length();
int pad = sizes[i] - utf8_length(item[header[i]["key"].string_value()].as_string());
if (header[i]["right"].bool_value())
{
// Align right

View File

@@ -13,7 +13,7 @@ struct image_changer_t
std::string image_name;
std::string new_name;
uint64_t new_size = 0;
bool force_size = false;
bool force_size = false, inc_size = false;
bool set_readonly = false, set_readwrite = false, force = false;
// interval between fsyncs
int fsync_interval = 128;
@@ -81,14 +81,14 @@ struct image_changer_t
}
if ((!set_readwrite || !cfg.readonly) &&
(!set_readonly || cfg.readonly) &&
(!new_size && !force_size || cfg.size == new_size) &&
(!new_size && !force_size || cfg.size == new_size || cfg.size >= new_size && inc_size) &&
(new_name == "" || new_name == image_name))
{
result = (cli_result_t){ .text = "No change" };
state = 100;
return;
}
if (new_size != 0 || force_size)
if ((new_size != 0 || force_size) && (cfg.size < new_size || !inc_size))
{
if (cfg.size >= new_size)
{
@@ -233,6 +233,7 @@ std::function<bool(cli_result_t &)> cli_tool_t::start_modify(json11::Json cfg)
changer->new_name = cfg["rename"].string_value();
changer->new_size = parse_size(cfg["resize"].as_string());
changer->force_size = cfg["force_size"].bool_value();
changer->inc_size = cfg["inc_size"].bool_value();
changer->force = cfg["force"].bool_value();
changer->set_readonly = cfg["readonly"].bool_value();
changer->set_readwrite = cfg["readwrite"].bool_value();

View File

@@ -147,7 +147,7 @@ int disk_tool_t::prepare_one(std::map<std::string, std::string> options, int is_
for (int i = 0; i < sizeof(allow_additional_params)/sizeof(allow_additional_params[0]); i++)
{
auto it = options.find(allow_additional_params[i]);
if (it != options.end())
if (it != options.end() && it->second != "")
{
sb[it->first] = it->second;
}
@@ -621,7 +621,7 @@ int disk_tool_t::prepare(std::vector<std::string> devices)
}
}
// Treat all disks as SSDs if not in the hybrid mode
prepare_one(options, hybrid && dev.is_hdd ? 1 : 0);
prepare_one(options, dev.is_hdd ? 1 : 0);
if (hybrid)
{
options.erase("journal_device");

View File

@@ -264,6 +264,7 @@ int write_zero(int fd, uint64_t offset, uint64_t size)
{
uint64_t buf_len = 1024*1024;
void *zero_buf = memalign_or_die(MEM_ALIGNMENT, buf_len);
memset(zero_buf, 0, buf_len);
ssize_t r;
while (size > 0)
{

View File

@@ -190,7 +190,15 @@ static int nfs3_setattr_proc(void *opaque, rpc_op_t *rop)
{
if (handle == "roothandle" || self->parent->dir_by_hash.find(handle) != self->parent->dir_by_hash.end())
{
*reply = (SETATTR3res){ .status = NFS3ERR_ISDIR };
if (args->new_attributes.size.set_it)
{
*reply = (SETATTR3res){ .status = NFS3ERR_ISDIR };
}
else
{
// Silently ignore mode, uid, gid, atime, mtime changes
*reply = (SETATTR3res){ .status = NFS3_OK };
}
}
else
{
@@ -358,7 +366,6 @@ static int nfs3_read_proc(void *opaque, rpc_op_t *rop)
}
static void nfs_resize_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint64_t new_size, uint64_t offset, uint64_t count, void *buf);
static void nfs_do_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint64_t offset, uint64_t count, void *buf);
static int nfs3_write_proc(void *opaque, rpc_op_t *rop)
{
@@ -392,7 +399,6 @@ static int nfs3_write_proc(void *opaque, rpc_op_t *rop)
.resok = (WRITE3resok){
//.file_wcc = ...,
.count = (unsigned)count,
.committed = args->stable,
},
};
if ((args->offset % alignment) != 0 || (count % alignment) != 0)
@@ -436,42 +442,101 @@ static int nfs3_write_proc(void *opaque, rpc_op_t *rop)
return 1;
}
static void nfs_resize_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint64_t new_size, uint64_t offset, uint64_t count, void *buf)
static void complete_extend_write(nfs_client_t *self, rpc_op_t *rop, inode_t inode, int res)
{
// Check if we have to resize the inode before writing
WRITE3args *args = (WRITE3args*)rop->request;
WRITE3res *reply = (WRITE3res*)rop->reply;
if (res < 0)
{
*reply = (WRITE3res){ .status = vitastor_nfs_map_err(res) };
rpc_queue_reply(rop);
return;
}
bool imm = self->parent->cli->get_immediate_commit(inode);
reply->resok.committed = args->stable != UNSTABLE || imm ? FILE_SYNC : UNSTABLE;
*(uint64_t*)reply->resok.verf = self->parent->server_id;
if (args->stable != UNSTABLE && !imm)
{
// Client requested a stable write. Add an fsync
auto op = new cluster_op_t;
op->opcode = OSD_OP_SYNC;
op->callback = [rop](cluster_op_t *op)
{
if (op->retval != 0)
{
WRITE3res *reply = (WRITE3res*)rop->reply;
*reply = (WRITE3res){ .status = vitastor_nfs_map_err(-op->retval) };
}
delete op;
rpc_queue_reply(rop);
};
self->parent->cli->execute(op);
}
else
{
rpc_queue_reply(rop);
}
}
static void complete_extend_inode(nfs_client_t *self, uint64_t inode, uint64_t new_size, int err)
{
auto ext_it = self->extend_writes.lower_bound((extend_size_t){ .inode = inode, .new_size = 0 });
while (ext_it != self->extend_writes.end() &&
ext_it->first.inode == inode &&
ext_it->first.new_size <= new_size)
{
ext_it->second.resize_res = err;
if (ext_it->second.write_res <= 0)
{
complete_extend_write(self, ext_it->second.rop, inode, ext_it->second.write_res < 0
? ext_it->second.write_res : ext_it->second.resize_res);
self->extend_writes.erase(ext_it++);
}
else
ext_it++;
}
}
static void extend_inode(nfs_client_t *self, uint64_t inode, uint64_t new_size)
{
// Send an extend request
auto & ext = self->extends[inode];
ext.cur_extend = new_size;
auto inode_it = self->parent->cli->st_cli.inode_config.find(inode);
if (inode_it != self->parent->cli->st_cli.inode_config.end() &&
inode_it->second.size < new_size)
{
self->parent->cmd->loop_and_wait(self->parent->cmd->start_modify(json11::Json::object {
// FIXME: Resizing by ID is probably more correct
{ "image", inode_it->second.name },
{ "resize", new_size },
{ "inc_size", true },
{ "force_size", true },
}), [=](const cli_result_t & r)
{
auto & ext = self->extends[inode];
if (r.err)
{
if (r.err == EAGAIN)
{
// Multiple concurrent resize requests received, try to repeat
nfs_resize_write(self, rop, inode, new_size, offset, count, buf);
return;
}
WRITE3res *reply = (WRITE3res*)rop->reply;
*reply = (WRITE3res){ .status = vitastor_nfs_map_err(r.err) };
rpc_queue_reply(rop);
fprintf(stderr, "Error extending inode %lu to %lu bytes: %s\n", inode, new_size, r.text.c_str());
}
if (r.err == EAGAIN || ext.next_extend > ext.cur_extend)
{
// Multiple concurrent resize requests received, try to repeat
extend_inode(self, inode, ext.next_extend > ext.cur_extend ? ext.next_extend : ext.cur_extend);
return;
}
nfs_do_write(self, rop, inode, offset, count, buf);
ext.cur_extend = ext.next_extend = 0;
complete_extend_inode(self, inode, new_size, r.err);
});
}
else
{
nfs_do_write(self, rop, inode, offset, count, buf);
complete_extend_inode(self, inode, new_size, 0);
}
}
static void nfs_do_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint64_t offset, uint64_t count, void *buf)
static void nfs_do_write(nfs_client_t *self, std::multimap<extend_size_t, extend_write_t>::iterator ewr_it,
rpc_op_t *rop, uint64_t inode, uint64_t offset, uint64_t count, void *buf)
{
cluster_op_t *op = new cluster_op_t;
op->opcode = OSD_OP_WRITE;
@@ -479,48 +544,61 @@ static void nfs_do_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint
op->offset = offset;
op->len = count;
op->iov.push_back(buf, count);
op->callback = [self, rop](cluster_op_t *op)
op->callback = [self, ewr_it, rop](cluster_op_t *op)
{
uint64_t inode = op->inode;
WRITE3args *args = (WRITE3args*)rop->request;
WRITE3res *reply = (WRITE3res*)rop->reply;
if (op->retval != op->len)
auto inode = op->inode;
int write_res = op->retval < 0 ? op->retval : (op->retval != op->len ? -ERANGE : 0);
if (ewr_it == self->extend_writes.end())
{
*reply = (WRITE3res){ .status = vitastor_nfs_map_err(-op->retval) };
delete op;
rpc_queue_reply(rop);
complete_extend_write(self, rop, inode, write_res);
}
else
{
*(uint64_t*)reply->resok.verf = self->parent->server_id;
delete op;
if (args->stable != UNSTABLE &&
!self->parent->cli->get_immediate_commit(inode))
ewr_it->second.write_res = write_res;
if (ewr_it->second.resize_res <= 0)
{
// Client requested a stable write. Add an fsync
op = new cluster_op_t;
op->opcode = OSD_OP_SYNC;
op->callback = [rop](cluster_op_t *op)
{
if (op->retval != 0)
{
WRITE3res *reply = (WRITE3res*)rop->reply;
*reply = (WRITE3res){ .status = vitastor_nfs_map_err(-op->retval) };
}
delete op;
rpc_queue_reply(rop);
};
self->parent->cli->execute(op);
}
else
{
rpc_queue_reply(rop);
complete_extend_write(self, rop, inode, write_res < 0 ? write_res : ewr_it->second.resize_res);
self->extend_writes.erase(ewr_it);
}
}
};
self->parent->cli->execute(op);
}
static void nfs_resize_write(nfs_client_t *self, rpc_op_t *rop, uint64_t inode, uint64_t new_size, uint64_t offset, uint64_t count, void *buf)
{
// Check if we have to resize the inode during write
auto inode_it = self->parent->cli->st_cli.inode_config.find(inode);
if (inode_it != self->parent->cli->st_cli.inode_config.end() &&
inode_it->second.size < new_size)
{
auto ewr_it = self->extend_writes.emplace((extend_size_t){
.inode = inode,
.new_size = new_size,
}, (extend_write_t){
.rop = rop,
.resize_res = 1,
.write_res = 1,
});
auto & ext = self->extends[inode];
if (ext.cur_extend > 0)
{
// Already resizing, just wait
if (ext.next_extend < new_size)
ext.next_extend = new_size;
}
else
{
extend_inode(self, inode, new_size);
}
nfs_do_write(self, ewr_it, rop, inode, offset, count, buf);
}
else
{
nfs_do_write(self, self->extend_writes.end(), rop, inode, offset, count, buf);
}
}
static int nfs3_create_proc(void *opaque, rpc_op_t *rop)
{
nfs_client_t *self = (nfs_client_t*)opaque;
@@ -881,6 +959,27 @@ static int nfs3_link_proc(void *opaque, rpc_op_t *rop)
return 0;
}
static void fill_dir_entry(nfs_client_t *self, rpc_op_t *rop,
std::map<std::string, nfs_dir_t>::iterator dir_id_it, struct entryplus3 *entry, bool is_plus)
{
if (dir_id_it == self->parent->dir_info.end())
{
return;
}
entry->fileid = dir_id_it->second.id;
if (is_plus)
{
entry->name_attributes = (post_op_attr){
.attributes_follow = 1,
.attributes = get_dir_attributes(self, dir_id_it->first),
};
entry->name_handle = (post_op_fh3){
.handle_follows = 1,
.handle = xdr_copy_string(rop->xdrs, "S"+base64_encode(sha256(dir_id_it->first))),
};
}
}
static void nfs3_readdir_common(void *opaque, rpc_op_t *rop, bool is_plus)
{
nfs_client_t *self = (nfs_client_t*)opaque;
@@ -958,17 +1057,17 @@ static void nfs3_readdir_common(void *opaque, rpc_op_t *rop, bool is_plus)
continue;
std::string subname = dir_id_it->first.substr(prefix.size());
// for directories, fileid changes when the user restarts proxy
entries[subname].fileid = dir_id_it->second.id;
if (is_plus)
fill_dir_entry(self, rop, dir_id_it, &entries[subname], is_plus);
}
// Add . and ..
{
auto dir_id_it = self->parent->dir_info.find(dir);
fill_dir_entry(self, rop, dir_id_it, &entries["."], is_plus);
auto sl = dir.rfind("/");
if (sl != std::string::npos)
{
entries[subname].name_attributes = (post_op_attr){
.attributes_follow = 1,
.attributes = get_dir_attributes(self, dir_id_it->first),
};
entries[subname].name_handle = (post_op_fh3){
.handle_follows = 1,
.handle = xdr_copy_string(rop->xdrs, "S"+base64_encode(sha256(dir_id_it->first))),
};
auto dir_id_it = self->parent->dir_info.find(dir.substr(0, sl));
fill_dir_entry(self, rop, dir_id_it, &entries[".."], is_plus);
}
}
// Offset results by the continuation cookie (equal to index in the listing)
@@ -1193,10 +1292,11 @@ static int nfs3_commit_proc(void *opaque, rpc_op_t *rop)
cluster_op_t *op = new cluster_op_t;
// fsync. we don't know how to fsync a single inode, so just fsync everything
op->opcode = OSD_OP_SYNC;
op->callback = [rop](cluster_op_t *op)
op->callback = [self, rop](cluster_op_t *op)
{
COMMIT3res *reply = (COMMIT3res*)rop->reply;
*reply = (COMMIT3res){ .status = vitastor_nfs_map_err(op->retval) };
*(uint64_t*)reply->resok.verf = self->parent->server_id;
rpc_queue_reply(rop);
};
self->parent->cli->execute(op);

View File

@@ -86,6 +86,28 @@ struct rpc_free_buffer_t
unsigned size;
};
struct extend_size_t
{
inode_t inode;
uint64_t new_size;
};
inline bool operator < (const extend_size_t &a, const extend_size_t &b)
{
return a.inode < b.inode || a.inode == b.inode && a.new_size < b.new_size;
}
struct extend_write_t
{
rpc_op_t *rop;
int resize_res, write_res; // 1 = started, 0 = completed OK, -errno = completed with error
};
struct extend_inode_t
{
uint64_t cur_extend = 0, next_extend = 0;
};
class nfs_client_t
{
public:
@@ -100,6 +122,8 @@ public:
rpc_cur_buffer_t cur_buffer = { 0 };
std::map<uint8_t*, rpc_used_buffer_t> used_buffers;
std::vector<rpc_free_buffer_t> free_buffers;
std::map<inode_t, extend_inode_t> extends;
std::multimap<extend_size_t, extend_write_t> extend_writes;
iovec read_iov;
msghdr read_msg = { 0 };

View File

@@ -234,15 +234,13 @@ out:
return;
}
#if defined VITASTOR_C_API_VERSION && VITASTOR_C_API_VERSION >= 2
static void vitastor_uring_handler(void *opaque)
{
VitastorClient *client = (VitastorClient*)opaque;
qemu_mutex_lock(&client->mutex);
client->bh_uring_scheduled = 0;
do
{
vitastor_c_uring_handle_events(client->proxy);
} while (vitastor_c_uring_has_work(client->proxy));
vitastor_c_uring_handle_events(client->proxy);
qemu_mutex_unlock(&client->mutex);
}
@@ -266,20 +264,23 @@ static void vitastor_schedule_uring_handler(VitastorClient *client)
replay_bh_schedule_oneshot_event(client->ctx, vitastor_uring_handler, opaque);
#elif QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 8
aio_bh_schedule_oneshot(client->ctx, vitastor_uring_handler, opaque);
#elif QEMU_VERSION_MAJOR >= 2
#else
VitastorBH *vbh = (VitastorBH*)malloc(sizeof(VitastorBH));
vbh->cli = client;
#if QEMU_VERSION_MAJOR >= 2
vbh->bh = aio_bh_new(bdrv_get_aio_context(task->bs), vitastor_bh_uring_handler, vbh);
qemu_bh_schedule(vbh->bh);
#else
client->bh_uring_scheduled = 0;
do
{
vitastor_c_uring_handle_events(client->proxy);
} while (vitastor_c_uring_has_work(client->proxy));
vbh->bh = qemu_bh_new(vitastor_bh_uring_handler, vbh);
#endif
qemu_bh_schedule(vbh->bh);
#endif
}
}
#else
static void vitastor_schedule_uring_handler(VitastorClient *client)
{
}
#endif
static void coroutine_fn vitastor_co_get_metadata(VitastorRPC *task)
{
@@ -406,20 +407,16 @@ static int vitastor_file_open(BlockDriverState *bs, QDict *options, int flags, E
vitastor_aio_set_fd_handler, client, client->config_path, client->etcd_host, client->etcd_prefix,
client->use_rdma, client->rdma_device, client->rdma_port_num, client->rdma_gid_index, client->rdma_mtu, 0
);
#else
client->proxy = vitastor_c_create_uring(
client->config_path, client->etcd_host, client->etcd_prefix,
client->use_rdma, client->rdma_device, client->rdma_port_num, client->rdma_gid_index, client->rdma_mtu, 0
);
#endif
if (!client->proxy)
{
fprintf(stderr, "vitastor: failed to create io_uring: %s - I/O will be slower\n", strerror(errno));
client->uring_eventfd = -1;
#endif
client->proxy = vitastor_c_create_qemu(
vitastor_aio_set_fd_handler, client, client->config_path, client->etcd_host, client->etcd_prefix,
client->use_rdma, client->rdma_device, client->rdma_port_num, client->rdma_gid_index, client->rdma_mtu, 0
);
#if defined VITASTOR_C_API_VERSION && VITASTOR_C_API_VERSION >= 2
}
else
{
@@ -433,6 +430,7 @@ static int vitastor_file_open(BlockDriverState *bs, QDict *options, int flags, E
}
universal_aio_set_fd_handler(client->ctx, client->uring_eventfd, vitastor_uring_handler, NULL, client);
}
#endif
image = client->image = g_strdup(qdict_get_try_str(options, "image"));
client->readonly = (flags & BDRV_O_RDWR) ? 1 : 0;
// Get image metadata (size and readonly flag) or just wait until the client is ready
@@ -664,7 +662,8 @@ static void vitastor_co_generic_cb(void *opaque, long retval)
task->bh = aio_bh_new(bdrv_get_aio_context(task->bs), vitastor_co_generic_bh_cb, opaque);
qemu_bh_schedule(task->bh);
#else
vitastor_co_generic_bh_cb(opaque);
task->bh = qemu_bh_new(vitastor_co_generic_bh_cb, opaque);
qemu_bh_schedule(task->bh);
#endif
}
@@ -741,7 +740,6 @@ static void vitastor_co_read_bitmap_cb(void *opaque, long retval, uint8_t *bitma
VitastorRPC *task = opaque;
VitastorClient *client = task->bs->opaque;
task->ret = retval;
task->complete = 1;
if (retval >= 0)
{
task->bitmap = bitmap;
@@ -753,15 +751,17 @@ static void vitastor_co_read_bitmap_cb(void *opaque, long retval, uint8_t *bitma
client->last_bitmap = bitmap;
}
}
if (qemu_coroutine_self() != task->co)
{
#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 8
aio_co_wake(task->co);
#if QEMU_VERSION_MAJOR > 4 || QEMU_VERSION_MAJOR == 4 && QEMU_VERSION_MINOR >= 2
replay_bh_schedule_oneshot_event(bdrv_get_aio_context(task->bs), vitastor_co_generic_bh_cb, opaque);
#elif QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 8
aio_bh_schedule_oneshot(bdrv_get_aio_context(task->bs), vitastor_co_generic_bh_cb, opaque);
#elif QEMU_VERSION_MAJOR >= 2
task->bh = aio_bh_new(bdrv_get_aio_context(task->bs), vitastor_co_generic_bh_cb, opaque);
qemu_bh_schedule(task->bh);
#else
qemu_coroutine_enter(task->co, NULL);
qemu_aio_release(task);
task->bh = qemu_bh_new(vitastor_co_generic_bh_cb, opaque);
qemu_bh_schedule(task->bh);
#endif
}
}
static int coroutine_fn vitastor_co_block_status(

View File

@@ -66,6 +66,16 @@ void ring_loop_t::unregister_consumer(ring_consumer_t *consumer)
void ring_loop_t::loop()
{
if (ring_eventfd >= 0)
{
// Reset eventfd counter
uint64_t ctr = 0;
int r = read(ring_eventfd, &ctr, 8);
if (r < 0 && errno != EAGAIN && errno != EINTR)
{
fprintf(stderr, "Error resetting eventfd: %s\n", strerror(errno));
}
}
struct io_uring_cqe *cqe;
while (!io_uring_peek_cqe(&ring, &cqe))
{
@@ -84,7 +94,7 @@ void ring_loop_t::loop()
}
else
{
printf("Warning: empty callback in SQE\n");
fprintf(stderr, "Warning: empty callback in SQE\n");
free_ring_data[free_ring_data_ptr++] = d - ring_datas;
}
io_uring_cqe_seen(&ring, cqe);

View File

@@ -308,3 +308,19 @@ std::string str_repeat(const std::string & str, int times)
r += str;
return r;
}
size_t utf8_length(const std::string & s)
{
size_t len = 0;
for (size_t i = 0; i < s.size(); i++)
len += (s[i] & 0xC0) != 0x80;
return len;
}
size_t utf8_length(const char *s)
{
size_t len = 0;
for (; *s; s++)
len += (*s & 0xC0) != 0x80;
return len;
}

View File

@@ -18,3 +18,5 @@ void print_help(const char *help_text, std::string exe_name, std::string cmd, bo
uint64_t parse_time(std::string time_str, bool *ok = NULL);
std::string read_all_fd(int fd);
std::string str_repeat(const std::string & str, int times);
size_t utf8_length(const std::string & s);
size_t utf8_length(const char *s);

View File

@@ -6,7 +6,7 @@ includedir=${prefix}/@CMAKE_INSTALL_INCLUDEDIR@
Name: Vitastor
Description: Vitastor client library
Version: 0.9.3
Version: 0.9.6
Libs: -L${libdir} -lvitastor_client
Cflags: -I${includedir}

View File

@@ -215,7 +215,7 @@ void vitastor_c_uring_wait_events(vitastor_c *client)
client->ringloop->wait();
}
bool vitastor_c_uring_has_work(vitastor_c *client)
int vitastor_c_uring_has_work(vitastor_c *client)
{
return client->ringloop->has_work();
}

View File

@@ -46,7 +46,7 @@ int vitastor_c_uring_register_eventfd(vitastor_c *client);
void vitastor_c_uring_wait_ready(vitastor_c *client);
void vitastor_c_uring_handle_events(vitastor_c *client);
void vitastor_c_uring_wait_events(vitastor_c *client);
bool vitastor_c_uring_has_work(vitastor_c *client);
int vitastor_c_uring_has_work(vitastor_c *client);
void vitastor_c_read(vitastor_c *client, uint64_t inode, uint64_t offset, uint64_t len,
struct iovec *iov, int iovcnt, VitastorReadHandler cb, void *opaque);
void vitastor_c_write(vitastor_c *client, uint64_t inode, uint64_t offset, uint64_t len, uint64_t check_version,