Skip to content

data integrity fixes V2

Christoph Hellwig qemu-devel 1 version New

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest5 patches 0 repliesNew

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/5block: add enable_write_cache flagblock.c block.h block_int.h +2 more+27 -30 replies
2/5block: use fdatasync instead of fsync if possibleblock/cow.c block/raw-posix.c cutils.c +1 more+19 -20 replies
3/5block: add aio_flush operationblock.c block.h block/raw-posix-aio.h +3 more+80 -30 replies
4/5ide: use bdrv_aio_flushhw/ide/core.c+13 -30 replies
5/5virtio-blk: add volatile writecache featurehw/virtio-blk.c hw/virtio-blk.h+27 -10 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [Qemu-devel] [PATCH 0/5] data integrity fixes V2

respin of the data integrity series. Major changes since the first version:

- split ide use of bdrv_aio_flush into a separate patch
- added a qemu_fdatasync helper that falls back to fsync if needed
- enable barriers for data=writeback

The last might be a bit controversial, but I can't find a workload where it really causes major regressions. And then you can still mount your guest filesystems with barrier=0/nobarrier (like the ext3 default) and this code won't ever be called.