diff options
| author | Fam Zheng <famz@redhat.com> | 2015-06-08 13:56:10 +0800 |
|---|---|---|
| committer | Stefan Hajnoczi <stefanha@redhat.com> | 2015-07-02 10:06:23 +0100 |
| commit | 508249952c0ea7472c62e17bf8132295dab4912d (patch) | |
| tree | 0097014b93cf0e14fe7208b1c1d2123f3201f07c /block/io.c | |
| parent | dcfb3beb5130694b76b57de109619fcbf9c7e5b5 (diff) | |
block: Fix dirty bitmap in bdrv_co_discard
Unsetting dirty globally with discard is not very correct. The discard may zero
out sectors (depending on can_write_zeroes_with_unmap), we should replicate
this change to destination side to make sure that the guest sees the same data.
Calling bdrv_reset_dirty also troubles mirror job because the hbitmap iterator
doesn't expect unsetting of bits after current position.
So let's do it the opposite way which fixes both problems: set the dirty bits
if we are to discard it.
Reported-by: wangxiaolong@ucloud.cn
Signed-off-by: Fam Zheng <famz@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Diffstat (limited to 'block/io.c')
| -rw-r--r-- | block/io.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/block/io.c b/block/io.c index ccf79c33c5..ad31822b08 100644 --- a/block/io.c +++ b/block/io.c @@ -2412,8 +2412,6 @@ int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num, return -EPERM; } - bdrv_reset_dirty(bs, sector_num, nb_sectors); - /* Do nothing if disabled. */ if (!(bs->open_flags & BDRV_O_UNMAP)) { return 0; @@ -2423,6 +2421,8 @@ int coroutine_fn bdrv_co_discard(BlockDriverState *bs, int64_t sector_num, return 0; } + bdrv_set_dirty(bs, sector_num, nb_sectors); + max_discard = MIN_NON_ZERO(bs->bl.max_discard, BDRV_REQUEST_MAX_SECTORS); while (nb_sectors > 0) { int ret; |