On Tue, Jan 26, 2021 at 11:25:10AM +0900, Naohiro Aota wrote:
> In ZONED, btrfs uses per-FS zoned_meta_io_lock to serialize the metadata
> write IOs.
>
> Even with these serialization, write bios sent from btree_write_cache_pages
> can be reordered by async checksum workers as these workers are per CPU and
> not per zone.
>
> To preserve write BIO ordering, we can disable async metadata checksum on
> ZONED. This does not result in lower performance with HDDs as a single CPU
> core is fast enough to do checksum for a single zone write stream with the
> maximum possible bandwidth of the device. If multiple zones are being
> written simultaneously, HDD seek overhead lowers the achievable maximum
> bandwidth, resulting again in a per zone checksum serialization not
> affecting performance.
>
> Reviewed-by: Josef Bacik <josef@xxxxxxxxxxxxxx>
> Signed-off-by: Naohiro Aota <naohiro.aota@xxxxxxx>
> ---
> fs/btrfs/disk-io.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
> index a41bdf9312d6..5d14100ecf72 100644
> --- a/fs/btrfs/disk-io.c
> +++ b/fs/btrfs/disk-io.c
> @@ -814,6 +814,8 @@ static blk_status_t btree_submit_bio_start(struct inode *inode, struct bio *bio,
> static int check_async_write(struct btrfs_fs_info *fs_info,
> struct btrfs_inode *bi)
> {
> + if (btrfs_is_zoned(fs_info))
> + return 0;
This check need to be after the other ones as zoned is a static per-fs
status, while other others depend on either current state or system
state (crypto implementation).
> if (atomic_read(&bi->sync_writers))
> return 0;
> if (test_bit(BTRFS_FS_CSUM_IMPL_FAST, &fs_info->flags))
> --
> 2.27.0
