On Fri, Feb 15, 2019 at 03:47:22PM +0100, Christoph Hellwig wrote:
> This way we can actually count how many bytes got converted and how many
> calls we need, unlike in the caller which doesn't have the detailed
> view.
>
> Note that this includes a slight change in behavior as the
> xs_xstrat_quick is now bumped for every allocation instead of just the
> one covering the requested writeback offset, which makes a lot more
> sense.
/me wonders if it's really supposed to be the case that we never touch
xs_xstrat_split and it stays zero, but as it hasn't done anything since
the start of linux.git I'll just set that aside for now and say...
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> Reviewed-by: Brian Foster <bfoster@xxxxxxxxxx>
...this by itself seems reasonable to me.
Reviewed-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
--D
> ---
> fs/xfs/libxfs/xfs_bmap.c | 3 +++
> fs/xfs/xfs_iomap.c | 4 ----
> 2 files changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
> index fc4f1d3145c4..4cf83475f0d0 100644
> --- a/fs/xfs/libxfs/xfs_bmap.c
> +++ b/fs/xfs/libxfs/xfs_bmap.c
> @@ -4516,6 +4516,9 @@ xfs_bmapi_convert_delalloc(
> if (WARN_ON_ONCE(!bma.got.br_startblock && !XFS_IS_REALTIME_INODE(ip)))
> goto out_finish;
>
> + XFS_STATS_ADD(mp, xs_xstrat_bytes, XFS_FSB_TO_B(mp, bma.length));
> + XFS_STATS_INC(mp, xs_xstrat_quick);
> +
> ASSERT(!isnullstartblock(bma.got.br_startblock));
> *imap = bma.got;
> *seq = READ_ONCE(ifp->if_seq);
> diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
> index 39be741cac5a..15da53b5fb53 100644
> --- a/fs/xfs/xfs_iomap.c
> +++ b/fs/xfs/xfs_iomap.c
> @@ -707,9 +707,6 @@ xfs_iomap_write_allocate(
> map_start_fsb = imap->br_startoff;
> map_count_fsb = imap->br_blockcount;
>
> - XFS_STATS_ADD(mp, xs_xstrat_bytes,
> - XFS_FSB_TO_B(mp, imap->br_blockcount));
> -
> while (true) {
> /*
> * Allocate in a loop because it may take several attempts to
> @@ -741,7 +738,6 @@ xfs_iomap_write_allocate(
> if ((offset_fsb >= imap->br_startoff) &&
> (offset_fsb < (imap->br_startoff +
> imap->br_blockcount))) {
> - XFS_STATS_INC(mp, xs_xstrat_quick);
> xfs_trim_extent(imap, map_start_fsb, map_count_fsb);
> ASSERT(offset_fsb >= imap->br_startoff &&
> offset_fsb < imap->br_startoff + imap->br_blockcount);
> --
> 2.20.1
>
