On Tue, Aug 16, 2022 at 9:19 AM Jeff Layton <jlayton@xxxxxxxxxx> wrote:
>
> The i_version in xfs_trans_log_inode is bumped for any inode update,
> including atime-only updates due to reads. We don't want to record those
> in the i_version, as they don't represent "real" changes. Remove that
> callsite.
>
> In xfs_vn_update_time, if S_VERSION is flagged, then attempt to bump the
> i_version and turn on XFS_ILOG_CORE if it happens. In
> xfs_trans_ichgtime, update the i_version if the mtime or ctime are being
> updated.
>
> Cc: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
> Cc: Dave Chinner <david@xxxxxxxxxxxxx>
> Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>
> ---
> fs/xfs/libxfs/xfs_trans_inode.c | 17 +++--------------
> fs/xfs/xfs_iops.c | 4 ++++
> 2 files changed, 7 insertions(+), 14 deletions(-)
>
> diff --git a/fs/xfs/libxfs/xfs_trans_inode.c b/fs/xfs/libxfs/xfs_trans_inode.c
> index 8b5547073379..78bf7f491462 100644
> --- a/fs/xfs/libxfs/xfs_trans_inode.c
> +++ b/fs/xfs/libxfs/xfs_trans_inode.c
> @@ -71,6 +71,8 @@ xfs_trans_ichgtime(
> inode->i_ctime = tv;
> if (flags & XFS_ICHGTIME_CREATE)
> ip->i_crtime = tv;
> + if (flags & (XFS_ICHGTIME_MOD|XFS_ICHGTIME_CHG))
> + inode_inc_iversion(inode);
> }
>
> /*
> @@ -116,20 +118,7 @@ xfs_trans_log_inode(
> spin_unlock(&inode->i_lock);
> }
>
> - /*
> - * First time we log the inode in a transaction, bump the inode change
> - * counter if it is configured for this to occur. While we have the
> - * inode locked exclusively for metadata modification, we can usually
> - * avoid setting XFS_ILOG_CORE if no one has queried the value since
> - * the last time it was incremented. If we have XFS_ILOG_CORE already
> - * set however, then go ahead and bump the i_version counter
> - * unconditionally.
> - */
> - if (!test_and_set_bit(XFS_LI_DIRTY, &iip->ili_item.li_flags)) {
> - if (IS_I_VERSION(inode) &&
> - inode_maybe_inc_iversion(inode, flags & XFS_ILOG_CORE))
> - iversion_flags = XFS_ILOG_CORE;
> - }
> + set_bit(XFS_LI_DIRTY, &iip->ili_item.li_flags);
>
> /*
> * If we're updating the inode core or the timestamps and it's possible
> diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
> index 45518b8c613c..162e044c7f56 100644
> --- a/fs/xfs/xfs_iops.c
> +++ b/fs/xfs/xfs_iops.c
> @@ -718,6 +718,7 @@ xfs_setattr_nonsize(
> }
>
> setattr_copy(mnt_userns, inode, iattr);
> + inode_inc_iversion(inode);
> xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
>
> XFS_STATS_INC(mp, xs_ig_attrchg);
> @@ -943,6 +944,7 @@ xfs_setattr_size(
>
> ASSERT(!(iattr->ia_valid & (ATTR_UID | ATTR_GID)));
> setattr_copy(mnt_userns, inode, iattr);
> + inode_inc_iversion(inode);
> xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
>
> XFS_STATS_INC(mp, xs_ig_attrchg);
> @@ -1047,6 +1049,8 @@ xfs_vn_update_time(
> inode->i_mtime = *now;
> if (flags & S_ATIME)
> inode->i_atime = *now;
> + if ((flags & S_VERSION) && inode_maybe_inc_iversion(inode, false))
> + log_flags |= XFS_ILOG_CORE;
>
> xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
> xfs_trans_log_inode(tp, ip, log_flags);
> --
> 2.37.2
>
I have a test (details below) that shows an open issue with NFSv4.x +
fscache where an xfs exported filesystem would trigger unnecessary
over the wire READs after a umount/mount cycle of the NFS mount. I
previously tracked this down to atime updates, but never followed
through on any patch. Now that Jeff worked it out and this patch is
under review, I built 5.19 vanilla, retested, then built 5.19 + this
patch and verified the problem is fixed.
You can add:
Tested-by: Dave Wysochanski <dwysocha@xxxxxxxxxx>
# ./t0_bz1913591.sh 4.1 xfs relatime
Setting NFS vers=4.1 filesystem to xfs and mount options relatime,rw
0. On NFS server, setup export with xfs filesystem on loop device
/dev/loop0 /export/dir1 xfs
rw,seclabel,relatime,attr2,inode64,logbufs=8,logbsize=32k,noquota 0 0
1. On NFS client, install and enable cachefilesd
2. On NFS client, mount -o vers=4.1,fsc 127.0.0.1:/export/dir1 /mnt
3. On NFS client, dd if=/dev/zero of=/mnt/file1.bin bs=4096 count=1
4. On NFS client, echo 3 > /proc/sys/vm/drop_caches
5. On NFS client, dd if=/mnt/file1.bin of=/dev/null (read into fscache)
6. On NFS client, umount /mnt
7. On NFS client, mount -o vers=4.1,fsc 127.0.0.1:/export/dir1 /mnt
8. On NFS client, repeat steps 4-5 (read from fscache)
9. On NFS client, check for READ ops (1st number) > 0 in /proc/self/mountstats
Found 4200 NFS READ and READ_PLUS ops in /proc/self/mountstats > 0
READ: 1 1 0 220 4200 0 0 1 0
READ_PLUS: 0 0 0 0 0 0 0 0 0
FAILED TEST ./t0_bz1913591.sh on kernel 5.19.0 with NFS vers=4.1
exported filesystem xfs options relatime,rw
# ./t0_bz1913591.sh 4.1 xfs relatime
Setting NFS vers=4.1 filesystem to xfs and mount options relatime,rw
0. On NFS server, setup export with xfs filesystem on loop device
/dev/loop0 /export/dir1 xfs
rw,seclabel,relatime,attr2,inode64,logbufs=8,logbsize=32k,noquota 0 0
1. On NFS client, install and enable cachefilesd
2. On NFS client, mount -o vers=4.1,fsc 127.0.0.1:/export/dir1 /mnt
3. On NFS client, dd if=/dev/zero of=/mnt/file1.bin bs=4096 count=1
4. On NFS client, echo 3 > /proc/sys/vm/drop_caches
5. On NFS client, dd if=/mnt/file1.bin of=/dev/null (read into fscache)
6. On NFS client, umount /mnt
7. On NFS client, mount -o vers=4.1,fsc 127.0.0.1:/export/dir1 /mnt
8. On NFS client, repeat steps 4-5 (read from fscache)
9. On NFS client, check for READ ops (1st number) > 0 in /proc/self/mountstats
10. On NFS client, check /proc/fs/fscache/stats fscache reads incrementing
PASSED TEST ./t0_bz1913591.sh on kernel 5.19.0i_version+ with NFS
vers=4.1 exported filesystem xfs options relatime,rw
