On Mon, Apr 22, 2019 at 08:45:40AM -0700, Darrick J. Wong wrote:
> From: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
>
> When we're flushing an inode log item, it is not necessary to mess with
> the inode cluster buffer's log item because the iflush code paths pass
> the inode log item directly. The unconditional reset causes us to leak
> buffer log items.
>
> Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
looks fine.
Reviewed-by: Bill O'Donnell <billodo@xxxxxxxxxx>
> ---
> libxfs/trans.c | 2 --
> libxfs/util.c | 1 -
> 2 files changed, 3 deletions(-)
>
>
> diff --git a/libxfs/trans.c b/libxfs/trans.c
> index 629501f8..d562cdc0 100644
> --- a/libxfs/trans.c
> +++ b/libxfs/trans.c
> @@ -826,10 +826,8 @@ inode_item_done(
> * of whether the flush succeed or not. If we fail the flush, make sure
> * we still release the buffer reference we currently hold.
> */
> - bp->b_log_item = iip;
> error = libxfs_iflush_int(ip, bp);
> ip->i_transp = NULL; /* disassociate from transaction */
> - bp->b_log_item = NULL; /* remove log item */
> bp->b_transp = NULL; /* remove xact ptr */
>
> if (error) {
> diff --git a/libxfs/util.c b/libxfs/util.c
> index ea75fa20..9fe9a367 100644
> --- a/libxfs/util.c
> +++ b/libxfs/util.c
> @@ -394,7 +394,6 @@ libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
> xfs_dinode_t *dip;
> xfs_mount_t *mp;
>
> - ASSERT(bp-b_log_item != NULL);
> ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
> ip->i_d.di_nextents > ip->i_df.if_ext_max);
> ASSERT(ip->i_d.di_version > 1);
>
