Re: [PATCH 12/17] xfs: compute maximum AG btree height for critical reservation calculation
From: Dave Chinner <david@fromorbit.com>
Date: 2021-10-14 22:54:51
On Thu, Oct 14, 2021 at 01:18:00PM -0700, Darrick J. Wong wrote:
From: Darrick J. Wong <djwong@kernel.org> Compute the actual maximum AG btree height for deciding if a per-AG block reservation is critically low. This only affects the sanity check condition, since we /generally/ will trigger on the 10% threshold. This is a long-winded way of saying that we're removing one more usage of XFS_BTREE_MAXLEVELS. Signed-off-by: Darrick J. Wong <djwong@kernel.org> --- fs/xfs/libxfs/xfs_ag_resv.c | 3 ++- fs/xfs/xfs_mount.c | 14 ++++++++++++++ fs/xfs/xfs_mount.h | 1 + 3 files changed, 17 insertions(+), 1 deletion(-)
One minor nit below, otherwise it looks good. Reviewed-by: Dave Chinner <redacted>
quoted hunk ↗ jump to hunk
diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c index 2aa2b3484c28..fe94058d4e9e 100644 --- a/fs/xfs/libxfs/xfs_ag_resv.c +++ b/fs/xfs/libxfs/xfs_ag_resv.c@@ -91,7 +91,8 @@ xfs_ag_resv_critical( trace_xfs_ag_resv_critical(pag, type, avail); /* Critically low if less than 10% or max btree height remains. */ - return XFS_TEST_ERROR(avail < orig / 10 || avail < XFS_BTREE_MAXLEVELS, + return XFS_TEST_ERROR(avail < orig / 10 || + avail < pag->pag_mount->m_agbtree_maxlevels, pag->pag_mount, XFS_ERRTAG_AG_RESV_CRITICAL); }diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 06dac09eddbd..5be1dd63fac5 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c@@ -567,6 +567,18 @@ xfs_mount_setup_inode_geom( xfs_ialloc_setup_geometry(mp); } +/* Compute maximum possible height for per-AG btree types for this fs. */ +static inline void +xfs_agbtree_compute_maxlevels( + struct xfs_mount *mp) +{ + unsigned int ret; + + ret = max(mp->m_alloc_maxlevels, M_IGEO(mp)->inobt_maxlevels); + ret = max(ret, mp->m_rmap_maxlevels); + mp->m_agbtree_maxlevels = max(ret, mp->m_refc_maxlevels); +}
"ret" should really be named "levels" here because it's not a return value anymore... Cheers, Dave. -- Dave Chinner david@fromorbit.com