From: Dave Chinner <dchinner@xxxxxxxxxx> xfs_ireclaim has to get and put te pag structure because it is only called with the inode to reclaim. The one caller of this function already has a reference on the pag and a pointer to is, so move the radix tree delete to the caller and remove xfs_ireclaim completely. This avoids a xfs_perag_get/put on every inode being reclaimed. The overhead was noticed in a bug report at: https://bugzilla.kernel.org/show_bug.cgi?id=16348 Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx> --- fs/xfs/linux-2.6/xfs_sync.c | 31 ++++++++++++++++++++++++- fs/xfs/xfs_iget.c | 53 +------------------------------------------ fs/xfs/xfs_inode.h | 2 +- 3 files changed, 32 insertions(+), 54 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 51da595..80ae18a 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -855,7 +855,36 @@ out: reclaim: xfs_ifunlock(ip); xfs_iunlock(ip, XFS_ILOCK_EXCL); - xfs_ireclaim(ip); + + XFS_STATS_INC(xs_ig_reclaims); + /* + * Remove the inode from the per-AG radix tree. + * + * Because radix_tree_delete won't complain even if the item was never + * added to the tree assert that it's been there before to catch + * problems with the inode life time early on. + */ + write_lock(&pag->pag_ici_lock); + if (!radix_tree_delete(&pag->pag_ici_root, + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino))) + ASSERT(0); + write_unlock(&pag->pag_ici_lock); + + /* + * Here we do an (almost) spurious inode lock in order to coordinate + * with inode cache radix tree lookups. This is because the lookup + * can reference the inodes in the cache without taking references. + * + * We make that OK here by ensuring that we wait until the inode is + * unlocked after the lookup before we go ahead and free it. We get + * both the ilock and the iolock because the code may need to drop the + * ilock one but will still hold the iolock. + */ + xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + xfs_qm_dqdetach(ip); + xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + + xfs_inode_free(ip); return error; } diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 9e86f21..eba5ae6 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -91,7 +91,7 @@ xfs_inode_alloc( return ip; } -STATIC void +void xfs_inode_free( struct xfs_inode *ip) { @@ -418,57 +418,6 @@ out_error_or_again: } /* - * This is called free all the memory associated with an inode. - * It must free the inode itself and any buffers allocated for - * if_extents/if_data and if_broot. It must also free the lock - * associated with the inode. - * - * Note: because we don't initialise everything on reallocation out - * of the zone, we must ensure we nullify everything correctly before - * freeing the structure. - */ -void -xfs_ireclaim( - struct xfs_inode *ip) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_perag *pag; - xfs_agino_t agino = XFS_INO_TO_AGINO(mp, ip->i_ino); - - XFS_STATS_INC(xs_ig_reclaims); - - /* - * Remove the inode from the per-AG radix tree. - * - * Because radix_tree_delete won't complain even if the item was never - * added to the tree assert that it's been there before to catch - * problems with the inode life time early on. - */ - pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); - write_lock(&pag->pag_ici_lock); - if (!radix_tree_delete(&pag->pag_ici_root, agino)) - ASSERT(0); - write_unlock(&pag->pag_ici_lock); - xfs_perag_put(pag); - - /* - * Here we do an (almost) spurious inode lock in order to coordinate - * with inode cache radix tree lookups. This is because the lookup - * can reference the inodes in the cache without taking references. - * - * We make that OK here by ensuring that we wait until the inode is - * unlocked after the lookup before we go ahead and free it. We get - * both the ilock and the iolock because the code may need to drop the - * ilock one but will still hold the iolock. - */ - xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - xfs_qm_dqdetach(ip); - xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - - xfs_inode_free(ip); -} - -/* * This is a wrapper routine around the xfs_ilock() routine * used to centralize some grungy code. It is used in places * that wish to lock the inode solely for reading the extents. diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index eb41559..0898c54 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -450,7 +450,7 @@ void xfs_ilock_demote(xfs_inode_t *, uint); int xfs_isilocked(xfs_inode_t *, uint); uint xfs_ilock_map_shared(xfs_inode_t *); void xfs_iunlock_map_shared(xfs_inode_t *, uint); -void xfs_ireclaim(xfs_inode_t *); +void xfs_inode_free(struct xfs_inode *ip); /* * xfs_inode.c prototypes. -- 1.7.1 _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
