From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id o6F0a5X2173213 for ; Wed, 14 Jul 2010 19:36:06 -0500 Received: from mail.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 460BF119503E for ; Wed, 14 Jul 2010 17:45:12 -0700 (PDT) Received: from mail.internode.on.net (bld-mail16.adl2.internode.on.net [150.101.137.101]) by cuda.sgi.com with ESMTP id uIYSlLETbgw9FZeJ for ; Wed, 14 Jul 2010 17:45:12 -0700 (PDT) Received: from dastard (unverified [121.44.38.238]) by mail.internode.on.net (SurgeMail 3.8f2) with ESMTP id 31478532-1927428 for ; Thu, 15 Jul 2010 10:08:58 +0930 (CST) Received: from disturbed ([192.168.1.9]) by dastard with esmtp (Exim 4.71) (envelope-from ) id 1OZCTF-0002zB-Nt for xfs@oss.sgi.com; Thu, 15 Jul 2010 10:38:37 +1000 Received: from dave by disturbed with local (Exim 4.71) (envelope-from ) id 1OZCT2-0000Xq-7d for xfs@oss.sgi.com; Thu, 15 Jul 2010 10:38:24 +1000 From: Dave Chinner Subject: [PATCH 1/5] xfs: track AGs with reclaimable inodes in per-ag radix tree Date: Thu, 15 Jul 2010 10:38:16 +1000 Message-Id: <1279154300-2018-2-git-send-email-david@fromorbit.com> In-Reply-To: <1279154300-2018-1-git-send-email-david@fromorbit.com> References: <1279154300-2018-1-git-send-email-david@fromorbit.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com From: Dave Chinner https://bugzilla.kernel.org/show_bug.cgi?id=16348 When the filesystem grows to a large number of allocation groups, the summing of recalimable inodes gets expensive. In many cases, most AGs won't have any reclaimable inodes and so we are wasting CPU time aggregating over these AGs. This is particularly important for the inode shrinker that gets called frequently under memory pressure. To avoid the overhead, track AGs with reclaimable inodes in the per-ag radix tree so that we can find all the AGs with reclaimable inodes via a simple gang tag lookup. This involves setting the tag when the first reclaimable inode is tracked in the AG, and removing the tag when the last reclaimable inode is removed from the tree. Then the summation process becomes a loop walking the radix tree summing AGs with the reclaim tag set. This significantly reduces the overhead of scanning - a 6400 AG filesystea now only uses about 25% of a cpu in kswapd while slab reclaim progresses instead of being permanently stuck at 100% CPU and making little progress. Clean filesystems filesystems will see no overhead and the overhead only increases linearly with the number of dirty AGs. Signed-off-by: Dave Chinner --- fs/xfs/linux-2.6/xfs_sync.c | 71 +++++++++++++++++++++++++++++++++++++---- fs/xfs/linux-2.6/xfs_trace.h | 3 ++ 2 files changed, 67 insertions(+), 7 deletions(-) diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 56fed91..51da595 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -133,6 +133,41 @@ restart: return last_error; } +/* + * Select the next per-ag structure to iterate during the walk. The reclaim + * walk is optimised only to walk AGs with reclaimable inodes in them. + */ +static struct xfs_perag * +xfs_inode_ag_iter_next_pag( + struct xfs_mount *mp, + xfs_agnumber_t *first, + int tag) +{ + struct xfs_perag *pag = NULL; + + if (tag == XFS_ICI_RECLAIM_TAG) { + int found; + int ref; + + spin_lock(&mp->m_perag_lock); + found = radix_tree_gang_lookup_tag(&mp->m_perag_tree, + (void **)&pag, *first, 1, tag); + if (found <= 0) { + spin_unlock(&mp->m_perag_lock); + return NULL; + } + *first = pag->pag_agno + 1; + /* open coded pag reference increment */ + ref = atomic_inc_return(&pag->pag_ref); + spin_unlock(&mp->m_perag_lock); + trace_xfs_perag_get_reclaim(mp, pag->pag_agno, ref, _RET_IP_); + } else { + pag = xfs_perag_get(mp, *first); + (*first)++; + } + return pag; +} + int xfs_inode_ag_iterator( struct xfs_mount *mp, @@ -143,16 +178,15 @@ xfs_inode_ag_iterator( int exclusive, int *nr_to_scan) { + struct xfs_perag *pag; int error = 0; int last_error = 0; xfs_agnumber_t ag; int nr; nr = nr_to_scan ? *nr_to_scan : INT_MAX; - for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) { - struct xfs_perag *pag; - - pag = xfs_perag_get(mp, ag); + ag = 0; + while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag, tag))) { error = xfs_inode_ag_walk(mp, pag, execute, flags, tag, exclusive, &nr); xfs_perag_put(pag); @@ -628,6 +662,17 @@ __xfs_inode_set_reclaim_tag( radix_tree_tag_set(&pag->pag_ici_root, XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), XFS_ICI_RECLAIM_TAG); + + if (!pag->pag_ici_reclaimable) { + /* propagate the reclaim tag up into the perag radix tree */ + spin_lock(&ip->i_mount->m_perag_lock); + radix_tree_tag_set(&ip->i_mount->m_perag_tree, + XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); + spin_unlock(&ip->i_mount->m_perag_lock); + trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno, + -1, _RET_IP_); + } pag->pag_ici_reclaimable++; } @@ -662,6 +707,16 @@ __xfs_inode_clear_reclaim_tag( radix_tree_tag_clear(&pag->pag_ici_root, XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); pag->pag_ici_reclaimable--; + if (!pag->pag_ici_reclaimable) { + /* clear the reclaim tag from the perag radix tree */ + spin_lock(&ip->i_mount->m_perag_lock); + radix_tree_tag_clear(&ip->i_mount->m_perag_tree, + XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); + spin_unlock(&ip->i_mount->m_perag_lock); + trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno, + -1, _RET_IP_); + } } /* @@ -832,7 +887,7 @@ xfs_reclaim_inode_shrink( struct xfs_mount *mp; struct xfs_perag *pag; xfs_agnumber_t ag; - int reclaimable = 0; + int reclaimable; if (nr_to_scan) { if (!(gfp_mask & __GFP_FS)) @@ -849,9 +904,11 @@ xfs_reclaim_inode_shrink( } down_read(&xfs_mount_list_lock); + reclaimable = 0; list_for_each_entry(mp, &xfs_mount_list, m_mplist) { - for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) { - pag = xfs_perag_get(mp, ag); + ag = 0; + while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag, + XFS_ICI_RECLAIM_TAG))) { reclaimable += pag->pag_ici_reclaimable; xfs_perag_put(pag); } diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h index d506753..24e5580 100644 --- a/fs/xfs/linux-2.6/xfs_trace.h +++ b/fs/xfs/linux-2.6/xfs_trace.h @@ -124,7 +124,10 @@ DEFINE_EVENT(xfs_perag_class, name, \ unsigned long caller_ip), \ TP_ARGS(mp, agno, refcount, caller_ip)) DEFINE_PERAG_REF_EVENT(xfs_perag_get); +DEFINE_PERAG_REF_EVENT(xfs_perag_get_reclaim); DEFINE_PERAG_REF_EVENT(xfs_perag_put); +DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim); +DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim); TRACE_EVENT(xfs_attr_list_node_descend, TP_PROTO(struct xfs_attr_list_context *ctx, -- 1.7.1 _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs