From efceab1d563153a2b1a6e7d35376241a48126989 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Aug 2010 11:44:56 +1000 Subject: xfs: handle negative wbc->nr_to_write during sync writeback During data integrity (WB_SYNC_ALL) writeback, wbc->nr_to_write will go negative on inodes with more than 1024 dirty pages due to implementation details of write_cache_pages(). Currently XFS will abort page clustering in writeback once nr_to_write drops below zero, and so for data integrity writeback we will do very inefficient page at a time allocation and IO submission for inodes with large numbers of dirty pages. Fix this by only aborting the page clustering code when wbc->nr_to_write is negative and the sync mode is WB_SYNC_NONE. Cc: Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/linux-2.6/xfs_aops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 15412fe15c3a..528be1ba1402 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -852,8 +852,8 @@ xfs_convert_page( SetPageUptodate(page); if (count) { - wbc->nr_to_write--; - if (wbc->nr_to_write <= 0) + if (--wbc->nr_to_write <= 0 && + wbc->sync_mode == WB_SYNC_NONE) done = 1; } xfs_start_page_writeback(page, !page_dirty, count); -- cgit v1.2.3 From 2fe33661fcd79d4c53022509f7223d526b5fa233 Mon Sep 17 00:00:00 2001 From: Stuart Brodsky Date: Tue, 24 Aug 2010 11:46:05 +1000 Subject: xfs: ensure f_ffree returned by statfs() is non-negative Because of delayed updates to sb_icount field in the super block, it is possible to allocate over maxicount number of inodes. This causes the arithmetic to calculate a negative number of free inodes in user commands like df or stat -f. Since maxicount is a somewhat arbitrary number, a slight over allocation is not critical but user commands should be displayed as 0 or greater and never go negative. To do this the value in the stats buffer f_ffree is capped to never go negative. [ Modified to use max_t as per Christoph's comment. ] Signed-off-by: Stu Brodsky Signed-off-by: Dave Chinner --- fs/xfs/linux-2.6/xfs_super.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 15c35b62ff14..c6b24e7c308a 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1226,6 +1226,7 @@ xfs_fs_statfs( struct xfs_inode *ip = XFS_I(dentry->d_inode); __uint64_t fakeinos, id; xfs_extlen_t lsize; + __int64_t ffree; statp->f_type = XFS_SB_MAGIC; statp->f_namelen = MAXNAMELEN - 1; @@ -1249,7 +1250,11 @@ xfs_fs_statfs( statp->f_files = min_t(typeof(statp->f_files), statp->f_files, mp->m_maxicount); - statp->f_ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); + + /* make sure statp->f_ffree does not underflow */ + ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); + statp->f_ffree = max_t(__int64_t, ffree, 0); + spin_unlock(&mp->m_sb_lock); if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) || -- cgit v1.2.3 From 1a387d3be2b30c90f20d49a3497a8fc0693a9d18 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 24 Aug 2010 11:46:31 +1000 Subject: xfs: dummy transactions should not dirty VFS state When we need to cover the log, we issue dummy transactions to ensure the current log tail is on disk. Unfortunately we currently use the root inode in the dummy transaction, and the act of committing the transaction dirties the inode at the VFS level. As a result, the VFS writeback of the dirty inode will prevent the filesystem from idling long enough for the log covering state machine to complete. The state machine gets stuck in a loop issuing new dummy transactions to cover the log and never makes progress. To avoid this problem, the dummy transactions should not cause externally visible state changes. To ensure this occurs, make sure that dummy transactions log an unchanging field in the superblock as it's state is never propagated outside the filesystem. This allows the log covering state machine to complete successfully and the filesystem now correctly enters a fully idle state about 90s after the last modification was made. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/linux-2.6/xfs_super.c | 2 +- fs/xfs/linux-2.6/xfs_sync.c | 42 ++++++------------------------------------ fs/xfs/xfs_fsops.c | 31 ++++++++++++++++++------------- fs/xfs/xfs_fsops.h | 2 +- 4 files changed, 26 insertions(+), 51 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index c6b24e7c308a..a4e07974955b 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1407,7 +1407,7 @@ xfs_fs_freeze( xfs_save_resvblks(mp); xfs_quiesce_attr(mp); - return -xfs_fs_log_dummy(mp); + return -xfs_fs_log_dummy(mp, SYNC_WAIT); } STATIC int diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index dfcbd98d1599..d59c4a65d492 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -34,6 +34,7 @@ #include "xfs_inode_item.h" #include "xfs_quota.h" #include "xfs_trace.h" +#include "xfs_fsops.h" #include #include @@ -340,38 +341,6 @@ xfs_sync_attr( XFS_ICI_NO_TAG, 0, NULL); } -STATIC int -xfs_commit_dummy_trans( - struct xfs_mount *mp, - uint flags) -{ - struct xfs_inode *ip = mp->m_rootip; - struct xfs_trans *tp; - int error; - - /* - * Put a dummy transaction in the log to tell recovery - * that all others are OK. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_DUMMY1); - error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); - if (error) { - xfs_trans_cancel(tp, 0); - return error; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - - xfs_trans_ijoin(tp, ip); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - error = xfs_trans_commit(tp, 0); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - /* the log force ensures this transaction is pushed to disk */ - xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0); - return error; -} - STATIC int xfs_sync_fsdata( struct xfs_mount *mp) @@ -432,7 +401,7 @@ xfs_quiesce_data( /* mark the log as covered if needed */ if (xfs_log_need_covered(mp)) - error2 = xfs_commit_dummy_trans(mp, SYNC_WAIT); + error2 = xfs_fs_log_dummy(mp, SYNC_WAIT); /* flush data-only devices */ if (mp->m_rtdev_targp) @@ -563,7 +532,7 @@ xfs_flush_inodes( /* * Every sync period we need to unpin all items, reclaim inodes and sync * disk quotas. We might need to cover the log to indicate that the - * filesystem is idle. + * filesystem is idle and not frozen. */ STATIC void xfs_sync_worker( @@ -577,8 +546,9 @@ xfs_sync_worker( xfs_reclaim_inodes(mp, 0); /* dgc: errors ignored here */ error = xfs_qm_sync(mp, SYNC_TRYLOCK); - if (xfs_log_need_covered(mp)) - error = xfs_commit_dummy_trans(mp, 0); + if (mp->m_super->s_frozen == SB_UNFROZEN && + xfs_log_need_covered(mp)) + error = xfs_fs_log_dummy(mp, 0); } mp->m_sync_seq++; wake_up(&mp->m_wait_single_sync_task); diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index dbca5f5c37ba..43b1d5699335 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -604,31 +604,36 @@ out: return 0; } +/* + * Dump a transaction into the log that contains no real change. This is needed + * to be able to make the log dirty or stamp the current tail LSN into the log + * during the covering operation. + * + * We cannot use an inode here for this - that will push dirty state back up + * into the VFS and then periodic inode flushing will prevent log covering from + * making progress. Hence we log a field in the superblock instead. + */ int xfs_fs_log_dummy( - xfs_mount_t *mp) + xfs_mount_t *mp, + int flags) { xfs_trans_t *tp; - xfs_inode_t *ip; int error; tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP); - error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0, + XFS_DEFAULT_LOG_COUNT); if (error) { xfs_trans_cancel(tp, 0); return error; } - ip = mp->m_rootip; - xfs_ilock(ip, XFS_ILOCK_EXCL); - - xfs_trans_ijoin(tp, ip); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - xfs_trans_set_sync(tp); - error = xfs_trans_commit(tp, 0); - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return error; + /* log the UUID because it is an unchanging field */ + xfs_mod_sb(tp, XFS_SB_UUID); + if (flags & SYNC_WAIT) + xfs_trans_set_sync(tp); + return xfs_trans_commit(tp, 0); } int diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h index 88435e0a77c9..a786c5212c1e 100644 --- a/fs/xfs/xfs_fsops.h +++ b/fs/xfs/xfs_fsops.h @@ -25,6 +25,6 @@ extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt); extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval, xfs_fsop_resblks_t *outval); extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags); -extern int xfs_fs_log_dummy(xfs_mount_t *mp); +extern int xfs_fs_log_dummy(xfs_mount_t *mp, int flags); #endif /* __XFS_FSOPS_H__ */ -- cgit v1.2.3 From b5420f235953448eeae615b3361584dc5e414f34 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 24 Aug 2010 11:47:51 +1000 Subject: xfs: do not discard page cache data on EAGAIN If xfs_map_blocks returns EAGAIN because of lock contention we must redirty the page and not disard the pagecache content and return an error from writepage. We used to do this correctly, but the logic got lost during the recent reshuffle of the writepage code. Signed-off-by: Christoph Hellwig Reported-by: Mike Gao Tested-by: Mike Gao Reviewed-by: Dave Chinner Signed-off-by: Dave Chinner --- fs/xfs/linux-2.6/xfs_aops.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 528be1ba1402..b552f816de15 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1068,7 +1068,7 @@ xfs_vm_writepage( * by themselves. */ if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC) - goto out_fail; + goto redirty; /* * We need a transaction if there are delalloc or unwritten buffers @@ -1080,7 +1080,7 @@ xfs_vm_writepage( */ xfs_count_page_state(page, &delalloc, &unwritten); if ((current->flags & PF_FSTRANS) && (delalloc || unwritten)) - goto out_fail; + goto redirty; /* Is this page beyond the end of the file? */ offset = i_size_read(inode); @@ -1245,12 +1245,15 @@ error: if (iohead) xfs_cancel_ioend(iohead); + if (err == -EAGAIN) + goto redirty; + xfs_aops_discard_page(page); ClearPageUptodate(page); unlock_page(page); return err; -out_fail: +redirty: redirty_page_for_writepage(wbc, page); unlock_page(page); return 0; -- cgit v1.2.3 From 9bc08a45fb117c696e4940cfa1208cb1cc7a2f25 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Thu, 2 Sep 2010 15:14:38 +1000 Subject: xfs: improve buffer cache hash scalability When doing large parallel file creates on a 16p machines, large amounts of time is being spent in _xfs_buf_find(). A system wide profile with perf top shows this: 1134740.00 19.3% _xfs_buf_find 733142.00 12.5% __ticket_spin_lock The problem is that the hash contains 45,000 buffers, and the hash table width is only 256 buffers. That means we've got around 200 buffers per chain, and searching it is quite expensive. The hash table size needs to increase. Secondly, every time we do a lookup, we promote the buffer we find to the head of the hash chain. This is causing cachelines to be dirtied and causes invalidation of cachelines across all CPUs that may have walked the hash chain recently. hence every walk of the hash chain is effectively a cold cache walk. Remove the promotion to avoid this invalidation. The results are: 1045043.00 21.2% __ticket_spin_lock 326184.00 6.6% _xfs_buf_find A 70% drop in the CPU usage when looking up buffers. Unfortunately that does not result in an increase in performance underthis workload as contention on the inode_lock soaks up most of the reduction in CPU usage. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/linux-2.6/xfs_buf.c | 8 +------- fs/xfs/linux-2.6/xfs_buf.h | 1 - 2 files changed, 1 insertion(+), 8 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index ea79072f5210..d72cf2bb054a 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -440,12 +440,7 @@ _xfs_buf_find( ASSERT(btp == bp->b_target); if (bp->b_file_offset == range_base && bp->b_buffer_length == range_length) { - /* - * If we look at something, bring it to the - * front of the list for next time. - */ atomic_inc(&bp->b_hold); - list_move(&bp->b_hash_list, &hash->bh_list); goto found; } } @@ -1443,8 +1438,7 @@ xfs_alloc_bufhash( { unsigned int i; - btp->bt_hashshift = external ? 3 : 8; /* 8 or 256 buckets */ - btp->bt_hashmask = (1 << btp->bt_hashshift) - 1; + btp->bt_hashshift = external ? 3 : 12; /* 8 or 4096 buckets */ btp->bt_hash = kmem_zalloc_large((1 << btp->bt_hashshift) * sizeof(xfs_bufhash_t)); for (i = 0; i < (1 << btp->bt_hashshift); i++) { diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index d072e5ff923b..2a05614f0b92 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -137,7 +137,6 @@ typedef struct xfs_buftarg { size_t bt_smask; /* per device buffer hash table */ - uint bt_hashmask; uint bt_hashshift; xfs_bufhash_t *bt_hash; -- cgit v1.2.3 From 23963e54ce187ca6e907c83176c15508b0f6e60d Mon Sep 17 00:00:00 2001 From: Arkadiusz Mi?kiewicz Date: Thu, 26 Aug 2010 10:19:43 +0000 Subject: xfs: Disallow 32bit project quota id Currently on-disk structure is able to keep only 16bit project quota id, so disallow 32bit ones. This fixes a problem where parts of kernel structures holding project quota id are 32bit while parts (on-disk) are 16bit variables which causes project quota member files to be inaccessible for some operations (like mv/rm). Signed-off-by: Arkadiusz Mi?kiewicz Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_ioctl.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 237f5ffb2ee8..4fec427b83ef 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -906,6 +906,13 @@ xfs_ioctl_setattr( if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); + /* + * Disallow 32bit project ids because on-disk structure + * is 16bit only. + */ + if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1)) + return XFS_ERROR(EINVAL); + /* * If disk quotas is on, we make sure that the dquots do exist on disk, * before we start any other transactions. Trying to do this later -- cgit v1.2.3 From 9af25465081480a75824fd7a16a37a5cfebeede9 Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Mon, 30 Aug 2010 02:44:03 +0000 Subject: xfs: Make fiemap work with sparse files In xfs_vn_fiemap, we set bvm_count to fi_extent_max + 1 and want to return fi_extent_max extents, but actually it won't work for a sparse file. The reason is that in xfs_getbmap we will calculate holes and set it in 'out', while out is malloced by bmv_count(fi_extent_max+1) which didn't consider holes. So in the worst case, if 'out' vector looks like [hole, extent, hole, extent, hole, ... hole, extent, hole], we will only return half of fi_extent_max extents. This patch add a new parameter BMV_IF_NO_HOLES for bvm_iflags. So with this flags, we don't use our 'out' in xfs_getbmap for a hole. The solution is a bit ugly by just don't increasing index of 'out' vector. I felt that it is not easy to skip it at the very beginning since we have the complicated check and some function like xfs_getbmapx_fix_eof_hole to adjust 'out'. Cc: Dave Chinner Signed-off-by: Tao Ma Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_iops.c | 2 +- fs/xfs/xfs_bmap.c | 14 +++++++++++++- fs/xfs/xfs_fs.h | 4 +++- 3 files changed, 17 insertions(+), 3 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 68be25dcd301..b1fc2a6bfe83 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -664,7 +664,7 @@ xfs_vn_fiemap( fieinfo->fi_extents_max + 1; bm.bmv_count = min_t(__s32, bm.bmv_count, (PAGE_SIZE * 16 / sizeof(struct getbmapx))); - bm.bmv_iflags = BMV_IF_PREALLOC; + bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES; if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) bm.bmv_iflags |= BMV_IF_ATTRFORK; if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC)) diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 23f14e595c18..f90dadd5a968 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -5533,12 +5533,24 @@ xfs_getbmap( map[i].br_startblock)) goto out_free_map; - nexleft--; bmv->bmv_offset = out[cur_ext].bmv_offset + out[cur_ext].bmv_length; bmv->bmv_length = max_t(__int64_t, 0, bmvend - bmv->bmv_offset); + + /* + * In case we don't want to return the hole, + * don't increase cur_ext so that we can reuse + * it in the next loop. + */ + if ((iflags & BMV_IF_NO_HOLES) && + map[i].br_startblock == HOLESTARTBLOCK) { + memset(&out[cur_ext], 0, sizeof(out[cur_ext])); + continue; + } + + nexleft--; bmv->bmv_entries++; cur_ext++; } diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 7cf7220e7d5f..87c2e9d02288 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -114,8 +114,10 @@ struct getbmapx { #define BMV_IF_NO_DMAPI_READ 0x2 /* Do not generate DMAPI read event */ #define BMV_IF_PREALLOC 0x4 /* rtn status BMV_OF_PREALLOC if req */ #define BMV_IF_DELALLOC 0x8 /* rtn status BMV_OF_DELALLOC if req */ +#define BMV_IF_NO_HOLES 0x10 /* Do not return holes */ #define BMV_IF_VALID \ - (BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC|BMV_IF_DELALLOC) + (BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC| \ + BMV_IF_DELALLOC|BMV_IF_NO_HOLES) /* bmv_oflags values - returned for each non-header segment */ #define BMV_OF_PREALLOC 0x1 /* segment = unwritten pre-allocation */ -- cgit v1.2.3 From a122eb2fdfd78b58c6dd992d6f4b1aaef667eef9 Mon Sep 17 00:00:00 2001 From: Dan Rosenberg Date: Mon, 6 Sep 2010 18:24:57 -0400 Subject: xfs: prevent reading uninitialized stack memory The XFS_IOC_FSGETXATTR ioctl allows unprivileged users to read 12 bytes of uninitialized stack memory, because the fsxattr struct declared on the stack in xfs_ioc_fsgetxattr() does not alter (or zero) the 12-byte fsx_pad member before copying it back to the user. This patch takes care of it. Signed-off-by: Dan Rosenberg Reviewed-by: Eric Sandeen Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_ioctl.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 4fec427b83ef..3b9e626f7cd1 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -785,6 +785,8 @@ xfs_ioc_fsgetxattr( { struct fsxattr fa; + memset(&fa, 0, sizeof(struct fsxattr)); + xfs_ilock(ip, XFS_ILOCK_SHARED); fa.fsx_xflags = xfs_ip2xflags(ip); fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog; -- cgit v1.2.3 From 51749e47e191db8e588ad5cebea731caf7b705d7 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 8 Sep 2010 09:00:22 +0000 Subject: xfs: log IO completion workqueue is a high priority queue The workqueue implementation in 2.6.36-rcX has changed, resulting in the workqueues no longer having dedicated threads for work processing. This has caused severe livelocks under heavy parallel create workloads because the log IO completions have been getting held up behind metadata IO completions. Hence log commits would stall, memory allocation would stall because pages could not be cleaned, and lock contention on the AIL during inode IO completion processing was being seen to slow everything down even further. By making the log Io completion workqueue a high priority workqueue, they are queued ahead of all data/metadata IO completions and processed before the data/metadata completions. Hence the log never gets stalled, and operations needed to clean memory can continue as quickly as possible. This avoids the livelock conditions and allos the system to keep running under heavy load as per normal. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_buf.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index d72cf2bb054a..286e36e21dae 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1932,7 +1932,8 @@ xfs_buf_init(void) if (!xfs_buf_zone) goto out; - xfslogd_workqueue = create_workqueue("xfslogd"); + xfslogd_workqueue = alloc_workqueue("xfslogd", + WQ_RESCUER | WQ_HIGHPRI, 1); if (!xfslogd_workqueue) goto out_free_buf_zone; -- cgit v1.2.3 From 081003fff467ea0e727f66d5d435b4f473a789b3 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Fri, 1 Oct 2010 07:43:54 +0000 Subject: xfs: properly account for reclaimed inodes When marking an inode reclaimable, a per-AG counter is increased, the inode is tagged reclaimable in its per-AG tree, and, when this is the first reclaimable inode in the AG, the AG entry in the per-mount tree is also tagged. When an inode is finally reclaimed, however, it is only deleted from the per-AG tree. Neither the counter is decreased, nor is the parent tree's AG entry untagged properly. Since the tags in the per-mount tree are not cleared, the inode shrinker iterates over all AGs that have had reclaimable inodes at one point in time. The counters on the other hand signal an increasing amount of slab objects to reclaim. Since "70e60ce xfs: convert inode shrinker to per-filesystem context" this is not a real issue anymore because the shrinker bails out after one iteration. But the problem was observable on a machine running v2.6.34, where the reclaimable work increased and each process going into direct reclaim eventually got stuck on the xfs inode shrinking path, trying to scan several million objects. Fix this by properly unwinding the reclaimable-state tracking of an inode when it is reclaimed. Signed-off-by: Johannes Weiner Cc: stable@kernel.org Reviewed-by: Dave Chinner Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_sync.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) (limited to 'fs/xfs/linux-2.6') diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index d59c4a65d492..81976ffed7d6 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -668,14 +668,11 @@ xfs_inode_set_reclaim_tag( xfs_perag_put(pag); } -void -__xfs_inode_clear_reclaim_tag( - xfs_mount_t *mp, +STATIC void +__xfs_inode_clear_reclaim( xfs_perag_t *pag, xfs_inode_t *ip) { - 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 */ @@ -689,6 +686,17 @@ __xfs_inode_clear_reclaim_tag( } } +void +__xfs_inode_clear_reclaim_tag( + xfs_mount_t *mp, + xfs_perag_t *pag, + xfs_inode_t *ip) +{ + radix_tree_tag_clear(&pag->pag_ici_root, + XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); + __xfs_inode_clear_reclaim(pag, ip); +} + /* * Inodes in different states need to be treated differently, and the return * value of xfs_iflush is not sufficient to get this right. The following table @@ -838,6 +846,7 @@ reclaim: if (!radix_tree_delete(&pag->pag_ici_root, XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino))) ASSERT(0); + __xfs_inode_clear_reclaim(pag, ip); write_unlock(&pag->pag_ici_lock); /* -- cgit v1.2.3