From 453eac8a9aa417878a38bdfbccafd5f7ce4e8e4e Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Mon, 11 Jan 2010 11:49:58 +0000 Subject: xfs: Don't wake the aild once per second Now that the AIL push algorithm is traversal safe, we don't need a watchdog function in the xfsaild to catch pushes that fail to make progress. Remove the watchdog timeout and make pushes purely driven by demand. This will remove the once-per-second wakeup that is seen when the filesystem is idle and make laptop power misers happy. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_super.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs/xfs/linux-2.6/xfs_super.c') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 77414db10dc2..9f2e398a5616 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -877,12 +877,11 @@ xfsaild( { struct xfs_ail *ailp = data; xfs_lsn_t last_pushed_lsn = 0; - long tout = 0; + long tout = 0; /* milliseconds */ while (!kthread_should_stop()) { - if (tout) - schedule_timeout_interruptible(msecs_to_jiffies(tout)); - tout = 1000; + schedule_timeout_interruptible(tout ? + msecs_to_jiffies(tout) : MAX_SCHEDULE_TIMEOUT); /* swsusp */ try_to_freeze(); -- cgit v1.2.3 From cbe132a8bdcff0f9afd9060948fb50597c7400b8 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Tue, 26 Jan 2010 15:08:49 +1100 Subject: xfs: don't hold onto reserved blocks on remount,ro If we hold onto reserved blocks when doing a remount,ro we end up writing the blocks used count to disk that includes the reserved blocks. Reserved blocks are not actually used, so this results in the values in the superblock being incorrect. Hence if we run xfs_check or xfs_repair -n while the filesystem is mounted remount,ro we end up with an inconsistent filesystem being reported. Also, running xfs_copy on the remount,ro filesystem will result in an inconsistent image being generated. To fix this, unreserve the blocks when doing the remount,ro, and reserved them again on remount,rw. This way a remount,ro filesystem will appear consistent on disk to all utilities. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/linux-2.6/xfs_super.c | 28 ++++++++++++++++++++++++++++ fs/xfs/xfs_mount.h | 1 + 2 files changed, 29 insertions(+) (limited to 'fs/xfs/linux-2.6/xfs_super.c') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 9f2e398a5616..e9c21454b9e2 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1318,6 +1318,8 @@ xfs_fs_remount( /* ro -> rw */ if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) { + __uint64_t resblks; + mp->m_flags &= ~XFS_MOUNT_RDONLY; if (mp->m_flags & XFS_MOUNT_BARRIER) xfs_mountfs_check_barriers(mp); @@ -1335,11 +1337,37 @@ xfs_fs_remount( } mp->m_update_flags = 0; } + + /* + * Fill out the reserve pool if it is empty. Use the stashed + * value if it is non-zero, otherwise go with the default. + */ + if (mp->m_resblks_save) { + resblks = mp->m_resblks_save; + mp->m_resblks_save = 0; + } else { + resblks = mp->m_sb.sb_dblocks; + do_div(resblks, 20); + resblks = min_t(__uint64_t, resblks, 1024); + } + xfs_reserve_blocks(mp, &resblks, NULL); } /* rw -> ro */ if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) { + /* + * After we have synced the data but before we sync the + * metadata, we need to free up the reserve block pool so that + * the used block count in the superblock on disk is correct at + * the end of the remount. Stash the current reserve pool size + * so that if we get remounted rw, we can return it to the same + * size. + */ + __uint64_t resblks = 0; + xfs_quiesce_data(mp); + mp->m_resblks_save = mp->m_resblks; + xfs_reserve_blocks(mp, &resblks, NULL); xfs_quiesce_attr(mp); mp->m_flags |= XFS_MOUNT_RDONLY; } diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index f4d1441f3f15..02d45f213e58 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -225,6 +225,7 @@ typedef struct xfs_mount { __uint64_t m_maxioffset; /* maximum inode offset */ __uint64_t m_resblks; /* total reserved blocks */ __uint64_t m_resblks_avail;/* available reserved blocks */ + __uint64_t m_resblks_save; /* reserved blks @ remount,ro */ int m_dalign; /* stripe unit */ int m_swidth; /* stripe width */ int m_sinoalign; /* stripe unit inode alignment */ -- cgit v1.2.3 From d5db0f97fbbeff11c88dec1aaf1536a975afbaeb Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Fri, 5 Feb 2010 22:59:53 +0000 Subject: xfs: more reserved blocks fixups This mangles the reserved blocks counts a little more. 1) add a helper function for the default reserved count 2) add helper functions to save/restore counts on ro/rw 3) save/restore reserved blocks on freeze/thaw 4) disallow changing reserved count while readonly V2: changed field name to match Dave's changes Signed-off-by: Eric Sandeen Signed-off-by: Alex Elder --- fs/xfs/linux-2.6/xfs_ioctl.c | 3 +++ fs/xfs/linux-2.6/xfs_super.c | 51 ++++++++++++++++++++++++++++++++------------ fs/xfs/xfs_mount.c | 34 +++++++++++++++++++---------- fs/xfs/xfs_mount.h | 1 + 4 files changed, 64 insertions(+), 25 deletions(-) (limited to 'fs/xfs/linux-2.6/xfs_super.c') diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 3906e85abfdc..4ea1ee18aded 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -1431,6 +1431,9 @@ xfs_file_ioctl( if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (mp->m_flags & XFS_MOUNT_RDONLY) + return -XFS_ERROR(EROFS); + if (copy_from_user(&inout, arg, sizeof(inout))) return -XFS_ERROR(EFAULT); diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index e9c21454b9e2..6ce828e0e17b 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1256,6 +1256,29 @@ xfs_fs_statfs( return 0; } +STATIC void +xfs_save_resvblks(struct xfs_mount *mp) +{ + __uint64_t resblks = 0; + + mp->m_resblks_save = mp->m_resblks; + xfs_reserve_blocks(mp, &resblks, NULL); +} + +STATIC void +xfs_restore_resvblks(struct xfs_mount *mp) +{ + __uint64_t resblks; + + if (mp->m_resblks_save) { + resblks = mp->m_resblks_save; + mp->m_resblks_save = 0; + } else + resblks = xfs_default_resblks(mp); + + xfs_reserve_blocks(mp, &resblks, NULL); +} + STATIC int xfs_fs_remount( struct super_block *sb, @@ -1318,8 +1341,6 @@ xfs_fs_remount( /* ro -> rw */ if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) { - __uint64_t resblks; - mp->m_flags &= ~XFS_MOUNT_RDONLY; if (mp->m_flags & XFS_MOUNT_BARRIER) xfs_mountfs_check_barriers(mp); @@ -1342,15 +1363,7 @@ xfs_fs_remount( * Fill out the reserve pool if it is empty. Use the stashed * value if it is non-zero, otherwise go with the default. */ - if (mp->m_resblks_save) { - resblks = mp->m_resblks_save; - mp->m_resblks_save = 0; - } else { - resblks = mp->m_sb.sb_dblocks; - do_div(resblks, 20); - resblks = min_t(__uint64_t, resblks, 1024); - } - xfs_reserve_blocks(mp, &resblks, NULL); + xfs_restore_resvblks(mp); } /* rw -> ro */ @@ -1363,11 +1376,9 @@ xfs_fs_remount( * so that if we get remounted rw, we can return it to the same * size. */ - __uint64_t resblks = 0; xfs_quiesce_data(mp); - mp->m_resblks_save = mp->m_resblks; - xfs_reserve_blocks(mp, &resblks, NULL); + xfs_save_resvblks(mp); xfs_quiesce_attr(mp); mp->m_flags |= XFS_MOUNT_RDONLY; } @@ -1386,10 +1397,21 @@ xfs_fs_freeze( { struct xfs_mount *mp = XFS_M(sb); + xfs_save_resvblks(mp); xfs_quiesce_attr(mp); return -xfs_fs_log_dummy(mp); } +STATIC int +xfs_fs_unfreeze( + struct super_block *sb) +{ + struct xfs_mount *mp = XFS_M(sb); + + xfs_restore_resvblks(mp); + return 0; +} + STATIC int xfs_fs_show_options( struct seq_file *m, @@ -1612,6 +1634,7 @@ static const struct super_operations xfs_super_operations = { .put_super = xfs_fs_put_super, .sync_fs = xfs_fs_sync_fs, .freeze_fs = xfs_fs_freeze, + .unfreeze_fs = xfs_fs_unfreeze, .statfs = xfs_fs_statfs, .remount_fs = xfs_fs_remount, .show_options = xfs_fs_show_options, diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 7f81ed72c875..5061149b2cc4 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1091,6 +1091,22 @@ xfs_mount_reset_sbqflags( return xfs_trans_commit(tp, 0); } +__uint64_t +xfs_default_resblks(xfs_mount_t *mp) +{ + __uint64_t resblks; + + /* + * We default to 5% or 1024 fsbs of space reserved, whichever is smaller. + * This may drive us straight to ENOSPC on mount, but that implies + * we were already there on the last unmount. Warn if this occurs. + */ + resblks = mp->m_sb.sb_dblocks; + do_div(resblks, 20); + resblks = min_t(__uint64_t, resblks, 1024); + return resblks; +} + /* * This function does the following on an initial mount of a file system: * - reads the superblock from disk and init the mount struct @@ -1401,18 +1417,14 @@ xfs_mountfs( * when at ENOSPC. This is needed for operations like create with * attr, unwritten extent conversion at ENOSPC, etc. Data allocations * are not allowed to use this reserved space. - * - * We default to 5% or 1024 fsbs of space reserved, whichever is smaller. - * This may drive us straight to ENOSPC on mount, but that implies - * we were already there on the last unmount. Warn if this occurs. */ - resblks = mp->m_sb.sb_dblocks; - do_div(resblks, 20); - resblks = min_t(__uint64_t, resblks, 1024); - error = xfs_reserve_blocks(mp, &resblks, NULL); - if (error) - cmn_err(CE_WARN, "XFS: Unable to allocate reserve blocks. " - "Continuing without a reserve pool."); + if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { + resblks = xfs_default_resblks(mp); + error = xfs_reserve_blocks(mp, &resblks, NULL); + if (error) + cmn_err(CE_WARN, "XFS: Unable to allocate reserve " + "blocks. Continuing without a reserve pool."); + } return 0; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 02d45f213e58..70504fcf14cd 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -421,6 +421,7 @@ typedef struct xfs_mod_sb { } xfs_mod_sb_t; extern int xfs_log_sbcount(xfs_mount_t *, uint); +extern __uint64_t xfs_default_resblks(xfs_mount_t *mp); extern int xfs_mountfs(xfs_mount_t *mp); extern void xfs_unmountfs(xfs_mount_t *); -- cgit v1.2.3 From c854363e80b49dd04a4de18ebc379eb8c8806674 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Sat, 6 Feb 2010 12:39:36 +1100 Subject: xfs: Use delayed write for inodes rather than async V2 We currently do background inode flush asynchronously, resulting in inodes being written in whatever order the background writeback issues them. Not only that, there are also blocking and non-blocking asynchronous inode flushes, depending on where the flush comes from. This patch completely removes asynchronous inode writeback. It removes all the strange writeback modes and replaces them with either a synchronous flush or a non-blocking delayed write flush. That is, inode flushes will only issue IO directly if they are synchronous, and background flushing may do nothing if the operation would block (e.g. on a pinned inode or buffer lock). Delayed write flushes will now result in the inode buffer sitting in the delwri queue of the buffer cache to be flushed by either an AIL push or by the xfsbufd timing out the buffer. This will allow accumulation of dirty inode buffers in memory and allow optimisation of inode cluster writeback at the xfsbufd level where we have much greater queue depths than the block layer elevators. We will also get adjacent inode cluster buffer IO merging for free when a later patch in the series allows sorting of the delayed write buffers before dispatch. This effectively means that any inode that is written back by background writeback will be seen as flush locked during AIL pushing, and will result in the buffers being pushed from there. This writeback path is currently non-optimal, but the next patch in the series will fix that problem. A side effect of this delayed write mechanism is that background inode reclaim will no longer directly flush inodes, nor can it wait on the flush lock. The result is that inode reclaim must leave the inode in the reclaimable state until it is clean. Hence attempts to reclaim a dirty inode in the background will simply skip the inode until it is clean and this allows other mechanisms (i.e. xfsbufd) to do more optimal writeback of the dirty buffers. As a result, the inode reclaim code has been rewritten so that it no longer relies on the ambiguous return values of xfs_iflush() to determine whether it is safe to reclaim an inode. Portions of this patch are derived from patches by Christoph Hellwig. Version 2: - cleanup reclaim code as suggested by Christoph - log background reclaim inode flush errors - just pass sync flags to xfs_iflush Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig --- fs/xfs/linux-2.6/xfs_super.c | 4 +- fs/xfs/linux-2.6/xfs_sync.c | 105 +++++++++++++++++++++++++++++++------------ fs/xfs/xfs_inode.c | 75 +++---------------------------- fs/xfs/xfs_inode.h | 10 ----- fs/xfs/xfs_inode_item.c | 10 +++-- fs/xfs/xfs_mount.c | 13 +++++- 6 files changed, 102 insertions(+), 115 deletions(-) (limited to 'fs/xfs/linux-2.6/xfs_super.c') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 6ce828e0e17b..3b5b46b8e3b9 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1064,7 +1064,7 @@ xfs_fs_write_inode( xfs_ilock(ip, XFS_ILOCK_SHARED); xfs_iflock(ip); - error = xfs_iflush(ip, XFS_IFLUSH_SYNC); + error = xfs_iflush(ip, SYNC_WAIT); } else { error = EAGAIN; if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) @@ -1072,7 +1072,7 @@ xfs_fs_write_inode( if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) goto out_unlock; - error = xfs_iflush(ip, XFS_IFLUSH_ASYNC_NOBLOCK); + error = xfs_iflush(ip, 0); } out_unlock: diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 525260c7617f..a9f6d20aff41 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -270,8 +270,7 @@ xfs_sync_inode_attr( goto out_unlock; } - error = xfs_iflush(ip, (flags & SYNC_WAIT) ? - XFS_IFLUSH_SYNC : XFS_IFLUSH_DELWRI); + error = xfs_iflush(ip, flags); out_unlock: xfs_iunlock(ip, XFS_ILOCK_SHARED); @@ -460,16 +459,18 @@ xfs_quiesce_fs( { int count = 0, pincount; + xfs_reclaim_inodes(mp, 0); xfs_flush_buftarg(mp->m_ddev_targp, 0); - xfs_reclaim_inodes(mp, XFS_IFLUSH_DELWRI_ELSE_ASYNC); /* * This loop must run at least twice. The first instance of the loop * will flush most meta data but that will generate more meta data * (typically directory updates). Which then must be flushed and - * logged before we can write the unmount record. + * logged before we can write the unmount record. We also so sync + * reclaim of inodes to catch any that the above delwri flush skipped. */ do { + xfs_reclaim_inodes(mp, SYNC_WAIT); xfs_sync_attr(mp, SYNC_WAIT); pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1); if (!pincount) { @@ -585,7 +586,7 @@ xfs_sync_worker( if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { xfs_log_force(mp, 0); - xfs_reclaim_inodes(mp, XFS_IFLUSH_DELWRI_ELSE_ASYNC); + xfs_reclaim_inodes(mp, 0); /* dgc: errors ignored here */ error = xfs_qm_sync(mp, SYNC_TRYLOCK); error = xfs_sync_fsdata(mp, SYNC_TRYLOCK); @@ -719,21 +720,42 @@ __xfs_inode_clear_reclaim_tag( * shutdown EIO unpin and reclaim * clean, unpinned 0 reclaim * stale, unpinned 0 reclaim - * clean, pinned(*) 0 unpin and reclaim - * stale, pinned 0 unpin and reclaim - * dirty, async 0 block on flush lock, reclaim - * dirty, sync flush 0 block on flush lock, reclaim + * clean, pinned(*) 0 requeue + * stale, pinned EAGAIN requeue + * dirty, delwri ok 0 requeue + * dirty, delwri blocked EAGAIN requeue + * dirty, sync flush 0 reclaim * * (*) dgc: I don't think the clean, pinned state is possible but it gets * handled anyway given the order of checks implemented. * + * As can be seen from the table, the return value of xfs_iflush() is not + * sufficient to correctly decide the reclaim action here. The checks in + * xfs_iflush() might look like duplicates, but they are not. + * + * Also, because we get the flush lock first, we know that any inode that has + * been flushed delwri has had the flush completed by the time we check that + * the inode is clean. The clean inode check needs to be done before flushing + * the inode delwri otherwise we would loop forever requeuing clean inodes as + * we cannot tell apart a successful delwri flush and a clean inode from the + * return value of xfs_iflush(). + * + * Note that because the inode is flushed delayed write by background + * writeback, the flush lock may already be held here and waiting on it can + * result in very long latencies. Hence for sync reclaims, where we wait on the + * flush lock, the caller should push out delayed write inodes first before + * trying to reclaim them to minimise the amount of time spent waiting. For + * background relaim, we just requeue the inode for the next pass. + * * Hence the order of actions after gaining the locks should be: * bad => reclaim * shutdown => unpin and reclaim - * pinned => unpin + * pinned, delwri => requeue + * pinned, sync => unpin * stale => reclaim * clean => reclaim - * dirty => flush, wait and reclaim + * dirty, delwri => flush and requeue + * dirty, sync => flush, wait and reclaim */ STATIC int xfs_reclaim_inode( @@ -741,7 +763,7 @@ xfs_reclaim_inode( struct xfs_perag *pag, int sync_mode) { - int error; + int error = 0; /* * The radix tree lock here protects a thread in xfs_iget from racing @@ -761,7 +783,11 @@ xfs_reclaim_inode( write_unlock(&pag->pag_ici_lock); xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_iflock(ip); + if (!xfs_iflock_nowait(ip)) { + if (!(sync_mode & SYNC_WAIT)) + goto out; + xfs_iflock(ip); + } if (is_bad_inode(VFS_I(ip))) goto reclaim; @@ -769,8 +795,13 @@ xfs_reclaim_inode( xfs_iunpin_wait(ip); goto reclaim; } - if (xfs_ipincount(ip)) + if (xfs_ipincount(ip)) { + if (!(sync_mode & SYNC_WAIT)) { + xfs_ifunlock(ip); + goto out; + } xfs_iunpin_wait(ip); + } if (xfs_iflags_test(ip, XFS_ISTALE)) goto reclaim; if (xfs_inode_clean(ip)) @@ -778,27 +809,43 @@ xfs_reclaim_inode( /* Now we have an inode that needs flushing */ error = xfs_iflush(ip, sync_mode); - if (!error) { - switch(sync_mode) { - case XFS_IFLUSH_DELWRI_ELSE_ASYNC: - case XFS_IFLUSH_DELWRI: - case XFS_IFLUSH_ASYNC: - case XFS_IFLUSH_DELWRI_ELSE_SYNC: - case XFS_IFLUSH_SYNC: - /* IO issued, synchronise with IO completion */ - xfs_iflock(ip); - break; - default: - ASSERT(0); - break; - } + if (sync_mode & SYNC_WAIT) { + xfs_iflock(ip); + goto reclaim; } + /* + * When we have to flush an inode but don't have SYNC_WAIT set, we + * flush the inode out using a delwri buffer and wait for the next + * call into reclaim to find it in a clean state instead of waiting for + * it now. We also don't return errors here - if the error is transient + * then the next reclaim pass will flush the inode, and if the error + * is permanent then the next sync reclaim will relcaim the inode and + * pass on the error. + */ + if (error && !XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_fs_cmn_err(CE_WARN, ip->i_mount, + "inode 0x%llx background reclaim flush failed with %d", + (long long)ip->i_ino, error); + } +out: + xfs_iflags_clear(ip, XFS_IRECLAIM); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + /* + * We could return EAGAIN here to make reclaim rescan the inode tree in + * a short while. However, this just burns CPU time scanning the tree + * waiting for IO to complete and xfssyncd never goes back to the idle + * state. Instead, return 0 to let the next scheduled background reclaim + * attempt to reclaim the inode again. + */ + return 0; + reclaim: xfs_ifunlock(ip); xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_ireclaim(ip); - return 0; + return error; + } int diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 8d0666dd170a..fa31360046d4 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2835,8 +2835,6 @@ xfs_iflush( xfs_dinode_t *dip; xfs_mount_t *mp; int error; - int noblock = (flags == XFS_IFLUSH_ASYNC_NOBLOCK); - enum { INT_DELWRI = (1 << 0), INT_ASYNC = (1 << 1) }; XFS_STATS_INC(xs_iflush_count); @@ -2859,7 +2857,7 @@ xfs_iflush( * in the same cluster are dirty, they will probably write the inode * out for us if they occur after the log force completes. */ - if (noblock && xfs_ipincount(ip)) { + if (!(flags & SYNC_WAIT) && xfs_ipincount(ip)) { xfs_iunpin_nowait(ip); xfs_ifunlock(ip); return EAGAIN; @@ -2892,61 +2890,11 @@ xfs_iflush( return XFS_ERROR(EIO); } - /* - * Decide how buffer will be flushed out. This is done before - * the call to xfs_iflush_int because this field is zeroed by it. - */ - if (iip != NULL && iip->ili_format.ilf_fields != 0) { - /* - * Flush out the inode buffer according to the directions - * of the caller. In the cases where the caller has given - * us a choice choose the non-delwri case. This is because - * the inode is in the AIL and we need to get it out soon. - */ - switch (flags) { - case XFS_IFLUSH_SYNC: - case XFS_IFLUSH_DELWRI_ELSE_SYNC: - flags = 0; - break; - case XFS_IFLUSH_ASYNC_NOBLOCK: - case XFS_IFLUSH_ASYNC: - case XFS_IFLUSH_DELWRI_ELSE_ASYNC: - flags = INT_ASYNC; - break; - case XFS_IFLUSH_DELWRI: - flags = INT_DELWRI; - break; - default: - ASSERT(0); - flags = 0; - break; - } - } else { - switch (flags) { - case XFS_IFLUSH_DELWRI_ELSE_SYNC: - case XFS_IFLUSH_DELWRI_ELSE_ASYNC: - case XFS_IFLUSH_DELWRI: - flags = INT_DELWRI; - break; - case XFS_IFLUSH_ASYNC_NOBLOCK: - case XFS_IFLUSH_ASYNC: - flags = INT_ASYNC; - break; - case XFS_IFLUSH_SYNC: - flags = 0; - break; - default: - ASSERT(0); - flags = 0; - break; - } - } - /* * Get the buffer containing the on-disk inode. */ error = xfs_itobp(mp, NULL, ip, &dip, &bp, - noblock ? XBF_TRYLOCK : XBF_LOCK); + (flags & SYNC_WAIT) ? XBF_LOCK : XBF_TRYLOCK); if (error || !bp) { xfs_ifunlock(ip); return error; @@ -2974,13 +2922,10 @@ xfs_iflush( if (error) goto cluster_corrupt_out; - if (flags & INT_DELWRI) { - xfs_bdwrite(mp, bp); - } else if (flags & INT_ASYNC) { - error = xfs_bawrite(mp, bp); - } else { + if (flags & SYNC_WAIT) error = xfs_bwrite(mp, bp); - } + else + xfs_bdwrite(mp, bp); return error; corrupt_out: @@ -3015,16 +2960,6 @@ xfs_iflush_int( iip = ip->i_itemp; mp = ip->i_mount; - - /* - * If the inode isn't dirty, then just release the inode - * flush lock and do nothing. - */ - if (xfs_inode_clean(ip)) { - xfs_ifunlock(ip); - return 0; - } - /* set *dip = inode's place in the buffer */ dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset); diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 8b618ea4d692..6c912b027596 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -419,16 +419,6 @@ static inline void xfs_ifunlock(xfs_inode_t *ip) #define XFS_IOLOCK_DEP(flags) (((flags) & XFS_IOLOCK_DEP_MASK) >> XFS_IOLOCK_SHIFT) #define XFS_ILOCK_DEP(flags) (((flags) & XFS_ILOCK_DEP_MASK) >> XFS_ILOCK_SHIFT) -/* - * Flags for xfs_iflush() - */ -#define XFS_IFLUSH_DELWRI_ELSE_SYNC 1 -#define XFS_IFLUSH_DELWRI_ELSE_ASYNC 2 -#define XFS_IFLUSH_SYNC 3 -#define XFS_IFLUSH_ASYNC 4 -#define XFS_IFLUSH_DELWRI 5 -#define XFS_IFLUSH_ASYNC_NOBLOCK 6 - /* * Flags for xfs_itruncate_start(). */ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 48ec1c0b23ce..207553e82954 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -866,10 +866,14 @@ xfs_inode_item_push( iip->ili_format.ilf_fields != 0); /* - * Write out the inode. The completion routine ('iflush_done') will - * pull it from the AIL, mark it clean, unlock the flush lock. + * Push the inode to it's backing buffer. This will not remove the + * inode from the AIL - a further push will be required to trigger a + * buffer push. However, this allows all the dirty inodes to be pushed + * to the buffer before it is pushed to disk. THe buffer IO completion + * will pull th einode from the AIL, mark it clean and unlock the flush + * lock. */ - (void) xfs_iflush(ip, XFS_IFLUSH_ASYNC); + (void) xfs_iflush(ip, 0); xfs_iunlock(ip, XFS_ILOCK_SHARED); return; diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 5061149b2cc4..6afaaeb2950a 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1468,7 +1468,18 @@ xfs_unmountfs( * need to force the log first. */ xfs_log_force(mp, XFS_LOG_SYNC); - xfs_reclaim_inodes(mp, XFS_IFLUSH_ASYNC); + + /* + * Do a delwri reclaim pass first so that as many dirty inodes are + * queued up for IO as possible. Then flush the buffers before making + * a synchronous path to catch all the remaining inodes are reclaimed. + * This makes the reclaim process as quick as possible by avoiding + * synchronous writeout and blocking on inodes already in the delwri + * state as much as possible. + */ + xfs_reclaim_inodes(mp, 0); + XFS_bflush(mp->m_ddev_targp); + xfs_reclaim_inodes(mp, SYNC_WAIT); xfs_qm_unmount(mp); -- cgit v1.2.3 From 07fec73625dc0db6f9aed68019918208a2ca53f5 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 9 Feb 2010 11:43:49 +1100 Subject: xfs: log changed inodes instead of writing them synchronously When an inode has already be flushed delayed write, xfs_inode_clean() returns true and hence xfs_fs_write_inode() can return on a synchronous inode write without having written the inode. Currently these sycnhronous writes only come sync(1), unmount, a sycnhronous NFS export and cachefiles so should be relatively rare and out of common performance paths. Realistically, a synchronous inode write is not necessary here; we can avoid writing the inode by logging any non-transactional changes that are pending. This needs to be done with synchronous transactions, but it avoids seeking between the log and inode clusters as we do now. We don't force the log if the inode is pinned, though, so this differs from the fsync case. For normal sys_sync and unmount behaviour this is fine because we do a synchronous log force in xfs_sync_data which is called from the ->sync_fs code. It does however break the NFS synchronous export guarantees for now, but work is under way to fix this at a higher level or for the higher level to provide an additional flag in the writeback control to tell us that a log force is needed. Portions of this patch are based on work from Dave Chinner. Signed-off-by: Christoph Hellwig Reviewed-by: Dave Chinner Reviewed-by: Alex Elder --- fs/xfs/linux-2.6/xfs_super.c | 111 ++++++++++++++++++++++++++++++++----------- 1 file changed, 82 insertions(+), 29 deletions(-) (limited to 'fs/xfs/linux-2.6/xfs_super.c') diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 3b5b46b8e3b9..25ea2408118f 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1021,12 +1021,45 @@ xfs_fs_dirty_inode( XFS_I(inode)->i_update_core = 1; } -/* - * Attempt to flush the inode, this will actually fail - * if the inode is pinned, but we dirty the inode again - * at the point when it is unpinned after a log write, - * since this is when the inode itself becomes flushable. - */ +STATIC int +xfs_log_inode( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error; + + xfs_iunlock(ip, XFS_ILOCK_SHARED); + tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); + error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + + if (error) { + xfs_trans_cancel(tp, 0); + /* we need to return with the lock hold shared */ + xfs_ilock(ip, XFS_ILOCK_SHARED); + return error; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * Note - it's possible that we might have pushed ourselves out of the + * way during trans_reserve which would flush the inode. But there's + * no guarantee that the inode buffer has actually gone out yet (it's + * delwri). Plus the buffer could be pinned anyway if it's part of + * an inode in another recent transaction. So we play it safe and + * fire off the transaction anyway. + */ + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + xfs_trans_ihold(tp, ip); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp, 0); + xfs_ilock_demote(ip, XFS_ILOCK_EXCL); + + return error; +} + STATIC int xfs_fs_write_inode( struct inode *inode, @@ -1034,7 +1067,7 @@ xfs_fs_write_inode( { struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; - int error = 0; + int error = EAGAIN; xfs_itrace_entry(ip); @@ -1045,35 +1078,55 @@ xfs_fs_write_inode( error = xfs_wait_on_pages(ip, 0, -1); if (error) goto out; - } - - /* - * Bypass inodes which have already been cleaned by - * the inode flush clustering code inside xfs_iflush - */ - if (xfs_inode_clean(ip)) - goto out; - /* - * We make this non-blocking if the inode is contended, return - * EAGAIN to indicate to the caller that they did not succeed. - * This prevents the flush path from blocking on inodes inside - * another operation right now, they get caught later by xfs_sync. - */ - if (sync) { + /* + * Make sure the inode has hit stable storage. By using the + * log and the fsync transactions we reduce the IOs we have + * to do here from two (log and inode) to just the log. + * + * Note: We still need to do a delwri write of the inode after + * this to flush it to the backing buffer so that bulkstat + * works properly if this is the first time the inode has been + * written. Because we hold the ilock atomically over the + * transaction commit and the inode flush we are guaranteed + * that the inode is not pinned when it returns. If the flush + * lock is already held, then the inode has already been + * flushed once and we don't need to flush it again. Hence + * the code will only flush the inode if it isn't already + * being flushed. + */ xfs_ilock(ip, XFS_ILOCK_SHARED); - xfs_iflock(ip); - - error = xfs_iflush(ip, SYNC_WAIT); + if (ip->i_update_core) { + error = xfs_log_inode(ip); + if (error) + goto out_unlock; + } } else { - error = EAGAIN; + /* + * We make this non-blocking if the inode is contended, return + * EAGAIN to indicate to the caller that they did not succeed. + * This prevents the flush path from blocking on inodes inside + * another operation right now, they get caught later by xfs_sync. + */ if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) goto out; - if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) - goto out_unlock; + } + + if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) + goto out_unlock; - error = xfs_iflush(ip, 0); + /* + * Now we have the flush lock and the inode is not pinned, we can check + * if the inode is really clean as we know that there are no pending + * transaction completions, it is not waiting on the delayed write + * queue and there is no IO in progress. + */ + if (xfs_inode_clean(ip)) { + xfs_ifunlock(ip); + error = 0; + goto out_unlock; } + error = xfs_iflush(ip, 0); out_unlock: xfs_iunlock(ip, XFS_ILOCK_SHARED); -- cgit v1.2.3