From a9e7f4472075fb6937c545af3f6329e9946bbe66 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Mon, 14 Dec 2009 15:21:14 +0300 Subject: ext4: Convert to generic reserved quota's space management. This patch also fixes write vs chown race condition. Acked-by: "Theodore Ts'o" Signed-off-by: Dmitry Monakhov Signed-off-by: Jan Kara --- fs/ext4/inode.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5352db1a3086..a31980dd9b86 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1003,17 +1003,12 @@ out: return err; } -qsize_t ext4_get_reserved_space(struct inode *inode) +#ifdef CONFIG_QUOTA +qsize_t *ext4_get_reserved_space(struct inode *inode) { - unsigned long long total; - - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - total = EXT4_I(inode)->i_reserved_data_blocks + - EXT4_I(inode)->i_reserved_meta_blocks; - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - - return (total << inode->i_blkbits); + return &EXT4_I(inode)->i_reserved_quota; } +#endif /* * Calculate the number of metadata blocks need to reserve * to allocate @blocks for non extent file based file @@ -4794,6 +4789,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; inode->i_size = ext4_isize(raw_inode); ei->i_disksize = inode->i_size; +#ifdef CONFIG_QUOTA + ei->i_reserved_quota = 0; +#endif inode->i_generation = le32_to_cpu(raw_inode->i_generation); ei->i_block_group = iloc.block_group; ei->i_last_alloc_group = ~0; -- cgit v1.2.1 From d21cd8f163ac44b15c465aab7306db931c606908 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Thu, 10 Dec 2009 03:31:45 +0000 Subject: ext4: Fix potential quota deadlock We have to delay vfs_dq_claim_space() until allocation context destruction. Currently we have following call-trace: ext4_mb_new_blocks() /* task is already holding ac->alloc_semp */ ->ext4_mb_mark_diskspace_used ->vfs_dq_claim_space() /* acquire dqptr_sem here. Possible deadlock */ ->ext4_mb_release_context() /* drop ac->alloc_semp here */ Let's move quota claiming to ext4_da_update_reserve_space() ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-rc7 #18 ------------------------------------------------------- write-truncate-/3465 is trying to acquire lock: (&s->s_dquot.dqptr_sem){++++..}, at: [] dquot_claim_space+0x3b/0x1b0 but task is already holding lock: (&meta_group_info[i]->alloc_sem){++++..}, at: [] ext4_mb_load_buddy+0xb2/0x370 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (&meta_group_info[i]->alloc_sem){++++..}: [] __lock_acquire+0xd7b/0x1260 [] lock_acquire+0xba/0xd0 [] down_read+0x51/0x90 [] ext4_mb_load_buddy+0xb2/0x370 [] ext4_mb_free_blocks+0x46c/0x870 [] ext4_free_blocks+0x73/0x130 [] ext4_ext_truncate+0x76c/0x8d0 [] ext4_truncate+0x187/0x5e0 [] vmtruncate+0x6b/0x70 [] inode_setattr+0x62/0x190 [] ext4_setattr+0x25a/0x370 [] notify_change+0x151/0x340 [] do_truncate+0x6d/0xa0 [] may_open+0x1d4/0x200 [] do_filp_open+0x1eb/0x910 [] do_sys_open+0x6d/0x140 [] sys_open+0x2e/0x40 [] sysenter_do_call+0x12/0x32 -> #2 (&ei->i_data_sem){++++..}: [] __lock_acquire+0xd7b/0x1260 [] lock_acquire+0xba/0xd0 [] down_read+0x51/0x90 [] ext4_get_blocks+0x47/0x450 [] ext4_getblk+0x61/0x1d0 [] ext4_bread+0x1f/0xa0 [] ext4_quota_write+0x12c/0x310 [] qtree_write_dquot+0x93/0x120 [] v2_write_dquot+0x28/0x30 [] dquot_commit+0xab/0xf0 [] ext4_write_dquot+0x77/0x90 [] ext4_mark_dquot_dirty+0x2f/0x50 [] dquot_alloc_inode+0x101/0x180 [] ext4_new_inode+0x602/0xf00 [] ext4_create+0x89/0x150 [] vfs_create+0xa2/0xc0 [] do_filp_open+0x7a7/0x910 [] do_sys_open+0x6d/0x140 [] sys_open+0x2e/0x40 [] sysenter_do_call+0x12/0x32 -> #1 (&sb->s_type->i_mutex_key#7/4){+.+...}: [] __lock_acquire+0xd7b/0x1260 [] lock_acquire+0xba/0xd0 [] mutex_lock_nested+0x65/0x2d0 [] vfs_load_quota_inode+0x4bd/0x5a0 [] vfs_quota_on_path+0x5f/0x70 [] ext4_quota_on+0x112/0x190 [] sys_quotactl+0x44a/0x8a0 [] sysenter_do_call+0x12/0x32 -> #0 (&s->s_dquot.dqptr_sem){++++..}: [] __lock_acquire+0x1091/0x1260 [] lock_acquire+0xba/0xd0 [] down_read+0x51/0x90 [] dquot_claim_space+0x3b/0x1b0 [] ext4_mb_mark_diskspace_used+0x36f/0x380 [] ext4_mb_new_blocks+0x34a/0x530 [] ext4_ext_get_blocks+0x122b/0x13c0 [] ext4_get_blocks+0x226/0x450 [] mpage_da_map_blocks+0xc3/0xaa0 [] ext4_da_writepages+0x506/0x790 [] do_writepages+0x22/0x50 [] __filemap_fdatawrite_range+0x6d/0x80 [] filemap_flush+0x2b/0x30 [] ext4_alloc_da_blocks+0x5c/0x60 [] ext4_release_file+0x75/0xb0 [] __fput+0xf9/0x210 [] fput+0x27/0x30 [] filp_close+0x4c/0x80 [] put_files_struct+0x6e/0xd0 [] exit_files+0x47/0x60 [] do_exit+0x144/0x710 [] do_group_exit+0x38/0xa0 [] get_signal_to_deliver+0x2ac/0x410 [] do_notify_resume+0xb9/0x890 [] work_notifysig+0x13/0x21 other info that might help us debug this: 3 locks held by write-truncate-/3465: #0: (jbd2_handle){+.+...}, at: [] start_this_handle+0x38f/0x5c0 #1: (&ei->i_data_sem){++++..}, at: [] ext4_get_blocks+0xb6/0x450 #2: (&meta_group_info[i]->alloc_sem){++++..}, at: [] ext4_mb_load_buddy+0xb2/0x370 stack backtrace: Pid: 3465, comm: write-truncate- Not tainted 2.6.32-rc7 #18 Call Trace: [] ? printk+0x1d/0x22 [] print_circular_bug+0xca/0xd0 [] __lock_acquire+0x1091/0x1260 [] ? sched_clock_local+0xd2/0x170 [] ? trace_hardirqs_off_caller+0x20/0xd0 [] lock_acquire+0xba/0xd0 [] ? dquot_claim_space+0x3b/0x1b0 [] down_read+0x51/0x90 [] ? dquot_claim_space+0x3b/0x1b0 [] dquot_claim_space+0x3b/0x1b0 [] ext4_mb_mark_diskspace_used+0x36f/0x380 [] ext4_mb_new_blocks+0x34a/0x530 [] ? ext4_ext_find_extent+0x25d/0x280 [] ext4_ext_get_blocks+0x122b/0x13c0 [] ? sched_clock_local+0xd2/0x170 [] ? sched_clock_cpu+0x120/0x160 [] ? cpu_clock+0x4f/0x60 [] ? trace_hardirqs_off_caller+0x20/0xd0 [] ? down_write+0x8c/0xa0 [] ext4_get_blocks+0x226/0x450 [] ? sched_clock_cpu+0x120/0x160 [] ? cpu_clock+0x4f/0x60 [] ? trace_hardirqs_off+0xb/0x10 [] mpage_da_map_blocks+0xc3/0xaa0 [] ? find_get_pages_tag+0x16c/0x180 [] ? find_get_pages_tag+0x0/0x180 [] ? __mpage_da_writepage+0x16d/0x1a0 [] ? pagevec_lookup_tag+0x2e/0x40 [] ? write_cache_pages+0xdb/0x3d0 [] ? __mpage_da_writepage+0x0/0x1a0 [] ext4_da_writepages+0x506/0x790 [] ? cpu_clock+0x4f/0x60 [] ? sched_clock_local+0xd2/0x170 [] ? sched_clock_cpu+0x120/0x160 [] ? sched_clock_cpu+0x120/0x160 [] ? ext4_da_writepages+0x0/0x790 [] do_writepages+0x22/0x50 [] __filemap_fdatawrite_range+0x6d/0x80 [] filemap_flush+0x2b/0x30 [] ext4_alloc_da_blocks+0x5c/0x60 [] ext4_release_file+0x75/0xb0 [] __fput+0xf9/0x210 [] fput+0x27/0x30 [] filp_close+0x4c/0x80 [] put_files_struct+0x6e/0xd0 [] exit_files+0x47/0x60 [] do_exit+0x144/0x710 [] ? lock_release_holdtime+0x33/0x210 [] ? _spin_unlock_irq+0x27/0x30 [] do_group_exit+0x38/0xa0 [] ? trace_hardirqs_on+0xb/0x10 [] get_signal_to_deliver+0x2ac/0x410 [] do_notify_resume+0xb9/0x890 [] ? trace_hardirqs_off_caller+0x20/0xd0 [] ? lock_release_holdtime+0x33/0x210 [] ? autoremove_wake_function+0x0/0x50 [] ? trace_hardirqs_on_caller+0x134/0x190 [] ? trace_hardirqs_on+0xb/0x10 [] ? security_file_permission+0x14/0x20 [] ? vfs_write+0x131/0x190 [] ? do_sync_write+0x0/0x120 [] ? sysenter_do_call+0x27/0x32 [] work_notifysig+0x13/0x21 CC: Theodore Ts'o Signed-off-by: Dmitry Monakhov Signed-off-by: Jan Kara --- fs/ext4/inode.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a31980dd9b86..ddd0a9c87d9e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1046,7 +1046,7 @@ static int ext4_calc_metadata_amount(struct inode *inode, int blocks) static void ext4_da_update_reserve_space(struct inode *inode, int used) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - int total, mdb, mdb_free; + int total, mdb, mdb_free, mdb_claim = 0; spin_lock(&EXT4_I(inode)->i_block_reservation_lock); /* recalculate the number of metablocks still need to be reserved */ @@ -1059,7 +1059,9 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) if (mdb_free) { /* Account for allocated meta_blocks */ - mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + mdb_claim = EXT4_I(inode)->i_allocated_meta_blocks; + BUG_ON(mdb_free < mdb_claim); + mdb_free -= mdb_claim; /* update fs dirty blocks counter */ percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); @@ -1070,8 +1072,11 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) /* update per-inode reservations */ BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); EXT4_I(inode)->i_reserved_data_blocks -= used; + percpu_counter_sub(&sbi->s_dirtyblocks_counter, used + mdb_claim); spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + vfs_dq_claim_block(inode, used + mdb_claim); + /* * free those over-booking quota for metadata blocks */ -- cgit v1.2.1 From 39bc680a8160bb9d6743f7873b535d553ff61058 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Thu, 10 Dec 2009 16:36:27 +0000 Subject: ext4: fix sleep inside spinlock issue with quota and dealloc (#14739) Unlock i_block_reservation_lock before vfs_dq_reserve_block(). This patch fixes http://bugzilla.kernel.org/show_bug.cgi?id=14739 CC: Theodore Ts'o Signed-off-by: Dmitry Monakhov Signed-off-by: Jan Kara --- fs/ext4/inode.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index ddd0a9c87d9e..ab807963a614 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1816,19 +1816,17 @@ repeat: md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; total = md_needed + nrblocks; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); /* * Make quota reservation here to prevent quota overflow * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, total)) { - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + if (vfs_dq_reserve_block(inode, total)) return -EDQUOT; - } if (ext4_claim_free_blocks(sbi, total)) { - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); vfs_dq_release_reservation_block(inode, total); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { yield(); @@ -1836,10 +1834,11 @@ repeat: } return -ENOSPC; } + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); EXT4_I(inode)->i_reserved_data_blocks += nrblocks; - EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; - + EXT4_I(inode)->i_reserved_meta_blocks += md_needed; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return 0; /* success */ } -- cgit v1.2.1 From c8afb44682fcef6273e8b8eb19fab13ddd05b386 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Wed, 23 Dec 2009 07:58:12 -0500 Subject: ext4: flush delalloc blocks when space is low Creating many small files in rapid succession on a small filesystem can lead to spurious ENOSPC; on a 104MB filesystem: for i in `seq 1 22500`; do echo -n > $SCRATCH_MNT/$i echo XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX > $SCRATCH_MNT/$i done leads to ENOSPC even though after a sync, 40% of the fs is free again. This is because we reserve worst-case metadata for delalloc writes, and when data is allocated that worst-case reservation is not usually needed. When freespace is low, kicking off an async writeback will start converting that worst-case space usage into something more realistic, almost always freeing up space to continue. This resolves the testcase for me, and survives all 4 generic ENOSPC tests in xfstests. We'll still need a hard synchronous sync to squeeze out the last bit, but this fixes things up to a large degree. Signed-off-by: Eric Sandeen Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index ab807963a614..282621f18c10 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2993,11 +2993,18 @@ static int ext4_nonda_switch(struct super_block *sb) if (2 * free_blocks < 3 * dirty_blocks || free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) { /* - * free block count is less that 150% of dirty blocks - * or free blocks is less that watermark + * free block count is less than 150% of dirty blocks + * or free blocks is less than watermark */ return 1; } + /* + * Even if we don't switch but are nearing capacity, + * start pushing delalloc when 1/2 of free blocks are dirty. + */ + if (free_blocks < 2 * dirty_blocks) + writeback_inodes_sb_if_idle(sb); + return 0; } -- cgit v1.2.1 From 2faf2e19dd0e060eeb32442858ef495ac3083277 Mon Sep 17 00:00:00 2001 From: Richard Kennedy Date: Fri, 25 Dec 2009 15:46:07 -0500 Subject: ext4: return correct wbc.nr_to_write in ext4_da_writepages When ext4_da_writepages increases the nr_to_write in writeback_control then it must always re-base the return value. Originally there was a (misguided) attempt prevent wbc.nr_to_write from going negative. In fact, it's necessary to allow nr_to_write to be negative so that wb_writeback() can correctly calculate how many pages were actually written. Signed-off-by: Richard Kennedy Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 282621f18c10..3e3b45458ef8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2967,8 +2967,7 @@ retry: out_writepages: if (!no_nrwrite_index_update) wbc->no_nrwrite_index_update = 0; - if (wbc->nr_to_write > nr_to_writebump) - wbc->nr_to_write -= nr_to_writebump; + wbc->nr_to_write -= nr_to_writebump; wbc->range_start = range_start; trace_ext4_da_writepages_result(inode, wbc, ret, pages_written); return ret; -- cgit v1.2.1 From 0637c6f4135f592f094207c7c21e7c0fc5557834 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 30 Dec 2009 14:20:45 -0500 Subject: ext4: Patch up how we claim metadata blocks for quota purposes As reported in Kernel Bugzilla #14936, commit d21cd8f triggered a BUG in the function ext4_da_update_reserve_space() found in fs/ext4/inode.c. The root cause of this BUG() was caused by the fact that ext4_calc_metadata_amount() can severely over-estimate how many metadata blocks will be needed, especially when using direct block-mapped files. In addition, it can also badly *under* estimate how much space is needed, since ext4_calc_metadata_amount() assumes that the blocks are contiguous, and this is not always true. If the application is writing blocks to a sparse file, the number of metadata blocks necessary can be severly underestimated by the functions ext4_da_reserve_space(), ext4_da_update_reserve_space() and ext4_da_release_space(). This was the cause of the dq_claim_space reports found on kerneloops.org. Unfortunately, doing this right means that we need to massively over-estimate the amount of free space needed. So in some cases we may need to force the inode to be written to disk asynchronously in to avoid spurious quota failures. http://bugzilla.kernel.org/show_bug.cgi?id=14936 Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 157 ++++++++++++++++++++++++++++++-------------------------- 1 file changed, 84 insertions(+), 73 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 3e3b45458ef8..84eeb8f515a3 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1043,43 +1043,47 @@ static int ext4_calc_metadata_amount(struct inode *inode, int blocks) return ext4_indirect_calc_metadata_amount(inode, blocks); } +/* + * Called with i_data_sem down, which is important since we can call + * ext4_discard_preallocations() from here. + */ static void ext4_da_update_reserve_space(struct inode *inode, int used) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - int total, mdb, mdb_free, mdb_claim = 0; - - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - /* recalculate the number of metablocks still need to be reserved */ - total = EXT4_I(inode)->i_reserved_data_blocks - used; - mdb = ext4_calc_metadata_amount(inode, total); - - /* figure out how many metablocks to release */ - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - - if (mdb_free) { - /* Account for allocated meta_blocks */ - mdb_claim = EXT4_I(inode)->i_allocated_meta_blocks; - BUG_ON(mdb_free < mdb_claim); - mdb_free -= mdb_claim; - - /* update fs dirty blocks counter */ + struct ext4_inode_info *ei = EXT4_I(inode); + int mdb_free = 0; + + spin_lock(&ei->i_block_reservation_lock); + if (unlikely(used > ei->i_reserved_data_blocks)) { + ext4_msg(inode->i_sb, KERN_NOTICE, "%s: ino %lu, used %d " + "with only %d reserved data blocks\n", + __func__, inode->i_ino, used, + ei->i_reserved_data_blocks); + WARN_ON(1); + used = ei->i_reserved_data_blocks; + } + + /* Update per-inode reservations */ + ei->i_reserved_data_blocks -= used; + used += ei->i_allocated_meta_blocks; + ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks; + ei->i_allocated_meta_blocks = 0; + percpu_counter_sub(&sbi->s_dirtyblocks_counter, used); + + if (ei->i_reserved_data_blocks == 0) { + /* + * We can release all of the reserved metadata blocks + * only when we have written all of the delayed + * allocation blocks. + */ + mdb_free = ei->i_allocated_meta_blocks; percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); - EXT4_I(inode)->i_allocated_meta_blocks = 0; - EXT4_I(inode)->i_reserved_meta_blocks = mdb; + ei->i_allocated_meta_blocks = 0; } - - /* update per-inode reservations */ - BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); - EXT4_I(inode)->i_reserved_data_blocks -= used; - percpu_counter_sub(&sbi->s_dirtyblocks_counter, used + mdb_claim); spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - vfs_dq_claim_block(inode, used + mdb_claim); - - /* - * free those over-booking quota for metadata blocks - */ + /* Update quota subsystem */ + vfs_dq_claim_block(inode, used); if (mdb_free) vfs_dq_release_reservation_block(inode, mdb_free); @@ -1088,7 +1092,8 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) * there aren't any writers on the inode, we can discard the * inode's preallocations. */ - if (!total && (atomic_read(&inode->i_writecount) == 0)) + if ((ei->i_reserved_data_blocks == 0) && + (atomic_read(&inode->i_writecount) == 0)) ext4_discard_preallocations(inode); } @@ -1801,7 +1806,8 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) { int retries = 0; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - unsigned long md_needed, mdblocks, total = 0; + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned long md_needed, md_reserved, total = 0; /* * recalculate the amount of metadata blocks to reserve @@ -1809,35 +1815,44 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) * worse case is one extent per block */ repeat: - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; - mdblocks = ext4_calc_metadata_amount(inode, total); - BUG_ON(mdblocks < EXT4_I(inode)->i_reserved_meta_blocks); - - md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; + spin_lock(&ei->i_block_reservation_lock); + md_reserved = ei->i_reserved_meta_blocks; + md_needed = ext4_calc_metadata_amount(inode, nrblocks); total = md_needed + nrblocks; - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + spin_unlock(&ei->i_block_reservation_lock); /* * Make quota reservation here to prevent quota overflow * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, total)) + if (vfs_dq_reserve_block(inode, total)) { + /* + * We tend to badly over-estimate the amount of + * metadata blocks which are needed, so if we have + * reserved any metadata blocks, try to force out the + * inode and see if we have any better luck. + */ + if (md_reserved && retries++ <= 3) + goto retry; return -EDQUOT; + } if (ext4_claim_free_blocks(sbi, total)) { vfs_dq_release_reservation_block(inode, total); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { + retry: + if (md_reserved) + write_inode_now(inode, (retries == 3)); yield(); goto repeat; } return -ENOSPC; } - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - EXT4_I(inode)->i_reserved_data_blocks += nrblocks; - EXT4_I(inode)->i_reserved_meta_blocks += md_needed; - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + spin_lock(&ei->i_block_reservation_lock); + ei->i_reserved_data_blocks += nrblocks; + ei->i_reserved_meta_blocks += md_needed; + spin_unlock(&ei->i_block_reservation_lock); return 0; /* success */ } @@ -1845,49 +1860,45 @@ repeat: static void ext4_da_release_space(struct inode *inode, int to_free) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - int total, mdb, mdb_free, release; + struct ext4_inode_info *ei = EXT4_I(inode); if (!to_free) return; /* Nothing to release, exit */ spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - if (!EXT4_I(inode)->i_reserved_data_blocks) { + if (unlikely(to_free > ei->i_reserved_data_blocks)) { /* - * if there is no reserved blocks, but we try to free some - * then the counter is messed up somewhere. - * but since this function is called from invalidate - * page, it's harmless to return without any action + * if there aren't enough reserved blocks, then the + * counter is messed up somewhere. Since this + * function is called from invalidate page, it's + * harmless to return without any action. */ - printk(KERN_INFO "ext4 delalloc try to release %d reserved " - "blocks for inode %lu, but there is no reserved " - "data blocks\n", to_free, inode->i_ino); - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - return; + ext4_msg(inode->i_sb, KERN_NOTICE, "ext4_da_release_space: " + "ino %lu, to_free %d with only %d reserved " + "data blocks\n", inode->i_ino, to_free, + ei->i_reserved_data_blocks); + WARN_ON(1); + to_free = ei->i_reserved_data_blocks; } + ei->i_reserved_data_blocks -= to_free; - /* recalculate the number of metablocks still need to be reserved */ - total = EXT4_I(inode)->i_reserved_data_blocks - to_free; - mdb = ext4_calc_metadata_amount(inode, total); - - /* figure out how many metablocks to release */ - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - - release = to_free + mdb_free; - - /* update fs dirty blocks counter for truncate case */ - percpu_counter_sub(&sbi->s_dirtyblocks_counter, release); + if (ei->i_reserved_data_blocks == 0) { + /* + * We can release all of the reserved metadata blocks + * only when we have written all of the delayed + * allocation blocks. + */ + to_free += ei->i_allocated_meta_blocks; + ei->i_allocated_meta_blocks = 0; + } - /* update per-inode reservations */ - BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks); - EXT4_I(inode)->i_reserved_data_blocks -= to_free; + /* update fs dirty blocks counter */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, to_free); - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - EXT4_I(inode)->i_reserved_meta_blocks = mdb; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - vfs_dq_release_reservation_block(inode, release); + vfs_dq_release_reservation_block(inode, to_free); } static void ext4_da_page_release_reservation(struct page *page, -- cgit v1.2.1 From ee5f4d9cdf32fd99172d11665c592a288c2b1ff4 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 1 Jan 2010 02:36:15 -0500 Subject: ext4: Fix accounting of reserved metadata blocks Commit 0637c6f had a typo which caused the reserved metadata blocks to not be released correctly. Fix this. Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 84eeb8f515a3..bdaa92a29e0e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1076,9 +1076,9 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) * only when we have written all of the delayed * allocation blocks. */ - mdb_free = ei->i_allocated_meta_blocks; + mdb_free = ei->i_reserved_meta_blocks; + ei->i_reserved_meta_blocks = 0; percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); - ei->i_allocated_meta_blocks = 0; } spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); @@ -1889,8 +1889,8 @@ static void ext4_da_release_space(struct inode *inode, int to_free) * only when we have written all of the delayed * allocation blocks. */ - to_free += ei->i_allocated_meta_blocks; - ei->i_allocated_meta_blocks = 0; + to_free += ei->i_reserved_meta_blocks; + ei->i_reserved_meta_blocks = 0; } /* update fs dirty blocks counter */ -- cgit v1.2.1 From 9d0be50230b333005635967f7ecd4897dbfd181b Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 1 Jan 2010 02:41:30 -0500 Subject: ext4: Calculate metadata requirements more accurately In the past, ext4_calc_metadata_amount(), and its sub-functions ext4_ext_calc_metadata_amount() and ext4_indirect_calc_metadata_amount() badly over-estimated the number of metadata blocks that might be required for delayed allocation blocks. This didn't matter as much when functions which managed the reserved metadata blocks were more aggressive about dropping reserved metadata blocks as delayed allocation blocks were written, but unfortunately they were too aggressive. This was fixed in commit 0637c6f, but as a result the over-estimation by ext4_calc_metadata_amount() would lead to reserving 2-3 times the number of pending delayed allocation blocks as potentially required metadata blocks. So if there are 1 megabytes of blocks which have been not yet been allocation, up to 3 megabytes of space would get reserved out of the user's quota and from the file system free space pool until all of the inode's data blocks have been allocated. This commit addresses this problem by much more accurately estimating the number of metadata blocks that will be required. It will still somewhat over-estimate the number of blocks needed, since it must make a worst case estimate not knowing which physical blocks will be needed, but it is much more accurate than before. Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 62 +++++++++++++++++++++++++++++++++------------------------ 1 file changed, 36 insertions(+), 26 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index bdaa92a29e0e..c818972c8302 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1009,38 +1009,44 @@ qsize_t *ext4_get_reserved_space(struct inode *inode) return &EXT4_I(inode)->i_reserved_quota; } #endif + /* * Calculate the number of metadata blocks need to reserve - * to allocate @blocks for non extent file based file + * to allocate a new block at @lblocks for non extent file based file */ -static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) +static int ext4_indirect_calc_metadata_amount(struct inode *inode, + sector_t lblock) { - int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); - int ind_blks, dind_blks, tind_blks; - - /* number of new indirect blocks needed */ - ind_blks = (blocks + icap - 1) / icap; + struct ext4_inode_info *ei = EXT4_I(inode); + int dind_mask = EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1; + int blk_bits; - dind_blks = (ind_blks + icap - 1) / icap; + if (lblock < EXT4_NDIR_BLOCKS) + return 0; - tind_blks = 1; + lblock -= EXT4_NDIR_BLOCKS; - return ind_blks + dind_blks + tind_blks; + if (ei->i_da_metadata_calc_len && + (lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) { + ei->i_da_metadata_calc_len++; + return 0; + } + ei->i_da_metadata_calc_last_lblock = lblock & dind_mask; + ei->i_da_metadata_calc_len = 1; + blk_bits = roundup_pow_of_two(lblock + 1); + return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1; } /* * Calculate the number of metadata blocks need to reserve - * to allocate given number of blocks + * to allocate a block located at @lblock */ -static int ext4_calc_metadata_amount(struct inode *inode, int blocks) +static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock) { - if (!blocks) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_calc_metadata_amount(inode, blocks); + return ext4_ext_calc_metadata_amount(inode, lblock); - return ext4_indirect_calc_metadata_amount(inode, blocks); + return ext4_indirect_calc_metadata_amount(inode, lblock); } /* @@ -1078,6 +1084,7 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) */ mdb_free = ei->i_reserved_meta_blocks; ei->i_reserved_meta_blocks = 0; + ei->i_da_metadata_calc_len = 0; percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); } spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); @@ -1802,12 +1809,15 @@ static int ext4_journalled_write_end(struct file *file, return ret ? ret : copied; } -static int ext4_da_reserve_space(struct inode *inode, int nrblocks) +/* + * Reserve a single block located at lblock + */ +static int ext4_da_reserve_space(struct inode *inode, sector_t lblock) { int retries = 0; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); struct ext4_inode_info *ei = EXT4_I(inode); - unsigned long md_needed, md_reserved, total = 0; + unsigned long md_needed, md_reserved; /* * recalculate the amount of metadata blocks to reserve @@ -1817,8 +1827,7 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) repeat: spin_lock(&ei->i_block_reservation_lock); md_reserved = ei->i_reserved_meta_blocks; - md_needed = ext4_calc_metadata_amount(inode, nrblocks); - total = md_needed + nrblocks; + md_needed = ext4_calc_metadata_amount(inode, lblock); spin_unlock(&ei->i_block_reservation_lock); /* @@ -1826,7 +1835,7 @@ repeat: * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, total)) { + if (vfs_dq_reserve_block(inode, md_needed + 1)) { /* * We tend to badly over-estimate the amount of * metadata blocks which are needed, so if we have @@ -1838,8 +1847,8 @@ repeat: return -EDQUOT; } - if (ext4_claim_free_blocks(sbi, total)) { - vfs_dq_release_reservation_block(inode, total); + if (ext4_claim_free_blocks(sbi, md_needed + 1)) { + vfs_dq_release_reservation_block(inode, md_needed + 1); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { retry: if (md_reserved) @@ -1850,7 +1859,7 @@ repeat: return -ENOSPC; } spin_lock(&ei->i_block_reservation_lock); - ei->i_reserved_data_blocks += nrblocks; + ei->i_reserved_data_blocks++; ei->i_reserved_meta_blocks += md_needed; spin_unlock(&ei->i_block_reservation_lock); @@ -1891,6 +1900,7 @@ static void ext4_da_release_space(struct inode *inode, int to_free) */ to_free += ei->i_reserved_meta_blocks; ei->i_reserved_meta_blocks = 0; + ei->i_da_metadata_calc_len = 0; } /* update fs dirty blocks counter */ @@ -2504,7 +2514,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, * XXX: __block_prepare_write() unmaps passed block, * is it OK? */ - ret = ext4_da_reserve_space(inode, 1); + ret = ext4_da_reserve_space(inode, iblock); if (ret) /* not enough space to reserve */ return ret; -- cgit v1.2.1 From 1db913823c0f8360fccbd24ca67eb073966a5ffd Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 22 Jan 2010 17:06:20 -0500 Subject: ext4: Handle -EDQUOT error on write We need to release the journal before we do a write_inode. Otherwise we could deadlock. Signed-off-by: Aneesh Kumar K.V --- fs/ext4/inode.c | 32 ++++++++++++++++++-------------- 1 file changed, 18 insertions(+), 14 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c818972c8302..60b3a19e9927 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1835,24 +1835,12 @@ repeat: * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, md_needed + 1)) { - /* - * We tend to badly over-estimate the amount of - * metadata blocks which are needed, so if we have - * reserved any metadata blocks, try to force out the - * inode and see if we have any better luck. - */ - if (md_reserved && retries++ <= 3) - goto retry; + if (vfs_dq_reserve_block(inode, md_needed + 1)) return -EDQUOT; - } if (ext4_claim_free_blocks(sbi, md_needed + 1)) { vfs_dq_release_reservation_block(inode, md_needed + 1); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { - retry: - if (md_reserved) - write_inode_now(inode, (retries == 3)); yield(); goto repeat; } @@ -3032,7 +3020,7 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { - int ret, retries = 0; + int ret, retries = 0, quota_retries = 0; struct page *page; pgoff_t index; unsigned from, to; @@ -3091,6 +3079,22 @@ retry: if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; + + if ((ret == -EDQUOT) && + EXT4_I(inode)->i_reserved_meta_blocks && + (quota_retries++ < 3)) { + /* + * Since we often over-estimate the number of meta + * data blocks required, we may sometimes get a + * spurios out of quota error even though there would + * be enough space once we write the data blocks and + * find out how many meta data blocks were _really_ + * required. So try forcing the inode write to see if + * that helps. + */ + write_inode_now(inode, (quota_retries == 3)); + goto retry; + } out: return ret; } -- cgit v1.2.1 From 5f634d064c709ea02c3cdaa850a08323a4a4bf28 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 25 Jan 2010 04:00:31 -0500 Subject: ext4: Fix quota accounting error with fallocate When we fallocate a region of the file which we had recently written, and which is still in the page cache marked as delayed allocated blocks we need to make sure we don't do the quota update on writepage path. This is because the needed quota updated would have already be done by fallocate. Signed-off-by: Aneesh Kumar K.V --- fs/ext4/inode.c | 44 +++++++++++++++++++++++++++++++------------- 1 file changed, 31 insertions(+), 13 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 60b3a19e9927..c955f6490b78 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1053,11 +1053,12 @@ static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock) * Called with i_data_sem down, which is important since we can call * ext4_discard_preallocations() from here. */ -static void ext4_da_update_reserve_space(struct inode *inode, int used) +void ext4_da_update_reserve_space(struct inode *inode, + int used, int quota_claim) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); struct ext4_inode_info *ei = EXT4_I(inode); - int mdb_free = 0; + int mdb_free = 0, allocated_meta_blocks = 0; spin_lock(&ei->i_block_reservation_lock); if (unlikely(used > ei->i_reserved_data_blocks)) { @@ -1073,6 +1074,7 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) ei->i_reserved_data_blocks -= used; used += ei->i_allocated_meta_blocks; ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks; + allocated_meta_blocks = ei->i_allocated_meta_blocks; ei->i_allocated_meta_blocks = 0; percpu_counter_sub(&sbi->s_dirtyblocks_counter, used); @@ -1090,9 +1092,23 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); /* Update quota subsystem */ - vfs_dq_claim_block(inode, used); - if (mdb_free) - vfs_dq_release_reservation_block(inode, mdb_free); + if (quota_claim) { + vfs_dq_claim_block(inode, used); + if (mdb_free) + vfs_dq_release_reservation_block(inode, mdb_free); + } else { + /* + * We did fallocate with an offset that is already delayed + * allocated. So on delayed allocated writeback we should + * not update the quota for allocated blocks. But then + * converting an fallocate region to initialized region would + * have caused a metadata allocation. So claim quota for + * that + */ + if (allocated_meta_blocks) + vfs_dq_claim_block(inode, allocated_meta_blocks); + vfs_dq_release_reservation_block(inode, mdb_free + used); + } /* * If we have done all the pending block allocations and if @@ -1292,18 +1308,20 @@ int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block, */ EXT4_I(inode)->i_state &= ~EXT4_STATE_EXT_MIGRATE; } - } + /* + * Update reserved blocks/metadata blocks after successful + * block allocation which had been deferred till now. We don't + * support fallocate for non extent files. So we can update + * reserve space here. + */ + if ((retval > 0) && + (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE)) + ext4_da_update_reserve_space(inode, retval, 1); + } if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) EXT4_I(inode)->i_delalloc_reserved_flag = 0; - /* - * Update reserved blocks/metadata blocks after successful - * block allocation which had been deferred till now. - */ - if ((retval > 0) && (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE)) - ext4_da_update_reserve_space(inode, retval); - up_write((&EXT4_I(inode)->i_data_sem)); if (retval > 0 && buffer_mapped(bh)) { int ret = check_block_validity(inode, "file system " -- cgit v1.2.1 From 1296cc85c26e94eb865d03f82140f27d598de467 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 15 Jan 2010 01:27:59 -0500 Subject: ext4: Drop EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE flag We should update reserve space if it is delalloc buffer and that is indicated by EXT4_GET_BLOCKS_DELALLOC_RESERVE flag. So use EXT4_GET_BLOCKS_DELALLOC_RESERVE in place of EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE Signed-off-by: Aneesh Kumar K.V --- fs/ext4/inode.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c955f6490b78..e11952404e02 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1316,7 +1316,7 @@ int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block, * reserve space here. */ if ((retval > 0) && - (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE)) + (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)) ext4_da_update_reserve_space(inode, retval, 1); } if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) @@ -2219,10 +2219,10 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd) * variables are updated after the blocks have been allocated. */ new.b_state = 0; - get_blocks_flags = (EXT4_GET_BLOCKS_CREATE | - EXT4_GET_BLOCKS_DELALLOC_RESERVE); + get_blocks_flags = EXT4_GET_BLOCKS_CREATE; if (mpd->b_state & (1 << BH_Delay)) - get_blocks_flags |= EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE; + get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE; + blks = ext4_get_blocks(handle, mpd->inode, next, max_blocks, &new, get_blocks_flags); if (blks < 0) { -- cgit v1.2.1