From 149feb00d7c3f4f06896b245533c957b54e3e052 Mon Sep 17 00:00:00 2001 From: Huang Weiyi Date: Mon, 14 Dec 2009 09:24:20 -0500 Subject: ext4: remove unused #include Remove unused #include ('s) in fs/ext4/block_validity.c fs/ext4/mballoc.h Signed-off-by: Huang Weiyi Signed-off-by: "Theodore Ts'o" --- fs/ext4/block_validity.c | 1 - fs/ext4/mballoc.h | 1 - 2 files changed, 2 deletions(-) diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c index 4df8621ec31c..a60ab9aad57d 100644 --- a/fs/ext4/block_validity.c +++ b/fs/ext4/block_validity.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include #include "ext4.h" diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 0ca811061bc7..436521cae456 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -17,7 +17,6 @@ #include #include #include -#include #include #include #include "ext4_jbd2.h" -- cgit v1.2.1 From 84c664730374248adaf420c0846a6158d64413c7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 21 Dec 2009 10:54:09 -0500 Subject: ext4: Don't ask about supporting ext2/3 in ext4 if ext4 is not configured Don't offer to build ext2/3 support into ext4 if ext4 itself is not configured on. Signed-off-by: David Howells Signed-off-by: "Theodore Ts'o" --- fs/ext4/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig index 9acf7e808139..9ed1bb1f319f 100644 --- a/fs/ext4/Kconfig +++ b/fs/ext4/Kconfig @@ -28,6 +28,7 @@ config EXT4_FS config EXT4_USE_FOR_EXT23 bool "Use ext4 for ext2/ext3 file systems" + depends on EXT4_FS depends on EXT3_FS=n || EXT2_FS=n default y help -- cgit v1.2.1 From 51b7e3c9fbe7d22d4e355101e9a73b44fc5c9feb Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 21 Dec 2009 10:56:09 -0500 Subject: ext4: add module aliases for ext2 and ext3 Add module aliases for ext2 and ext3 when CONFIG_EXT4_USE_FOR_EXT23 is set. This makes the existing user-space stuff like mkinitrd working as is. Signed-off-by: Takashi Iwai Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 6ed9aa91f27d..0a06fe6f2bc2 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4005,6 +4005,7 @@ static inline void unregister_as_ext2(void) { unregister_filesystem(&ext2_fs_type); } +MODULE_ALIAS("ext2"); #else static inline void register_as_ext2(void) { } static inline void unregister_as_ext2(void) { } @@ -4031,6 +4032,7 @@ static inline void unregister_as_ext3(void) { unregister_filesystem(&ext3_fs_type); } +MODULE_ALIAS("ext3"); #else static inline void register_as_ext3(void) { } static inline void unregister_as_ext3(void) { } -- cgit v1.2.1 From 034fb4c95fc0fed4ec4a50778127b92c6f2aec01 Mon Sep 17 00:00:00 2001 From: Surbhi Palande Date: Mon, 14 Dec 2009 09:53:52 -0500 Subject: ext4: replace BUG() with return -EIO in ext4_ext_get_blocks This patch fixes the Kernel BZ #14286. When the address of an extent corresponding to a valid block is corrupted, a -EIO should be reported instead of a BUG(). This situation should not normally not occur except in the case of a corrupted filesystem. If however it does, then the system should not panic directly but depending on the mount time options appropriate action should be taken. If the mount options so permit, the I/O should be gracefully aborted by returning a -EIO. http://bugzilla.kernel.org/show_bug.cgi?id=14286 Signed-off-by: Surbhi Palande Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 3a7928f825e4..8fd6c567964a 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3190,7 +3190,13 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, * this situation is possible, though, _during_ tree modification; * this is why assert can't be put in ext4_ext_find_extent() */ - BUG_ON(path[depth].p_ext == NULL && depth != 0); + if (path[depth].p_ext == NULL && depth != 0) { + ext4_error(inode->i_sb, __func__, "bad extent address " + "inode: %lu, iblock: %d, depth: %d", + inode->i_ino, iblock, depth); + err = -EIO; + goto out2; + } eh = path[depth].p_hdr; ex = path[depth].p_ext; -- cgit v1.2.1 From cc3e1bea5d87635c519da657303690f5538bb4eb Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 23 Dec 2009 06:52:08 -0500 Subject: ext4, jbd2: Add barriers for file systems with exernal journals This is a bit complicated because we are trying to optimize when we send barriers to the fs data disk. We could just throw in an extra barrier to the data disk whenever we send a barrier to the journal disk, but that's not always strictly necessary. We only need to send a barrier during a commit when there are data blocks which are must be written out due to an inode written in ordered mode, or if fsync() depends on the commit to force data blocks to disk. Finally, before we drop transactions from the beginning of the journal during a checkpoint operation, we need to guarantee that any blocks that were flushed out to the data disk are firmly on the rust platter before we drop the transaction from the journal. Thanks to Oleg Drokin for pointing out this flaw in ext3/ext4. Signed-off-by: "Theodore Ts'o" --- fs/ext4/fsync.c | 16 ++++++++++++++-- fs/jbd2/checkpoint.c | 15 +++++++++++++++ fs/jbd2/commit.c | 19 +++++++++++-------- include/linux/jbd2.h | 1 + 4 files changed, 41 insertions(+), 10 deletions(-) diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 0b22497d92e1..98bd140aad01 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -88,9 +88,21 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) return ext4_force_commit(inode->i_sb); commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid; - if (jbd2_log_start_commit(journal, commit_tid)) + if (jbd2_log_start_commit(journal, commit_tid)) { + /* + * When the journal is on a different device than the + * fs data disk, we need to issue the barrier in + * writeback mode. (In ordered mode, the jbd2 layer + * will take care of issuing the barrier. In + * data=journal, all of the data blocks are written to + * the journal device.) + */ + if (ext4_should_writeback_data(inode) && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(inode->i_sb->s_bdev, NULL); jbd2_log_wait_commit(journal, commit_tid); - else if (journal->j_flags & JBD2_BARRIER) + } else if (journal->j_flags & JBD2_BARRIER) blkdev_issue_flush(inode->i_sb->s_bdev, NULL); return ret; } diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index ca0f5eb62b20..886849370950 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -22,6 +22,7 @@ #include #include #include +#include #include /* @@ -515,6 +516,20 @@ int jbd2_cleanup_journal_tail(journal_t *journal) journal->j_tail_sequence = first_tid; journal->j_tail = blocknr; spin_unlock(&journal->j_state_lock); + + /* + * If there is an external journal, we need to make sure that + * any data blocks that were recently written out --- perhaps + * by jbd2_log_do_checkpoint() --- are flushed out before we + * drop the transactions from the external journal. It's + * unlikely this will be necessary, especially with a + * appropriately sized journal, but we need this to guarantee + * correctness. Fortunately jbd2_cleanup_journal_tail() + * doesn't get called all that often. + */ + if ((journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, NULL); if (!(journal->j_flags & JBD2_ABORT)) jbd2_journal_update_superblock(journal, 1); return 0; diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6a10238d2c63..1bc74b6f26d2 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -259,6 +259,7 @@ static int journal_submit_data_buffers(journal_t *journal, ret = err; spin_lock(&journal->j_list_lock); J_ASSERT(jinode->i_transaction == commit_transaction); + commit_transaction->t_flushed_data_blocks = 1; jinode->i_flags &= ~JI_COMMIT_RUNNING; wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); } @@ -708,8 +709,17 @@ start_journal_io: } } - /* Done it all: now write the commit record asynchronously. */ + /* + * If the journal is not located on the file system device, + * then we must flush the file system device before we issue + * the commit record + */ + if (commit_transaction->t_flushed_data_blocks && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, NULL); + /* Done it all: now write the commit record asynchronously. */ if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { err = journal_submit_commit_record(journal, commit_transaction, @@ -720,13 +730,6 @@ start_journal_io: blkdev_issue_flush(journal->j_dev, NULL); } - /* - * This is the right place to wait for data buffers both for ASYNC - * and !ASYNC commit. If commit is ASYNC, we need to wait only after - * the commit block went to disk (which happens above). If commit is - * SYNC, we need to wait for data buffers before we start writing - * commit block, which happens below in such setting. - */ err = journal_finish_inode_data_buffers(journal, commit_transaction); if (err) { printk(KERN_WARNING diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index f1011f7f3d41..638ce4554c76 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -653,6 +653,7 @@ struct transaction_s * waiting for it to finish. */ unsigned int t_synchronous_commit:1; + unsigned int t_flushed_data_blocks:1; /* * For use by the filesystem to store fs-specific data -- cgit v1.2.1 From a6b43e382568a49cc64291bfaddf896652d44f70 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 23 Dec 2009 07:48:08 -0500 Subject: ext4: fix unsigned long long printk warning in super.c sparc64 allmodconfig: fs/ext4/super.c: In function `lifetime_write_kbytes_show': fs/ext4/super.c:2174: warning: long long unsigned int format, long unsigned int arg (arg 4) fs/ext4/super.c:2174: warning: long long unsigned int format, long unsigned int arg (arg 4) Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 0a06fe6f2bc2..7cccb35c0f4d 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2174,9 +2174,9 @@ static ssize_t lifetime_write_kbytes_show(struct ext4_attr *a, struct super_block *sb = sbi->s_buddy_cache->i_sb; return snprintf(buf, PAGE_SIZE, "%llu\n", - sbi->s_kbytes_written + + (unsigned long long)(sbi->s_kbytes_written + ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) - - EXT4_SB(sb)->s_sectors_written_start) >> 1)); + EXT4_SB(sb)->s_sectors_written_start) >> 1))); } static ssize_t inode_readahead_blks_store(struct ext4_attr *a, -- cgit v1.2.1 From d3533d72e7478a61a3e1936956fc825289a2acf4 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Wed, 23 Dec 2009 07:52:31 -0500 Subject: ext4: Eliminate potential double free on error path b_entry_name and buffer are initially NULL, are initialized within a loop to the result of calling kmalloc, and are freed at the bottom of this loop. The loop contains gotos to cleanup, which also frees b_entry_name and buffer. Some of these gotos are before the reinitializations of b_entry_name and buffer. To maintain the invariant that b_entry_name and buffer are NULL at the top of the loop, and thus acceptable arguments to kfree, these variables are now set to NULL after the kfrees. This seems to be the simplest solution. A more complicated solution would be to introduce more labels in the error handling code at the end of the function. A simplified version of the semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @r@ identifier E; expression E1; iterator I; statement S; @@ *kfree(E); ... when != E = E1 when != I(E,...) S when != &E *kfree(E); // Signed-off-by: Julia Lawall Signed-off-by: "Theodore Ts'o" --- fs/ext4/xattr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 83218bebbc7c..f3a2f7ed45aa 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1332,6 +1332,8 @@ retry: goto cleanup; kfree(b_entry_name); kfree(buffer); + b_entry_name = NULL; + buffer = NULL; brelse(is->iloc.bh); kfree(is); kfree(bs); -- cgit v1.2.1 From 17bd55d037a02b04d9119511cfd1a4b985d20f63 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Wed, 23 Dec 2009 07:57:07 -0500 Subject: fs-writeback: Add helper function to start writeback if idle ext4, at least, would like to start pushing on writeback if it starts to get close to ENOSPC when reserving worst-case blocks for delalloc writes. Writing out delalloc data will convert those worst-case predictions into usually smaller actual usage, freeing up space before we hit ENOSPC based on this speculation. Thanks to Jens for the suggestion for the helper function, & the naming help. I've made the helper return status on whether writeback was started even though I don't plan to use it in the ext4 patch; it seems like it would be potentially useful to test this in some cases. Signed-off-by: Eric Sandeen Acked-by: Jan Kara --- fs/fs-writeback.c | 17 +++++++++++++++++ include/linux/writeback.h | 1 + 2 files changed, 18 insertions(+) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 49bc1b8e8f19..f6c2155e0026 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1186,6 +1186,23 @@ void writeback_inodes_sb(struct super_block *sb) } EXPORT_SYMBOL(writeback_inodes_sb); +/** + * writeback_inodes_sb_if_idle - start writeback if none underway + * @sb: the superblock + * + * Invoke writeback_inodes_sb if no writeback is currently underway. + * Returns 1 if writeback was started, 0 if not. + */ +int writeback_inodes_sb_if_idle(struct super_block *sb) +{ + if (!writeback_in_progress(sb->s_bdi)) { + writeback_inodes_sb(sb); + return 1; + } else + return 0; +} +EXPORT_SYMBOL(writeback_inodes_sb_if_idle); + /** * sync_inodes_sb - sync sb inode pages * @sb: the superblock diff --git a/include/linux/writeback.h b/include/linux/writeback.h index c18c008f4bbf..76e8903cd204 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -70,6 +70,7 @@ struct writeback_control { struct bdi_writeback; int inode_wait(void *); void writeback_inodes_sb(struct super_block *); +int writeback_inodes_sb_if_idle(struct super_block *); void sync_inodes_sb(struct super_block *); void writeback_inodes_wbc(struct writeback_control *wbc); long wb_do_writeback(struct bdi_writeback *wb, int force_wait); -- 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(-) 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 3ebfdf885acedafcadceb4007e77438479094135 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 23 Dec 2009 08:05:15 -0500 Subject: jbd2: don't use __GFP_NOFAIL in journal_init_common() It triggers the warning in get_page_from_freelist(), and it isn't appropriate to use __GFP_NOFAIL here anyway. Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14843 Reported-by: Christian Casteyde Signed-off-by: Andrew Morton Signed-off-by: "Theodore Ts'o" --- fs/jbd2/journal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 17af879e6e9e..ac0d027595d0 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -814,7 +814,7 @@ static journal_t * journal_init_common (void) journal_t *journal; int err; - journal = kzalloc(sizeof(*journal), GFP_KERNEL|__GFP_NOFAIL); + journal = kzalloc(sizeof(*journal), GFP_KERNEL); if (!journal) goto fail; -- cgit v1.2.1 From 6d3b82f2d31f22085e5711b28dddcb9fb3d97a25 Mon Sep 17 00:00:00 2001 From: Fang Wenqi Date: Thu, 24 Dec 2009 17:51:42 -0500 Subject: ext4: Update documentation to correct the inode_readahead_blks option name Per commit 240799cd, the option name for readahead should be inode_readahead_blks, not inode_readahead. Signed-off-by: Fang Wenqi Signed-off-by: "Theodore Ts'o" --- Documentation/filesystems/ext4.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index af6885c3c821..e1def1786e50 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -196,7 +196,7 @@ nobarrier This also requires an IO stack which can support also be used to enable or disable barriers, for consistency with other ext4 mount options. -inode_readahead=n This tuning parameter controls the maximum +inode_readahead_blks=n This tuning parameter controls the maximum number of inode table blocks that ext4's inode table readahead algorithm will pre-read into the buffer cache. The default value is 32 blocks. -- 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(-) 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 515f41c33a9d44a964264c9511ad2c869af1fac3 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Tue, 29 Dec 2009 23:39:06 -0500 Subject: ext4: Ensure zeroout blocks have no dirty metadata This fixes a bug (found by Curt Wohlgemuth) in which new blocks returned from an extent created with ext4_ext_zeroout() can have dirty metadata still associated with them. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Curt Wohlgemuth Signed-off-by: "Theodore Ts'o" --- fs/ext4/extents.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 8fd6c567964a..91ae46098ea4 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3023,6 +3023,14 @@ out: return err; } +static void unmap_underlying_metadata_blocks(struct block_device *bdev, + sector_t block, int count) +{ + int i; + for (i = 0; i < count; i++) + unmap_underlying_metadata(bdev, block + i); +} + static int ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned int max_blocks, @@ -3098,6 +3106,18 @@ out: } else allocated = ret; set_buffer_new(bh_result); + /* + * if we allocated more blocks than requested + * we need to make sure we unmap the extra block + * allocated. The actual needed block will get + * unmapped later when we find the buffer_head marked + * new. + */ + if (allocated > max_blocks) { + unmap_underlying_metadata_blocks(inode->i_sb->s_bdev, + newblock + max_blocks, + allocated - max_blocks); + } map_out: set_buffer_mapped(bh_result); out1: -- 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(-) 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