From 729f52c6be51013c9268e5fc85acbc1091286fdb Mon Sep 17 00:00:00 2001 From: Zheng Liu Date: Mon, 9 Jul 2012 16:29:29 -0400 Subject: ext4: add a new nolock flag in ext4_map_blocks EXT4_GET_BLOCKS_NO_LOCK flag is added to indicate that we don't need to acquire i_data_sem lock in ext4_map_blocks. Meanwhile, it changes ext4_get_block() to not start a new journal because when we do a overwrite dio, there is no any metadata that needs to be modified. We define a new function called ext4_get_block_write_nolock, which is used in dio overwrite nolock. In this function, it doesn't try to acquire i_data_sem lock and doesn't start a new journal as it does a lookup. CC: Tao Ma CC: Eric Sandeen CC: Robin Dong Signed-off-by: Zheng Liu Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 49 insertions(+), 10 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 02bc8cbe728..76cb3b1ad78 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -544,7 +544,8 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, * Try to see if we can get the block without requesting a new * file system block. */ - down_read((&EXT4_I(inode)->i_data_sem)); + if (!(flags & EXT4_GET_BLOCKS_NO_LOCK)) + down_read((&EXT4_I(inode)->i_data_sem)); if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { retval = ext4_ext_map_blocks(handle, inode, map, flags & EXT4_GET_BLOCKS_KEEP_SIZE); @@ -552,7 +553,8 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, retval = ext4_ind_map_blocks(handle, inode, map, flags & EXT4_GET_BLOCKS_KEEP_SIZE); } - up_read((&EXT4_I(inode)->i_data_sem)); + if (!(flags & EXT4_GET_BLOCKS_NO_LOCK)) + up_read((&EXT4_I(inode)->i_data_sem)); if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) { int ret = check_block_validity(inode, map); @@ -2818,6 +2820,32 @@ static int ext4_get_block_write(struct inode *inode, sector_t iblock, EXT4_GET_BLOCKS_IO_CREATE_EXT); } +static int ext4_get_block_write_nolock(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int flags) +{ + handle_t *handle = ext4_journal_current_handle(); + struct ext4_map_blocks map; + int ret = 0; + + ext4_debug("ext4_get_block_write_nolock: inode %lu, flag %d\n", + inode->i_ino, flags); + + flags = EXT4_GET_BLOCKS_NO_LOCK; + + map.m_lblk = iblock; + map.m_len = bh_result->b_size >> inode->i_blkbits; + + ret = ext4_map_blocks(handle, inode, &map, flags); + if (ret > 0) { + map_bh(bh_result, inode->i_sb, map.m_pblk); + bh_result->b_state = (bh_result->b_state & ~EXT4_MAP_FLAGS) | + map.m_flags; + bh_result->b_size = inode->i_sb->s_blocksize * map.m_len; + ret = 0; + } + return ret; +} + static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, ssize_t size, void *private, int ret, bool is_async) @@ -2966,6 +2994,8 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, loff_t final_size = offset + count; if (rw == WRITE && final_size <= inode->i_size) { + int overwrite = 0; + /* * We could direct write to holes and fallocate. * @@ -3005,13 +3035,22 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, EXT4_I(inode)->cur_aio_dio = iocb->private; } - ret = __blockdev_direct_IO(rw, iocb, inode, - inode->i_sb->s_bdev, iov, - offset, nr_segs, - ext4_get_block_write, - ext4_end_io_dio, - NULL, - DIO_LOCKING); + if (overwrite) + ret = __blockdev_direct_IO(rw, iocb, inode, + inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block_write_nolock, + ext4_end_io_dio, + NULL, + 0); + else + ret = __blockdev_direct_IO(rw, iocb, inode, + inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block_write, + ext4_end_io_dio, + NULL, + DIO_LOCKING); if (iocb->private) EXT4_I(inode)->cur_aio_dio = NULL; /* @@ -3031,7 +3070,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, if (ret != -EIOCBQUEUED && ret <= 0 && iocb->private) { ext4_free_io_end(iocb->private); iocb->private = NULL; - } else if (ret > 0 && ext4_test_inode_state(inode, + } else if (ret > 0 && !overwrite && ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN)) { int err; /* -- cgit v1.2.3-70-g09d2 From 4bd809dbbf177ad0c450d702466b1da63e1b4b7e Mon Sep 17 00:00:00 2001 From: Zheng Liu Date: Sun, 22 Jul 2012 20:19:31 -0400 Subject: ext4: don't take the i_mutex lock when doing DIO overwrites Aligned and overwrite direct I/O can be parallelized. In ext4_file_dio_write, we first check whether these conditions are satisfied or not. If so, we take i_data_sem and release i_mutex lock directly. Meanwhile iocb->private is set to indicate that this is a dio overwrite, and it will be handled in ext4_ext_direct_IO. [ Added fix from Dan Carpenter to fix locking bug on the error path. ] CC: Tao Ma CC: Eric Sandeen CC: Robin Dong Signed-off-by: Zheng Liu Signed-off-by: "Theodore Ts'o" Signed-off-by: Dan Carpenter --- fs/ext4/file.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++-- fs/ext4/inode.c | 24 ++++++++++++++++++++++-- 2 files changed, 71 insertions(+), 4 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/file.c b/fs/ext4/file.c index a10dc7742ae..1c81509f5bd 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -93,9 +93,13 @@ static ssize_t ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { - struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode; + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + struct blk_plug plug; int unaligned_aio = 0; ssize_t ret; + int overwrite = 0; + size_t length = iov_length(iov, nr_segs); if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) && !is_sync_kiocb(iocb)) @@ -115,7 +119,50 @@ ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, ext4_aiodio_wait(inode); } - ret = generic_file_aio_write(iocb, iov, nr_segs, pos); + BUG_ON(iocb->ki_pos != pos); + + mutex_lock(&inode->i_mutex); + blk_start_plug(&plug); + + iocb->private = &overwrite; + + /* check whether we do a DIO overwrite or not */ + if (ext4_should_dioread_nolock(inode) && !unaligned_aio && + !file->f_mapping->nrpages && pos + length <= i_size_read(inode)) { + struct ext4_map_blocks map; + unsigned int blkbits = inode->i_blkbits; + int err, len; + + map.m_lblk = pos >> blkbits; + map.m_len = (EXT4_BLOCK_ALIGN(pos + length, blkbits) >> blkbits) + - map.m_lblk; + len = map.m_len; + + err = ext4_map_blocks(NULL, inode, &map, 0); + /* + * 'err==len' means that all of blocks has been preallocated no + * matter they are initialized or not. For excluding + * uninitialized extents, we need to check m_flags. There are + * two conditions that indicate for initialized extents. + * 1) If we hit extent cache, EXT4_MAP_MAPPED flag is returned; + * 2) If we do a real lookup, non-flags are returned. + * So we should check these two conditions. + */ + if (err == len && (map.m_flags & EXT4_MAP_MAPPED)) + overwrite = 1; + } + + ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); + mutex_unlock(&inode->i_mutex); + + if (ret > 0 || ret == -EIOCBQUEUED) { + ssize_t err; + + err = generic_write_sync(file, pos, ret); + if (err < 0 && ret > 0) + ret = err; + } + blk_finish_plug(&plug); if (unaligned_aio) mutex_unlock(ext4_aio_mutex(inode)); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 76cb3b1ad78..bed574dd4c2 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2996,6 +2996,16 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, if (rw == WRITE && final_size <= inode->i_size) { int overwrite = 0; + BUG_ON(iocb->private == NULL); + + /* If we do a overwrite dio, i_mutex locking can be released */ + overwrite = *((int *)iocb->private); + + if (overwrite) { + down_read(&EXT4_I(inode)->i_data_sem); + mutex_unlock(&inode->i_mutex); + } + /* * We could direct write to holes and fallocate. * @@ -3021,8 +3031,10 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, if (!is_sync_kiocb(iocb)) { ext4_io_end_t *io_end = ext4_init_io_end(inode, GFP_NOFS); - if (!io_end) - return -ENOMEM; + if (!io_end) { + ret = -ENOMEM; + goto retake_lock; + } io_end->flag |= EXT4_IO_END_DIRECT; iocb->private = io_end; /* @@ -3083,6 +3095,14 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, ret = err; ext4_clear_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN); } + + retake_lock: + /* take i_mutex locking again if we do a ovewrite dio */ + if (overwrite) { + up_read(&EXT4_I(inode)->i_data_sem); + mutex_lock(&inode->i_mutex); + } + return ret; } -- cgit v1.2.3-70-g09d2 From b50924c2c606eccfe0caef39beb0929dfa9a1a81 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Sun, 22 Jul 2012 20:37:31 -0400 Subject: ext4: remove unnecessary argument from __ext4_handle_dirty_metadata() The '__ext4_handle_dirty_metadata()' does not need the 'now' argument anymore and we can kill it. Signed-off-by: Artem Bityutskiy Signed-off-by: "Theodore Ts'o" Reviewed-by: Jan Kara --- fs/ext4/ext4_jbd2.c | 3 +-- fs/ext4/ext4_jbd2.h | 7 ++----- fs/ext4/inode.c | 2 +- fs/ext4/namei.c | 4 ++-- fs/ext4/resize.c | 2 +- 5 files changed, 7 insertions(+), 11 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index c19ab6addb2..bfa65b49d42 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -138,8 +138,7 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line, } int __ext4_handle_dirty_super(const char *where, unsigned int line, - handle_t *handle, struct super_block *sb, - int now) + handle_t *handle, struct super_block *sb) { struct buffer_head *bh = EXT4_SB(sb)->s_sbh; int err = 0; diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index 1393c830411..56d258c1830 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -219,8 +219,7 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line, struct buffer_head *bh); int __ext4_handle_dirty_super(const char *where, unsigned int line, - handle_t *handle, struct super_block *sb, - int now); + handle_t *handle, struct super_block *sb); #define ext4_journal_get_write_access(handle, bh) \ __ext4_journal_get_write_access(__func__, __LINE__, (handle), (bh)) @@ -232,10 +231,8 @@ int __ext4_handle_dirty_super(const char *where, unsigned int line, #define ext4_handle_dirty_metadata(handle, inode, bh) \ __ext4_handle_dirty_metadata(__func__, __LINE__, (handle), (inode), \ (bh)) -#define ext4_handle_dirty_super_now(handle, sb) \ - __ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb), 1) #define ext4_handle_dirty_super(handle, sb) \ - __ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb), 0) + __ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb)) handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index bed574dd4c2..a533a18de98 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4093,7 +4093,7 @@ static int ext4_do_update_inode(handle_t *handle, EXT4_SET_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_LARGE_FILE); ext4_handle_sync(handle); - err = ext4_handle_dirty_super_now(handle, sb); + err = ext4_handle_dirty_super(handle, sb); } } raw_inode->i_generation = cpu_to_le32(inode->i_generation); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 0edaf18d843..37faf56e558 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2397,7 +2397,7 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode) /* Insert this inode at the head of the on-disk orphan list... */ NEXT_ORPHAN(inode) = le32_to_cpu(EXT4_SB(sb)->s_es->s_last_orphan); EXT4_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); - err = ext4_handle_dirty_super_now(handle, sb); + err = ext4_handle_dirty_super(handle, sb); rc = ext4_mark_iloc_dirty(handle, inode, &iloc); if (!err) err = rc; @@ -2470,7 +2470,7 @@ int ext4_orphan_del(handle_t *handle, struct inode *inode) if (err) goto out_brelse; sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); - err = ext4_handle_dirty_super_now(handle, inode->i_sb); + err = ext4_handle_dirty_super(handle, inode->i_sb); } else { struct ext4_iloc iloc2; struct inode *i_prev = diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 50530bdbc02..41f6ef68e2e 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -798,7 +798,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, ext4_kvfree(o_group_desc); le16_add_cpu(&es->s_reserved_gdt_blocks, -1); - err = ext4_handle_dirty_super_now(handle, sb); + err = ext4_handle_dirty_super(handle, sb); if (err) ext4_std_error(sb, err); -- cgit v1.2.3-70-g09d2 From 97795d2a5b8d3c8dc4365d4bd3404191840453ba Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Sun, 22 Jul 2012 23:59:40 -0400 Subject: ext4: don't let i_reserved_meta_blocks go negative If we hit a condition where we have allocated metadata blocks that were not appropriately reserved, we risk underflow of ei->i_reserved_meta_blocks. In turn, this can throw sbi->s_dirtyclusters_counter significantly out of whack and undermine the nondelalloc fallback logic in ext4_nonda_switch(). Warn if this occurs and set i_allocated_meta_blocks to avoid this problem. This condition is reproduced by xfstests 270 against ext2 with delalloc enabled: Mar 28 08:58:02 localhost kernel: [ 171.526344] EXT4-fs (loop1): delayed block allocation failed for inode 14 at logical offset 64486 with max blocks 64 with error -28 Mar 28 08:58:02 localhost kernel: [ 171.526346] EXT4-fs (loop1): This should not happen!! Data will be lost 270 ultimately fails with an inconsistent filesystem and requires an fsck to repair. The cause of the error is an underflow in ext4_da_update_reserve_space() due to an unreserved meta block allocation. Signed-off-by: Brian Foster Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- fs/ext4/inode.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a533a18de98..25f809dc45a 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -346,6 +346,15 @@ void ext4_da_update_reserve_space(struct inode *inode, used = ei->i_reserved_data_blocks; } + if (unlikely(ei->i_allocated_meta_blocks > ei->i_reserved_meta_blocks)) { + ext4_msg(inode->i_sb, KERN_NOTICE, "%s: ino %lu, allocated %d " + "with only %d reserved metadata blocks\n", __func__, + inode->i_ino, ei->i_allocated_meta_blocks, + ei->i_reserved_meta_blocks); + WARN_ON(1); + ei->i_allocated_meta_blocks = ei->i_reserved_meta_blocks; + } + /* Update per-inode reservations */ ei->i_reserved_data_blocks -= used; ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks; -- cgit v1.2.3-70-g09d2 From 03179fe92318e7934c180d96f12eff2cb36ef7b6 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 23 Jul 2012 00:00:20 -0400 Subject: ext4: undo ext4_calc_metadata_amount if we fail to claim space The function ext4_calc_metadata_amount() has side effects, although it's not obvious from its function name. So if we fail to claim space, regardless of whether we retry to claim the space again, or return an error, we need to undo these side effects. Otherwise we can end up incorrectly calculating the number of metadata blocks needed for the operation, which was responsible for an xfstests failure for test #271 when using an ext2 file system with delalloc enabled. Reported-by: Brian Foster Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- fs/ext4/inode.c | 32 +++++++++++++++++++++----------- 1 file changed, 21 insertions(+), 11 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 25f809dc45a..89b59cb7f9b 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1182,6 +1182,17 @@ static int ext4_da_reserve_space(struct inode *inode, ext4_lblk_t lblock) struct ext4_inode_info *ei = EXT4_I(inode); unsigned int md_needed; int ret; + ext4_lblk_t save_last_lblock; + int save_len; + + /* + * We will charge metadata quota at writeout time; this saves + * us from metadata over-estimation, though we may go over by + * a small amount in the end. Here we just reserve for data. + */ + ret = dquot_reserve_block(inode, EXT4_C2B(sbi, 1)); + if (ret) + return ret; /* * recalculate the amount of metadata blocks to reserve @@ -1190,32 +1201,31 @@ static int ext4_da_reserve_space(struct inode *inode, ext4_lblk_t lblock) */ repeat: spin_lock(&ei->i_block_reservation_lock); + /* + * ext4_calc_metadata_amount() has side effects, which we have + * to be prepared undo if we fail to claim space. + */ + save_len = ei->i_da_metadata_calc_len; + save_last_lblock = ei->i_da_metadata_calc_last_lblock; md_needed = EXT4_NUM_B2C(sbi, ext4_calc_metadata_amount(inode, lblock)); trace_ext4_da_reserve_space(inode, md_needed); - spin_unlock(&ei->i_block_reservation_lock); - /* - * We will charge metadata quota at writeout time; this saves - * us from metadata over-estimation, though we may go over by - * a small amount in the end. Here we just reserve for data. - */ - ret = dquot_reserve_block(inode, EXT4_C2B(sbi, 1)); - if (ret) - return ret; /* * We do still charge estimated metadata to the sb though; * we cannot afford to run out of free blocks. */ if (ext4_claim_free_clusters(sbi, md_needed + 1, 0)) { - dquot_release_reservation_block(inode, EXT4_C2B(sbi, 1)); + ei->i_da_metadata_calc_len = save_len; + ei->i_da_metadata_calc_last_lblock = save_last_lblock; + spin_unlock(&ei->i_block_reservation_lock); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { yield(); goto repeat; } + dquot_release_reservation_block(inode, EXT4_C2B(sbi, 1)); return -ENOSPC; } - spin_lock(&ei->i_block_reservation_lock); ei->i_reserved_data_blocks++; ei->i_reserved_meta_blocks += md_needed; spin_unlock(&ei->i_block_reservation_lock); -- cgit v1.2.3-70-g09d2