From a1de02dccf906faba2ee2d99cac56799bda3b96a Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Thu, 4 Feb 2010 23:58:38 -0500 Subject: ext4: fix async i/o writes beyond 4GB to a sparse file The "offset" member in ext4_io_end holds bytes, not blocks, so ext4_lblk_t is wrong - and too small (u32). This caused the async i/o writes to sparse files beyond 4GB to fail when they wrapped around to 0. Also fix up the type of arguments to ext4_convert_unwritten_extents(), it gets ssize_t from ext4_end_aio_dio_nolock() and ext4_ext_direct_IO(). Reported-by: Giel de Nijs Signed-off-by: Eric Sandeen --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index e11952404e02..2059c34ac4c8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3551,7 +3551,7 @@ static int ext4_end_aio_dio_nolock(ext4_io_end_t *io) { struct inode *inode = io->inode; loff_t offset = io->offset; - size_t size = io->size; + ssize_t size = io->size; int ret = 0; ext4_debug("end_aio_dio_onlock: io 0x%p from inode %lu,list->next 0x%p," -- cgit v1.2.1 From 1f2acb6017d8528135ec3b01ab7cd2be6ea0630b Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 22 Jan 2010 17:40:42 -0500 Subject: ext4: Add block validity check when truncating indirect block mapped inodes Add checks to ext4_free_branches() to make sure a block number found in an indirect block are valid before trying to free it. If a bad block number is found, stop freeing the indirect block immediately, since the file system is corrupt and we will need to run fsck anyway. This also avoids spamming the logs, and specifically avoids driver-level "attempt to access beyond end of device" errors obscure what is really going on. If you get *really*, *really*, *really* unlucky, without this patch, a supposed indirect block containing garbage might contain a reference to a primary block group descriptor, in which case ext4_free_branches() could end up zero'ing out a block group descriptor block, and if then one of the block bitmaps for a block group described by that bg descriptor block is not in memory, and is read in by ext4_read_block_bitmap(). This function calls ext4_valid_block_bitmap(), which assumes that bg_inode_table() was validated at mount time and hasn't been modified since. Since this assumption is no longer valid, it's possible for the value (ext4_inode_table(sb, desc) - group_first_block) to go negative, which will cause ext4_find_next_zero_bit() to trigger a kernel GPF. Addresses-Google-Bug: #2220436 Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 39 ++++++++++++++++++++++++++++++--------- 1 file changed, 30 insertions(+), 9 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2059c34ac4c8..3e8afd969236 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4130,18 +4130,27 @@ no_top: * We release `count' blocks on disk, but (last - first) may be greater * than `count' because there can be holes in there. */ -static void ext4_clear_blocks(handle_t *handle, struct inode *inode, - struct buffer_head *bh, - ext4_fsblk_t block_to_free, - unsigned long count, __le32 *first, - __le32 *last) +static int ext4_clear_blocks(handle_t *handle, struct inode *inode, + struct buffer_head *bh, + ext4_fsblk_t block_to_free, + unsigned long count, __le32 *first, + __le32 *last) { __le32 *p; - int flags = EXT4_FREE_BLOCKS_FORGET; + int flags = EXT4_FREE_BLOCKS_FORGET | EXT4_FREE_BLOCKS_VALIDATED; if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) flags |= EXT4_FREE_BLOCKS_METADATA; + if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), block_to_free, + count)) { + ext4_error(inode->i_sb, __func__, "inode #%lu: " + "attempt to clear blocks %llu len %lu, invalid", + inode->i_ino, (unsigned long long) block_to_free, + count); + return 1; + } + if (try_to_extend_transaction(handle, inode)) { if (bh) { BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); @@ -4160,6 +4169,7 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode, *p = 0; ext4_free_blocks(handle, inode, 0, block_to_free, count, flags); + return 0; } /** @@ -4215,9 +4225,10 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, } else if (nr == block_to_free + count) { count++; } else { - ext4_clear_blocks(handle, inode, this_bh, - block_to_free, - count, block_to_free_p, p); + if (ext4_clear_blocks(handle, inode, this_bh, + block_to_free, count, + block_to_free_p, p)) + break; block_to_free = nr; block_to_free_p = p; count = 1; @@ -4281,6 +4292,16 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, if (!nr) continue; /* A hole */ + if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), + nr, 1)) { + ext4_error(inode->i_sb, __func__, + "indirect mapped block in inode " + "#%lu invalid (level %d, blk #%lu)", + inode->i_ino, depth, + (unsigned long) nr); + break; + } + /* Go read the buffer for the next level down */ bh = sb_bread(inode->i_sb, nr); -- cgit v1.2.1 From f8ec9d6837241865cf99bed97bb99f4399fd5a03 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Fri, 1 Jan 2010 01:00:21 -0500 Subject: ext4: Add new tracepoints to debug delayed allocation space functions Add tracepoints for ext4_da_reserve_space(), ext4_da_update_reserve_space(), and ext4_da_release_space(). Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 3e8afd969236..1a3d7b232cd7 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1061,6 +1061,7 @@ void ext4_da_update_reserve_space(struct inode *inode, int mdb_free = 0, allocated_meta_blocks = 0; spin_lock(&ei->i_block_reservation_lock); + trace_ext4_da_update_reserve_space(inode, used); 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", @@ -1846,6 +1847,7 @@ repeat: spin_lock(&ei->i_block_reservation_lock); md_reserved = ei->i_reserved_meta_blocks; md_needed = ext4_calc_metadata_amount(inode, lblock); + trace_ext4_da_reserve_space(inode, md_needed); spin_unlock(&ei->i_block_reservation_lock); /* -- cgit v1.2.1 From 19f5fb7ad679bb361222c7916086435020c37cce Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sun, 24 Jan 2010 14:34:07 -0500 Subject: ext4: Use bitops to read/modify EXT4_I(inode)->i_state At several places we modify EXT4_I(inode)->i_state without holding i_mutex (ext4_release_file, ext4_bmap, ext4_journalled_writepage, ext4_do_update_inode, ...). These modifications are racy and we can lose updates to i_state. So convert handling of i_state to use bitops which are atomic. Cc: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 38 ++++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 18 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 1a3d7b232cd7..3e530119d7f0 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1307,7 +1307,7 @@ int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block, * i_data's format changing. Force the migrate * to fail by clearing migrate flags */ - EXT4_I(inode)->i_state &= ~EXT4_STATE_EXT_MIGRATE; + ext4_clear_inode_state(inode, EXT4_STATE_EXT_MIGRATE); } /* @@ -1794,7 +1794,7 @@ static int ext4_journalled_write_end(struct file *file, new_i_size = pos + copied; if (new_i_size > inode->i_size) i_size_write(inode, pos+copied); - EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + ext4_set_inode_state(inode, EXT4_STATE_JDATA); if (new_i_size > EXT4_I(inode)->i_disksize) { ext4_update_i_disksize(inode, new_i_size); ret2 = ext4_mark_inode_dirty(handle, inode); @@ -2632,7 +2632,7 @@ static int __ext4_journalled_writepage(struct page *page, ret = err; walk_page_buffers(handle, page_bufs, 0, len, NULL, bput_one); - EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + ext4_set_inode_state(inode, EXT4_STATE_JDATA); out: return ret; } @@ -3303,7 +3303,8 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) filemap_write_and_wait(mapping); } - if (EXT4_JOURNAL(inode) && EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { + if (EXT4_JOURNAL(inode) && + ext4_test_inode_state(inode, EXT4_STATE_JDATA)) { /* * This is a REALLY heavyweight approach, but the use of * bmap on dirty files is expected to be extremely rare: @@ -3322,7 +3323,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) * everything they get. */ - EXT4_I(inode)->i_state &= ~EXT4_STATE_JDATA; + ext4_clear_inode_state(inode, EXT4_STATE_JDATA); journal = EXT4_JOURNAL(inode); jbd2_journal_lock_updates(journal); err = jbd2_journal_flush(journal); @@ -3790,8 +3791,8 @@ 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_I(inode)->i_state & - EXT4_STATE_DIO_UNWRITTEN)) { + } else if (ret > 0 && ext4_test_inode_state(inode, + EXT4_STATE_DIO_UNWRITTEN)) { int err; /* * for non AIO case, since the IO is already @@ -3801,7 +3802,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, offset, ret); if (err < 0) ret = err; - EXT4_I(inode)->i_state &= ~EXT4_STATE_DIO_UNWRITTEN; + ext4_clear_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN); } return ret; } @@ -4457,7 +4458,7 @@ void ext4_truncate(struct inode *inode) return; if (inode->i_size == 0 && !test_opt(inode->i_sb, NO_AUTO_DA_ALLOC)) - ei->i_state |= EXT4_STATE_DA_ALLOC_CLOSE; + ext4_set_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE); if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { ext4_ext_truncate(inode); @@ -4743,7 +4744,7 @@ int ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc) { /* We have all inode data except xattrs in memory here. */ return __ext4_get_inode_loc(inode, iloc, - !(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)); + !ext4_test_inode_state(inode, EXT4_STATE_XATTR)); } void ext4_set_inode_flags(struct inode *inode) @@ -4837,7 +4838,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) } inode->i_nlink = le16_to_cpu(raw_inode->i_links_count); - ei->i_state = 0; + ei->i_state_flags = 0; ei->i_dir_start_lookup = 0; ei->i_dtime = le32_to_cpu(raw_inode->i_dtime); /* We now have enough fields to check if the inode was active or not. @@ -4920,7 +4921,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize; if (*magic == cpu_to_le32(EXT4_XATTR_MAGIC)) - ei->i_state |= EXT4_STATE_XATTR; + ext4_set_inode_state(inode, EXT4_STATE_XATTR); } } else ei->i_extra_isize = 0; @@ -5060,7 +5061,7 @@ static int ext4_do_update_inode(handle_t *handle, /* For fields not not tracking in the in-memory inode, * initialise them to zero for new inodes. */ - if (ei->i_state & EXT4_STATE_NEW) + if (ext4_test_inode_state(inode, EXT4_STATE_NEW)) memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size); ext4_get_inode_flags(ei); @@ -5156,7 +5157,7 @@ static int ext4_do_update_inode(handle_t *handle, rc = ext4_handle_dirty_metadata(handle, inode, bh); if (!err) err = rc; - ei->i_state &= ~EXT4_STATE_NEW; + ext4_clear_inode_state(inode, EXT4_STATE_NEW); ext4_update_inode_fsync_trans(handle, inode, 0); out_brelse: @@ -5580,8 +5581,8 @@ static int ext4_expand_extra_isize(struct inode *inode, entry = IFIRST(header); /* No extended attributes present */ - if (!(EXT4_I(inode)->i_state & EXT4_STATE_XATTR) || - header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)) { + if (!ext4_test_inode_state(inode, EXT4_STATE_XATTR) || + header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)) { memset((void *)raw_inode + EXT4_GOOD_OLD_INODE_SIZE, 0, new_extra_isize); EXT4_I(inode)->i_extra_isize = new_extra_isize; @@ -5625,7 +5626,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) err = ext4_reserve_inode_write(handle, inode, &iloc); if (ext4_handle_valid(handle) && EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize && - !(EXT4_I(inode)->i_state & EXT4_STATE_NO_EXPAND)) { + !ext4_test_inode_state(inode, EXT4_STATE_NO_EXPAND)) { /* * We need extra buffer credits since we may write into EA block * with this same handle. If journal_extend fails, then it will @@ -5639,7 +5640,8 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) sbi->s_want_extra_isize, iloc, handle); if (ret) { - EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND; + ext4_set_inode_state(inode, + EXT4_STATE_NO_EXPAND); if (mnt_count != le16_to_cpu(sbi->s_es->s_mnt_count)) { ext4_warning(inode->i_sb, __func__, -- cgit v1.2.1 From 12062dddda450976b129dcb1bacd91acaf4d8030 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Mon, 15 Feb 2010 14:19:27 -0500 Subject: ext4: move __func__ into a macro for ext4_warning, ext4_error Just a pet peeve of mine; we had a mishash of calls with either __func__ or "function_name" and the latter tends to get out of sync. I think it's easier to just hide the __func__ in a macro, and it'll be consistent from then on. Signed-off-by: Eric Sandeen Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 43 ++++++++++++++++++------------------------- 1 file changed, 18 insertions(+), 25 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 3e530119d7f0..536067bcf75b 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -194,7 +194,7 @@ void ext4_delete_inode(struct inode *inode) inode->i_size = 0; err = ext4_mark_inode_dirty(handle, inode); if (err) { - ext4_warning(inode->i_sb, __func__, + ext4_warning(inode->i_sb, "couldn't mark inode dirty (err %d)", err); goto stop_handle; } @@ -212,7 +212,7 @@ void ext4_delete_inode(struct inode *inode) if (err > 0) err = ext4_journal_restart(handle, 3); if (err != 0) { - ext4_warning(inode->i_sb, __func__, + ext4_warning(inode->i_sb, "couldn't extend journal (err %d)", err); stop_handle: ext4_journal_stop(handle); @@ -323,8 +323,7 @@ static int ext4_block_to_path(struct inode *inode, offsets[n++] = i_block & (ptrs - 1); final = ptrs; } else { - ext4_warning(inode->i_sb, "ext4_block_to_path", - "block %lu > max in inode %lu", + ext4_warning(inode->i_sb, "block %lu > max in inode %lu", i_block + direct_blocks + indirect_blocks + double_blocks, inode->i_ino); } @@ -344,7 +343,7 @@ static int __ext4_check_blockref(const char *function, struct inode *inode, if (blk && unlikely(!ext4_data_block_valid(EXT4_SB(inode->i_sb), blk, 1))) { - ext4_error(inode->i_sb, function, + __ext4_error(inode->i_sb, function, "invalid block reference %u " "in inode #%lu", blk, inode->i_ino); return -EIO; @@ -1125,7 +1124,7 @@ static int check_block_validity(struct inode *inode, const char *msg, sector_t logical, sector_t phys, int len) { if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), phys, len)) { - ext4_error(inode->i_sb, msg, + __ext4_error(inode->i_sb, msg, "inode #%lu logical block %llu mapped to %llu " "(size %d)", inode->i_ino, (unsigned long long) logical, @@ -4147,7 +4146,7 @@ static int ext4_clear_blocks(handle_t *handle, struct inode *inode, if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), block_to_free, count)) { - ext4_error(inode->i_sb, __func__, "inode #%lu: " + ext4_error(inode->i_sb, "inode #%lu: " "attempt to clear blocks %llu len %lu, invalid", inode->i_ino, (unsigned long long) block_to_free, count); @@ -4255,7 +4254,7 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, if ((EXT4_JOURNAL(inode) == NULL) || bh2jh(this_bh)) ext4_handle_dirty_metadata(handle, inode, this_bh); else - ext4_error(inode->i_sb, __func__, + ext4_error(inode->i_sb, "circular indirect block detected, " "inode=%lu, block=%llu", inode->i_ino, @@ -4297,7 +4296,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), nr, 1)) { - ext4_error(inode->i_sb, __func__, + ext4_error(inode->i_sb, "indirect mapped block in inode " "#%lu invalid (level %d, blk #%lu)", inode->i_ino, depth, @@ -4313,7 +4312,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, * (should be rare). */ if (!bh) { - ext4_error(inode->i_sb, "ext4_free_branches", + ext4_error(inode->i_sb, "Read failure, inode=%lu, block=%llu", inode->i_ino, nr); continue; @@ -4628,9 +4627,8 @@ static int __ext4_get_inode_loc(struct inode *inode, bh = sb_getblk(sb, block); if (!bh) { - ext4_error(sb, "ext4_get_inode_loc", "unable to read " - "inode block - inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, "unable to read inode block - " + "inode=%lu, block=%llu", inode->i_ino, block); return -EIO; } if (!buffer_uptodate(bh)) { @@ -4728,9 +4726,8 @@ make_io: submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ext4_error(sb, __func__, - "unable to read inode block - inode=%lu, " - "block=%llu", inode->i_ino, block); + ext4_error(sb, "unable to read inode block - inode=%lu," + " block=%llu", inode->i_ino, block); brelse(bh); return -EIO; } @@ -4941,8 +4938,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ret = 0; if (ei->i_file_acl && !ext4_data_block_valid(EXT4_SB(sb), ei->i_file_acl, 1)) { - ext4_error(sb, __func__, - "bad extended attribute block %llu in inode #%lu", + ext4_error(sb, "bad extended attribute block %llu inode #%lu", ei->i_file_acl, inode->i_ino); ret = -EIO; goto bad_inode; @@ -4988,8 +4984,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } else { ret = -EIO; - ext4_error(inode->i_sb, __func__, - "bogus i_mode (%o) for inode=%lu", + ext4_error(inode->i_sb, "bogus i_mode (%o) for inode=%lu", inode->i_mode, inode->i_ino); goto bad_inode; } @@ -5228,10 +5223,8 @@ int ext4_write_inode(struct inode *inode, int wait) if (wait) sync_dirty_buffer(iloc.bh); if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) { - ext4_error(inode->i_sb, __func__, - "IO error syncing inode, " - "inode=%lu, block=%llu", - inode->i_ino, + ext4_error(inode->i_sb, "IO error syncing inode, " + "inode=%lu, block=%llu", inode->i_ino, (unsigned long long)iloc.bh->b_blocknr); err = -EIO; } @@ -5644,7 +5637,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) EXT4_STATE_NO_EXPAND); if (mnt_count != le16_to_cpu(sbi->s_es->s_mnt_count)) { - ext4_warning(inode->i_sb, __func__, + ext4_warning(inode->i_sb, "Unable to expand inode %lu. Delete" " some EAs or run e2fsck.", inode->i_ino); -- cgit v1.2.1 From 73b50c1c92666d326b5fa2c945d46509f2f6d91f Mon Sep 17 00:00:00 2001 From: Curt Wohlgemuth Date: Tue, 16 Feb 2010 15:06:29 -0500 Subject: ext4: Fix BUG_ON at fs/buffer.c:652 in no journal mode Calls to ext4_handle_dirty_metadata should only pass in an inode pointer for inode-specific metadata, and not for shared metadata blocks such as inode table blocks, block group descriptors, the superblock, etc. The BUG_ON can get tripped when updating a special device (such as a block device) that is opened (so that i_mapping is set in fs/block_dev.c) and the file system is mounted in no journal mode. Addresses-Google-Bug: #2404870 Signed-off-by: Curt Wohlgemuth Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 536067bcf75b..ecac8c5a6f5c 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5120,7 +5120,7 @@ static int ext4_do_update_inode(handle_t *handle, EXT4_FEATURE_RO_COMPAT_LARGE_FILE); sb->s_dirt = 1; ext4_handle_sync(handle); - err = ext4_handle_dirty_metadata(handle, inode, + err = ext4_handle_dirty_metadata(handle, NULL, EXT4_SB(sb)->s_sbh); } } @@ -5149,7 +5149,7 @@ static int ext4_do_update_inode(handle_t *handle, } BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); - rc = ext4_handle_dirty_metadata(handle, inode, bh); + rc = ext4_handle_dirty_metadata(handle, NULL, bh); if (!err) err = rc; ext4_clear_inode_state(inode, EXT4_STATE_NEW); @@ -5701,7 +5701,7 @@ static int ext4_pin_inode(handle_t *handle, struct inode *inode) err = jbd2_journal_get_write_access(handle, iloc.bh); if (!err) err = ext4_handle_dirty_metadata(handle, - inode, + NULL, iloc.bh); brelse(iloc.bh); } -- cgit v1.2.1 From c8d46e41bc744c8fa0092112af3942fcd46c8b18 Mon Sep 17 00:00:00 2001 From: Jiaying Zhang Date: Wed, 24 Feb 2010 09:52:53 -0500 Subject: ext4: Add flag to files with blocks intentionally past EOF fallocate() may potentially instantiate blocks past EOF, depending on the flags used when it is called. e2fsck currently has a test for blocks past i_size, and it sometimes trips up - noticeably on xfstests 013 which runs fsstress. This patch from Jiayang does fix it up - it (along with e2fsprogs updates and other patches recently from Aneesh) has survived many fsstress runs in a row. Signed-off-by: Eric Sandeen Signed-off-by: Jiaying Zhang Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index ecac8c5a6f5c..edb7edc99f71 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4456,6 +4456,8 @@ void ext4_truncate(struct inode *inode) if (!ext4_can_truncate(inode)) return; + EXT4_I(inode)->i_flags &= ~EXT4_EOFBLOCKS_FL; + if (inode->i_size == 0 && !test_opt(inode->i_sb, NO_AUTO_DA_ALLOC)) ext4_set_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE); @@ -5305,7 +5307,9 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) } if (S_ISREG(inode->i_mode) && - attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) { + attr->ia_valid & ATTR_SIZE && + (attr->ia_size < inode->i_size || + (EXT4_I(inode)->i_flags & EXT4_EOFBLOCKS_FL))) { handle_t *handle; handle = ext4_journal_start(inode, 3); @@ -5336,6 +5340,9 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) goto err_out; } } + /* ext4_truncate will clear the flag */ + if ((EXT4_I(inode)->i_flags & EXT4_EOFBLOCKS_FL)) + ext4_truncate(inode); } rc = inode_setattr(inode, attr); -- cgit v1.2.1 From da1dafca84413145f5ac59998b4cdd06fb89f721 Mon Sep 17 00:00:00 2001 From: Dmitry Monakhov Date: Mon, 1 Mar 2010 23:15:02 -0500 Subject: ext4: explicitly remove inode from orphan list after failed direct io Otherwise non-empty orphan list will be triggered on umount. Signed-off-by: Dmitry Monakhov Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index edb7edc99f71..427f4690ad6d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3438,6 +3438,9 @@ retry: * but cannot extend i_size. Bail out and pretend * the write failed... */ ret = PTR_ERR(handle); + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + goto out; } if (inode->i_nlink) -- cgit v1.2.1 From c7064ef13b2181a489836349f9baf87df0dab28f Mon Sep 17 00:00:00 2001 From: Jiaying Zhang Date: Tue, 2 Mar 2010 13:28:44 -0500 Subject: ext4: mechanical rename some of the direct I/O get_block's identifiers This commit renames some of the direct I/O's block allocation flags, variables, and functions introduced in Mingming's "Direct IO for holes and fallocate" patches so that they can be used by ext4's buffered write path as well. Also changed the related function comments accordingly to cover both direct write and buffered write cases. Signed-off-by: Jiaying Zhang Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 84 ++++++++++++++++++++++++++------------------------------- 1 file changed, 38 insertions(+), 46 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 427f4690ad6d..28f116bdc405 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3468,7 +3468,7 @@ out: return ret; } -static int ext4_get_block_dio_write(struct inode *inode, sector_t iblock, +static int ext4_get_block_write(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { handle_t *handle = NULL; @@ -3476,28 +3476,14 @@ static int ext4_get_block_dio_write(struct inode *inode, sector_t iblock, unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; int dio_credits; - ext4_debug("ext4_get_block_dio_write: inode %lu, create flag %d\n", + ext4_debug("ext4_get_block_write: inode %lu, create flag %d\n", inode->i_ino, create); /* - * DIO VFS code passes create = 0 flag for write to - * the middle of file. It does this to avoid block - * allocation for holes, to prevent expose stale data - * out when there is parallel buffered read (which does - * not hold the i_mutex lock) while direct IO write has - * not completed. DIO request on holes finally falls back - * to buffered IO for this reason. - * - * For ext4 extent based file, since we support fallocate, - * new allocated extent as uninitialized, for holes, we - * could fallocate blocks for holes, thus parallel - * buffered IO read will zero out the page when read on - * a hole while parallel DIO write to the hole has not completed. - * - * when we come here, we know it's a direct IO write to - * to the middle of file ( DIO_MAX_BLOCKS) max_blocks = DIO_MAX_BLOCKS; @@ -3524,19 +3510,20 @@ static void ext4_free_io_end(ext4_io_end_t *io) iput(io->inode); kfree(io); } -static void dump_aio_dio_list(struct inode * inode) + +static void dump_completed_IO(struct inode * inode) { #ifdef EXT4_DEBUG struct list_head *cur, *before, *after; ext4_io_end_t *io, *io0, *io1; - if (list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)){ - ext4_debug("inode %lu aio dio list is empty\n", inode->i_ino); + if (list_empty(&EXT4_I(inode)->i_completed_io_list)){ + ext4_debug("inode %lu completed_io list is empty\n", inode->i_ino); return; } - ext4_debug("Dump inode %lu aio_dio_completed_IO list \n", inode->i_ino); - list_for_each_entry(io, &EXT4_I(inode)->i_aio_dio_complete_list, list){ + ext4_debug("Dump inode %lu completed_io list \n", inode->i_ino); + list_for_each_entry(io, &EXT4_I(inode)->i_completed_io_list, list){ cur = &io->list; before = cur->prev; io0 = container_of(before, ext4_io_end_t, list); @@ -3552,21 +3539,21 @@ static void dump_aio_dio_list(struct inode * inode) /* * check a range of space and convert unwritten extents to written. */ -static int ext4_end_aio_dio_nolock(ext4_io_end_t *io) +static int ext4_end_io_nolock(ext4_io_end_t *io) { struct inode *inode = io->inode; loff_t offset = io->offset; ssize_t size = io->size; int ret = 0; - ext4_debug("end_aio_dio_onlock: io 0x%p from inode %lu,list->next 0x%p," + ext4_debug("ext4_end_io_nolock: io 0x%p from inode %lu,list->next 0x%p," "list->prev 0x%p\n", io, inode->i_ino, io->list.next, io->list.prev); if (list_empty(&io->list)) return ret; - if (io->flag != DIO_AIO_UNWRITTEN) + if (io->flag != EXT4_IO_UNWRITTEN) return ret; if (offset + size <= i_size_read(inode)) @@ -3584,17 +3571,18 @@ static int ext4_end_aio_dio_nolock(ext4_io_end_t *io) io->flag = 0; return ret; } + /* * work on completed aio dio IO, to convert unwritten extents to extents */ -static void ext4_end_aio_dio_work(struct work_struct *work) +static void ext4_end_io_work(struct work_struct *work) { ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); struct inode *inode = io->inode; int ret = 0; mutex_lock(&inode->i_mutex); - ret = ext4_end_aio_dio_nolock(io); + ret = ext4_end_io_nolock(io); if (ret >= 0) { if (!list_empty(&io->list)) list_del_init(&io->list); @@ -3602,32 +3590,35 @@ static void ext4_end_aio_dio_work(struct work_struct *work) } mutex_unlock(&inode->i_mutex); } + /* * This function is called from ext4_sync_file(). * - * When AIO DIO IO is completed, the work to convert unwritten - * extents to written is queued on workqueue but may not get immediately + * When IO is completed, the work to convert unwritten extents to + * written is queued on workqueue but may not get immediately * scheduled. When fsync is called, we need to ensure the * conversion is complete before fsync returns. - * The inode keeps track of a list of completed AIO from DIO path - * that might needs to do the conversion. This function walks through - * the list and convert the related unwritten extents to written. + * The inode keeps track of a list of pending/completed IO that + * might needs to do the conversion. This function walks through + * the list and convert the related unwritten extents for completed IO + * to written. + * The function return the number of pending IOs on success. */ -int flush_aio_dio_completed_IO(struct inode *inode) +int flush_completed_IO(struct inode *inode) { ext4_io_end_t *io; int ret = 0; int ret2 = 0; - if (list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)) + if (list_empty(&EXT4_I(inode)->i_completed_io_list)) return ret; - dump_aio_dio_list(inode); - while (!list_empty(&EXT4_I(inode)->i_aio_dio_complete_list)){ - io = list_entry(EXT4_I(inode)->i_aio_dio_complete_list.next, + dump_completed_IO(inode); + while (!list_empty(&EXT4_I(inode)->i_completed_io_list)){ + io = list_entry(EXT4_I(inode)->i_completed_io_list.next, ext4_io_end_t, list); /* - * Calling ext4_end_aio_dio_nolock() to convert completed + * Calling ext4_end_io_nolock() to convert completed * IO to written. * * When ext4_sync_file() is called, run_queue() may already @@ -3640,7 +3631,7 @@ int flush_aio_dio_completed_IO(struct inode *inode) * avoid double converting from both fsync and background work * queue work. */ - ret = ext4_end_aio_dio_nolock(io); + ret = ext4_end_io_nolock(io); if (ret < 0) ret2 = ret; else @@ -3662,7 +3653,7 @@ static ext4_io_end_t *ext4_init_io_end (struct inode *inode) io->offset = 0; io->size = 0; io->error = 0; - INIT_WORK(&io->work, ext4_end_aio_dio_work); + INIT_WORK(&io->work, ext4_end_io_work); INIT_LIST_HEAD(&io->list); } @@ -3685,7 +3676,7 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, size); /* if not aio dio with unwritten extents, just free io and return */ - if (io_end->flag != DIO_AIO_UNWRITTEN){ + if (io_end->flag != EXT4_IO_UNWRITTEN){ ext4_free_io_end(io_end); iocb->private = NULL; return; @@ -3700,9 +3691,10 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, /* Add the io_end to per-inode completed aio dio list*/ list_add_tail(&io_end->list, - &EXT4_I(io_end->inode)->i_aio_dio_complete_list); + &EXT4_I(io_end->inode)->i_completed_io_list); iocb->private = NULL; } + /* * For ext4 extent files, ext4 will do direct-io write to holes, * preallocated extents, and those write extend the file, no need to @@ -3772,7 +3764,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, offset, nr_segs, - ext4_get_block_dio_write, + ext4_get_block_write, ext4_end_io_dio); if (iocb->private) EXT4_I(inode)->cur_aio_dio = NULL; -- cgit v1.2.1 From 744692dc059845b2a3022119871846e74d4f6e11 Mon Sep 17 00:00:00 2001 From: Jiaying Zhang Date: Thu, 4 Mar 2010 16:14:02 -0500 Subject: ext4: use ext4_get_block_write in buffer write Allocate uninitialized extent before ext4 buffer write and convert the extent to initialized after io completes. The purpose is to make sure an extent can only be marked initialized after it has been written with new data so we can safely drop the i_mutex lock in ext4 DIO read without exposing stale data. This helps to improve multi-thread DIO read performance on high-speed disks. Skip the nobh and data=journal mount cases to make things simple for now. Signed-off-by: Jiaying Zhang Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 213 +++++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 174 insertions(+), 39 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 28f116bdc405..d291310aef6b 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -38,6 +38,7 @@ #include #include #include +#include #include "ext4_jbd2.h" #include "xattr.h" @@ -1534,6 +1535,8 @@ static void ext4_truncate_failed_write(struct inode *inode) ext4_truncate(inode); } +static int ext4_get_block_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create); static int ext4_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -1575,8 +1578,12 @@ retry: } *pagep = page; - ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, - ext4_get_block); + if (ext4_should_dioread_nolock(inode)) + ret = block_write_begin(file, mapping, pos, len, flags, pagep, + fsdata, ext4_get_block_write); + else + ret = block_write_begin(file, mapping, pos, len, flags, pagep, + fsdata, ext4_get_block); if (!ret && ext4_should_journal_data(inode)) { ret = walk_page_buffers(handle, page_buffers(page), @@ -2092,6 +2099,8 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, } else if (buffer_mapped(bh)) BUG_ON(bh->b_blocknr != pblock); + if (buffer_uninit(exbh)) + set_buffer_uninit(bh); cur_logical++; pblock++; } while ((bh = bh->b_this_page) != head); @@ -2221,6 +2230,8 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd) */ new.b_state = 0; get_blocks_flags = EXT4_GET_BLOCKS_CREATE; + if (ext4_should_dioread_nolock(mpd->inode)) + get_blocks_flags |= EXT4_GET_BLOCKS_IO_CREATE_EXT; if (mpd->b_state & (1 << BH_Delay)) get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE; @@ -2636,6 +2647,9 @@ out: return ret; } +static int ext4_set_bh_endio(struct buffer_head *bh, struct inode *inode); +static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate); + /* * Note that we don't need to start a transaction unless we're journaling data * because we should have holes filled from ext4_page_mkwrite(). We even don't @@ -2683,7 +2697,7 @@ static int ext4_writepage(struct page *page, int ret = 0; loff_t size; unsigned int len; - struct buffer_head *page_bufs; + struct buffer_head *page_bufs = NULL; struct inode *inode = page->mapping->host; trace_ext4_writepage(inode, page); @@ -2759,7 +2773,11 @@ static int ext4_writepage(struct page *page, if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) ret = nobh_writepage(page, noalloc_get_block_write, wbc); - else + else if (page_bufs && buffer_uninit(page_bufs)) { + ext4_set_bh_endio(page_bufs, inode); + ret = block_write_full_page_endio(page, noalloc_get_block_write, + wbc, ext4_end_io_buffer_write); + } else ret = block_write_full_page(page, noalloc_get_block_write, wbc); @@ -3347,10 +3365,44 @@ ext4_readpages(struct file *file, struct address_space *mapping, return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); } +static void ext4_free_io_end(ext4_io_end_t *io) +{ + BUG_ON(!io); + if (io->page) + put_page(io->page); + iput(io->inode); + kfree(io); +} + +static void ext4_invalidatepage_free_endio(struct page *page, unsigned long offset) +{ + struct buffer_head *head, *bh; + unsigned int curr_off = 0; + + if (!page_has_buffers(page)) + return; + head = bh = page_buffers(page); + do { + if (offset <= curr_off && test_clear_buffer_uninit(bh) + && bh->b_private) { + ext4_free_io_end(bh->b_private); + bh->b_private = NULL; + bh->b_end_io = NULL; + } + curr_off = curr_off + bh->b_size; + bh = bh->b_this_page; + } while (bh != head); +} + static void ext4_invalidatepage(struct page *page, unsigned long offset) { journal_t *journal = EXT4_JOURNAL(page->mapping->host); + /* + * free any io_end structure allocated for buffers to be discarded + */ + if (ext4_should_dioread_nolock(page->mapping->host)) + ext4_invalidatepage_free_endio(page, offset); /* * If it's a full truncate we just forget about the pending dirtying */ @@ -3471,10 +3523,11 @@ out: static int ext4_get_block_write(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { - handle_t *handle = NULL; + handle_t *handle = ext4_journal_current_handle(); int ret = 0; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; int dio_credits; + int started = 0; ext4_debug("ext4_get_block_write: inode %lu, create flag %d\n", inode->i_ino, create); @@ -3485,37 +3538,36 @@ static int ext4_get_block_write(struct inode *inode, sector_t iblock, */ create = EXT4_GET_BLOCKS_IO_CREATE_EXT; - if (max_blocks > DIO_MAX_BLOCKS) - max_blocks = DIO_MAX_BLOCKS; - dio_credits = ext4_chunk_trans_blocks(inode, max_blocks); - handle = ext4_journal_start(inode, dio_credits); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; + if (!handle) { + if (max_blocks > DIO_MAX_BLOCKS) + max_blocks = DIO_MAX_BLOCKS; + dio_credits = ext4_chunk_trans_blocks(inode, max_blocks); + handle = ext4_journal_start(inode, dio_credits); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + started = 1; } + ret = ext4_get_blocks(handle, inode, iblock, max_blocks, bh_result, create); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); ret = 0; } - ext4_journal_stop(handle); + if (started) + ext4_journal_stop(handle); out: return ret; } -static void ext4_free_io_end(ext4_io_end_t *io) -{ - BUG_ON(!io); - iput(io->inode); - kfree(io); -} - static void dump_completed_IO(struct inode * inode) { #ifdef EXT4_DEBUG struct list_head *cur, *before, *after; ext4_io_end_t *io, *io0, *io1; + unsigned long flags; if (list_empty(&EXT4_I(inode)->i_completed_io_list)){ ext4_debug("inode %lu completed_io list is empty\n", inode->i_ino); @@ -3523,6 +3575,7 @@ static void dump_completed_IO(struct inode * inode) } ext4_debug("Dump inode %lu completed_io list \n", inode->i_ino); + spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); list_for_each_entry(io, &EXT4_I(inode)->i_completed_io_list, list){ cur = &io->list; before = cur->prev; @@ -3533,6 +3586,7 @@ static void dump_completed_IO(struct inode * inode) ext4_debug("io 0x%p from inode %lu,prev 0x%p,next 0x%p\n", io, inode->i_ino, io0, io1); } + spin_unlock_irqrestore(&EXT4_I(inode)->i_completed_io_lock, flags); #endif } @@ -3556,9 +3610,7 @@ static int ext4_end_io_nolock(ext4_io_end_t *io) if (io->flag != EXT4_IO_UNWRITTEN) return ret; - if (offset + size <= i_size_read(inode)) - ret = ext4_convert_unwritten_extents(inode, offset, size); - + ret = ext4_convert_unwritten_extents(inode, offset, size); if (ret < 0) { printk(KERN_EMERG "%s: failed to convert unwritten" "extents to written extents, error is %d" @@ -3577,18 +3629,25 @@ static int ext4_end_io_nolock(ext4_io_end_t *io) */ static void ext4_end_io_work(struct work_struct *work) { - ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); - struct inode *inode = io->inode; - int ret = 0; + ext4_io_end_t *io = container_of(work, ext4_io_end_t, work); + struct inode *inode = io->inode; + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned long flags; + int ret; mutex_lock(&inode->i_mutex); ret = ext4_end_io_nolock(io); - if (ret >= 0) { - if (!list_empty(&io->list)) - list_del_init(&io->list); - ext4_free_io_end(io); + if (ret < 0) { + mutex_unlock(&inode->i_mutex); + return; } + + spin_lock_irqsave(&ei->i_completed_io_lock, flags); + if (!list_empty(&io->list)) + list_del_init(&io->list); + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); mutex_unlock(&inode->i_mutex); + ext4_free_io_end(io); } /* @@ -3607,15 +3666,18 @@ static void ext4_end_io_work(struct work_struct *work) int flush_completed_IO(struct inode *inode) { ext4_io_end_t *io; + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned long flags; int ret = 0; int ret2 = 0; - if (list_empty(&EXT4_I(inode)->i_completed_io_list)) + if (list_empty(&ei->i_completed_io_list)) return ret; dump_completed_IO(inode); - while (!list_empty(&EXT4_I(inode)->i_completed_io_list)){ - io = list_entry(EXT4_I(inode)->i_completed_io_list.next, + spin_lock_irqsave(&ei->i_completed_io_lock, flags); + while (!list_empty(&ei->i_completed_io_list)){ + io = list_entry(ei->i_completed_io_list.next, ext4_io_end_t, list); /* * Calling ext4_end_io_nolock() to convert completed @@ -3631,20 +3693,23 @@ int flush_completed_IO(struct inode *inode) * avoid double converting from both fsync and background work * queue work. */ + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); ret = ext4_end_io_nolock(io); + spin_lock_irqsave(&ei->i_completed_io_lock, flags); if (ret < 0) ret2 = ret; else list_del_init(&io->list); } + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); return (ret2 < 0) ? ret2 : 0; } -static ext4_io_end_t *ext4_init_io_end (struct inode *inode) +static ext4_io_end_t *ext4_init_io_end (struct inode *inode, gfp_t flags) { ext4_io_end_t *io = NULL; - io = kmalloc(sizeof(*io), GFP_NOFS); + io = kmalloc(sizeof(*io), flags); if (io) { igrab(inode); @@ -3652,7 +3717,7 @@ static ext4_io_end_t *ext4_init_io_end (struct inode *inode) io->flag = 0; io->offset = 0; io->size = 0; - io->error = 0; + io->page = NULL; INIT_WORK(&io->work, ext4_end_io_work); INIT_LIST_HEAD(&io->list); } @@ -3665,6 +3730,8 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, { ext4_io_end_t *io_end = iocb->private; struct workqueue_struct *wq; + unsigned long flags; + struct ext4_inode_info *ei; /* if not async direct IO or dio with 0 bytes write, just return */ if (!io_end || !size) @@ -3684,17 +3751,85 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, io_end->offset = offset; io_end->size = size; + io_end->flag = EXT4_IO_UNWRITTEN; wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq; /* queue the work to convert unwritten extents to written */ queue_work(wq, &io_end->work); /* Add the io_end to per-inode completed aio dio list*/ - list_add_tail(&io_end->list, - &EXT4_I(io_end->inode)->i_completed_io_list); + ei = EXT4_I(io_end->inode); + spin_lock_irqsave(&ei->i_completed_io_lock, flags); + list_add_tail(&io_end->list, &ei->i_completed_io_list); + spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); iocb->private = NULL; } +static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate) +{ + ext4_io_end_t *io_end = bh->b_private; + struct workqueue_struct *wq; + struct inode *inode; + unsigned long flags; + + if (!test_clear_buffer_uninit(bh) || !io_end) + goto out; + + if (!(io_end->inode->i_sb->s_flags & MS_ACTIVE)) { + printk("sb umounted, discard end_io request for inode %lu\n", + io_end->inode->i_ino); + ext4_free_io_end(io_end); + goto out; + } + + io_end->flag = EXT4_IO_UNWRITTEN; + inode = io_end->inode; + + /* Add the io_end to per-inode completed io list*/ + spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); + list_add_tail(&io_end->list, &EXT4_I(inode)->i_completed_io_list); + spin_unlock_irqrestore(&EXT4_I(inode)->i_completed_io_lock, flags); + + wq = EXT4_SB(inode->i_sb)->dio_unwritten_wq; + /* queue the work to convert unwritten extents to written */ + queue_work(wq, &io_end->work); +out: + bh->b_private = NULL; + bh->b_end_io = NULL; + clear_buffer_uninit(bh); + end_buffer_async_write(bh, uptodate); +} + +static int ext4_set_bh_endio(struct buffer_head *bh, struct inode *inode) +{ + ext4_io_end_t *io_end; + struct page *page = bh->b_page; + loff_t offset = (sector_t)page->index << PAGE_CACHE_SHIFT; + size_t size = bh->b_size; + +retry: + io_end = ext4_init_io_end(inode, GFP_ATOMIC); + if (!io_end) { + if (printk_ratelimit()) + printk(KERN_WARNING "%s: allocation fail\n", __func__); + schedule(); + goto retry; + } + io_end->offset = offset; + io_end->size = size; + /* + * We need to hold a reference to the page to make sure it + * doesn't get evicted before ext4_end_io_work() has a chance + * to convert the extent from written to unwritten. + */ + io_end->page = page; + get_page(io_end->page); + + bh->b_private = io_end; + bh->b_end_io = ext4_end_io_buffer_write; + return 0; +} + /* * For ext4 extent files, ext4 will do direct-io write to holes, * preallocated extents, and those write extend the file, no need to @@ -3748,7 +3883,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, iocb->private = NULL; EXT4_I(inode)->cur_aio_dio = NULL; if (!is_sync_kiocb(iocb)) { - iocb->private = ext4_init_io_end(inode); + iocb->private = ext4_init_io_end(inode, GFP_NOFS); if (!iocb->private) return -ENOMEM; /* -- cgit v1.2.1 From b7adc1f363e72e9131a582cc2cb00eaf83f51a39 Mon Sep 17 00:00:00 2001 From: Jiaying Zhang Date: Tue, 2 Mar 2010 13:26:36 -0500 Subject: ext4: Use direct_IO_no_locking in ext4 dio read Signed-off-by: Jiaying Zhang Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index d291310aef6b..92214d4e5afa 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3474,7 +3474,14 @@ static ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, } retry: - ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + if (rw == READ && ext4_should_dioread_nolock(inode)) + ret = blockdev_direct_IO_no_locking(rw, iocb, inode, + inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block, NULL); + else + ret = blockdev_direct_IO(rw, iocb, inode, + inode->i_sb->s_bdev, iov, offset, nr_segs, ext4_get_block, NULL); if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) -- cgit v1.2.1 From 273df556b6ee2065bfe96edab5888d3dc9b108d8 Mon Sep 17 00:00:00 2001 From: Frank Mayhar Date: Tue, 2 Mar 2010 11:46:09 -0500 Subject: ext4: Convert BUG_ON checks to use ext4_error() instead Convert a bunch of BUG_ONs to emit a ext4_error() message and return EIO. This is a first pass and most notably does _not_ cover mballoc.c, which is a morass of void functions. Signed-off-by: Frank Mayhar Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 92214d4e5afa..c717a74f2178 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -607,7 +607,14 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, if (*err) goto failed_out; - BUG_ON(current_block + count > EXT4_MAX_BLOCK_FILE_PHYS); + if (unlikely(current_block + count > EXT4_MAX_BLOCK_FILE_PHYS)) { + EXT4_ERROR_INODE(inode, + "current_block %llu + count %lu > %d!", + current_block, count, + EXT4_MAX_BLOCK_FILE_PHYS); + *err = -EIO; + goto failed_out; + } target -= count; /* allocate blocks for indirect blocks */ @@ -643,7 +650,14 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, ar.flags = EXT4_MB_HINT_DATA; current_block = ext4_mb_new_blocks(handle, &ar, err); - BUG_ON(current_block + ar.len > EXT4_MAX_BLOCK_FILE_PHYS); + if (unlikely(current_block + ar.len > EXT4_MAX_BLOCK_FILE_PHYS)) { + EXT4_ERROR_INODE(inode, + "current_block %llu + ar.len %d > %d!", + current_block, ar.len, + EXT4_MAX_BLOCK_FILE_PHYS); + *err = -EIO; + goto failed_out; + } if (*err && (target == blks)) { /* -- cgit v1.2.1 From 9b1d0998d24f9c207d5fbdd0b8bac07284e0eda7 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 3 Mar 2010 16:19:32 -0500 Subject: ext4: Release page references acquired in ext4_da_block_invalidatepages We forget to release page references we acquire in ext4_da_block_invalidatepages. Luckily, this function gets called only if we are not able to allocate blocks for delay-allocated data so that function should better never be called. Also cleanup handling of index variable. Reported-by: Wu Fengguang Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c717a74f2178..f55df7192b95 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2157,17 +2157,16 @@ static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd, break; for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; - index = page->index; - if (index > end) + if (page->index > end) break; - index++; - BUG_ON(!PageLocked(page)); BUG_ON(PageWriteback(page)); block_invalidatepage(page, 0); ClearPageUptodate(page); unlock_page(page); } + index = pvec.pages[nr_pages - 1]->index + 1; + pagevec_release(&pvec); } return; } -- cgit v1.2.1 From 5dd4056db84387975140ff2568eaa0406f07985e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 3 Mar 2010 09:05:00 -0500 Subject: dquot: cleanup space allocation / freeing routines Get rid of the alloc_space, free_space, reserve_space, claim_space and release_rsv dquot operations - they are always called from the filesystem and if a filesystem really needs their own (which none currently does) it can just call into it's own routine directly. Move shared logic into the common __dquot_alloc_space, dquot_claim_space_nodirty and __dquot_free_space low-level methods, and rationalize the wrappers around it to move as much as possible code into the common block for CONFIG_QUOTA vs not. Also rename all these helpers to be named dquot_* instead of vfs_dq_*. Signed-off-by: Christoph Hellwig Signed-off-by: Jan Kara --- fs/ext4/inode.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index e11952404e02..9f607ea411c8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1093,9 +1093,9 @@ void ext4_da_update_reserve_space(struct inode *inode, /* Update quota subsystem */ if (quota_claim) { - vfs_dq_claim_block(inode, used); + dquot_claim_block(inode, used); if (mdb_free) - vfs_dq_release_reservation_block(inode, mdb_free); + dquot_release_reservation_block(inode, mdb_free); } else { /* * We did fallocate with an offset that is already delayed @@ -1106,8 +1106,8 @@ void ext4_da_update_reserve_space(struct inode *inode, * that */ if (allocated_meta_blocks) - vfs_dq_claim_block(inode, allocated_meta_blocks); - vfs_dq_release_reservation_block(inode, mdb_free + used); + dquot_claim_block(inode, allocated_meta_blocks); + dquot_release_reservation_block(inode, mdb_free + used); } /* @@ -1836,6 +1836,7 @@ static int ext4_da_reserve_space(struct inode *inode, sector_t lblock) struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); struct ext4_inode_info *ei = EXT4_I(inode); unsigned long md_needed, md_reserved; + int ret; /* * recalculate the amount of metadata blocks to reserve @@ -1853,11 +1854,12 @@ repeat: * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, md_needed + 1)) - return -EDQUOT; + ret = dquot_reserve_block(inode, md_needed + 1); + if (ret) + return ret; if (ext4_claim_free_blocks(sbi, md_needed + 1)) { - vfs_dq_release_reservation_block(inode, md_needed + 1); + dquot_release_reservation_block(inode, md_needed + 1); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { yield(); goto repeat; @@ -1914,7 +1916,7 @@ static void ext4_da_release_space(struct inode *inode, int to_free) spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - vfs_dq_release_reservation_block(inode, to_free); + dquot_release_reservation_block(inode, to_free); } static void ext4_da_page_release_reservation(struct page *page, @@ -5641,7 +5643,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) * i_size has been changed by generic_commit_write() and we thus need * to include the updated inode in the current transaction. * - * Also, vfs_dq_alloc_block() will always dirty the inode when blocks + * Also, dquot_alloc_block() will always dirty the inode when blocks * are allocated to the file. * * If the inode is marked synchronous, we don't honour that here - doing -- cgit v1.2.1 From b43fa8284d7790d9cca32c9c55e24f29be2fa33b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 3 Mar 2010 09:05:03 -0500 Subject: dquot: cleanup dquot transfer routine Get rid of the transfer dquot operation - it is now always called from the filesystem and if a filesystem really needs it's own (which none currently does) it can just call into it's own routine directly. Rename the now static low-level dquot_transfer helper to __dquot_transfer and vfs_dq_transfer to dquot_transfer to have a consistent namespace, and make the new dquot_transfer return a normal negative errno value which all callers expect. Signed-off-by: Christoph Hellwig Signed-off-by: Jan Kara --- fs/ext4/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9f607ea411c8..6a002a6d0624 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5263,7 +5263,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) error = PTR_ERR(handle); goto err_out; } - error = vfs_dq_transfer(inode, attr) ? -EDQUOT : 0; + error = dquot_transfer(inode, attr); if (error) { ext4_journal_stop(handle); return error; -- cgit v1.2.1 From 907f4554e2521cb28b0009d17167760650a9561c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 3 Mar 2010 09:05:06 -0500 Subject: dquot: move dquot initialization responsibility into the filesystem Currently various places in the VFS call vfs_dq_init directly. This means we tie the quota code into the VFS. Get rid of that and make the filesystem responsible for the initialization. For most metadata operations this is a straight forward move into the methods, but for truncate and open it's a bit more complicated. For truncate we currently only call vfs_dq_init for the sys_truncate case because open already takes care of it for ftruncate and open(O_TRUNC) - the new code causes an additional vfs_dq_init for those which is harmless. For open the initialization is moved from do_filp_open into the open method, which means it happens slightly earlier now, and only for regular files. The latter is fine because we don't need to initialize it for operations on special files, and we already do it as part of the namespace operations for directories. Add a dquot_file_open helper that filesystems that support generic quotas can use to fill in ->open. Signed-off-by: Christoph Hellwig Signed-off-by: Jan Kara --- fs/ext4/inode.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 6a002a6d0624..eaa22ae9f1f6 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -170,6 +170,9 @@ void ext4_delete_inode(struct inode *inode) handle_t *handle; int err; + if (!is_bad_inode(inode)) + vfs_dq_init(inode); + if (ext4_should_order_data(inode)) ext4_begin_ordered_truncate(inode, 0); truncate_inode_pages(&inode->i_data, 0); @@ -5251,6 +5254,8 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) if (error) return error; + if (ia_valid & ATTR_SIZE) + vfs_dq_init(inode); if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) || (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) { handle_t *handle; -- cgit v1.2.1 From 871a293155a24554e153538d36e3a80fa169aefb Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 3 Mar 2010 09:05:07 -0500 Subject: dquot: cleanup dquot initialize routine Get rid of the initialize dquot operation - it is now always called from the filesystem and if a filesystem really needs it's own (which none currently does) it can just call into it's own routine directly. Rename the now static low-level dquot_initialize helper to __dquot_initialize and vfs_dq_init to dquot_initialize to have a consistent namespace. Signed-off-by: Christoph Hellwig Signed-off-by: Jan Kara --- fs/ext4/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index eaa22ae9f1f6..bec222ca9ba4 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -171,7 +171,7 @@ void ext4_delete_inode(struct inode *inode) int err; if (!is_bad_inode(inode)) - vfs_dq_init(inode); + dquot_initialize(inode); if (ext4_should_order_data(inode)) ext4_begin_ordered_truncate(inode, 0); @@ -5255,7 +5255,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) return error; if (ia_valid & ATTR_SIZE) - vfs_dq_init(inode); + dquot_initialize(inode); if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) || (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) { handle_t *handle; -- cgit v1.2.1 From a9185b41a4f84971b930c519f0c63bd450c4810d Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 5 Mar 2010 09:21:37 +0100 Subject: pass writeback_control to ->write_inode This gives the filesystem more information about the writeback that is happening. Trond requested this for the NFS unstable write handling, and other filesystems might benefit from this too by beeing able to distinguish between the different callers in more detail. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- fs/ext4/inode.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index e11952404e02..d01a6cdbf854 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5177,7 +5177,7 @@ out_brelse: * `stuff()' is running, and the new i_size will be lost. Plus the inode * will no longer be on the superblock's dirty inode list. */ -int ext4_write_inode(struct inode *inode, int wait) +int ext4_write_inode(struct inode *inode, struct writeback_control *wbc) { int err; @@ -5191,7 +5191,7 @@ int ext4_write_inode(struct inode *inode, int wait) return -EIO; } - if (!wait) + if (wbc->sync_mode != WB_SYNC_ALL) return 0; err = ext4_force_commit(inode->i_sb); @@ -5201,7 +5201,7 @@ int ext4_write_inode(struct inode *inode, int wait) err = ext4_get_inode_loc(inode, &iloc); if (err) return err; - if (wait) + if (wbc->sync_mode == WB_SYNC_ALL) sync_dirty_buffer(iloc.bh); if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) { ext4_error(inode->i_sb, __func__, -- cgit v1.2.1