From 4df031ff5876d94b48dd9ee486ba5522382a06b2 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 16 Jul 2021 20:20:21 +0800 Subject: ext4: check and update i_disksize properly After commit 3da40c7b0898 ("ext4: only call ext4_truncate when size <= isize"), i_disksize could always be updated to i_size in ext4_setattr(), and we could sure that i_disksize <= i_size since holding inode lock and if i_disksize < i_size there are delalloc writes pending in the range upto i_size. If the end of the current write is <= i_size, there's no need to touch i_disksize since writeback will push i_disksize upto i_size eventually. So we can switch to check i_size instead of i_disksize in ext4_da_write_end() when write to the end of the file. we also could remove ext4_mark_inode_dirty() together because we defer inode dirtying to generic_write_end() or ext4_da_write_inline_data_end(). Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210716122024.1105856-2-yi.zhang@huawei.com --- fs/ext4/inode.c | 34 ++++++++++++++++++---------------- 1 file changed, 18 insertions(+), 16 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2c33c795c4a7..76d9e25fe62c 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3084,35 +3084,37 @@ static int ext4_da_write_end(struct file *file, end = start + copied - 1; /* - * generic_write_end() will run mark_inode_dirty() if i_size - * changes. So let's piggyback the i_disksize mark_inode_dirty - * into that. + * Since we are holding inode lock, we are sure i_disksize <= + * i_size. We also know that if i_disksize < i_size, there are + * delalloc writes pending in the range upto i_size. If the end of + * the current write is <= i_size, there's no need to touch + * i_disksize since writeback will push i_disksize upto i_size + * eventually. If the end of the current write is > i_size and + * inside an allocated block (ext4_da_should_update_i_disksize() + * check), we need to update i_disksize here as neither + * ext4_writepage() nor certain ext4_writepages() paths not + * allocating blocks update i_disksize. + * + * Note that we defer inode dirtying to generic_write_end() / + * ext4_da_write_inline_data_end(). */ new_i_size = pos + copied; - if (copied && new_i_size > EXT4_I(inode)->i_disksize) { + if (copied && new_i_size > inode->i_size) { if (ext4_has_inline_data(inode) || - ext4_da_should_update_i_disksize(page, end)) { + ext4_da_should_update_i_disksize(page, end)) ext4_update_i_disksize(inode, new_i_size); - /* We need to mark inode dirty even if - * new_i_size is less that inode->i_size - * bu greater than i_disksize.(hint delalloc) - */ - ret = ext4_mark_inode_dirty(handle, inode); - } } if (write_mode != CONVERT_INLINE_DATA && ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && ext4_has_inline_data(inode)) - ret2 = ext4_da_write_inline_data_end(inode, pos, len, copied, + ret = ext4_da_write_inline_data_end(inode, pos, len, copied, page); else - ret2 = generic_write_end(file, mapping, pos, len, copied, + ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - copied = ret2; - if (ret2 < 0) - ret = ret2; + copied = ret; ret2 = ext4_journal_stop(handle); if (unlikely(ret2 && !ret)) ret = ret2; -- cgit v1.2.3 From 55ce2f649b9e88111270333a8127e23f4f8f42d7 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 16 Jul 2021 20:20:22 +0800 Subject: ext4: correct the error path of ext4_write_inline_data_end() Current error path of ext4_write_inline_data_end() is not correct. Firstly, it should pass out the error value if ext4_get_inode_loc() return fail, or else it could trigger infinite loop if we inject error here. And then it's better to add inode to orphan list if it return fail in ext4_journal_stop(), otherwise we could not restore inline xattr entry after power failure. Finally, we need to reset the 'ret' value if ext4_write_inline_data_end() return success in ext4_write_end() and ext4_journalled_write_end(), otherwise we could not get the error return value of ext4_journal_stop(). Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210716122024.1105856-3-yi.zhang@huawei.com --- fs/ext4/inline.c | 15 +++++---------- fs/ext4/inode.c | 7 +++++-- 2 files changed, 10 insertions(+), 12 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 70cb64db33f7..28b666f25ac2 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -733,25 +733,20 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, void *kaddr; struct ext4_iloc iloc; - if (unlikely(copied < len)) { - if (!PageUptodate(page)) { - copied = 0; - goto out; - } - } + if (unlikely(copied < len) && !PageUptodate(page)) + return 0; ret = ext4_get_inode_loc(inode, &iloc); if (ret) { ext4_std_error(inode->i_sb, ret); - copied = 0; - goto out; + return ret; } ext4_write_lock_xattr(inode, &no_expand); BUG_ON(!ext4_has_inline_data(inode)); kaddr = kmap_atomic(page); - ext4_write_inline_data(inode, &iloc, kaddr, pos, len); + ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); kunmap_atomic(kaddr); SetPageUptodate(page); /* clear page dirty so that writepages wouldn't work for us. */ @@ -760,7 +755,7 @@ int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, ext4_write_unlock_xattr(inode, &no_expand); brelse(iloc.bh); mark_inode_dirty(inode); -out: + return copied; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 76d9e25fe62c..2baa17285096 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1295,6 +1295,7 @@ static int ext4_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1321,13 +1322,14 @@ static int ext4_write_end(struct file *file, if (i_size_changed || inline_data) ret = ext4_mark_inode_dirty(handle, inode); +errout: if (pos + len > inode->i_size && !verity && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside * inode->i_size. So truncate them */ ext4_orphan_add(handle, inode); -errout: + ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; @@ -1410,6 +1412,7 @@ static int ext4_journalled_write_end(struct file *file, goto errout; } copied = ret; + ret = 0; } else if (unlikely(copied < len) && !PageUptodate(page)) { copied = 0; ext4_journalled_zero_new_buffers(handle, page, from, to); @@ -1439,6 +1442,7 @@ static int ext4_journalled_write_end(struct file *file, ret = ret2; } +errout: if (pos + len > inode->i_size && !verity && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1446,7 +1450,6 @@ static int ext4_journalled_write_end(struct file *file, */ ext4_orphan_add(handle, inode); -errout: ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; -- cgit v1.2.3 From 6984aef59814fb5c47b0e30c56e101186b5ebf8c Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 16 Jul 2021 20:20:23 +0800 Subject: ext4: factor out write end code of inline file Now that the inline_data file write end procedure are falled into the common write end functions, it is not clear. Factor them out and do some cleanup. This patch also drop ext4_da_write_inline_data_end() and switch to use ext4_write_inline_data_end() instead because we also need to do the same error processing if we failed to write data into inline entry. Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210716122024.1105856-4-yi.zhang@huawei.com --- fs/ext4/ext4.h | 3 -- fs/ext4/inline.c | 117 +++++++++++++++++++++++++++++-------------------------- fs/ext4/inode.c | 73 +++++++++++----------------------- 3 files changed, 84 insertions(+), 109 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index d71dcac3b97f..ea6c0aae8f1b 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -3533,9 +3533,6 @@ extern int ext4_da_write_inline_data_begin(struct address_space *mapping, unsigned flags, struct page **pagep, void **fsdata); -extern int ext4_da_write_inline_data_end(struct inode *inode, loff_t pos, - unsigned len, unsigned copied, - struct page *page); extern int ext4_try_add_inline_entry(handle_t *handle, struct ext4_filename *fname, struct inode *dir, struct inode *inode); diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 28b666f25ac2..d30709d42a27 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -729,34 +729,76 @@ convert: int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len, unsigned copied, struct page *page) { - int ret, no_expand; + handle_t *handle = ext4_journal_current_handle(); + int no_expand; void *kaddr; struct ext4_iloc iloc; + int ret = 0, ret2; if (unlikely(copied < len) && !PageUptodate(page)) - return 0; + copied = 0; - ret = ext4_get_inode_loc(inode, &iloc); - if (ret) { - ext4_std_error(inode->i_sb, ret); - return ret; - } + if (likely(copied)) { + ret = ext4_get_inode_loc(inode, &iloc); + if (ret) { + unlock_page(page); + put_page(page); + ext4_std_error(inode->i_sb, ret); + goto out; + } + ext4_write_lock_xattr(inode, &no_expand); + BUG_ON(!ext4_has_inline_data(inode)); - ext4_write_lock_xattr(inode, &no_expand); - BUG_ON(!ext4_has_inline_data(inode)); + kaddr = kmap_atomic(page); + ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); + kunmap_atomic(kaddr); + SetPageUptodate(page); + /* clear page dirty so that writepages wouldn't work for us. */ + ClearPageDirty(page); - kaddr = kmap_atomic(page); - ext4_write_inline_data(inode, &iloc, kaddr, pos, copied); - kunmap_atomic(kaddr); - SetPageUptodate(page); - /* clear page dirty so that writepages wouldn't work for us. */ - ClearPageDirty(page); + ext4_write_unlock_xattr(inode, &no_expand); + brelse(iloc.bh); - ext4_write_unlock_xattr(inode, &no_expand); - brelse(iloc.bh); - mark_inode_dirty(inode); + /* + * It's important to update i_size while still holding page + * lock: page writeout could otherwise come in and zero + * beyond i_size. + */ + ext4_update_inode_size(inode, pos + copied); + } + unlock_page(page); + put_page(page); + + /* + * Don't mark the inode dirty under page lock. First, it unnecessarily + * makes the holding time of page lock longer. Second, it forces lock + * ordering of page lock and transaction start for journaling + * filesystems. + */ + if (likely(copied)) + mark_inode_dirty(inode); +out: + /* + * If we didn't copy as much data as expected, we need to trim back + * size of xattr containing inline data. + */ + if (pos + len > inode->i_size && ext4_can_truncate(inode)) + ext4_orphan_add(handle, inode); - return copied; + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + if (pos + len > inode->i_size) { + ext4_truncate_failed_write(inode); + /* + * If truncate failed early the inode might still be + * on the orphan list; we need to make sure the inode + * is removed from the orphan list in that case. + */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + } + return ret ? ret : copied; } struct buffer_head * @@ -937,43 +979,6 @@ out: return ret; } -int ext4_da_write_inline_data_end(struct inode *inode, loff_t pos, - unsigned len, unsigned copied, - struct page *page) -{ - int ret; - - ret = ext4_write_inline_data_end(inode, pos, len, copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - return ret; - } - copied = ret; - - /* - * No need to use i_size_read() here, the i_size - * cannot change under us because we hold i_mutex. - * - * But it's important to update i_size while still holding page lock: - * page writeout could otherwise come in and zero beyond i_size. - */ - if (pos+copied > inode->i_size) - i_size_write(inode, pos+copied); - unlock_page(page); - put_page(page); - - /* - * Don't mark the inode dirty under page lock. First, it unnecessarily - * makes the holding time of page lock longer. Second, it forces lock - * ordering of page lock and transaction start for journaling - * filesystems. - */ - mark_inode_dirty(inode); - - return copied; -} - #ifdef INLINE_DIR_DEBUG void ext4_show_inline_dir(struct inode *dir, struct buffer_head *bh, void *inline_start, int inline_size) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2baa17285096..b80f15bba727 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1282,23 +1282,14 @@ static int ext4_write_end(struct file *file, loff_t old_size = inode->i_size; int ret = 0, ret2; int i_size_changed = 0; - int inline_data = ext4_has_inline_data(inode); bool verity = ext4_verity_in_progress(inode); trace_ext4_write_end(inode, pos, len, copied); - if (inline_data) { - ret = ext4_write_inline_data_end(inode, pos, len, - copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - goto errout; - } - copied = ret; - ret = 0; - } else - copied = block_write_end(file, mapping, pos, - len, copied, page, fsdata); + + if (ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + + copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); /* * it's important to update i_size while still holding page lock: * page writeout could otherwise come in and zero beyond i_size. @@ -1319,10 +1310,9 @@ static int ext4_write_end(struct file *file, * ordering of page lock and transaction start for journaling * filesystems. */ - if (i_size_changed || inline_data) + if (i_size_changed) ret = ext4_mark_inode_dirty(handle, inode); -errout: if (pos + len > inode->i_size && !verity && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1394,7 +1384,6 @@ static int ext4_journalled_write_end(struct file *file, int partial = 0; unsigned from, to; int size_changed = 0; - int inline_data = ext4_has_inline_data(inode); bool verity = ext4_verity_in_progress(inode); trace_ext4_journalled_write_end(inode, pos, len, copied); @@ -1403,17 +1392,10 @@ static int ext4_journalled_write_end(struct file *file, BUG_ON(!ext4_handle_valid(handle)); - if (inline_data) { - ret = ext4_write_inline_data_end(inode, pos, len, - copied, page); - if (ret < 0) { - unlock_page(page); - put_page(page); - goto errout; - } - copied = ret; - ret = 0; - } else if (unlikely(copied < len) && !PageUptodate(page)) { + if (ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + + if (unlikely(copied < len) && !PageUptodate(page)) { copied = 0; ext4_journalled_zero_new_buffers(handle, page, from, to); } else { @@ -1436,13 +1418,12 @@ static int ext4_journalled_write_end(struct file *file, if (old_size < pos && !verity) pagecache_isize_extended(inode, old_size, pos); - if (size_changed || inline_data) { + if (size_changed) { ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; } -errout: if (pos + len > inode->i_size && !verity && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -3072,7 +3053,7 @@ static int ext4_da_write_end(struct file *file, struct page *page, void *fsdata) { struct inode *inode = mapping->host; - int ret = 0, ret2; + int ret; handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; @@ -3083,6 +3064,12 @@ static int ext4_da_write_end(struct file *file, len, copied, page, fsdata); trace_ext4_da_write_end(inode, pos, len, copied); + + if (write_mode != CONVERT_INLINE_DATA && + ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && + ext4_has_inline_data(inode)) + return ext4_write_inline_data_end(inode, pos, len, copied, page); + start = pos & (PAGE_SIZE - 1); end = start + copied - 1; @@ -3102,26 +3089,12 @@ static int ext4_da_write_end(struct file *file, * ext4_da_write_inline_data_end(). */ new_i_size = pos + copied; - if (copied && new_i_size > inode->i_size) { - if (ext4_has_inline_data(inode) || - ext4_da_should_update_i_disksize(page, end)) - ext4_update_i_disksize(inode, new_i_size); - } - - if (write_mode != CONVERT_INLINE_DATA && - ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) && - ext4_has_inline_data(inode)) - ret = ext4_da_write_inline_data_end(inode, pos, len, copied, - page); - else - ret = generic_write_end(file, mapping, pos, len, copied, - page, fsdata); - - copied = ret; - ret2 = ext4_journal_stop(handle); - if (unlikely(ret2 && !ret)) - ret = ret2; + if (copied && new_i_size > inode->i_size && + ext4_da_should_update_i_disksize(page, end)) + ext4_update_i_disksize(inode, new_i_size); + copied = generic_write_end(file, mapping, pos, len, copied, page, fsdata); + ret = ext4_journal_stop(handle); return ret ? ret : copied; } -- cgit v1.2.3 From cc883236b79297f6266ca6f4e7f24f3fd3c736c1 Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 16 Jul 2021 20:20:24 +0800 Subject: ext4: drop unnecessary journal handle in delalloc write After we factor out the inline data write procedure from ext4_da_write_end(), we don't need to start journal handle for the cases of both buffer overwrite and append-write. If we need to update i_disksize, mark_inode_dirty() do start handle and update inode buffer. So we could just remove all the journal handle codes in the delalloc write procedure. After this patch, we could get a lot of performance improvement. Below is the Unixbench comparison data test on my machine with 'Intel Xeon Gold 5120' CPU and nvme SSD backend. Test cmd: ./Run -c 56 -i 3 fstime fsbuffer fsdisk Before this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 422965.0 1068.1 File Copy 256 bufsize 500 maxblocks 1655.0 105077.0 634.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 1429092.0 2464.0 ====== System Benchmarks Index Score (Partial Only) 1186.6 After this patch: System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 732716.0 1850.3 File Copy 256 bufsize 500 maxblocks 1655.0 184940.0 1117.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 2427152.0 4184.7 ====== System Benchmarks Index Score (Partial Only) 2053.0 Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210716122024.1105856-5-yi.zhang@huawei.com --- fs/ext4/inode.c | 60 +++++---------------------------------------------------- 1 file changed, 5 insertions(+), 55 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index b80f15bba727..502f60621bad 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2910,19 +2910,6 @@ static int ext4_nonda_switch(struct super_block *sb) return 0; } -/* We always reserve for an inode update; the superblock could be there too */ -static int ext4_da_write_credits(struct inode *inode, loff_t pos, unsigned len) -{ - if (likely(ext4_has_feature_large_file(inode->i_sb))) - return 1; - - if (pos + len <= 0x7fffffffULL) - return 1; - - /* We might need to update the superblock to set LARGE_FILE */ - return 2; -} - static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -2931,7 +2918,6 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, struct page *page; pgoff_t index; struct inode *inode = mapping->host; - handle_t *handle; if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; @@ -2957,41 +2943,11 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, return 0; } - /* - * grab_cache_page_write_begin() can take a long time if the - * system is thrashing due to memory pressure, or if the page - * is being written back. So grab it first before we start - * the transaction handle. This also allows us to allocate - * the page (if needed) without using GFP_NOFS. - */ -retry_grab: +retry: page = grab_cache_page_write_begin(mapping, index, flags); if (!page) return -ENOMEM; - unlock_page(page); - - /* - * With delayed allocation, we don't log the i_disksize update - * if there is delayed block allocation. But we still need - * to journalling the i_disksize update if writes to the end - * of file which has an already mapped buffer. - */ -retry_journal: - handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, - ext4_da_write_credits(inode, pos, len)); - if (IS_ERR(handle)) { - put_page(page); - return PTR_ERR(handle); - } - lock_page(page); - if (page->mapping != mapping) { - /* The page got truncated from under us */ - unlock_page(page); - put_page(page); - ext4_journal_stop(handle); - goto retry_grab; - } /* In case writeback began while the page was unlocked */ wait_for_stable_page(page); @@ -3003,20 +2959,18 @@ retry_journal: #endif if (ret < 0) { unlock_page(page); - ext4_journal_stop(handle); + put_page(page); /* * block_write_begin may have instantiated a few blocks * outside i_size. Trim these off again. Don't need - * i_size_read because we hold i_mutex. + * i_size_read because we hold inode lock. */ if (pos + len > inode->i_size) ext4_truncate_failed_write(inode); if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) - goto retry_journal; - - put_page(page); + goto retry; return ret; } @@ -3053,8 +3007,6 @@ static int ext4_da_write_end(struct file *file, struct page *page, void *fsdata) { struct inode *inode = mapping->host; - int ret; - handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; int write_mode = (int)(unsigned long)fsdata; @@ -3093,9 +3045,7 @@ static int ext4_da_write_end(struct file *file, ext4_da_should_update_i_disksize(page, end)) ext4_update_i_disksize(inode, new_i_size); - copied = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - ret = ext4_journal_stop(handle); - return ret ? ret : copied; + return generic_write_end(file, mapping, pos, len, copied, page, fsdata); } /* -- cgit v1.2.3 From 948ca5f30e1df0c11eb5b0f410b9ceb97fa77ad9 Mon Sep 17 00:00:00 2001 From: Eric Whitney Date: Thu, 19 Aug 2021 10:49:27 -0400 Subject: ext4: enforce buffer head state assertion in ext4_da_map_blocks Remove the code that re-initializes a buffer head with an invalid block number and BH_New and BH_Delay bits when a matching delayed and unwritten block has been found in the extent status cache. Replace it with assertions that verify the buffer head already has this state correctly set. The current code masked an inline data truncation bug that left stale entries in the extent status cache. With this change, generic/130 can be used to reproduce and detect that bug. Signed-off-by: Eric Whitney Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210819144927.25163-3-enwlinux@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8204176256c8..2a076d236ba1 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1706,13 +1706,16 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock, } /* - * Delayed extent could be allocated by fallocate. - * So we need to check it. + * the buffer head associated with a delayed and not unwritten + * block found in the extent status cache must contain an + * invalid block number and have its BH_New and BH_Delay bits + * set, reflecting the state assigned when the block was + * initially delayed allocated */ - if (ext4_es_is_delayed(&es) && !ext4_es_is_unwritten(&es)) { - map_bh(bh, inode->i_sb, invalid_block); - set_buffer_new(bh); - set_buffer_delay(bh); + if (ext4_es_is_delonly(&es)) { + BUG_ON(bh->b_blocknr != invalid_block); + BUG_ON(!buffer_new(bh)); + BUG_ON(!buffer_delay(bh)); return 0; } -- cgit v1.2.3 From 6fed83957f21eff11c8496e9f24253b03d2bc1dc Mon Sep 17 00:00:00 2001 From: Jeffle Xu Date: Mon, 23 Aug 2021 14:13:58 +0800 Subject: ext4: fix reserved space counter leakage When ext4_insert_delayed block receives and recovers from an error from ext4_es_insert_delayed_block(), e.g., ENOMEM, it does not release the space it has reserved for that block insertion as it should. One effect of this bug is that s_dirtyclusters_counter is not decremented and remains incorrectly elevated until the file system has been unmounted. This can result in premature ENOSPC returns and apparent loss of free space. Another effect of this bug is that /sys/fs/ext4//delayed_allocation_blocks can remain non-zero even after syncfs has been executed on the filesystem. Besides, add check for s_dirtyclusters_counter when inode is going to be evicted and freed. s_dirtyclusters_counter can still keep non-zero until inode is written back in .evict_inode(), and thus the check is delayed to .destroy_inode(). Fixes: 51865fda28e5 ("ext4: let ext4 maintain extent status tree") Cc: stable@kernel.org Suggested-by: Gao Xiang Signed-off-by: Jeffle Xu Reviewed-by: Eric Whitney Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210823061358.84473-1-jefflexu@linux.alibaba.com --- fs/ext4/inode.c | 5 +++++ fs/ext4/super.c | 6 ++++++ 2 files changed, 11 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 2a076d236ba1..9df1ab070fa5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1628,6 +1628,7 @@ static int ext4_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk) struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); int ret; bool allocated = false; + bool reserved = false; /* * If the cluster containing lblk is shared with a delayed, @@ -1644,6 +1645,7 @@ static int ext4_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk) ret = ext4_da_reserve_space(inode); if (ret != 0) /* ENOSPC */ goto errout; + reserved = true; } else { /* bigalloc */ if (!ext4_es_scan_clu(inode, &ext4_es_is_delonly, lblk)) { if (!ext4_es_scan_clu(inode, @@ -1656,6 +1658,7 @@ static int ext4_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk) ret = ext4_da_reserve_space(inode); if (ret != 0) /* ENOSPC */ goto errout; + reserved = true; } else { allocated = true; } @@ -1666,6 +1669,8 @@ static int ext4_insert_delayed_block(struct inode *inode, ext4_lblk_t lblk) } ret = ext4_es_insert_delayed_block(inode, lblk, allocated); + if (ret && reserved) + ext4_da_release_space(inode, 1); errout: return ret; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index feca816b6bf3..a52f1572daa5 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1352,6 +1352,12 @@ static void ext4_destroy_inode(struct inode *inode) true); dump_stack(); } + + if (EXT4_I(inode)->i_reserved_data_blocks) + ext4_msg(inode->i_sb, KERN_ERR, + "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!", + inode->i_ino, EXT4_I(inode), + EXT4_I(inode)->i_reserved_data_blocks); } static void init_once(void *foo) -- cgit v1.2.3 From f2c77973507fd116c3657df1dc048864a2b16a1c Mon Sep 17 00:00:00 2001 From: Zhang Yi Date: Fri, 10 Sep 2021 16:03:16 +0800 Subject: ext4: recheck buffer uptodate bit under buffer lock Commit 8e33fadf945a ("ext4: remove an unnecessary if statement in __ext4_get_inode_loc()") forget to recheck buffer's uptodate bit again under buffer lock, which may overwrite the buffer if someone else have already brought it uptodate and changed it. Fixes: 8e33fadf945a ("ext4: remove an unnecessary if statement in __ext4_get_inode_loc()") Cc: stable@kernel.org Signed-off-by: Zhang Yi Reviewed-by: Jan Kara Signed-off-by: Theodore Ts'o Link: https://lore.kernel.org/r/20210910080316.70421-1-yi.zhang@huawei.com --- fs/ext4/inode.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'fs/ext4/inode.c') diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9df1ab070fa5..b7c561ad38dd 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4277,6 +4277,12 @@ static int __ext4_get_inode_loc(struct super_block *sb, unsigned long ino, goto has_buffer; lock_buffer(bh); + if (ext4_buffer_uptodate(bh)) { + /* Someone brought it uptodate while we waited */ + unlock_buffer(bh); + goto has_buffer; + } + /* * If we have all information of the inode in memory and this * is the only valid inode in the block, we need not read the -- cgit v1.2.3