From 42c16da6d684391db83788eb680accd84f6c2083 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 1 Jul 2019 05:12:46 +0000 Subject: [PATCH 1/4] btrfs: inode: Don't compress if NODATASUM or NODATACOW set As btrfs(5) specified: Note If nodatacow or nodatasum are enabled, compression is disabled. If NODATASUM or NODATACOW set, we should not compress the extent. Normally NODATACOW is detected properly in run_delalloc_range() so compression won't happen for NODATACOW. However for NODATASUM we don't have any check, and it can cause compressed extent without csum pretty easily, just by: mkfs.btrfs -f $dev mount $dev $mnt -o nodatasum touch $mnt/foobar mount -o remount,datasum,compress $mnt xfs_io -f -c "pwrite 0 128K" $mnt/foobar And in fact, we have a bug report about corrupted compressed extent without proper data checksum so even RAID1 can't recover the corruption. (https://bugzilla.kernel.org/show_bug.cgi?id=199707) Running compression without proper checksum could cause more damage when corruption happens, as compressed data could make the whole extent unreadable, so there is no need to allow compression for NODATACSUM. The fix will refactor the inode compression check into two parts: - inode_can_compress() As the hard requirement, checked at btrfs_run_delalloc_range(), so no compression will happen for NODATASUM inode at all. - inode_need_compress() As the soft requirement, checked at btrfs_run_delalloc_range() and compress_file_range(). Reported-by: James Harvey CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/inode.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 1af069a9a0c7..ee582a36653d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -395,10 +395,31 @@ static noinline int add_async_extent(struct async_chunk *cow, return 0; } +/* + * Check if the inode has flags compatible with compression + */ +static inline bool inode_can_compress(struct inode *inode) +{ + if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW || + BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) + return false; + return true; +} + +/* + * Check if the inode needs to be submitted to compression, based on mount + * options, defragmentation, properties or heuristics. + */ static inline int inode_need_compress(struct inode *inode, u64 start, u64 end) { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + if (!inode_can_compress(inode)) { + WARN(IS_ENABLED(CONFIG_BTRFS_DEBUG), + KERN_ERR "BTRFS: unexpected compression for ino %llu\n", + btrfs_ino(BTRFS_I(inode))); + return 0; + } /* force compress */ if (btrfs_test_opt(fs_info, FORCE_COMPRESS)) return 1; @@ -1631,7 +1652,8 @@ int btrfs_run_delalloc_range(struct inode *inode, struct page *locked_page, } else if (BTRFS_I(inode)->flags & BTRFS_INODE_PREALLOC && !force_cow) { ret = run_delalloc_nocow(inode, locked_page, start, end, page_started, 0, nr_written); - } else if (!inode_need_compress(inode, start, end)) { + } else if (!inode_can_compress(inode) || + !inode_need_compress(inode, start, end)) { ret = cow_file_range(inode, locked_page, start, end, end, page_started, nr_written, 1, NULL); } else { From 314c4cd6d9e60b9412dcd1b1783a66532f91ea2d Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Tue, 2 Jul 2019 22:39:03 +0800 Subject: [PATCH 2/4] btrfs: Fix build error while LIBCRC32C is module If CONFIG_BTRFS_FS is y and CONFIG_LIBCRC32C is m, building fails: fs/btrfs/super.o: In function `btrfs_mount_root': super.c:(.text+0xb7f9): undefined reference to `crc32c_impl' fs/btrfs/super.o: In function `init_btrfs_fs': super.c:(.init.text+0x3465): undefined reference to `crc32c_impl' fs/btrfs/extent-tree.o: In function `hash_extent_data_ref': extent-tree.c:(.text+0xe60): undefined reference to `crc32c' extent-tree.c:(.text+0xe78): undefined reference to `crc32c' extent-tree.c:(.text+0xe8b): undefined reference to `crc32c' fs/btrfs/dir-item.o: In function `btrfs_insert_xattr_item': dir-item.c:(.text+0x291): undefined reference to `crc32c' fs/btrfs/dir-item.o: In function `btrfs_insert_dir_item': dir-item.c:(.text+0x429): undefined reference to `crc32c' Select LIBCRC32C to fix it. Reported-by: Hulk Robot Fixes: d5178578bcd4 ("btrfs: directly call into crypto framework for checksumming") Reviewed-by: Johannes Thumshirn Signed-off-by: YueHaibing Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig index 212b4a854f2c..38651fae7f21 100644 --- a/fs/btrfs/Kconfig +++ b/fs/btrfs/Kconfig @@ -4,6 +4,7 @@ config BTRFS_FS tristate "Btrfs filesystem support" select CRYPTO select CRYPTO_CRC32C + select LIBCRC32C select ZLIB_INFLATE select ZLIB_DEFLATE select LZO_COMPRESS From bfcea1c661eddabd7b5fee85dd3399bb16712c5a Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Thu, 11 Jul 2019 17:23:04 +0200 Subject: [PATCH 3/4] btrfs: free checksum hash on in close_ctree fs_info::csum_hash gets initialized in btrfs_init_csum_hash() which is called by open_ctree(). But it only gets freed if open_ctree() fails, not on normal operation. This leads to a memory leak like the following found by kmemleak: unreferenced object 0xffff888132cb8720 (size 96): comm "mount", pid 450, jiffies 4294912436 (age 17.584s) hex dump (first 32 bytes): 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<000000000c9643d4>] crypto_create_tfm+0x2d/0xd0 [<00000000ae577f68>] crypto_alloc_tfm+0x4b/0xb0 [<000000002b5cdf30>] open_ctree+0xb84/0x2060 [btrfs] [<0000000043204297>] btrfs_mount_root+0x552/0x640 [btrfs] [<00000000c99b10ea>] legacy_get_tree+0x22/0x40 [<0000000071a6495f>] vfs_get_tree+0x1f/0xc0 [<00000000f180080e>] fc_mount+0x9/0x30 [<000000009e36cebd>] vfs_kern_mount.part.11+0x6a/0x80 [<0000000004594c05>] btrfs_mount+0x174/0x910 [btrfs] [<00000000c99b10ea>] legacy_get_tree+0x22/0x40 [<0000000071a6495f>] vfs_get_tree+0x1f/0xc0 [<00000000b86e92c5>] do_mount+0x6b0/0x940 [<0000000097464494>] ksys_mount+0x7b/0xd0 [<0000000057213c80>] __x64_sys_mount+0x1c/0x20 [<00000000cb689b5e>] do_syscall_64+0x43/0x130 [<000000002194e289>] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Free fs_info::csum_hash in close_ctree() to avoid the memory leak. Fixes: 6d97c6e31b55 ("btrfs: add boilerplate code for directly including the crypto framework") Reviewed-by: Qu Wenruo Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 41a2bd2e0c56..5f7ee70b3d1a 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4106,6 +4106,7 @@ void close_ctree(struct btrfs_fs_info *fs_info) percpu_counter_destroy(&fs_info->dev_replace.bio_counter); cleanup_srcu_struct(&fs_info->subvol_srcu); + btrfs_free_csum_hash(fs_info); btrfs_free_stripe_hash_table(fs_info); btrfs_free_ref_cache(fs_info); } From 373c3b80e459cb57c34381b928588a3794eb5bbd Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Mon, 15 Jul 2019 15:16:12 +0200 Subject: [PATCH 4/4] btrfs: don't leak extent_map in btrfs_get_io_geometry() btrfs_get_io_geometry() calls btrfs_get_chunk_map() to acquire a reference on a extent_map, but on normal operation it does not drop this reference anymore. This leads to excessive kmemleak reports. Always call free_extent_map(), not just in the error case. Fixes: 5f1411265e16 ("btrfs: Introduce btrfs_io_geometry infrastructure") Reviewed-by: Nikolay Borisov Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index a13ddba1ebc3..d74b74ca07af 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -5941,6 +5941,7 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, u64 stripe_len; u64 raid56_full_stripe_start = (u64)-1; int data_stripes; + int ret = 0; ASSERT(op != BTRFS_MAP_DISCARD); @@ -5961,8 +5962,8 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, btrfs_crit(fs_info, "stripe math has gone wrong, stripe_offset=%llu offset=%llu start=%llu logical=%llu stripe_len=%llu", stripe_offset, offset, em->start, logical, stripe_len); - free_extent_map(em); - return -EINVAL; + ret = -EINVAL; + goto out; } /* stripe_offset is the offset of this block in its stripe */ @@ -6009,7 +6010,10 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, io_geom->stripe_offset = stripe_offset; io_geom->raid56_stripe_offset = raid56_full_stripe_start; - return 0; +out: + /* once for us */ + free_extent_map(em); + return ret; } static int __btrfs_map_block(struct btrfs_fs_info *fs_info,