hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I97HJA CVE: NA
--------------------------------
In ext4 continue & no-journal mode, physical blocks could be allocated more than once (caused by writing extent entries failed & reclaiming extent cache) in preallocation process, which could trigger a BUG_ON (pa->pa_free < len) in ext4_mb_use_inode_pa().
kernel BUG at fs/ext4/mballoc.c:4681! invalid opcode: 0000 [#1] PREEMPT SMP CPU: 3 PID: 97 Comm: kworker/u8:3 Not tainted 6.8.0-rc7 RIP: 0010:ext4_mb_use_inode_pa+0x1b6/0x1e0 Call Trace: ext4_mb_use_preallocated.constprop.0+0x19e/0x540 ext4_mb_new_blocks+0x220/0x1f30 ext4_ext_map_blocks+0xf3c/0x2900 ext4_map_blocks+0x264/0xa40 ext4_do_writepages+0xb15/0x1400 do_writepages+0x8c/0x260 writeback_sb_inodes+0x224/0x720 wb_writeback+0xd8/0x580 wb_workfn+0x148/0x820
Details are shown as following:
0. Given a file with i_size=4096 with one mapped block 1. Write block no 1, blocks 1~3 are preallocated. ext4_ext_map_blocks ext4_mb_normalize_request size = 16 * 1024 size = end - start // Allocate 3 blocks (bs = 4096) ext4_mb_regular_allocator ext4_mb_regular_allocator ext4_mb_regular_allocator ext4_mb_use_inode_pa pa->pa_free -= len // 3 - 1 = 2 2. Extent buffer head is written failed, es cache and buffer head are reclaimed. 3. Write blocks 1~3 ext4_ext_map_blocks newex.ee_len = 3 ext4_ext_check_overlap // Find nothing, there should have been block 1 allocated = map->m_len // 3 ext4_mb_new_blocks ext4_mb_use_preallocated ext4_mb_use_inode_pa BUG_ON(pa->pa_free < len) // 2 < 3!
Fix it by adding validation checking for inode pa. If invalid pa is detected, stop using inode preallocation, drop invalid pa to avoid it being used again, mark group block bitmap as corrupted to avoid allocating from the erroneous group.
Fetch a reproducer in Link.
Cc: stable@vger.kernel.org Link: https://bugzilla.kernel.org/show_bug.cgi?id=218576 Signed-off-by: Zhihao Cheng chengzhihao1@huawei.com Signed-off-by: Zhang Yi yi.zhang@huawei.com --- fs/ext4/mballoc.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 62 insertions(+)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index c4f018dc0b59..2d33c0123b72 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -352,6 +352,9 @@ static void ext4_mb_generate_from_freelist(struct super_block *sb, void *bitmap, ext4_group_t group); static void ext4_mb_new_preallocation(struct ext4_allocation_context *ac);
+static void ext4_mb_put_pa(struct ext4_allocation_context *ac, + struct super_block *sb, struct ext4_prealloc_space *pa); + /* * The algorithm using this percpu seq counter goes below: * 1. We sample the percpu discard_pa_seq counter before trying for block @@ -3812,6 +3815,47 @@ static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac) pa->pa_free += ac->ac_b_ex.fe_len; }
+/* + * check if found pa is valid + */ +static bool ext4_mb_pa_is_valid(struct ext4_allocation_context *ac, + struct ext4_prealloc_space *pa) +{ + struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb); + ext4_fsblk_t start; + ext4_fsblk_t end; + int len; + + start = pa->pa_pstart + (ac->ac_o_ex.fe_logical - pa->pa_lstart); + end = min(pa->pa_pstart + EXT4_C2B(sbi, pa->pa_len), + start + EXT4_C2B(sbi, ac->ac_o_ex.fe_len)); + len = EXT4_NUM_B2C(sbi, end - start); + + if (unlikely(start < pa->pa_pstart)) { + ext4_msg(ac->ac_sb, KERN_ERR, + "invalid pa, start(%llu) < pa_pstart(%llu)", + start, pa->pa_pstart); + return false; + } + if (unlikely(end > pa->pa_pstart + EXT4_C2B(sbi, pa->pa_len))) { + ext4_msg(ac->ac_sb, KERN_ERR, + "invalid pa, end(%llu) > pa_pstart(%llu) + pa_len(%d)", + end, pa->pa_pstart, EXT4_C2B(sbi, pa->pa_len)); + return false; + } + if (unlikely(pa->pa_free < len)) { + ext4_msg(ac->ac_sb, KERN_ERR, + "invalid pa, pa_free(%d) < len(%d)", pa->pa_free, len); + return false; + } + if (unlikely(len <= 0)) { + ext4_msg(ac->ac_sb, KERN_ERR, "invalid pa, len(%d) <= 0", len); + return false; + } + + return true; +} + /* * use blocks preallocated to inode */ @@ -3932,6 +3976,23 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
/* found preallocated blocks, use them */ spin_lock(&pa->pa_lock); + if (unlikely(!ext4_mb_pa_is_valid(ac, pa))) { + ext4_group_t group; + + pa->pa_free = 0; + atomic_inc(&pa->pa_count); + spin_unlock(&pa->pa_lock); + rcu_read_unlock(); + ext4_mb_put_pa(ac, ac->ac_sb, pa); + group = ext4_get_group_number(ac->ac_sb, pa->pa_pstart); + ext4_lock_group(ac->ac_sb, group); + ext4_mark_group_bitmap_corrupted(ac->ac_sb, group, + EXT4_GROUP_INFO_BBITMAP_CORRUPT); + ext4_unlock_group(ac->ac_sb, group); + ext4_error(ac->ac_sb, "drop pa and mark group %u block bitmap corrupted", + group); + goto try_group_pa; + } if (pa->pa_deleted == 0 && pa->pa_free) { atomic_inc(&pa->pa_count); ext4_mb_use_inode_pa(ac, pa); @@ -3944,6 +4005,7 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac) } rcu_read_unlock();
+try_group_pa: /* can we use group allocation? */ if (!(ac->ac_flags & EXT4_MB_HINT_GROUP_ALLOC)) return false;