linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] f2fs: fix to do sanity check on segment bitmap of LFS curseg
@ 2019-05-16  6:36 Chao Yu
  0 siblings, 0 replies; only message in thread
From: Chao Yu @ 2019-05-16  6:36 UTC (permalink / raw)
  To: jaegeuk; +Cc: linux-f2fs-devel, linux-kernel, chao, Chao Yu

As Jungyeon Reported in bugzilla:

https://bugzilla.kernel.org/show_bug.cgi?id=203233

- Reproduces
gcc poc_13.c
./run.sh f2fs

- Kernel messages
 F2FS-fs (sdb): Bitmap was wrongly set, blk:4608
 kernel BUG at fs/f2fs/segment.c:2133!
 RIP: 0010:update_sit_entry+0x35d/0x3e0
 Call Trace:
  f2fs_allocate_data_block+0x16c/0x5a0
  do_write_page+0x57/0x100
  f2fs_do_write_node_page+0x33/0xa0
  __write_node_page+0x270/0x4e0
  f2fs_sync_node_pages+0x5df/0x670
  f2fs_write_checkpoint+0x364/0x13a0
  f2fs_sync_fs+0xa3/0x130
  f2fs_do_sync_file+0x1a6/0x810
  do_fsync+0x33/0x60
  __x64_sys_fsync+0xb/0x10
  do_syscall_64+0x43/0x110
  entry_SYSCALL_64_after_hwframe+0x44/0xa9

The testcase fails because that, in fuzzed image, current segment was
allocated with LFS type, its .next_blkoff should point to an unused
block address, but actually, its bitmap shows it's not. So during
allocation, f2fs crash when setting bitmap.

Introducing sanity_check_curseg() to check such inconsistence of
current in-used segment.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/segment.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 8dee063c833f..4a25fb12bdb1 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -4232,6 +4232,35 @@ static int build_dirty_segmap(struct f2fs_sb_info *sbi)
 	return init_victim_secmap(sbi);
 }
 
+int sanity_check_curseg(struct f2fs_sb_info *sbi)
+{
+	int i, j;
+
+	/*
+	 * In current segment with LFS allocation type, all space after
+	 * .next_blkoff position should be all valid.
+	 */
+	for (i = 0; i < NO_CHECK_TYPE; i++) {
+		struct curseg_info *curseg = CURSEG_I(sbi, i);
+		struct seg_entry *se = get_seg_entry(sbi, curseg->segno);
+
+		if (curseg->alloc_type == SSR)
+			continue;
+
+		for (j = curseg->next_blkoff; j < sbi->blocks_per_seg; j++) {
+			if (!f2fs_test_bit(j, se->cur_valid_map))
+				continue;
+
+			f2fs_msg(sbi->sb, KERN_ERR,
+				"Current segment:%u, segno:%u, "
+				"next_blkoff:%u, cur:%u",
+				i, curseg->segno, curseg->next_blkoff, j);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
 /*
  * Update min, max modified time for cost-benefit GC algorithm
  */
@@ -4327,6 +4356,10 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi)
 	if (err)
 		return err;
 
+	err = sanity_check_curseg(sbi);
+	if (err)
+		return err;
+
 	init_min_max_mtime(sbi);
 	return 0;
 }
-- 
2.18.0.rc1


^ permalink raw reply related	[flat|nested] only message in thread

only message in thread, other threads:[~2019-05-16  6:37 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-16  6:36 [PATCH] f2fs: fix to do sanity check on segment bitmap of LFS curseg Chao Yu

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).