From: Ryusuke Konishi <konishi.ryusuke-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: linux-nilfs <linux-nilfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
syzbot
<syzbot+ede796cecd5296353515-Pl5Pbv+GP7P466ipTTIvnc23WoclnBCfAL8bYrjMMd8@public.gmane.org>,
syzkaller-bugs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org,
LKML <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: [PATCH] nilfs2: fix general protection fault in nilfs_btree_insert()
Date: Thu, 5 Jan 2023 14:53:56 +0900 [thread overview]
Message-ID: <20230105055356.8811-1-konishi.ryusuke@gmail.com> (raw)
In-Reply-To: <000000000000bd89e205f0e38355-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
If nilfs2 reads a corrupted disk image and tries to reads a b-tree node
block by calling __nilfs_btree_get_block() against an invalid virtual
block address, it returns -ENOENT because conversion of the virtual
block address to a disk block address fails. However, this return
value is the same as the internal code that b-tree lookup routines
return to indicate that the block being searched does not exist, so
functions that operate on that b-tree may misbehave.
When nilfs_btree_insert() receives this spurious 'not found' code from
nilfs_btree_do_lookup(), it misunderstands that the 'not found' check
was successful and continues the insert operation using incomplete
lookup path data, causing the following crash:
general protection fault, probably for non-canonical address
0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
...
RIP: 0010:nilfs_btree_get_nonroot_node fs/nilfs2/btree.c:418 [inline]
RIP: 0010:nilfs_btree_prepare_insert fs/nilfs2/btree.c:1077 [inline]
RIP: 0010:nilfs_btree_insert+0x6d3/0x1c10 fs/nilfs2/btree.c:1238
Code: bc 24 80 00 00 00 4c 89 f8 48 c1 e8 03 42 80 3c 28 00 74 08 4c 89
ff e8 4b 02 92 fe 4d 8b 3f 49 83 c7 28 4c 89 f8 48 c1 e8 03 <42> 80 3c
28 00 74 08 4c 89 ff e8 2e 02 92 fe 4d 8b 3f 49 83 c7 02
...
Call Trace:
<TASK>
nilfs_bmap_do_insert fs/nilfs2/bmap.c:121 [inline]
nilfs_bmap_insert+0x20d/0x360 fs/nilfs2/bmap.c:147
nilfs_get_block+0x414/0x8d0 fs/nilfs2/inode.c:101
__block_write_begin_int+0x54c/0x1a80 fs/buffer.c:1991
__block_write_begin fs/buffer.c:2041 [inline]
block_write_begin+0x93/0x1e0 fs/buffer.c:2102
nilfs_write_begin+0x9c/0x110 fs/nilfs2/inode.c:261
generic_perform_write+0x2e4/0x5e0 mm/filemap.c:3772
__generic_file_write_iter+0x176/0x400 mm/filemap.c:3900
generic_file_write_iter+0xab/0x310 mm/filemap.c:3932
call_write_iter include/linux/fs.h:2186 [inline]
new_sync_write fs/read_write.c:491 [inline]
vfs_write+0x7dc/0xc50 fs/read_write.c:584
ksys_write+0x177/0x2a0 fs/read_write.c:637
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x3d/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
...
</TASK>
This patch fixes the root cause of this problem by replacing the error
code that __nilfs_btree_get_block() returns on block address conversion
failure from -ENOENT to another internal code -EINVAL which means that
the b-tree metadata is corrupted.
By returning -EINVAL, it propagates without glitches, and for all
relevant b-tree operations, functions in the upper bmap layer output
an error message indicating corrupted b-tree metadata via
nilfs_bmap_convert_error(), and code -EIO will be eventually returned
as it should be.
Link: https://lkml.kernel.org/r/000000000000bd89e205f0e38355-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org
Reported-by: syzbot+ede796cecd5296353515-Pl5Pbv+GP7P466ipTTIvnc23WoclnBCfAL8bYrjMMd8@public.gmane.org
Signed-off-by: Ryusuke Konishi <konishi.ryusuke-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Tested-by: Ryusuke Konishi <konishi.ryusuke-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
---
Hi Andrew, please apply this bugfix.
This fixes the kernel crash above for corrupted disk images like
the one syzbot produced, and presumably fixes some unexpected WARN_ONs
on nilfs2 as well (which I will continue to investigate).
Thanks,
Ryusuke Konishi
fs/nilfs2/btree.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c
index b9d15c3df3cc..40ce92a332fe 100644
--- a/fs/nilfs2/btree.c
+++ b/fs/nilfs2/btree.c
@@ -480,9 +480,18 @@ static int __nilfs_btree_get_block(const struct nilfs_bmap *btree, __u64 ptr,
ret = nilfs_btnode_submit_block(btnc, ptr, 0, REQ_OP_READ, &bh,
&submit_ptr);
if (ret) {
- if (ret != -EEXIST)
- return ret;
- goto out_check;
+ if (likely(ret == -EEXIST))
+ goto out_check;
+ if (ret == -ENOENT) {
+ /*
+ * Block address translation failed due to invalid
+ * value of 'ptr'. In this case, return internal code
+ * -EINVAL (broken bmap) to notify bmap layer of fatal
+ * metadata corruption.
+ */
+ ret = -EINVAL;
+ }
+ return ret;
}
if (ra) {
--
2.34.1
prev parent reply other threads:[~2023-01-05 5:53 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-28 13:40 [syzbot] [nilfs2?] general protection fault in nilfs_btree_insert syzbot
[not found] ` <000000000000bd89e205f0e38355-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2022-12-28 19:20 ` Ryusuke Konishi
2023-01-05 5:53 ` Ryusuke Konishi [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230105055356.8811-1-konishi.ryusuke@gmail.com \
--to=konishi.ryusuke-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-nilfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=syzbot+ede796cecd5296353515-Pl5Pbv+GP7P466ipTTIvnc23WoclnBCfAL8bYrjMMd8@public.gmane.org \
--cc=syzkaller-bugs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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 read-only IMAP folder(s) and NNTP newsgroup(s).