Re: general protection fault in relay_open_buf

From: Kees Cook
Date: Thu Jan 31 2019 - 04:54:37 EST


On Thu, Jan 31, 2019 at 7:53 AM syzbot
<syzbot+16c3a70e1e9b29346c43@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>
> Hello,
>
> syzbot found the following crash on:
>
> HEAD commit: 02495e76ded5 Add linux-next specific files for 20190130
> git tree: linux-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=12cf10df400000
> kernel config: https://syzkaller.appspot.com/x/.config?x=a2b2e9c0bc43c14d
> dashboard link: https://syzkaller.appspot.com/bug?extid=16c3a70e1e9b29346c43
> compiler: gcc (GCC) 9.0.0 20181231 (experimental)
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=13266698c00000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1715bb64c00000
>
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+16c3a70e1e9b29346c43@xxxxxxxxxxxxxxxxxxxxxxxxx
>
> kasan: CONFIG_KASAN_INLINE enabled
> kasan: GPF could be caused by NULL-ptr deref or user memory access
> general protection fault: 0000 [#1] PREEMPT SMP KASAN
> CPU: 0 PID: 8092 Comm: syz-executor405 Not tainted 5.0.0-rc4-next-20190130
> #22
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
> Google 01/01/2011
> RIP: 0010:relay_set_buf_dentry kernel/relay.c:412 [inline]

static inline void relay_set_buf_dentry(struct rchan_buf *buf,
struct dentry *dentry)
{
buf->dentry = dentry;
d_inode(buf->dentry)->i_size = buf->early_bytes; <--
}

Doing a bisect landed on this:

ff9fb72bc07705c00795ca48631f7fffe24d2c6b ("debugfs: return error
values, not NULL")

If I revert this patch, I can't reproduce any more. I don't see a
relationship, though...

My crash appears as:
[ 121.934378] BUG: unable to handle kernel NULL pointer dereference
at 0000000000000047
[ 121.937187] #PF error: [normal kernel read fault]
[ 121.938824] PGD 800000041f699067 P4D 800000041f699067 PUD 42d08f067 PMD 0
[ 121.941166] Oops: 0000 [#1] SMP PTI
[ 121.942381] CPU: 2 PID: 3134 Comm: relay Not tainted
5.0.0-rc4-next-20190130 #1020
[ 121.943873] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
BIOS 1.10.2-1ubuntu1 04/01/2014
[ 121.945395] RIP: 0010:relay_open_buf.part.10+0x2b8/0x330
...
[ 121.960021] Call Trace:
[ 121.960453] relay_open+0x18e/0x2c0
[ 121.961070] __blk_trace_setup+0x1af/0x350
[ 121.961777] blk_trace_ioctl+0x93/0x100


$ ./scripts/faddr2line vmlinux relay_open_buf.part.10+0x2b8/0x330
relay_open_buf.part.10+0x2b8/0x330:
relay_set_buf_dentry at kernel/relay.c:412
(inlined by) relay_open_buf at kernel/relay.c:458

So it's the same location, but not sure about 0x47 offset. d_inode is
0x58 from dentry. And i_size is 0x50 from inode. If this isn't NULL,
but rather an ERR_PTR, the errno is either:

EBADF 9 Bad file descriptor
EEXIST 17 File exists

Neither are used in the debugfs patch, but debugfs is clearly used in
do_blk_trace_setup():

if (!blk_debugfs_root)
return -ENOENT;
...
dir = debugfs_lookup(buts->name, blk_debugfs_root);
if (!dir)
bt->dir = dir = debugfs_create_dir(buts->name,
blk_debugfs_root);
if (!dir)
goto err;
...
bt->rchan = relay_open("trace", dir, buts->buf_size,
buts->buf_nr, &blk_relay_callbacks, bt);

Which is confirmed by the next line in my traceback:

$ ./scripts/faddr2line vmlinux __blk_trace_setup+0x1af/0x350
__blk_trace_setup+0x1af/0x350:
do_blk_trace_setup at kernel/trace/blktrace.c:534
(inlined by) __blk_trace_setup at kernel/trace/blktrace.c:577

-Kees

> RIP: 0010:relay_open_buf.part.0+0x7cb/0xb40 kernel/relay.c:458
> Code: c1 ea 03 80 3c 02 00 0f 85 4c 03 00 00 49 8d 7d 58 4d 89 ac 24 90 00
> 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f
> 85 1b 03 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8b
> RSP: 0018:ffff888088897670 EFLAGS: 00010203
> RAX: dffffc0000000000 RBX: ffff88809f2b3080 RCX: 1ffff110110c6919
> RDX: 0000000000000008 RSI: ffffffff81833b80 RDI: 0000000000000047
> RBP: ffff8880888976e8 R08: 0000000000000006 R09: ffff8880886348c8
> R10: ffff888088634000 R11: 0000000000000000 R12: ffff88809f5ef580
> R13: ffffffffffffffef R14: 0000000000000000 R15: 0000000000000000
> FS: 00000000021c4880(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00000000006cf090 CR3: 00000000a4ccd000 CR4: 00000000001406f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> relay_open_buf kernel/relay.c:447 [inline]
> relay_open kernel/relay.c:597 [inline]
> relay_open+0x5f3/0xaf0 kernel/relay.c:561
> do_blk_trace_setup+0x50d/0xdb0 kernel/trace/blktrace.c:532
> __blk_trace_setup+0xe3/0x190 kernel/trace/blktrace.c:577
> blk_trace_ioctl+0x170/0x300 kernel/trace/blktrace.c:716
> blkdev_ioctl+0x141/0x2120 block/ioctl.c:591
> block_ioctl+0xee/0x130 fs/block_dev.c:1914
> vfs_ioctl fs/ioctl.c:46 [inline]
> file_ioctl fs/ioctl.c:509 [inline]
> do_vfs_ioctl+0x107b/0x17d0 fs/ioctl.c:696
> ksys_ioctl+0xab/0xd0 fs/ioctl.c:713
> __do_sys_ioctl fs/ioctl.c:720 [inline]
> __se_sys_ioctl fs/ioctl.c:718 [inline]
> __x64_sys_ioctl+0x73/0xb0 fs/ioctl.c:718
> do_syscall_64+0x1a3/0x800 arch/x86/entry/common.c:290
> entry_SYSCALL_64_after_hwframe+0x49/0xbe
> RIP: 0033:0x444af9
> Code: e8 cc ab 02 00 48 83 c4 18 c3 0f 1f 80 00 00 00 00 48 89 f8 48 89 f7
> 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff
> ff 0f 83 db ce fb ff c3 66 2e 0f 1f 84 00 00 00 00
> RSP: 002b:00007ffd22903da8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
> RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 0000000000444af9
> RDX: 0000000020000040 RSI: 00000000c0481273 RDI: 0000000000000003
> RBP: 000000000000d8fa R08: 00000000021c4880 R09: 00000000004002e0
> R10: 000000000000000f R11: 0000000000000246 R12: 0000000000401e60
> R13: 0000000000401ef0 R14: 0000000000000000 R15: 0000000000000000
> Modules linked in:
> ---[ end trace d79fd6b85e15805c ]---
> RIP: 0010:relay_set_buf_dentry kernel/relay.c:412 [inline]
> RIP: 0010:relay_open_buf.part.0+0x7cb/0xb40 kernel/relay.c:458
> Code: c1 ea 03 80 3c 02 00 0f 85 4c 03 00 00 49 8d 7d 58 4d 89 ac 24 90 00
> 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f
> 85 1b 03 00 00 48 b8 00 00 00 00 00 fc ff df 4d 8b
> RSP: 0018:ffff888088897670 EFLAGS: 00010203
> RAX: dffffc0000000000 RBX: ffff88809f2b3080 RCX: 1ffff110110c6919
> RDX: 0000000000000008 RSI: ffffffff81833b80 RDI: 0000000000000047
> RBP: ffff8880888976e8 R08: 0000000000000006 R09: ffff8880886348c8
> R10: ffff888088634000 R11: 0000000000000000 R12: ffff88809f5ef580
> R13: ffffffffffffffef R14: 0000000000000000 R15: 0000000000000000
> FS: 00000000021c4880(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00000000006cf090 CR3: 00000000a4ccd000 CR4: 00000000001406f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>
>
> ---
> This bug is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@xxxxxxxxxxxxxxxxx
>
> syzbot will keep track of this bug report. See:
> https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with
> syzbot.
> syzbot can test patches for this bug, for details see:
> https://goo.gl/tpsmEJ#testing-patches



--
Kees Cook