On Tue, Jan 31, 2017 at 2:17 PM, Andrey Konovalov <[email protected]> wrote:
> Hi,
>
> I've got the following error report while running the syzkaller fuzzer.
>
> On commit 566cf877a1fcb6d6dc0126b076aad062054c2637 (4.10-rc6).
>
> The fuzzer hits this issue quite often, but I don't have a working reproducer.
I still see this on 5a7ad1146caa895ad718a534399e38bd2ba721b7 (4.11-rc8).
I also have a reproducer now, attached.
It takes around 10 seconds for the warning to trigger.
>
> WARNING: CPU: 3 PID: 7091 at net/core/skbuff.c:4331
> skb_try_coalesce+0x14b1/0x1f80 net/core/skbuff.c:4331
> Kernel panic - not syncing: panic_on_warn set ...
>
> CPU: 3 PID: 7091 Comm: syz-executor4 Not tainted 4.10.0-rc6+ #116
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
> Call Trace:
> __dump_stack lib/dump_stack.c:15 [inline]
> dump_stack+0x292/0x398 lib/dump_stack.c:51
> panic+0x1cb/0x3a9 kernel/panic.c:179
> __warn+0x1c4/0x1e0 kernel/panic.c:539
> warn_slowpath_null+0x2c/0x40 kernel/panic.c:582
> skb_try_coalesce+0x14b1/0x1f80 net/core/skbuff.c:4331
> tcp_try_coalesce+0x178/0x560 net/ipv4/tcp_input.c:4337
> tcp_queue_rcv+0x13c/0x7a0 net/ipv4/tcp_input.c:4572
> tcp_data_queue+0x97e/0x46b0 net/ipv4/tcp_input.c:4681
> tcp_rcv_state_process+0xd92/0x42d0 net/ipv4/tcp_input.c:6130
> tcp_v4_do_rcv+0x571/0x940 net/ipv4/tcp_ipv4.c:1426
> sk_backlog_rcv include/net/sock.h:893 [inline]
> __release_sock+0x127/0x3a0 net/core/sock.c:2053
> release_sock+0xa5/0x2b0 net/core/sock.c:2540
> inet_shutdown+0x1a8/0x350 net/ipv4/af_inet.c:833
> SYSC_shutdown net/socket.c:1840 [inline]
> SyS_shutdown+0x253/0x2f0 net/socket.c:1831
> entry_SYSCALL_64_fastpath+0x1f/0xc2
> RIP: 0033:0x4457d9
> RSP: 002b:00007f73b41a5b58 EFLAGS: 00000282 ORIG_RAX: 0000000000000030
> RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 00000000004457d9
> RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000005
> RBP: 00000000006e2170 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000282 R12: 0000000000708000
> R13: 0000000000000005 R14: 0000000000000029 R15: 000000000000003f
> Dumping ftrace buffer:
> (ftrace buffer empty)
> Kernel Offset: disabled
> Rebooting in 86400 seconds..
On Wed, Apr 26, 2017 at 5:08 AM, Andrey Konovalov <[email protected]> wrote:
> On Tue, Jan 31, 2017 at 2:17 PM, Andrey Konovalov <[email protected]> wrote:
>> Hi,
>>
>> I've got the following error report while running the syzkaller fuzzer.
>>
>> On commit 566cf877a1fcb6d6dc0126b076aad062054c2637 (4.10-rc6).
>>
>> The fuzzer hits this issue quite often, but I don't have a working reproducer.
>
> I still see this on 5a7ad1146caa895ad718a534399e38bd2ba721b7 (4.11-rc8).
>
> I also have a reproducer now, attached.
>
> It takes around 10 seconds for the warning to trigger.
>
It does not trigger for me, but seeing that you use SO_ATTACH_FILTER,
I have a pretty good idea how to remove this splat.
A call to skb_condense() from ___pskb_trim() will likely help, I will
post a patch after my commute.
Thanks.