Thread (1 message) 1 message, 1 author, 2024-07-23

Re: [syzbot] [net?] WARNING in skb_warn_bad_offload (5)

From: Jakub Sitnicki <jakub@cloudflare.com>
Date: 2024-07-23 20:04:39
Also in: lkml

On Fri, Jul 19, 2024 at 09:34 PM +02, Jakub Sitnicki wrote:
On Tue, Jul 16, 2024 at 07:54 AM -07, Eric Dumazet wrote:
quoted
On Tue, Jul 16, 2024 at 3:17 AM Jakub Sitnicki [off-list ref] wrote:
quoted
On Tue, Jul 16, 2024 at 12:04 PM +02, Paolo Abeni wrote:
quoted
On 7/16/24 03:23, syzbot wrote:
quoted
syzbot found the following issue on:
HEAD commit:    80ab5445da62 Merge tag 'wireless-next-2024-07-11' of git:/..
git tree:       net-next
console+strace: https://syzkaller.appspot.com/x/log.txt?x=175fb821980000
kernel config:  https://syzkaller.appspot.com/x/.config?x=2dbcdd8641c4638f
dashboard link: https://syzkaller.appspot.com/bug?extid=e15b7e15b8a751a91d9a
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=172bf566980000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=12fff535980000
Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/184da3869c30/disk-80ab5445.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/85bfe9b60f21/vmlinux-80ab5445.xz
kernel image: https://storage.googleapis.com/syzbot-assets/06064623a948/bzImage-80ab5445.xz
The issue was bisected to:
commit 10154dbded6d6a2fecaebdfda206609de0f121a9
Author: Jakub Sitnicki [off-list ref]
Date:   Wed Jun 26 17:51:26 2024 +0000
     udp: Allow GSO transmit from devices with no checksum offload
bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=142ccbed980000
final oops:     https://syzkaller.appspot.com/x/report.txt?x=162ccbed980000
console output: https://syzkaller.appspot.com/x/log.txt?x=122ccbed980000
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+e15b7e15b8a751a91d9a@syzkaller.appspotmail.com
Fixes: 10154dbded6d ("udp: Allow GSO transmit from devices with no checksum offload")
skb frag:     00000080: 62 3f 77 e4 0e 82 0d 2f 85 cc 44 ea 25 5a 99 76
skb frag:     00000090: f2 53
------------[ cut here ]------------
ip6tnl0: caps=(0x00000006401d7869, 0x00000006401d7869)
WARNING: CPU: 0 PID: 5112 at net/core/dev.c:3293 skb_warn_bad_offload+0x166/0x1a0 net/core/dev.c:3291
Modules linked in:
CPU: 0 PID: 5112 Comm: syz-executor391 Not tainted 6.10.0-rc7-syzkaller-01603-g80ab5445da62 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 06/07/2024
RIP: 0010:skb_warn_bad_offload+0x166/0x1a0 net/core/dev.c:3291
Code: e8 5f 94 a3 f8 49 8b 04 24 48 8d 88 a0 03 00 00 48 85 c0 48 0f 44 cd 48 c7 c7 00 cc c5 8c 4c 89 f6 48 89 da e8 fb 92 ff f7 90 <0f> 0b 90 90 5b 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 44 89 f9
RSP: 0018:ffffc900034bedc8 EFLAGS: 00010246
RAX: 7d287cad4185da00 RBX: ffff888040cdc0b8 RCX: ffff888023d1bc00
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffffffff8cc5cbc0 R08: ffffffff815857b2 R09: fffffbfff1c39994
R10: dffffc0000000000 R11: fffffbfff1c39994 R12: ffff888022880518
R13: dffffc0000000000 R14: ffff888040cdc130 R15: ffff888040cdc130
FS:  000055556e9e9380(0000) GS:ffff8880b9400000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020001180 CR3: 000000007c876000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
  <TASK>
  __skb_gso_segment+0x3be/0x4c0 net/core/gso.c:127
  skb_gso_segment include/net/gso.h:83 [inline]
  validate_xmit_skb+0x585/0x1120 net/core/dev.c:3661
  __dev_queue_xmit+0x17a4/0x3e90 net/core/dev.c:4415
  neigh_output include/net/neighbour.h:542 [inline]
  ip6_finish_output2+0xffa/0x1680 net/ipv6/ip6_output.c:137
  ip6_finish_output+0x41e/0x810 net/ipv6/ip6_output.c:222
  ip6_send_skb+0x112/0x230 net/ipv6/ip6_output.c:1958
  udp_v6_send_skb+0xbf5/0x1870 net/ipv6/udp.c:1292
  udpv6_sendmsg+0x23b3/0x3270 net/ipv6/udp.c:1588
  sock_sendmsg_nosec net/socket.c:730 [inline]
  __sock_sendmsg+0xef/0x270 net/socket.c:745
  ____sys_sendmsg+0x525/0x7d0 net/socket.c:2585
  ___sys_sendmsg net/socket.c:2639 [inline]
  __sys_sendmmsg+0x3b2/0x740 net/socket.c:2725
  __do_sys_sendmmsg net/socket.c:2754 [inline]
  __se_sys_sendmmsg net/socket.c:2751 [inline]
  __x64_sys_sendmmsg+0xa0/0xb0 net/socket.c:2751
  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
  do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
  entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f04f688fe89
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 01 1a 00 00 90 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 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffeebc526e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000133
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f04f688fe89
RDX: 0000000000000001 RSI: 0000000020003cc0 RDI: 0000000000000003
RBP: 00000000000f4240 R08: 0000000000000000 R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffeebc52740
R13: 00007f04f68dd406 R14: 0000000000000003 R15: 00007ffeebc52720
  </TASK>
Looking at the console log, the the relevant GSO packet is an UFO one with
CSUM_NONE. commit 10154dbded6d6a2fecaebdfda206609de0f121a9 only adjust the skb
csum for USO packets. @Jakub S. could you please have a look?
Will do. Thanks for the hint.
The trigger for the bug is the following :

setsockopt(3, SOL_IPV6, IPV6_HOPOPTS,
"\0\3\0\0\0\0\0\0\5\2\0\0\0\1\0\302\4\200\0\0\0\5\2\0\6\302\4\0\0\0\1\302"...,
40) = 0

Some random IPV6_HOPTS, with multiple IPV6_TLV_JUMBO options

Non GSO path sends a malformed packet just fine, but GSO complains loudly.

(flowlabel 0x754ca, hlim 64, next-header Options (0) payload length:
186) localhost > localhost: HBH
(pad1)(pad1)(pad1)(pad1)(pad1)(pad1)(rtalert: 0x0000)
(pad1)(padn)(jumbo: 2147483648 - payload len != 0) (rtalert: 0x0006)
(jumbo: 1 - already seen)  [|hbhopt]
Thank you for the hint. Extracted a reproducer. Fix will follow.
Thanks for the patience.

I've got a fix to propose which pulls the gso_skb->ip_summed tweak added
to __udp_gso_segment() in commit 10154dbded6d ("udp: Allow GSO transmit
from devices with no checksum offload") from gso/udp code up to the udp
layer (udp[_v6]_send_skb()).

This warning can also be triggered for an ipv4 tunnel by turning off
csum offload (ethtool -K $tnl tx-checksum-ip-generic off). Will extend
udpgso.sh test to cover both v4 and v6.

[...]
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help