[syzbot] WARNING in handle_exception_nmi (2)

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 80d47f5de5e3 mm: don't try to NUMA-migrate COW pages that ..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=12a7324c700000
kernel config: https://syzkaller.appspot.com/x/.config?x=f6a069ed94a1ed1d
dashboard link: https://syzkaller.appspot.com/bug?extid=4688c50a9c8e68e7aaa1
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2

Unfortunately, I don't have any reproducer for this issue yet.

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+4688c5...@syzkaller.appspotmail.com

------------[ cut here ]------------
WARNING: CPU: 0 PID: 7975 at arch/x86/kvm/vmx/vmx.c:4906 handle_exception_nmi+0xfdc/0x1190 arch/x86/kvm/vmx/vmx.c:4906
Modules linked in:
CPU: 0 PID: 7975 Comm: syz-executor.0 Not tainted 5.17.0-rc4-syzkaller-00055-g80d47f5de5e3 #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
RIP: 0010:handle_exception_nmi+0xfdc/0x1190 arch/x86/kvm/vmx/vmx.c:4906
Code: 0f 84 c8 f3 ff ff e8 03 a8 56 00 48 89 ef c7 85 a4 0d 00 00 00 00 00 00 e8 51 1a ec ff 41 89 c4 e9 af f3 ff ff e8 e4 a7 56 00 <0f> 0b e9 69 f6 ff ff e8 d8 a7 56 00 be f5 ff ff ff bf 01 00 00 00
RSP: 0018:ffffc9000615fad8 EFLAGS: 00010216
RAX: 000000000001e2fd RBX: 0000000000000000 RCX: ffffc90002b71000
RDX: 0000000000040000 RSI: ffffffff81216b3c RDI: 0000000000000003
RBP: ffff888025308040 R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff812161a4 R11: 0000000000000000 R12: fffffffffffffc4d
R13: ffff888025308288 R14: 0000000000000000 R15: 0000000080000300
FS: 00007f47cabb1700(0000) GS:ffff88802cd00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000111c960 CR3: 000000001bb69000 CR4: 0000000000152ee0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
__vmx_handle_exit arch/x86/kvm/vmx/vmx.c:6179 [inline]
vmx_handle_exit+0x498/0x1950 arch/x86/kvm/vmx/vmx.c:6196
vcpu_enter_guest+0x29dd/0x4450 arch/x86/kvm/x86.c:10163
vcpu_run arch/x86/kvm/x86.c:10245 [inline]
kvm_arch_vcpu_ioctl_run+0x521/0x21a0 arch/x86/kvm/x86.c:10451
kvm_vcpu_ioctl+0x570/0xf30 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3908
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:874 [inline]
__se_sys_ioctl fs/ioctl.c:860 [inline]
__x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f47cc23c059
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 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 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f47cabb1168 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007f47cc34ef60 RCX: 00007f47cc23c059
RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000005
RBP: 00007f47cc29608d R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ffd6f43c1df R14: 00007f47cabb1300 R15: 0000000000022000
</TASK>


---
This report 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 syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

[syzbot]

syzbot has found a reproducer for the following issue on:

HEAD commit: 7a68065eb9cd Merge tag 'gpio-fixes-for-v5.19-rc2' of git:/..
git tree: upstream
console+strace: https://syzkaller.appspot.com/x/log.txt?x=177df408080000
kernel config: https://syzkaller.appspot.com/x/.config?x=20ac3e0ebf0db3bd

dashboard link: https://syzkaller.appspot.com/bug?extid=4688c50a9c8e68e7aaa1
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2

syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12087173f00000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=16529343f00000

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+4688c5...@syzkaller.appspotmail.com

------------[ cut here ]------------

WARNING: CPU: 0 PID: 3609 at arch/x86/kvm/vmx/vmx.c:4896 handle_exception_nmi+0xfdc/0x1190 arch/x86/kvm/vmx/vmx.c:4896
Modules linked in:
CPU: 0 PID: 3609 Comm: syz-executor169 Not tainted 5.19.0-rc1-syzkaller-00303-g7a68065eb9cd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:handle_exception_nmi+0xfdc/0x1190 arch/x86/kvm/vmx/vmx.c:4896
Code: 0f 84 c8 f3 ff ff e8 33 5c 58 00 48 89 ef c7 85 84 0d 00 00 00 00 00 00 e8 21 35 ec ff 41 89 c4 e9 af f3 ff ff e8 14 5c 58 00 <0f> 0b e9 69 f6 ff ff e8 08 5c 58 00 be f5 ff ff ff bf 01 00 00 00
RSP: 0018:ffffc9000309faf8 EFLAGS: 00010293
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff88801ba1d880 RSI: ffffffff8122171c RDI: 0000000000000001
RBP: ffff88807cd88000 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000001 R12: 00000000a0000975
R13: ffff88807cd88248 R14: 0000000000000000 R15: 0000000080000300
FS: 0000555556c8d300(0000) GS:ffff8880b9b00000(0000) knlGS:0000000000000000

CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033

CR2: 0000000000000000 CR3: 00000000229ca000 CR4: 00000000003526e0

DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000

DR3: 00000000b8fecd19 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
__vmx_handle_exit arch/x86/kvm/vmx/vmx.c:6174 [inline]
vmx_handle_exit+0x498/0x1950 arch/x86/kvm/vmx/vmx.c:6191
vcpu_enter_guest arch/x86/kvm/x86.c:10361 [inline]
vcpu_run arch/x86/kvm/x86.c:10450 [inline]
kvm_arch_vcpu_ioctl_run+0x4208/0x66f0 arch/x86/kvm/x86.c:10654
kvm_vcpu_ioctl+0x570/0xf30 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3944
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:870 [inline]
__se_sys_ioctl fs/ioctl.c:856 [inline]
__x64_sys_ioctl+0x193/0x200 fs/ioctl.c:856

do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80

entry_SYSCALL_64_after_hwframe+0x46/0xb0
RIP: 0033:0x7f56efaee199
Code: 28 c3 e8 2a 14 00 00 66 2e 0f 1f 84 00 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 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc37353158 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f56efaee199

RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000005

RBP: 00007f56efab1bf0 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f56efab1c80
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
</TASK>

[Sean Christopherson]

On Sun, Jun 12, 2022, syzbot wrote:
> syzbot has found a reproducer for the following issue on:
>
> HEAD commit: 7a68065eb9cd Merge tag 'gpio-fixes-for-v5.19-rc2' of git:/..
> git tree: upstream
> console+strace: https://syzkaller.appspot.com/x/log.txt?x=177df408080000
> kernel config: https://syzkaller.appspot.com/x/.config?x=20ac3e0ebf0db3bd
> dashboard link: https://syzkaller.appspot.com/bug?extid=4688c50a9c8e68e7aaa1
> compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12087173f00000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=16529343f00000
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+4688c5...@syzkaller.appspotmail.com
>
> ------------[ cut here ]------------
> WARNING: CPU: 0 PID: 3609 at arch/x86/kvm/vmx/vmx.c:4896 handle_exception_nmi+0xfdc/0x1190 arch/x86/kvm/vmx/vmx.c:4896
> Modules linked in:
> CPU: 0 PID: 3609 Comm: syz-executor169 Not tainted 5.19.0-rc1-syzkaller-00303-g7a68065eb9cd #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011

I'm pretty sure this is a bug in GCE's L0 KVM. Per syzbot's bisection, the WARN
repros all the way back to v5.11 on GCE. The test is doing VMWRITE from its L1
(effective L2) with a memory operand that takes a #PF. The #PF ends up in L1,
which triggers the splat as KVM (real L1) isn't intercepting #PFs.

I've repro'd this on a GCE host running a GCE kernel, and verified the same host
running a v5.18 kernel does _not_ trigger the splat.

I'll route this to someone internally.