general protection fault in vcpu_enter_guest (2)

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit: 76c938fc Add linux-next specific files for 20190423
git tree: linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=16863d28a00000
kernel config: https://syzkaller.appspot.com/x/.config?x=62738d35457c577d
dashboard link: https://syzkaller.appspot.com/bug?extid=f7e65445a40d3e0e4ebf
compiler: gcc (GCC) 9.0.0 20181231 (experimental)
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1364a6d4a00000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=17c25dc4a00000

Bisection is inconclusive: the first bad commit could be any of:

57bf67e7 KVM: lapic: Disable timer advancement if adaptive tuning goes
haywire
39497d76 KVM: lapic: Track lapic timer advance per vCPU

bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=17a32fc4a00000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+f7e654...@syzkaller.appspotmail.com

L1TF CPU bug present and SMT on, data leak possible. See CVE-2018-3646 and
https://www.kernel.org/doc/html/latest/admin-guide/l1tf.html for details.
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: 7872 Comm: syz-executor233 Not tainted 5.1.0-rc6-next-20190423
#29
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
RIP: 0010:vcpu_enter_guest+0xbcd/0x5fb0 arch/x86/kvm/x86.c:7895
Code: 48 c1 ea 03 80 3c 02 00 0f 85 6f 48 00 00 49 8b 9f b0 03 00 00 48 b8
00 00 00 00 00 fc ff df 48 8d 7b 78 48 89 fa 48 c1 ea 03 <0f> b6 04 02 84
c0 74 08 3c 03 0f 8e 39 48 00 00 8b 5b 78 31 ff 89
RSP: 0018:ffff88808bd4fa00 EFLAGS: 00010006
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff810d1ec9
RDX: 000000000000000f RSI: ffffffff810ceef2 RDI: 0000000000000078
RBP: ffff88808bd4fb10 R08: ffff8880a912e580 R09: ffffed1015d05dd0
R10: ffffed1015d05dcf R11: ffff8880ae82ee7b R12: ffff88808f3e006c
R13: 0000000000000001 R14: ffff88808f3e0070 R15: ffff88808f3e0040
FS: 00007f7adb927700(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fa0427eb000 CR3: 00000000939e4000 CR4: 00000000001426f0
Call Trace:
vcpu_run arch/x86/kvm/x86.c:8037 [inline]
kvm_arch_vcpu_ioctl_run+0x425/0x1750 arch/x86/kvm/x86.c:8245
kvm_vcpu_ioctl+0x4dc/0xf90 arch/x86/kvm/../../../virt/kvm/kvm_main.c:2684
vfs_ioctl fs/ioctl.c:46 [inline]
file_ioctl fs/ioctl.c:509 [inline]
do_vfs_ioctl+0xd6e/0x1390 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+0x103/0x670 arch/x86/entry/common.c:298
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x446779
Code: e8 5c b3 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 0b 08 fc ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007f7adb926db8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00000000006dbc48 RCX: 0000000000446779
RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000005
RBP: 00000000006dbc40 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00000000006dbc4c
R13: 00007fff1cfa7c1f R14: 00007f7adb9279c0 R15: 20c49ba5e353f7cf
Modules linked in:
---[ end trace da6276bd16a5f99d ]---
RIP: 0010:vcpu_enter_guest+0xbcd/0x5fb0 arch/x86/kvm/x86.c:7895
Code: 48 c1 ea 03 80 3c 02 00 0f 85 6f 48 00 00 49 8b 9f b0 03 00 00 48 b8
00 00 00 00 00 fc ff df 48 8d 7b 78 48 89 fa 48 c1 ea 03 <0f> b6 04 02 84
c0 74 08 3c 03 0f 8e 39 48 00 00 8b 5b 78 31 ff 89
RSP: 0018:ffff88808bd4fa00 EFLAGS: 00010006
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff810d1ec9
RDX: 000000000000000f RSI: ffffffff810ceef2 RDI: 0000000000000078
RBP: ffff88808bd4fb10 R08: ffff8880a912e580 R09: ffffed1015d05dd0
R10: ffffed1015d05dcf R11: ffff8880ae82ee7b R12: ffff88808f3e006c
R13: 0000000000000001 R14: ffff88808f3e0070 R15: ffff88808f3e0040
FS: 00007f7adb927700(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fa0427eb000 CR3: 00000000939e4000 CR4: 00000000001426f0


---
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 syzk...@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches

[Sean Christopherson]

On Thu, Apr 25, 2019 at 05:07:05PM -0700, syzbot wrote:
> Hello,
>
> syzbot found the following crash on:
>
> HEAD commit: 76c938fc Add linux-next specific files for 20190423
> git tree: linux-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=16863d28a00000
> kernel config: https://syzkaller.appspot.com/x/.config?x=62738d35457c577d
> dashboard link: https://syzkaller.appspot.com/bug?extid=f7e65445a40d3e0e4ebf
> compiler: gcc (GCC) 9.0.0 20181231 (experimental)
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1364a6d4a00000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=17c25dc4a00000
>
> Bisection is inconclusive: the first bad commit could be any of:
>
> 57bf67e7 KVM: lapic: Disable timer advancement if adaptive tuning goes
> haywire
> 39497d76 KVM: lapic: Track lapic timer advance per vCPU

Well that was depressingly easy to reproduce. The per-vCPU commit is at
fault, it fails to ensure the lapic is in kernel before dereferencing the
associated pointer. Patch incoming...