kvm: GPF in kvm_pic_clear_all

49 views
Skip to first unread message

Dmitry Vyukov

unread,
Feb 15, 2016, 8:51:18 AM2/15/16
to Gleb Natapov, Paolo Bonzini, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x...@kernel.org, k...@vger.kernel.org, LKML, Eric Northup, Andrew Honig, Steve Rutherford, Robert Swiecki, Kostya Serebryany, Alexander Potapenko, syzkaller
Hello,

The following program triggers GPF in kvm_pic_clear_all if run in a
tight parallel loop:

general protection fault: 0000 [#71] SMP DEBUG_PAGEALLOC KASAN
Modules linked in:
CPU: 0 PID: 7107 Comm: a2.out Tainted: G D 4.5.0-rc3+ #326
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
task: ffff88002144df00 ti: ffff8800219a0000 task.ti: ffff8800219a0000
RIP: 0010:[<ffffffff81467875>] [<ffffffff81467875>] do_raw_spin_lock+0x25/0x2b0
RSP: 0018:ffff8800219a7840 EFLAGS: 00010247
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000004
RBP: ffff8800219a7868 R08: 0000000000000001 R09: 0000000000000000
R10: ffff88002144df00 R11: 0000000000000018 R12: 0000000000000002
R13: 0000000000000130 R14: 0000000000000000 R15: ffff880022f874e0
FS: 00007f6181c19700(0000) GS:ffff88003ec00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 000000c82071a140 CR3: 0000000027104000 CR4: 00000000000026f0
Stack:
0000000000000000 0000000000000002 0000000000000130 0000000000000000
ffff880022f874e0 ffff8800219a7888 ffffffff8666031b ffffffff810cc11c
00000000000000b0 ffff8800219a78b8 ffffffff810cc11c ffff88004ea70000
Call Trace:
[< inline >] __raw_spin_lock include/linux/spinlock_api_smp.h:145
[<ffffffff8666031b>] _raw_spin_lock+0x3b/0x50 kernel/locking/spinlock.c:151
[< inline >] spin_lock include/linux/spinlock.h:302
[< inline >] pic_lock arch/x86/kvm/i8259.c:45
[<ffffffff810cc11c>] kvm_pic_clear_all+0x2c/0x60 arch/x86/kvm/i8259.c:213
[<ffffffff810e31d9>] kvm_free_irq_source_id+0xa9/0xe0
arch/x86/kvm/irq_comm.c:186
[<ffffffff810de6cf>] kvm_create_pit+0x51f/0x6d0 arch/x86/kvm/i8254.c:711
[<ffffffff81061122>] kvm_arch_vm_ioctl+0xaa2/0x1dc0 arch/x86/kvm/x86.c:3856
[<ffffffff8102013c>] kvm_vm_ioctl+0x17c/0xfe0
arch/x86/kvm/../../../virt/kvm/kvm_main.c:2891
[< inline >] vfs_ioctl fs/ioctl.c:43
[<ffffffff817fc26c>] do_vfs_ioctl+0x18c/0xfb0 fs/ioctl.c:674
[< inline >] SYSC_ioctl fs/ioctl.c:689
[<ffffffff817fd11f>] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:680
[<ffffffff866613f6>] entry_SYSCALL_64_fastpath+0x16/0x7a
arch/x86/entry/entry_64.S:185
Code: 84 00 00 00 00 00 55 48 b8 00 00 00 00 00 fc ff df 48 89 e5 41
57 41 56 41 55 41 54 53 48 89 fb 48 83 c7 04 48 89 fa 48 c1 ea 03 <0f>
b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85
RIP [< inline >] debug_spin_lock_before
kernel/locking/spinlock_debug.c:83
RIP [<ffffffff81467875>] do_raw_spin_lock+0x25/0x2b0
kernel/locking/spinlock_debug.c:135
RSP <ffff8800219a7840>
---[ end trace a9bcc650712efe82 ]---

BUG: sleeping function called from invalid context at include/linux/sched.h:2805
in_atomic(): 1, irqs_disabled(): 0, pid: 7107, name: a2.out
lockdep is turned off.
CPU: 0 PID: 7107 Comm: a2.out Tainted: G D 4.5.0-rc3+ #326
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
ffffffff87b05080 ffff8800219a7470 ffffffff82be461f ffffffff00000016
fffffbfff0f60a10 ffff88002144df00 0000000000001bc3 0000000000000000
0000000000000000 dffffc0000000000 ffff8800219a7498 ffffffff813cd42b
Call Trace:
[< inline >] __dump_stack lib/dump_stack.c:15
[<ffffffff82be461f>] dump_stack+0x12e/0x18f lib/dump_stack.c:51
[<ffffffff813cd42b>] ___might_sleep+0x27b/0x3a0 kernel/sched/core.c:7703
[<ffffffff813cd5e0>] __might_sleep+0x90/0x1a0 kernel/sched/core.c:7665
[< inline >] threadgroup_change_begin include/linux/sched.h:2805
[<ffffffff81384b11>] exit_signals+0x81/0x430 kernel/signal.c:2392
[<ffffffff8135de3c>] do_exit+0x23c/0x2cb0 kernel/exit.c:701
[<ffffffff811aa28f>] oops_end+0x9f/0xd0 arch/x86/kernel/dumpstack.c:250
[<ffffffff811aa686>] die+0x46/0x60 arch/x86/kernel/dumpstack.c:316
[<ffffffff811a4725>] do_general_protection+0x235/0x3e0
arch/x86/kernel/traps.c:463
[<ffffffff866636a8>] general_protection+0x28/0x30 arch/x86/entry/entry_64.S:982
[< inline >] __raw_spin_lock include/linux/spinlock_api_smp.h:145
[<ffffffff8666031b>] _raw_spin_lock+0x3b/0x50 kernel/locking/spinlock.c:151
[< inline >] spin_lock include/linux/spinlock.h:302
[< inline >] pic_lock arch/x86/kvm/i8259.c:45
[<ffffffff810cc11c>] kvm_pic_clear_all+0x2c/0x60 arch/x86/kvm/i8259.c:213
[<ffffffff810e31d9>] kvm_free_irq_source_id+0xa9/0xe0
arch/x86/kvm/irq_comm.c:186
[<ffffffff810de6cf>] kvm_create_pit+0x51f/0x6d0 arch/x86/kvm/i8254.c:711
[<ffffffff81061122>] kvm_arch_vm_ioctl+0xaa2/0x1dc0 arch/x86/kvm/x86.c:3856
[<ffffffff8102013c>] kvm_vm_ioctl+0x17c/0xfe0
arch/x86/kvm/../../../virt/kvm/kvm_main.c:2891
[< inline >] vfs_ioctl fs/ioctl.c:43
[<ffffffff817fc26c>] do_vfs_ioctl+0x18c/0xfb0 fs/ioctl.c:674
[< inline >] SYSC_ioctl fs/ioctl.c:689
[<ffffffff817fd11f>] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:680
[<ffffffff866613f6>] entry_SYSCALL_64_fastpath+0x16/0x7a
arch/x86/entry/entry_64.S:185
note: a2.out[7107] exited with preempt_count 1



// autogenerated by syzkaller (http://github.com/google/syzkaller)
#include <pthread.h>
#include <stdint.h>
#include <string.h>
#include <sys/syscall.h>
#include <unistd.h>

long r[23];

void* thr(void* arg)
{
switch ((long)arg) {
case 0:
r[0] = syscall(SYS_mmap, 0x20000000ul, 0xe000ul, 0x3ul, 0x32ul,
0xfffffffffffffffful, 0x0ul);
break;
case 1:
memcpy((void*)0x20001000, "\x2f\x64\x65\x76\x2f\x6b\x76\x6d\x00",
9);
r[2] = syscall(SYS_open, "/dev/kvm", 0x2000ul, 0, 0, 0);
break;
case 2:
r[3] = syscall(SYS_ioctl, r[2], 0xae01ul, 0x0ul, 0, 0, 0);
break;
case 3:
r[4] = syscall(SYS_ioctl, r[3], 0xae60ul, 0, 0, 0, 0);
break;
case 4:
r[5] = syscall(SYS_prctl, 0x23ul, 0x4ul, 0x20000000ul, 0, 0, 0);
break;
case 5:
*(uint32_t*)0x20009000 = (uint32_t)0x4;
*(uint32_t*)0x20009004 = (uint32_t)0x0;
*(uint32_t*)0x20009008 = (uint32_t)0x0;
*(uint32_t*)0x2000900c = (uint32_t)0x0;
*(uint32_t*)0x20009010 = (uint32_t)0x0;
*(uint32_t*)0x20009014 = (uint32_t)0x0;
*(uint32_t*)0x20009018 = (uint32_t)0x0;
*(uint32_t*)0x2000901c = (uint32_t)0x0;
*(uint32_t*)0x20009020 = (uint32_t)0x0;
*(uint32_t*)0x20009024 = (uint32_t)0x0;
*(uint32_t*)0x20009028 = (uint32_t)0x0;
*(uint32_t*)0x2000902c = (uint32_t)0x0;
*(uint32_t*)0x20009030 = (uint32_t)0x0;
*(uint32_t*)0x20009034 = (uint32_t)0x0;
*(uint32_t*)0x20009038 = (uint32_t)0x0;
*(uint32_t*)0x2000903c = (uint32_t)0x0;
r[22] =
syscall(SYS_ioctl, r[3], 0x4040ae77ul, 0x20009000ul, 0, 0, 0);
break;
}
return 0;
}

int main()
{
long i;
pthread_t th[6];

srand(getpid());
memset(r, -1, sizeof(r));
for (i = 0; i < 6; i++) {
pthread_create(&th[i], 0, thr, (void*)i);
usleep(1000+rand()%1000);
}
for (i = 0; i < 6; i++) {
pthread_create(&th[i], 0, thr, (void*)i);
if (rand()%2)
usleep(rand()%1000);
}
usleep(20000);
return 0;
}

On commit 388f7b1d6e8ca06762e2454d28d6c3c55ad0fe95 (4.5-rc3)

Steve Rutherford

unread,
Feb 25, 2016, 5:48:00 PM2/25/16
to Dmitry Vyukov, Gleb Natapov, Paolo Bonzini, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x...@kernel.org, KVM list, LKML, Eric Northup, Andrew Honig, Robert Swiecki, Kostya Serebryany, Alexander Potapenko, syzkaller
Hey Dmitry,

I ran this against 4.5 rc3 and couldn't get a crash on bare metal (overnight one machine bunch of processes looping). Are you running these tests with nested?

--Steve

Dmitry Vyukov

unread,
Feb 28, 2016, 6:39:02 AM2/28/16
to Steve Rutherford, Gleb Natapov, Paolo Bonzini, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x...@kernel.org, KVM list, LKML, Eric Northup, Andrew Honig, Robert Swiecki, Kostya Serebryany, Alexander Potapenko, syzkaller
On Thu, Feb 25, 2016 at 11:47 PM, Steve Rutherford
<sruth...@google.com> wrote:
> Hey Dmitry,
>
> I ran this against 4.5 rc3 and couldn't get a crash on bare metal (overnight
> one machine bunch of processes looping). Are you running these tests with
> nested?


Yes, I run it in qemu (host has Intel processor).
Reply all
Reply to author
Forward
0 new messages