Re: general protection fault in start_creating

[Vitaly Kuznetsov <vkuznets@redhat.com>]

syzbot <syzbot+705f4401d5a93a59b87d@syzkaller.appspotmail.com> writes:

> syzbot has found a reproducer for the following crash on:
>
> HEAD commit:    cb8e59cc Merge git://git.kernel.org/pub/scm/linux/kernel/g..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=170f49de100000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=a16ddbc78955e3a9
> dashboard link: https://syzkaller.appspot.com/bug?extid=705f4401d5a93a59b87d
> compiler:       gcc (GCC) 9.0.0 20181231 (experimental)
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1367410e100000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=10e07e0e100000
>
> The bug was bisected to:
>
> commit 63d04348371b7ea4a134bcf47c79763d969e9168
> Author: Paolo Bonzini <pbonzini@redhat.com>
> Date:   Tue Mar 31 22:42:22 2020 +0000
>
>     KVM: x86: move kvm_create_vcpu_debugfs after last failure point
>
> bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=1366069a100000
> final crash:    https://syzkaller.appspot.com/x/report.txt?x=10e6069a100000
> console output: https://syzkaller.appspot.com/x/log.txt?x=1766069a100000
>
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+705f4401d5a93a59b87d@syzkaller.appspotmail.com
> Fixes: 63d04348371b ("KVM: x86: move kvm_create_vcpu_debugfs after last failure point")
>
> general protection fault, probably for non-canonical address 0xdffffc000000002a: 0000 [#1] PREEMPT SMP KASAN
> KASAN: null-ptr-deref in range [0x0000000000000150-0x0000000000000157]
> CPU: 0 PID: 19367 Comm: syz-executor088 Not tainted 5.7.0-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:__lock_acquire+0xe1b/0x4a70 kernel/locking/lockdep.c:4250
> Code: 91 0a 41 be 01 00 00 00 0f 86 ce 0b 00 00 89 05 4b 9a 91 0a e9 c3 0b 00 00 48 b8 00 00 00 00 00 fc ff df 4c 89 d2 48 c1 ea 03 <80> 3c 02 00 0f 85 57 2e 00 00 49 81 3a c0 74 c0 8b 0f 84 b0 f2 ff
> RSP: 0018:ffffc90004b477b8 EFLAGS: 00010002
> RAX: dffffc0000000000 RBX: 0000000000000001 RCX: 0000000000000000
> RDX: 000000000000002a RSI: 0000000000000000 RDI: 0000000000000150
> RBP: 0000000000000001 R08: 0000000000000001 R09: 0000000000000000
> R10: 0000000000000150 R11: 0000000000000001 R12: 0000000000000000
> R13: ffff8880a77c2200 R14: 0000000000000000 R15: 0000000000000000
> FS:  00007f9f3c6e5700(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00000000004d5bb0 CR3: 00000000821f9000 CR4: 00000000001426f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
>  lock_acquire+0x1f2/0x8f0 kernel/locking/lockdep.c:4959
>  down_write+0x8d/0x150 kernel/locking/rwsem.c:1531
>  inode_lock include/linux/fs.h:799 [inline]
>  start_creating+0xa8/0x250 fs/debugfs/inode.c:334
>  __debugfs_create_file+0x62/0x400 fs/debugfs/inode.c:383
>  kvm_arch_create_vcpu_debugfs+0x9f/0x200 arch/x86/kvm/debugfs.c:52
>  kvm_create_vcpu_debugfs arch/x86/kvm/../../../virt/kvm/kvm_main.c:3012 [inline]

...

I think what happens here is one thread does kvm_vm_ioctl_create_vcpu()
and another tries to delete the VM. The problem was probably present
even before the commit as both kvm_create_vcpu_debugfs() and
kvm_destroy_vm_debugfs() happen outside of kvm_lock but maybe it was
harder to trigger. Is there a reason to not put all this under kvm_lock?
I.e. the following:

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 7fa1e38e1659..d53784cb920f 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -793,9 +793,9 @@ static void kvm_destroy_vm(struct kvm *kvm)
        struct mm_struct *mm = kvm->mm;
 
        kvm_uevent_notify_change(KVM_EVENT_DESTROY_VM, kvm);
-       kvm_destroy_vm_debugfs(kvm);
        kvm_arch_sync_events(kvm);
        mutex_lock(&kvm_lock);
+       kvm_destroy_vm_debugfs(kvm);
        list_del(&kvm->vm_list);
        mutex_unlock(&kvm_lock);
        kvm_arch_pre_destroy_vm(kvm);
@@ -3084,9 +3084,9 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
        smp_wmb();
        atomic_inc(&kvm->online_vcpus);
 
+       kvm_create_vcpu_debugfs(vcpu);
        mutex_unlock(&kvm->lock);
        kvm_arch_vcpu_postcreate(vcpu);
-       kvm_create_vcpu_debugfs(vcpu);
        return r;
 
 unlock_vcpu_destroy:

should probably do. The reproducer doesn't work for me (or just takes
too long so I gave up), unfortunately. Or I may have misunderstood
everything :-)

-- 
Vitaly