2017-08-23 0:09 GMT+08:00 Paolo Bonzini <pbonzini@xxxxxxxxxx>:
> On 22/08/2017 01:08, Wanpeng Li wrote:
>> From: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
>>
>> ------------[ cut here ]------------
>> WARNING: CPU: 7 PID: 3861 at /home/kernel/ssd/kvm/arch/x86/kvm//vmx.c:11299 nested_vmx_vmexit+0x176e/0x1980 [kvm_intel]
>> CPU: 7 PID: 3861 Comm: qemu-system-x86 Tainted: G W OE 4.13.0-rc4+ #11
>> RIP: 0010:nested_vmx_vmexit+0x176e/0x1980 [kvm_intel]
>> Call Trace:
>> ? kvm_multiple_exception+0x149/0x170 [kvm]
>> ? handle_emulation_failure+0x79/0x230 [kvm]
>> ? load_vmcs12_host_state+0xa80/0xa80 [kvm_intel]
>> ? check_chain_key+0x137/0x1e0
>> ? reexecute_instruction.part.168+0x130/0x130 [kvm]
>> nested_vmx_inject_exception_vmexit+0xb7/0x100 [kvm_intel]
>> ? nested_vmx_inject_exception_vmexit+0xb7/0x100 [kvm_intel]
>> vmx_queue_exception+0x197/0x300 [kvm_intel]
>> kvm_arch_vcpu_ioctl_run+0x1b0c/0x2c90 [kvm]
>> ? kvm_arch_vcpu_runnable+0x220/0x220 [kvm]
>> ? preempt_count_sub+0x18/0xc0
>> ? restart_apic_timer+0x17d/0x300 [kvm]
>> ? kvm_lapic_restart_hv_timer+0x37/0x50 [kvm]
>> ? kvm_arch_vcpu_load+0x1d8/0x350 [kvm]
>> kvm_vcpu_ioctl+0x4e4/0x910 [kvm]
>> ? kvm_vcpu_ioctl+0x4e4/0x910 [kvm]
>> ? kvm_dev_ioctl+0xbe0/0xbe0 [kvm]
>>
>> The flag "nested_run_pending", which can override the decision of which should run
>> next, L1 or L2. nested_run_pending=1 means that we *must* run L2 next, not L1. This
>> is necessary in particular when L1 did a VMLAUNCH of L2 and therefore expects L2 to
>> be run (and perhaps be injected with an event it specified, etc.). Nested_run_pending
>> is especially intended to avoid switching to L1 in the injection decision-point.
>>
>> I catch this in the queue exception path, this patch fixes it by running L2 next
>> instead of L1 in the queue exception path and injecting the pending exception to
>> L2 directly.
>>
>> Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
>> Cc: Radim Krčmář <rkrcmar@xxxxxxxxxx>
>> Signed-off-by: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
>> ---
>> v2 -> v3:
>> * move the nested_run_pending to the else branch
>> v1 -> v2:
>> * request an immediate VM exit from L2 and keep the exception for
>> L1 pending for a subsequent nested VM exit
>>
>> arch/x86/kvm/vmx.c | 4 ++++
>> 1 file changed, 4 insertions(+)
>>
>> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
>> index e398946..685f51e 100644
>> --- a/arch/x86/kvm/vmx.c
>> +++ b/arch/x86/kvm/vmx.c
>> @@ -2488,6 +2488,10 @@ static int nested_vmx_check_exception(struct kvm_vcpu *vcpu)
>> }
>> } else {
>> unsigned long exit_qual = 0;
>> +
>> + if (to_vmx(vcpu)->nested.nested_run_pending)
>> + return 0;
>> +
>> if (nr == DB_VECTOR)
>> exit_qual = vcpu->arch.dr6;
>>
>>
>
> Hmm, why would this not apply to page faults? It doesn't make much sense...
Agreed, I do this in v1, please have a look.
Regards,
Wanpeng Li
