Re: [PATCH 2/2] KVM: x86/pmu: Protect kvm->arch.pmu_event_filter with SRCU

[Paolo Bonzini <pbonzini@redhat.com>]

On 2/17/22 09:36, Like Xu wrote:
> From: Like Xu<likexu@tencent.com>
> 
> Fix the following positive warning:
> 
>   =============================
>   WARNING: suspicious RCU usage
>   arch/x86/kvm/pmu.c:190 suspicious rcu_dereference_check() usage!
>   other info that might help us debug this:
>   rcu_scheduler_active = 2, debug_locks = 1
>   1 lock held by CPU 28/KVM/370841:
>   #0: ff11004089f280b8 (&vcpu->mutex){+.+.}-{3:3}, at: kvm_vcpu_ioctl+0x87/0x730 [kvm]
>   Call Trace:
>    <TASK>
>    dump_stack_lvl+0x59/0x73
>    reprogram_fixed_counter+0x15d/0x1a0 [kvm]
>    kvm_pmu_trigger_event+0x1a3/0x260 [kvm]
>    ? free_moved_vector+0x1b4/0x1e0
>    complete_fast_pio_in+0x8a/0xd0 [kvm]
>    [...]

I think the right fix is to add SRCU protection to complete_userspace_io 
in kvm_arch_vcpu_ioctl_run.  Most calls of complete_userspace_io can 
execute similar code to vmexits.

> Fixes: 66bb8a065f5a ("KVM: x86: PMU Event Filter")

It fixes 9cd803d496e7 ("KVM: x86: Update vPMCs when retiring 
instructions", 2022-01-07), actually.  That is when the PMU filter was 
added to kvm_skip_emulated_instruction (called by kvm_fast_pio_in).

Thanks,

Paolo

> It's possible to call KVM_SET_PMU_EVENT_FILTER ioctl with the vCPU running.
> Similar to "kvm->arch.msr_filter", KVM should guarantee that vCPUs will
> see either the previous filter or the new filter so that guest pmu events
> with identical settings in both the old and new filter have deterministic
> behavior.