Re: [PATCH v2] KVM: x86/intr: Explicitly check NMI from guest to eliminate false positives

[Like Xu]

On 13/12/2023 3:28 pm, Dongli Zhang wrote:
> Hi Like,
>
> On 12/5/23 19:20, Like Xu wrote:
> > From: Like Xu <likexu@xxxxxxxxxxx>
> >
> > Explicitly checking the source of external interrupt is indeed NMI and not
> > other types in the kvm_arch_pmi_in_guest(), which prevents perf-kvm false
> > positive samples generated in perf/core NMI mode after vm-exit but before
> > kvm_before_interrupt() from being incorrectly labelled as guest samples:
>
> About the before kvm_before_interrupt() ...
>
> > # test: perf-record + cpu-cycles:HP (which collects host-only precise samples)
> > # Symbol                                   Overhead       sys       usr  guest sys  guest usr
> > # .......................................  ........  ........  ........  .........  .........
> > #
> > # Before:
> >    [g] entry_SYSCALL_64                       24.63%     0.00%     0.00%     24.63%      0.00%
> >    [g] syscall_return_via_sysret              23.23%     0.00%     0.00%     23.23%      0.00%
> >    [g] files_lookup_fd_raw                     6.35%     0.00%     0.00%      6.35%      0.00%
> > # After:
> >    [k] perf_adjust_freq_unthr_context         57.23%    57.23%     0.00%      0.00%      0.00%
> >    [k] __vmx_vcpu_run                          4.09%     4.09%     0.00%      0.00%      0.00%
> >    [k] vmx_update_host_rsp                     3.17%     3.17%     0.00%      0.00%      0.00%
> >
> > In the above case, perf records the samples labelled '[g]', the RIPs behind
> > the weird samples are actually being queried by perf_instruction_pointer()
> > after determining whether it's in GUEST state or not, and here's the issue:
> >
> > If vm-exit is caused by a non-NMI interrupt (such as hrtimer_interrupt) and
> > at least one PMU counter is enabled on host, the kvm_arch_pmi_in_guest()
> > will remain true (KVM_HANDLING_IRQ is set) until kvm_before_interrupt().
>
> ... and here.
>
> Would you mind helping why kvm_arch_pmi_in_guest() remains true before
> *kvm_before_interrupt()*.
>
> According to the source code, the vcpu->arch.handling_intr_from_guest
> is set to non-zero only at kvm_before_interrupt(), and cleared at
> kvm_after_interrupt().
>
> Or would you mean kvm_after_interrupt()?

Oops, it should refer to kvm_after_interrupt() as the code fixed. Thank you.

> Thank you very much!
>
> Dongli Zhang
>
> > During this window, if a PMI occurs on host (since the KVM instructions on
> > host are being executed), the control flow, with the help of the host NMI
> > context, will be transferred to perf/core to generate performance samples,
> > thus perf_instruction_pointer() and perf_guest_get_ip() is called.
> >
> > Since kvm_arch_pmi_in_guest() only checks if there is an interrupt, it may
> > cause perf/core to mistakenly assume that the source RIP of the host NMI
> > belongs to the guest world and use perf_guest_get_ip() to get the RIP of
> > a vCPU that has already exited by a non-NMI interrupt.
> >
> > Error samples are recorded and presented to the end-user via perf-report.
> > Such false positive samples could be eliminated by explicitly determining
> > if the exit reason is KVM_HANDLING_NMI.
> >
> > Note that when vm-exit is indeed triggered by PMI and before HANDLING_NMI
> > is cleared, it's also still possible that another PMI is generated on host.
> > Also for perf/core timer mode, the false positives are still possible since
> > that non-NMI sources of interrupts are not always being used by perf/core.
> > In both cases above, perf/core should correctly distinguish between real
> > RIP sources or even need to generate two samples, belonging to host and
> > guest separately, but that's perf/core's story for interested warriors.
> >
> > Fixes: dd60d217062f ("KVM: x86: Fix perf timer mode IP reporting")
> > Signed-off-by: Like Xu <likexu@xxxxxxxxxxx>
> > ---
> > V1 -> V2 Changelog:
> > - Refine commit message to cover both perf/core timer and NMI modes;
> > - Use in_nmi() to distinguish whether it's NMI mode or not; (Sean)
> > V1: https://urldefense.com/v3/__https://lore.kernel.org/kvm/20231204074535.9567-1-likexu@xxxxxxxxxxx/__;!!ACWV5N9M2RV99hQ!MQ8FetD27SVKN34CS_P-K3qrhspFnpf_Mqb0McFN9y5vSUeScc5b0TlZ3ZMDvt4Cn4b3g0h9ci6EO9k3PBEQXpePrg$
> >   arch/x86/include/asm/kvm_host.h | 10 +++++++++-
> >   arch/x86/kvm/x86.h              |  6 ------
> >   2 files changed, 9 insertions(+), 7 deletions(-)
> >
> > diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> > index c8c7e2475a18..167d592e08d0 100644
> > --- a/arch/x86/include/asm/kvm_host.h
> > +++ b/arch/x86/include/asm/kvm_host.h
> > @@ -1868,8 +1868,16 @@ static inline int kvm_arch_flush_remote_tlbs_range(struct kvm *kvm, gfn_t gfn,
> >   }
> >   #endif /* CONFIG_HYPERV */
> >   
> > +enum kvm_intr_type {
> > +	/* Values are arbitrary, but must be non-zero. */
> > +	KVM_HANDLING_IRQ = 1,
> > +	KVM_HANDLING_NMI,
> > +};
> > +
> > +/* Enable perf NMI and timer modes to work, and minimise false positives. */
> >   #define kvm_arch_pmi_in_guest(vcpu) \
> > -	((vcpu) && (vcpu)->arch.handling_intr_from_guest)
> > +	((vcpu) && (vcpu)->arch.handling_intr_from_guest && \
> > +	 (in_nmi() == ((vcpu)->arch.handling_intr_from_guest == KVM_HANDLING_NMI)))
> >   
> >   void __init kvm_mmu_x86_module_init(void);
> >   int kvm_mmu_vendor_module_init(void);
> > diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
> > index 2f7e19166658..4dc38092d599 100644
> > --- a/arch/x86/kvm/x86.h
> > +++ b/arch/x86/kvm/x86.h
> > @@ -431,12 +431,6 @@ static inline bool kvm_notify_vmexit_enabled(struct kvm *kvm)
> >   	return kvm->arch.notify_vmexit_flags & KVM_X86_NOTIFY_VMEXIT_ENABLED;
> >   }
> >   
> > -enum kvm_intr_type {
> > -	/* Values are arbitrary, but must be non-zero. */
> > -	KVM_HANDLING_IRQ = 1,
> > -	KVM_HANDLING_NMI,
> > -};
> > -
> >   static __always_inline void kvm_before_interrupt(struct kvm_vcpu *vcpu,
> >   						 enum kvm_intr_type intr)
> >   {
> >
> > base-commit: 1ab097653e4dd8d23272d028a61352c23486fd4a