Re: [PATCH v2 11/43] KVM: Don't block+unblock when halt-polling is successful

From: Maxim Levitsky
Date: Wed Oct 27 2021 - 09:41:04 EST


On Fri, 2021-10-08 at 19:12 -0700, Sean Christopherson wrote:
> Invoke the arch hooks for block+unblock if and only if KVM actually
> attempts to block the vCPU. The only non-nop implementation is on x86,
> specifically SVM's AVIC, and there is no need to put the AVIC prior to
> halt-polling as KVM x86's kvm_vcpu_has_events() will scour the full vIRR
> to find pending IRQs regardless of whether the AVIC is loaded/"running".
>
> The primary motivation is to allow future cleanup to split out "block"
> from "halt", but this is also likely a small performance boost on x86 SVM
> when halt-polling is successful.
>
> Adjust the post-block path to update "cur" after unblocking, i.e. include
> AVIC load time in halt_wait_ns and halt_wait_hist, so that the behavior
> is consistent. Moving just the pre-block arch hook would result in only
> the AVIC put latency being included in the halt_wait stats. There is no
> obvious evidence that one way or the other is correct, so just ensure KVM
> is consistent.
>
> Note, x86 has two separate paths for handling APICv with respect to vCPU
> blocking. VMX uses hooks in x86's vcpu_block(), while SVM uses the arch
> hooks in kvm_vcpu_block(). Prior to this path, the two paths were more
> or less functionally identical. That is very much not the case after
> this patch, as the hooks used by VMX _must_ fire before halt-polling.
> x86's entire mess will be cleaned up in future patches.
>
> Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
> ---
> virt/kvm/kvm_main.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index f90b3ed05628..227f6bbe0716 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -3235,8 +3235,6 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
> bool waited = false;
> u64 block_ns;
>
> - kvm_arch_vcpu_blocking(vcpu);
> -
> start = cur = poll_end = ktime_get();
> if (do_halt_poll) {
> ktime_t stop = ktime_add_ns(ktime_get(), vcpu->halt_poll_ns);
> @@ -3253,6 +3251,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
> } while (kvm_vcpu_can_poll(cur, stop));
> }
>
> + kvm_arch_vcpu_blocking(vcpu);
>
> prepare_to_rcuwait(wait);
> for (;;) {
> @@ -3265,6 +3264,9 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
> schedule();
> }
> finish_rcuwait(wait);
> +
> + kvm_arch_vcpu_unblocking(vcpu);
> +
> cur = ktime_get();
> if (waited) {
> vcpu->stat.generic.halt_wait_ns +=
> @@ -3273,7 +3275,6 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
> ktime_to_ns(cur) - ktime_to_ns(poll_end));
> }
> out:
> - kvm_arch_vcpu_unblocking(vcpu);
> block_ns = ktime_to_ns(cur) - ktime_to_ns(start);
>
> /*

Makes sense.

Reviewed-by: Maxim Levitsky <mlevitsk@xxxxxxxxxx>

Best regards,
Maxim Levitsky