Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <1620871189-4763-1-git-send-email-wanpengli@tencent.com>
In-Reply-To: <1620871189-4763-1-git-send-email-wanpengli@tencent.com>
From:   David Matlack <dmatlack@google.com>
Date:   Fri, 14 May 2021 14:32:42 -0700
Message-ID: <CALzav=e98KRgG+z5oezPmENKDt+NqtEA57ijYh3kMBZyduQUZg@mail.gmail.com>
Subject: Re: [PATCH v2 2/4] KVM: X86: Bail out of direct yield in case of
 under-comitted scenarios
To:     Wanpeng Li <kernellwp@gmail.com>
Cc:     linux-kernel@vger.kernel.org, kvm list <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Sean Christopherson <seanjc@google.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, May 12, 2021 at 7:01 PM Wanpeng Li <kernellwp@gmail.com> wrote:
>
> From: Wanpeng Li <wanpengli@tencent.com>
>
> In case of under-comitted scenarios, vCPU can get scheduling easily,
> kvm_vcpu_yield_to adds extra overhead, we can observe a lot of race
> between vcpu->ready is true and yield fails due to p->state is
> TASK_RUNNING. Let's bail out in such scenarios by checking the length
> of current cpu runqueue, it can be treated as a hint of under-committed
> instead of guarantee of accuracy.
>
> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> ---
> v1 -> v2:
>  * move the check after attempted counting
>  * update patch description
>
>  arch/x86/kvm/x86.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 9b6bca6..dfb7c32 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -8360,6 +8360,9 @@ static void kvm_sched_yield(struct kvm_vcpu *vcpu, unsigned long dest_id)
>
>         vcpu->stat.directed_yield_attempted++;
>
> +       if (single_task_running())
> +               goto no_yield;

Since this is a heuristic, do you have any experimental or real world
results that show the benefit?

> +
>         rcu_read_lock();
>         map = rcu_dereference(vcpu->kvm->arch.apic_map);
>
> --
> 2.7.4
>