[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191101211623.GB20061@amt.cnet>
Date: Fri, 1 Nov 2019 19:16:26 -0200
From: Marcelo Tosatti <mtosatti@...hat.com>
To: Zhenzhong Duan <zhenzhong.duan@...cle.com>
Cc: linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
joao.m.martins@...cle.com, rafael.j.wysocki@...el.com,
rkrcmar@...hat.com, pbonzini@...hat.com
Subject: Re: [PATCH 3/5] KVM: ensure pool time is longer than block_ns
On Sat, Oct 26, 2019 at 11:23:57AM +0800, Zhenzhong Duan wrote:
> When (block_ns == vcpu->halt_poll_ns), there is not a margin so that
> vCPU may still get into block state unnecessorily.
>
> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@...cle.com>
> ---
> virt/kvm/kvm_main.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 1b6fe3b..48a1f1a 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -2371,7 +2371,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
> if (!vcpu_valid_wakeup(vcpu)) {
> shrink_halt_poll_ns(vcpu);
> } else if (halt_poll_ns) {
> - if (block_ns <= vcpu->halt_poll_ns)
> + if (block_ns < vcpu->halt_poll_ns)
> ;
> /* we had a short halt and our poll time is too small */
> else if (block_ns < halt_poll_ns)
> --
> 1.8.3.1
Makes sense.
Powered by blists - more mailing lists