All of lore.kernel.org
 help / color / mirror / Atom feed
From: Wanpeng Li <kernellwp@gmail.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
	"Radim Krčmář" <rkrcmar@redhat.com>,
	"Sean Christopherson" <sean.j.christopherson@intel.com>,
	"Vitaly Kuznetsov" <vkuznets@redhat.com>,
	"Wanpeng Li" <wanpengli@tencent.com>,
	"Jim Mattson" <jmattson@google.com>,
	"Joerg Roedel" <joro@8bytes.org>
Subject: Re: [PATCH 1/2] KVM: X86: Single target IPI fastpath
Date: Thu, 14 Nov 2019 11:12:17 +0800	[thread overview]
Message-ID: <CANRm+CzJ+kiVohGE=nPVw3fGUqHWyeW0hkSs2nA-Quwok8qn1w@mail.gmail.com> (raw)
In-Reply-To: <CANRm+CzK_h2E9XWFipkNpAALLCBcM2vrUkdBpumwmT9AP09hfA@mail.gmail.com>

On Tue, 12 Nov 2019 at 09:33, Wanpeng Li <kernellwp@gmail.com> wrote:
>
> On Tue, 12 Nov 2019 at 05:59, Paolo Bonzini <pbonzini@redhat.com> wrote:
> >
> > On 09/11/19 08:05, Wanpeng Li wrote:
> > > From: Wanpeng Li <wanpengli@tencent.com>
> > >
> > > This patch tries to optimize x2apic physical destination mode, fixed delivery
> > > mode single target IPI by delivering IPI to receiver immediately after sender
> > > writes ICR vmexit to avoid various checks when possible.
> > >
> > > Testing on Xeon Skylake server:
> > >
> > > The virtual IPI latency from sender send to receiver receive reduces more than
> > > 330+ cpu cycles.
> > >
> > > Running hackbench(reschedule ipi) in the guest, the avg handle time of MSR_WRITE
> > > caused vmexit reduces more than 1000+ cpu cycles:
> > >
> > > Before patch:
> > >
> > >   VM-EXIT    Samples  Samples%     Time%    Min Time    Max Time   Avg time
> > > MSR_WRITE    5417390    90.01%    16.31%      0.69us    159.60us    1.08us
> > >
> > > After patch:
> > >
> > >   VM-EXIT    Samples  Samples%     Time%    Min Time    Max Time   Avg time
> > > MSR_WRITE    6726109    90.73%    62.18%      0.48us    191.27us    0.58us
> >
> > Do you have retpolines enabled?  The bulk of the speedup might come just
> > from the indirect jump.
>
> Adding 'mitigations=off' to the host grub parameter:
>
> Before patch:
>
>     VM-EXIT    Samples  Samples%     Time%    Min Time    Max Time   Avg time
> MSR_WRITE    2681713    92.98%    77.52%      0.38us     18.54us
> 0.73us ( +-   0.02% )
>
> After patch:
>
>     VM-EXIT    Samples  Samples%     Time%    Min Time    Max Time   Avg time
> MSR_WRITE    2953447    92.48%    62.47%      0.30us     59.09us
> 0.40us ( +-   0.02% )

Hmm, sender side less vmexit time is due to kvm_exit tracepoint is
still left in vmx_handle_exit, and ICR wrmsr is moved ahead, that is
why the time between kvm_exit tracepoint and kvm_entry tracepoint is
reduced.  But the virtual IPI latency still can reduce 330+ cycles.

    Wanpeng

  parent reply	other threads:[~2019-11-14  3:12 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-09  7:05 [PATCH 1/2] KVM: X86: Single target IPI fastpath Wanpeng Li
2019-11-09  7:05 ` [PATCH 2/2] KVM: LAPIC: micro-optimize fixed mode ipi delivery Wanpeng Li
2019-11-11 21:59   ` Paolo Bonzini
2019-11-12  1:34     ` Wanpeng Li
2019-11-10  1:59 ` [PATCH 1/2] KVM: X86: Single target IPI fastpath kbuild test robot
2019-11-10  1:59   ` kbuild test robot
2019-11-10  1:59 ` [PATCH] KVM: X86: fix semicolon.cocci warnings kbuild test robot
2019-11-10  1:59   ` kbuild test robot
2019-11-11 13:06 ` [PATCH 1/2] KVM: X86: Single target IPI fastpath Vitaly Kuznetsov
2019-11-12  1:18   ` Wanpeng Li
2019-11-11 21:59 ` Paolo Bonzini
2019-11-12  1:33   ` Wanpeng Li
2019-11-13  6:05     ` Wanpeng Li
2019-11-14  3:12     ` Wanpeng Li [this message]
2019-11-14 11:58 ` Paolo Bonzini
2019-11-14 15:22   ` Sean Christopherson
2019-11-14 15:44     ` Paolo Bonzini
2019-11-14 16:34       ` Sean Christopherson
2019-11-15  5:56         ` Wanpeng Li

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CANRm+CzJ+kiVohGE=nPVw3fGUqHWyeW0hkSs2nA-Quwok8qn1w@mail.gmail.com' \
    --to=kernellwp@gmail.com \
    --cc=jmattson@google.com \
    --cc=joro@8bytes.org \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=rkrcmar@redhat.com \
    --cc=sean.j.christopherson@intel.com \
    --cc=vkuznets@redhat.com \
    --cc=wanpengli@tencent.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.