[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANRm+Cw6vKHnfq4G0wmKmbs63vXbVGtPBPpxZBzr1JL+HTaYGQ@mail.gmail.com>
Date: Wed, 20 Nov 2019 08:24:45 +0800
From: Wanpeng Li <kernellwp@...il.com>
To: Sean Christopherson <sean.j.christopherson@...el.com>
Cc: LKML <linux-kernel@...r.kernel.org>, kvm <kvm@...r.kernel.org>,
Paolo Bonzini <pbonzini@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>,
Vitaly Kuznetsov <vkuznets@...hat.com>,
Wanpeng Li <wanpengli@...cent.com>,
Jim Mattson <jmattson@...gle.com>,
Joerg Roedel <joro@...tes.org>
Subject: Re: [PATCH v2 1/2] KVM: VMX: FIXED+PHYSICAL mode single target IPI fastpath
On Wed, 20 Nov 2019 at 02:36, Sean Christopherson
<sean.j.christopherson@...el.com> wrote:
[...]
>
> From 1ea8ff1aa766928c869ef7c1eb437fe4f7b8daf9 Mon Sep 17 00:00:00 2001
> From: Sean Christopherson <sean.j.christopherson@...el.com>
> Date: Tue, 19 Nov 2019 09:50:42 -0800
> Subject: [PATCH] KVM: x86: Add a fast path for sending virtual IPIs in x2APIC
> mode
>
> Add a fast path to handle writes to the ICR when the local APIC is
> emulated in the kernel and x2APIC is enabled. The fast path is invoked
> at ->handle_exit_irqoff() to emulate only the effect of the ICR write
> itself, i.e. the sending of IPIs. Sending IPIs early in the VM-Exit
> flow reduces the latency of virtual IPIs by avoiding the expensive bits
> of transitioning from guest to host, e.g. reacquiring KVM's SRCU lock.
>
> Suggested-by: Wanpeng Li <wanpengli@...cent.com>
> Signed-off-by: Sean Christopherson <sean.j.christopherson@...el.com>
Hmm, I welcome the idea to improve the original patch, but this is too
much for me. :(
Powered by blists - more mailing lists