Re: What does KVM_HINTS_REALTIME do?

From: Wanpeng Li <kernellwp@gmail.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: kvm <kvm@vger.kernel.org>, Jenifer Abrams <jhopper@redhat.com>,
	atheurer@redhat.com, jmario@redhat.com,
	Paolo Bonzini <pbonzini@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Wanpeng Li <wanpengli@tencent.com>,
	Eduardo Habkost <ehabkost@redhat.com>,
	Maxim Levitsky <mlevitsk@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: What does KVM_HINTS_REALTIME do?
Date: Wed, 4 Aug 2021 08:38:50 +0800	[thread overview]
Message-ID: <CANRm+Cy6CN2hF=BxZagkm0amUoQ8UcuD0JGvCV_Sj7bJFit0ow@mail.gmail.com> (raw)
In-Reply-To: <YQlOeGxhor3wJM6i@stefanha-x1.localdomain>

On Tue, 3 Aug 2021 at 22:14, Stefan Hajnoczi <stefanha@redhat.com> wrote:
>
> Hi,
> I was just in a discussion where we realized KVM_HINTS_REALTIME is a
> little underdocumented. Here is attempt to address that. Please correct
> me if there are inaccuracies or reply if you have additional questions:
>
> KVM_HINTS_REALTIME (aka QEMU kvm-hint-dedicated) is defined as follows
> in Documentation/virt/kvm/cpuid.rst:
>
>   guest checks this feature bit to determine that vCPUs are never
>   preempted for an unlimited time allowing optimizations
>
> Users or management tools set this flag themselves (it is not set
> automatically). This raises the question of what effects this flag has
> and when it should be set.
>
> When should I set KVM_HINTS_REALTIME?
> -------------------------------------
> When vCPUs are pinned to dedicated pCPUs. Even better if the isolcpus=
> kernel parameter is used on the host so there are no disturbances.
>
> Is the flag guest-wide or per-vCPU?
> -----------------------------------
> This flag is guest-wide so all vCPUs should be dedicated, not just some
> of them.
>
> Which Linux guest features are affected?
> ----------------------------------------
> PV spinlocks, PV TLB flush, and PV sched yield are disabled by
> KVM_HINTS_REALTIME. This is because no other vCPUs or host tasks will be
> running on the pCPUs, so there is no benefit in involving the host.
>
> The cpuidle-haltpoll driver is enabled by KVM_HINTS_REALTIME. This
> driver performs busy waiting inside the guest before halting the CPU in
> order to avoid the vCPU's wakeup latency. This driver also has a boolean
> "force" module parameter if you wish to enable it without setting
> KVM_HINTS_REALTIME.
>
> When KVM_HINTS_REALTIME is set, the KVM_CAP_X86_DISABLE_EXITS capability
> can also be used to disable MWAIT/HLT/PAUSE/CSTATE exits. This improves
> the latency of these operations. The user or management tools need to
> disable these exits themselves, e.g. with QEMU's -overcommit cpu-pm=on.

Looks good to me, too. :)

    Wanpeng