From: Paolo Bonzini <pbonzini@redhat.com>
To: Wanpeng Li <kernellwp@gmail.com>,
linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: "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 v5 3/3] KVM: LAPIC: Tune lapic_timer_advance_ns smoothly
Date: Tue, 17 Sep 2019 19:07:15 +0200 [thread overview]
Message-ID: <879b1950-3ca7-0a53-9e4f-508fd5db4bd4@redhat.com> (raw)
In-Reply-To: <1568708186-20260-3-git-send-email-wanpengli@tencent.com>
On 17/09/19 10:16, Wanpeng Li wrote:
> From: Wanpeng Li <wanpengli@tencent.com>
>
> Filter out drastic fluctuation and random fluctuation, remove
> timer_advance_adjust_done altogether, the adjustment would be
> continuous.
>
> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Queued, thanks (I renamed the new variable to lapic_timer_advance_dynamic).
Thanks,
Paolo
> ---
> arch/x86/kvm/lapic.c | 28 ++++++++++++++--------------
> arch/x86/kvm/lapic.h | 1 -
> 2 files changed, 14 insertions(+), 15 deletions(-)
>
> diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> index dbbe478..323bdca 100644
> --- a/arch/x86/kvm/lapic.c
> +++ b/arch/x86/kvm/lapic.c
> @@ -65,7 +65,9 @@
> #define APIC_BROADCAST 0xFF
> #define X2APIC_BROADCAST 0xFFFFFFFFul
>
> -#define LAPIC_TIMER_ADVANCE_ADJUST_DONE 100
> +static bool dynamically_adjust_timer_advance __read_mostly;
> +#define LAPIC_TIMER_ADVANCE_ADJUST_MIN 100
> +#define LAPIC_TIMER_ADVANCE_ADJUST_MAX 5000
> #define LAPIC_TIMER_ADVANCE_ADJUST_INIT 1000
> /* step-by-step approximation to mitigate fluctuation */
> #define LAPIC_TIMER_ADVANCE_ADJUST_STEP 8
> @@ -1485,26 +1487,25 @@ static inline void adjust_lapic_timer_advance(struct kvm_vcpu *vcpu,
> u32 timer_advance_ns = apic->lapic_timer.timer_advance_ns;
> u64 ns;
>
> + /* Do not adjust for tiny fluctuations or large random spikes. */
> + if (abs(advance_expire_delta) > LAPIC_TIMER_ADVANCE_ADJUST_MAX ||
> + abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_MIN)
> + return;
> +
> /* too early */
> if (advance_expire_delta < 0) {
> ns = -advance_expire_delta * 1000000ULL;
> do_div(ns, vcpu->arch.virtual_tsc_khz);
> - timer_advance_ns -= min((u32)ns,
> - timer_advance_ns / LAPIC_TIMER_ADVANCE_ADJUST_STEP);
> + timer_advance_ns -= ns/LAPIC_TIMER_ADVANCE_ADJUST_STEP;
> } else {
> /* too late */
> ns = advance_expire_delta * 1000000ULL;
> do_div(ns, vcpu->arch.virtual_tsc_khz);
> - timer_advance_ns += min((u32)ns,
> - timer_advance_ns / LAPIC_TIMER_ADVANCE_ADJUST_STEP);
> + timer_advance_ns += ns/LAPIC_TIMER_ADVANCE_ADJUST_STEP;
> }
>
> - if (abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_DONE)
> - apic->lapic_timer.timer_advance_adjust_done = true;
> - if (unlikely(timer_advance_ns > 5000)) {
> + if (unlikely(timer_advance_ns > LAPIC_TIMER_ADVANCE_ADJUST_MAX))
> timer_advance_ns = LAPIC_TIMER_ADVANCE_ADJUST_INIT;
> - apic->lapic_timer.timer_advance_adjust_done = false;
> - }
> apic->lapic_timer.timer_advance_ns = timer_advance_ns;
> }
>
> @@ -1524,7 +1525,7 @@ static void __kvm_wait_lapic_expire(struct kvm_vcpu *vcpu)
> if (guest_tsc < tsc_deadline)
> __wait_lapic_expire(vcpu, tsc_deadline - guest_tsc);
>
> - if (unlikely(!apic->lapic_timer.timer_advance_adjust_done))
> + if (dynamically_adjust_timer_advance)
> adjust_lapic_timer_advance(vcpu, apic->lapic_timer.advance_expire_delta);
> }
>
> @@ -2302,13 +2303,12 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu, int timer_advance_ns)
> apic->lapic_timer.timer.function = apic_timer_fn;
> if (timer_advance_ns == -1) {
> apic->lapic_timer.timer_advance_ns = LAPIC_TIMER_ADVANCE_ADJUST_INIT;
> - apic->lapic_timer.timer_advance_adjust_done = false;
> + dynamically_adjust_timer_advance = true;
> } else {
> apic->lapic_timer.timer_advance_ns = timer_advance_ns;
> - apic->lapic_timer.timer_advance_adjust_done = true;
> + dynamically_adjust_timer_advance = false;
> }
>
> -
> /*
> * APIC is created enabled. This will prevent kvm_lapic_set_base from
> * thinking that APIC state has changed.
> diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
> index 50053d2..2aad7e2 100644
> --- a/arch/x86/kvm/lapic.h
> +++ b/arch/x86/kvm/lapic.h
> @@ -35,7 +35,6 @@ struct kvm_timer {
> s64 advance_expire_delta;
> atomic_t pending; /* accumulated triggered timers */
> bool hv_timer_in_use;
> - bool timer_advance_adjust_done;
> };
>
> struct kvm_lapic {
>
next prev parent reply other threads:[~2019-09-17 17:07 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-17 8:16 [PATCH 1/3] KVM: Fix coalesced mmio ring buffer out-of-bounds access Wanpeng Li
2019-09-17 8:16 ` [PATCH v2 2/3] KVM: X86: Fix userspace set broken combinations of CPUID and CR4 Wanpeng Li
2019-09-17 17:32 ` Sean Christopherson
2019-09-18 9:56 ` Wanpeng Li
2019-09-24 13:55 ` Paolo Bonzini
2019-09-17 8:16 ` [PATCH v5 3/3] KVM: LAPIC: Tune lapic_timer_advance_ns smoothly Wanpeng Li
2019-09-17 17:07 ` Paolo Bonzini [this message]
2019-09-17 8:18 ` [PATCH 1/3] KVM: Fix coalesced mmio ring buffer out-of-bounds access Paolo Bonzini
2019-09-17 14:58 ` Jim Mattson
2019-09-17 15:18 ` Matt Delco
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=879b1950-3ca7-0a53-9e4f-508fd5db4bd4@redhat.com \
--to=pbonzini@redhat.com \
--cc=jmattson@google.com \
--cc=joro@8bytes.org \
--cc=kernellwp@gmail.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®