From: Wanpeng Li <kernellwp@gmail.com>
To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Cc: "Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>,
"Sean Christopherson" <sean.j.christopherson@intel.com>,
"Liran Alon" <liran.alon@oracle.com>
Subject: [PATCH v3 4/5] KVM: LAPIC: Delay trace advance expire delta
Date: Thu, 16 May 2019 11:06:19 +0800 [thread overview]
Message-ID: <1557975980-9875-5-git-send-email-wanpengli@tencent.com> (raw)
In-Reply-To: <1557975980-9875-1-git-send-email-wanpengli@tencent.com>
From: Wanpeng Li <wanpengli@tencent.com>
wait_lapic_expire() call was moved above guest_enter_irqoff() because of
its tracepoint, which violated the RCU extended quiescent state invoked
by guest_enter_irqoff()[1][2]. This patch simply moves the tracepoint
below guest_exit_irqoff() in vcpu_enter_guest(). Snapshot the delta before
VM-Enter, but trace it after VM-Exit. This can help us to move
wait_lapic_expire() just before vmentry in the later patch.
[1] Commit 8b89fe1f6c43 ("kvm: x86: move tracepoints outside extended quiescent state")
[2] https://patchwork.kernel.org/patch/7821111/
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Sean Christopherson <sean.j.christopherson@intel.com>
Cc: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
---
arch/x86/kvm/lapic.c | 16 ++++++++--------
arch/x86/kvm/lapic.h | 1 +
arch/x86/kvm/x86.c | 2 ++
3 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 2f364fe..af38ece 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -1502,27 +1502,27 @@ static inline void __wait_lapic_expire(struct kvm_vcpu *vcpu, u64 guest_cycles)
}
static inline void adaptive_tune_timer_advancement(struct kvm_vcpu *vcpu,
- u64 guest_tsc, u64 tsc_deadline)
+ s64 advance_expire_delta)
{
struct kvm_lapic *apic = vcpu->arch.apic;
u32 timer_advance_ns = apic->lapic_timer.timer_advance_ns;
u64 ns;
/* too early */
- if (guest_tsc < tsc_deadline) {
- ns = (tsc_deadline - guest_tsc) * 1000000ULL;
+ 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);
} else {
/* too late */
- ns = (guest_tsc - tsc_deadline) * 1000000ULL;
+ 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);
}
- if (abs(guest_tsc - tsc_deadline) < LAPIC_TIMER_ADVANCE_ADJUST_DONE)
+ if (abs(advance_expire_delta) < LAPIC_TIMER_ADVANCE_ADJUST_DONE)
apic->lapic_timer.timer_advance_adjust_done = true;
if (unlikely(timer_advance_ns > 5000)) {
timer_advance_ns = 0;
@@ -1545,13 +1545,13 @@ void wait_lapic_expire(struct kvm_vcpu *vcpu)
tsc_deadline = apic->lapic_timer.expired_tscdeadline;
apic->lapic_timer.expired_tscdeadline = 0;
guest_tsc = kvm_read_l1_tsc(vcpu, rdtsc());
- trace_kvm_wait_lapic_expire(vcpu->vcpu_id, guest_tsc - tsc_deadline);
+ apic->lapic_timer.advance_expire_delta = guest_tsc - tsc_deadline;
- if (guest_tsc < tsc_deadline)
+ if (apic->lapic_timer.advance_expire_delta < 0)
__wait_lapic_expire(vcpu, tsc_deadline - guest_tsc);
if (unlikely(!apic->lapic_timer.timer_advance_adjust_done))
- adaptive_tune_timer_advancement(vcpu, guest_tsc, tsc_deadline);
+ adaptive_tune_timer_advancement(vcpu, apic->lapic_timer.advance_expire_delta);
}
static void start_sw_tscdeadline(struct kvm_lapic *apic)
diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
index d6d049b..3e72a25 100644
--- a/arch/x86/kvm/lapic.h
+++ b/arch/x86/kvm/lapic.h
@@ -32,6 +32,7 @@ struct kvm_timer {
u64 tscdeadline;
u64 expired_tscdeadline;
u32 timer_advance_ns;
+ s64 advance_expire_delta;
atomic_t pending; /* accumulated triggered timers */
bool hv_timer_in_use;
bool timer_advance_adjust_done;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index f2e3847..4a7b00c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7961,6 +7961,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
++vcpu->stat.exits;
guest_exit_irqoff();
+ trace_kvm_wait_lapic_expire(vcpu->vcpu_id,
+ vcpu->arch.apic->lapic_timer.advance_expire_delta);
local_irq_enable();
preempt_enable();
--
2.7.4
next prev parent reply other threads:[~2019-05-16 3:06 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-16 3:06 [PATCH v3 0/5] KVM: LAPIC: Optimize timer latency further Wanpeng Li
2019-05-16 3:06 ` [PATCH v3 1/5] KVM: LAPIC: Extract adaptive tune timer advancement logic Wanpeng Li
2019-05-16 3:06 ` [PATCH v3 2/5] KVM: LAPIC: Fix lapic_timer_advance_ns parameter overflow Wanpeng Li
2019-05-16 3:06 ` [PATCH v3 3/5] KVM: LAPIC: Expose per-vCPU timer_advance_ns to userspace Wanpeng Li
2019-05-17 20:05 ` Sean Christopherson
2019-05-20 8:43 ` Wanpeng Li
2019-05-16 3:06 ` Wanpeng Li [this message]
2019-05-17 19:44 ` [PATCH v3 4/5] KVM: LAPIC: Delay trace advance expire delta Sean Christopherson
2019-05-20 6:38 ` Wanpeng Li
2019-05-20 14:56 ` Sean Christopherson
2019-05-16 3:06 ` [PATCH v3 5/5] KVM: LAPIC: Optimize timer latency further Wanpeng Li
2019-05-17 19:50 ` Sean Christopherson
2019-05-20 8:19 ` Wanpeng Li
2019-05-20 15:08 ` Sean Christopherson
2019-06-12 9:35 [PATCH v3 0/5] " Wanpeng Li
2019-06-12 9:35 ` [PATCH v3 4/5] KVM: LAPIC: Delay trace advance expire delta 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=1557975980-9875-5-git-send-email-wanpengli@tencent.com \
--to=kernellwp@gmail.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=liran.alon@oracle.com \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=sean.j.christopherson@intel.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;
as well as URLs for NNTP newsgroup(s).