From: Like Xu <like.xu.linux@gmail.com>
To: Paolo Bonzini <pbonzini@redhat.com>, Jim Mattson <jmattson@google.com>
Cc: Sean Christopherson <seanjc@google.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Wanpeng Li <wanpengli@tencent.com>,
Joerg Roedel <joro@8bytes.org>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH 4/4] KVM: x86/pmu: Refactoring kvm_perf_overflow{_intr}()
Date: Tue, 16 Nov 2021 20:20:30 +0800 [thread overview]
Message-ID: <20211116122030.4698-5-likexu@tencent.com> (raw)
In-Reply-To: <20211116122030.4698-1-likexu@tencent.com>
From: Like Xu <likexu@tencent.com>
Depending on whether intr should be triggered or not, KVM registers
two different event overflow callbacks in the perf_event context.
The code skeleton of these two functions is very similar, so
need_overflow_intr() is introduced to increase the potential code reuse.
There is a trade-off between extra cycles in the irq context and a
smaller instructions footprint against the u-architecture branch predictor.
Signed-off-by: Like Xu <likexu@tencent.com>
---
arch/x86/kvm/pmu.c | 64 +++++++++++++++++++++++++---------------------
1 file changed, 35 insertions(+), 29 deletions(-)
diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index 3c45467b4275..ef4bba8be7f7 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -55,43 +55,50 @@ static void kvm_pmi_trigger_fn(struct irq_work *irq_work)
kvm_pmu_deliver_pmi(vcpu);
}
-static void kvm_perf_overflow(struct perf_event *perf_event,
- struct perf_sample_data *data,
- struct pt_regs *regs)
+static inline bool need_overflow_intr(struct kvm_pmc *pmc)
{
- struct kvm_pmc *pmc = perf_event->overflow_handler_context;
struct kvm_pmu *pmu = pmc_to_pmu(pmc);
- if (!test_and_set_bit(pmc->idx, pmu->reprogram_pmi)) {
- __set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
- kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
- }
+ if (pmc_is_gp(pmc))
+ return (pmc->eventsel & ARCH_PERFMON_EVENTSEL_INT);
+ else
+ return fixed_ctrl_field(pmu->fixed_ctr_ctrl,
+ pmc->idx - INTEL_PMC_IDX_FIXED) & 0x8;
}
-static void kvm_perf_overflow_intr(struct perf_event *perf_event,
- struct perf_sample_data *data,
- struct pt_regs *regs)
+static inline void kvm_pmu_counter_overflow(struct kvm_pmc *pmc, bool intr)
{
- struct kvm_pmc *pmc = perf_event->overflow_handler_context;
struct kvm_pmu *pmu = pmc_to_pmu(pmc);
- if (!test_and_set_bit(pmc->idx, pmu->reprogram_pmi)) {
- __set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
- kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
+ __set_bit(pmc->idx, (unsigned long *)&pmu->global_status);
+ kvm_make_request(KVM_REQ_PMU, pmc->vcpu);
- /*
- * Inject PMI. If vcpu was in a guest mode during NMI PMI
- * can be ejected on a guest mode re-entry. Otherwise we can't
- * be sure that vcpu wasn't executing hlt instruction at the
- * time of vmexit and is not going to re-enter guest mode until
- * woken up. So we should wake it, but this is impossible from
- * NMI context. Do it from irq work instead.
- */
- if (!kvm_is_in_guest())
- irq_work_queue(&pmc_to_pmu(pmc)->irq_work);
- else
- kvm_make_request(KVM_REQ_PMI, pmc->vcpu);
- }
+ if (!intr)
+ return;
+
+ /*
+ * Inject PMI. If vcpu was in a guest mode during NMI PMI
+ * can be ejected on a guest mode re-entry. Otherwise we can't
+ * be sure that vcpu wasn't executing hlt instruction at the
+ * time of vmexit and is not going to re-enter guest mode until
+ * woken up. So we should wake it, but this is impossible from
+ * NMI context. Do it from irq work instead.
+ */
+ if (!kvm_is_in_guest())
+ irq_work_queue(&pmc_to_pmu(pmc)->irq_work);
+ else
+ kvm_make_request(KVM_REQ_PMI, pmc->vcpu);
+}
+
+static void kvm_perf_overflow(struct perf_event *perf_event,
+ struct perf_sample_data *data,
+ struct pt_regs *regs)
+{
+ struct kvm_pmc *pmc = perf_event->overflow_handler_context;
+ struct kvm_pmu *pmu = pmc_to_pmu(pmc);
+
+ if (!test_and_set_bit(pmc->idx, pmu->reprogram_pmi))
+ kvm_pmu_counter_overflow(pmc, need_overflow_intr(pmc));
}
static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
@@ -126,7 +133,6 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
}
event = perf_event_create_kernel_counter(&attr, -1, current,
- intr ? kvm_perf_overflow_intr :
kvm_perf_overflow, pmc);
if (IS_ERR(event)) {
pr_debug_ratelimited("kvm_pmu: event creation failed %ld for pmc->idx = %d\n",
--
2.33.1
next prev parent reply other threads:[~2021-11-16 12:21 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-16 12:20 [PATCH 0/4] KVM: x86/pmu: An insightful refactoring of vPMU code Like Xu
2021-11-16 12:20 ` [PATCH 1/4] KVM: x86/pmu: Setup pmc->eventsel for fixed PMCs Like Xu
2021-11-16 12:20 ` [PATCH 2/4] KVM: x86/pmu: Refactoring find_arch_event() to find_perf_hw_id() Like Xu
2021-11-18 13:29 ` Paolo Bonzini
2021-11-16 12:20 ` [PATCH 3/4] KVM: x86/pmu: Reuse find_perf_hw_id() and drop find_fixed_event() Like Xu
2021-11-18 14:43 ` Paolo Bonzini
2021-11-18 15:00 ` Paolo Bonzini
2021-11-19 7:16 ` Like Xu
2021-11-19 10:29 ` Paolo Bonzini
2021-11-19 11:10 ` Like Xu
2021-11-16 12:20 ` Like Xu [this message]
2021-11-18 14:53 ` [PATCH 4/4] KVM: x86/pmu: Refactoring kvm_perf_overflow{_intr}() Paolo Bonzini
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=20211116122030.4698-5-likexu@tencent.com \
--to=like.xu.linux@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=seanjc@google.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;
as well as URLs for NNTP newsgroup(s).