From: Oliver Upton <oliver.upton@linux.dev>
To: kvmarm@lists.linux.dev
Cc: Marc Zyngier <maz@kernel.org>, Joey Gouly <joey.gouly@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Zenghui Yu <yuzenghui@huawei.com>,
Mingwei Zhang <mizhang@google.com>,
Colton Lewis <coltonlewis@google.com>,
Raghavendra Rao Ananta <rananta@google.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, Janne Grunau <j@jannau.net>,
Oliver Upton <oliver.upton@linux.dev>
Subject: [PATCH v3 11/14] KVM: arm64: Remap PMUv3 events onto hardware
Date: Wed, 5 Mar 2025 12:26:38 -0800 [thread overview]
Message-ID: <20250305202641.428114-12-oliver.upton@linux.dev> (raw)
In-Reply-To: <20250305202641.428114-1-oliver.upton@linux.dev>
Map PMUv3 event IDs onto hardware, if the driver exposes such a helper.
This is expected to be quite rare, and only useful for non-PMUv3 hardware.
Tested-by: Janne Grunau <j@jannau.net>
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
---
arch/arm64/kvm/pmu-emul.c | 25 ++++++++++++++++++++++++-
include/linux/perf/arm_pmu.h | 4 ++++
2 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index 38d9490c17fd..5f4e9be8aa8b 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -677,6 +677,20 @@ static bool kvm_pmc_counts_at_el2(struct kvm_pmc *pmc)
return kvm_pmc_read_evtreg(pmc) & ARMV8_PMU_INCLUDE_EL2;
}
+static int kvm_map_pmu_event(struct kvm *kvm, unsigned int eventsel)
+{
+ struct arm_pmu *pmu = kvm->arch.arm_pmu;
+
+ /*
+ * The CPU PMU likely isn't PMUv3; let the driver provide a mapping
+ * for the guest's PMUv3 event ID.
+ */
+ if (unlikely(pmu->map_pmuv3_event))
+ return pmu->map_pmuv3_event(eventsel);
+
+ return eventsel;
+}
+
/**
* kvm_pmu_create_perf_event - create a perf event for a counter
* @pmc: Counter context
@@ -687,7 +701,8 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
struct arm_pmu *arm_pmu = vcpu->kvm->arch.arm_pmu;
struct perf_event *event;
struct perf_event_attr attr;
- u64 eventsel, evtreg;
+ int eventsel;
+ u64 evtreg;
evtreg = kvm_pmc_read_evtreg(pmc);
@@ -713,6 +728,14 @@ static void kvm_pmu_create_perf_event(struct kvm_pmc *pmc)
!test_bit(eventsel, vcpu->kvm->arch.pmu_filter))
return;
+ /*
+ * Don't create an event if we're running on hardware that requires
+ * PMUv3 event translation and we couldn't find a valid mapping.
+ */
+ eventsel = kvm_map_pmu_event(vcpu->kvm, eventsel);
+ if (eventsel < 0)
+ return;
+
memset(&attr, 0, sizeof(struct perf_event_attr));
attr.type = arm_pmu->pmu.type;
attr.size = sizeof(attr);
diff --git a/include/linux/perf/arm_pmu.h b/include/linux/perf/arm_pmu.h
index 4b5b83677e3f..7ce6dea5bfa9 100644
--- a/include/linux/perf/arm_pmu.h
+++ b/include/linux/perf/arm_pmu.h
@@ -100,6 +100,10 @@ struct arm_pmu {
void (*stop)(struct arm_pmu *);
void (*reset)(void *);
int (*map_event)(struct perf_event *event);
+ /*
+ * Called by KVM to map the PMUv3 event space onto non-PMUv3 hardware.
+ */
+ int (*map_pmuv3_event)(unsigned int eventsel);
DECLARE_BITMAP(cntr_mask, ARMPMU_MAX_HWEVENTS);
bool secure_access; /* 32-bit ARM only */
#define ARMV8_PMUV3_MAX_COMMON_EVENTS 0x40
--
2.39.5
next prev parent reply other threads:[~2025-03-05 20:49 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-05 20:26 [PATCH v3 00/14] KVM: arm64: FEAT_PMUv3 on Apple hardware Oliver Upton
2025-03-05 20:26 ` [PATCH v3 01/14] drivers/perf: apple_m1: Refactor event select/filter configuration Oliver Upton
2025-03-05 20:26 ` [PATCH v3 02/14] drivers/perf: apple_m1: Support host/guest event filtering Oliver Upton
2025-03-05 20:26 ` [PATCH v3 03/14] KVM: arm64: Compute PMCEID from arm_pmu's event bitmaps Oliver Upton
2025-03-05 20:26 ` [PATCH v3 04/14] KVM: arm64: Always support SW_INCR PMU event Oliver Upton
2025-03-05 20:26 ` [PATCH v3 05/14] KVM: arm64: Use a cpucap to determine if system supports FEAT_PMUv3 Oliver Upton
2025-03-05 20:26 ` [PATCH v3 06/14] KVM: arm64: Drop kvm_arm_pmu_available static key Oliver Upton
2025-03-05 20:26 ` [PATCH v3 07/14] KVM: arm64: Use guard() to cleanup usage of arm_pmus_lock Oliver Upton
2025-03-05 20:26 ` [PATCH v3 08/14] KVM: arm64: Move PMUVer filtering into KVM code Oliver Upton
2025-03-05 20:26 ` [PATCH v3 09/14] KVM: arm64: Compute synthetic sysreg ESR for Apple PMUv3 traps Oliver Upton
2025-03-05 20:26 ` [PATCH v3 10/14] KVM: arm64: Advertise PMUv3 if IMPDEF traps are present Oliver Upton
2025-03-05 20:26 ` Oliver Upton [this message]
2025-03-05 20:26 ` [PATCH v3 12/14] drivers/perf: apple_m1: Provide helper for mapping PMUv3 events Oliver Upton
2025-03-05 20:30 ` [PATCH v3 13/14] KVM: arm64: Provide 1 event counter on IMPDEF hardware Oliver Upton
2025-03-05 20:30 ` [PATCH v3 14/14] arm64: Enable IMP DEF PMUv3 traps on Apple M* Oliver Upton
2025-03-10 13:13 ` [PATCH v3 00/14] KVM: arm64: FEAT_PMUv3 on Apple hardware Marc Zyngier
2025-03-11 23:44 ` Oliver Upton
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=20250305202641.428114-12-oliver.upton@linux.dev \
--to=oliver.upton@linux.dev \
--cc=catalin.marinas@arm.com \
--cc=coltonlewis@google.com \
--cc=j@jannau.net \
--cc=joey.gouly@arm.com \
--cc=kvmarm@lists.linux.dev \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=maz@kernel.org \
--cc=mizhang@google.com \
--cc=rananta@google.com \
--cc=suzuki.poulose@arm.com \
--cc=will@kernel.org \
--cc=yuzenghui@huawei.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).