public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: Yicong Yang <yangyicong@huawei.com>
To: <will@kernel.org>, <mark.rutland@arm.com>,
	<linux-arm-kernel@lists.infradead.org>
Cc: <sudeep.holla@arm.com>, <james.clark@linaro.org>,
	<robh@kernel.org>, <anshuman.khandual@arm.com>,
	<jonathan.cameron@huawei.com>, <hejunhao3@huawei.com>,
	<linuxarm@huawei.com>, <prime.zeng@hisilicon.com>,
	<xuwei5@huawei.com>, <wangyushan12@huawei.com>,
	<yangyicong@hisilicon.com>
Subject: [PATCH v2 2/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores
Date: Wed, 20 Aug 2025 16:45:34 +0800	[thread overview]
Message-ID: <20250820084534.28037-3-yangyicong@huawei.com> (raw)
In-Reply-To: <20250820084534.28037-1-yangyicong@huawei.com>

From: Yicong Yang <yangyicong@hisilicon.com>

CPU_CYCLES is expected to count the logical CPU (PE) clock. Currently it's
preferred to use PMCCNTR_EL0 for counting CPU_CYCLES, but it'll count
processor clock rather than the PE clock (ARM DDI0487 L.b D13.1.3) if
one of the SMT siblings is not idle on a multi-threaded implementation.
So don't use it on SMT cores.

Introduce topology_core_has_smt() for knowing the SMT implementation and
cached it in arm_pmu::has_smt during allocation.

When counting cycles on SMT CPU 2-3 and CPU 3 is idle, without this
patch we'll get:
[root@client1 tmp]# perf stat -e cycles -A -C 2-3 -- stress-ng -c 1
--taskset 2 --timeout 1
[...]
 Performance counter stats for 'CPU(s) 2-3':

CPU2           2880457316      cycles
CPU3           2880459810      cycles
       1.254688470 seconds time elapsed

With this patch the idle state of CPU3 is observed as expected:
[root@client1 ~]#  perf stat -e cycles -A -C 2-3 -- stress-ng -c 1
--taskset 2 --timeout 1
[...]
 Performance counter stats for 'CPU(s) 2-3':

CPU2           2558580492      cycles
CPU3               305749      cycles
       1.113626410 seconds time elapsed

Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
---
 drivers/perf/arm_pmu.c        |  3 +++
 drivers/perf/arm_pmuv3.c      | 10 ++++++++++
 include/linux/arch_topology.h | 11 +++++++++++
 include/linux/perf/arm_pmu.h  |  1 +
 4 files changed, 25 insertions(+)

diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
index 5c310e803dd7..137ef55d6973 100644
--- a/drivers/perf/arm_pmu.c
+++ b/drivers/perf/arm_pmu.c
@@ -901,6 +901,9 @@ struct arm_pmu *armpmu_alloc(void)
 
 		events = per_cpu_ptr(pmu->hw_events, cpu);
 		events->percpu_pmu = pmu;
+
+		if (!pmu->has_smt && topology_core_has_smt(cpu))
+			pmu->has_smt = true;
 	}
 
 	return pmu;
diff --git a/drivers/perf/arm_pmuv3.c b/drivers/perf/arm_pmuv3.c
index 69c5cc8f5606..32b58a0feb33 100644
--- a/drivers/perf/arm_pmuv3.c
+++ b/drivers/perf/arm_pmuv3.c
@@ -981,6 +981,7 @@ static int armv8pmu_get_chain_idx(struct pmu_hw_events *cpuc,
 static bool armv8pmu_can_use_pmccntr(struct pmu_hw_events *cpuc,
 				     struct perf_event *event)
 {
+	struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
 	struct hw_perf_event *hwc = &event->hw;
 	unsigned long evtype = hwc->config_base & ARMV8_PMU_EVTYPE_EVENT;
 
@@ -1001,6 +1002,15 @@ static bool armv8pmu_can_use_pmccntr(struct pmu_hw_events *cpuc,
 	if (has_branch_stack(event))
 		return false;
 
+	/*
+	 * The PMCCNTR_EL0 increments from the processor clock rather than
+	 * the PE clock (ARM DDI0487 L.b D13.1.3) which means it'll continue
+	 * counting on a WFI PE if one of its SMT silbing is not idle on a
+	 * multi-threaded implementation. So don't use it on SMT cores.
+	 */
+	if (cpu_pmu->has_smt)
+		return false;
+
 	return true;
 }
 
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index d72d6e5aa200..daa1af2e8204 100644
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -89,6 +89,17 @@ void remove_cpu_topology(unsigned int cpuid);
 void reset_cpu_topology(void);
 int parse_acpi_topology(void);
 void freq_inv_set_max_ratio(int cpu, u64 max_rate);
+
+/*
+ * Architectures like ARM64 don't have reliable architectural way to get SMT
+ * information and depend on the firmware (ACPI/OF) report. Non-SMT core won't
+ * initialize thread_id so we can use this to detect the SMT implementation.
+ */
+static inline bool topology_core_has_smt(int cpu)
+{
+	return cpu_topology[cpu].thread_id != -1;
+}
+
 #endif
 
 #endif /* _LINUX_ARCH_TOPOLOGY_H_ */
diff --git a/include/linux/perf/arm_pmu.h b/include/linux/perf/arm_pmu.h
index 93c9a26492fc..2d39322c40c4 100644
--- a/include/linux/perf/arm_pmu.h
+++ b/include/linux/perf/arm_pmu.h
@@ -119,6 +119,7 @@ struct arm_pmu {
 
 	/* PMUv3 only */
 	int		pmuver;
+	bool		has_smt;
 	u64		reg_pmmir;
 	u64		reg_brbidr;
 #define ARMV8_PMUV3_MAX_COMMON_EVENTS		0x40
-- 
2.24.0



  parent reply	other threads:[~2025-08-20  9:30 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-20  8:45 [PATCH v2 0/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores Yicong Yang
2025-08-20  8:45 ` [PATCH v2 1/2] perf: arm_pmuv3: Factor out PMCCNTR_EL0 use conditions Yicong Yang
2025-08-20  8:45 ` Yicong Yang [this message]
2025-09-18 13:32   ` [PATCH v2 2/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores Will Deacon
2025-09-19  8:56     ` Yicong Yang
2025-09-19  9:16       ` Mark Rutland
2025-09-19 10:27         ` Yicong Yang
2025-09-19 11:17           ` Will Deacon
2025-09-19  9:37     ` Sudeep Holla
2025-09-19 11:18       ` Will Deacon
2025-09-08  7:09 ` [PATCH v2 0/2] " Yicong Yang
2025-09-18 16:43 ` Will Deacon

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=20250820084534.28037-3-yangyicong@huawei.com \
    --to=yangyicong@huawei.com \
    --cc=anshuman.khandual@arm.com \
    --cc=hejunhao3@huawei.com \
    --cc=james.clark@linaro.org \
    --cc=jonathan.cameron@huawei.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linuxarm@huawei.com \
    --cc=mark.rutland@arm.com \
    --cc=prime.zeng@hisilicon.com \
    --cc=robh@kernel.org \
    --cc=sudeep.holla@arm.com \
    --cc=wangyushan12@huawei.com \
    --cc=will@kernel.org \
    --cc=xuwei5@huawei.com \
    --cc=yangyicong@hisilicon.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