From: mark.rutland@arm.com (Mark Rutland)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 04/11] arm: perf: filter unschedulable events
Date: Fri, 7 Nov 2014 16:25:29 +0000 [thread overview]
Message-ID: <1415377536-12841-5-git-send-email-mark.rutland@arm.com> (raw)
In-Reply-To: <1415377536-12841-1-git-send-email-mark.rutland@arm.com>
Different CPU microarchitectures implement different PMU events, and
thus events which can be scheduled on one microarchitecture cannot be
scheduled on another, and vice-versa. Some archicted events behave
differently across microarchitectures, and thus cannot be meaningfully
summed. Due to this, we reject the scheduling of an event on a CPU of a
different microarchitecture to that the event targets.
When the core perf code is scheduling events and encounters an event
which cannot be scheduled, it stops attempting to schedule events. As
the perf core periodically rotates the list of events, for some
proportion of the time events which are unschedulable will block events
which are schedulable, resulting in low utilisation of the hardware
counters.
This patch implements a pmu::filter_match callback such that we can
detect and skip such events while scheduling early, before they can
block the schedulable events. This prevents the low HW counter
utilisation issue.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
---
arch/arm/kernel/perf_event.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index 9ad21ab..b00f6aa 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -509,6 +509,18 @@ static void armpmu_disable(struct pmu *pmu)
armpmu->stop(armpmu);
}
+/*
+ * In heterogeneous systems, events are specific to a particular
+ * microarchitecture, and aren't suitable for another. Thus, only match CPUs of
+ * the same microarchitecture.
+ */
+static int armpmu_filter_match(struct perf_event *event)
+{
+ struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+ unsigned int cpu = smp_processor_id();
+ return cpumask_test_cpu(cpu, &armpmu->supported_cpus);
+}
+
#ifdef CONFIG_PM_RUNTIME
static int armpmu_runtime_resume(struct device *dev)
{
@@ -549,6 +561,7 @@ static void armpmu_init(struct arm_pmu *armpmu)
.start = armpmu_start,
.stop = armpmu_stop,
.read = armpmu_read,
+ .filter_match = armpmu_filter_match,
};
}
--
1.9.1
next prev parent reply other threads:[~2014-11-07 16:25 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-07 16:25 [PATCH 00/11] arm: perf: add support for heterogeneous PMUs Mark Rutland
2014-11-07 16:25 ` [PATCH 01/11] of: Add empty of_get_next_parent stub Mark Rutland
2014-11-07 16:25 ` [PATCH 02/11] perf: allow for PMU-specific event filtering Mark Rutland
2014-11-07 16:25 ` [PATCH 03/11] arm: perf: treat PMUs as CPU affine Mark Rutland
2014-11-07 16:25 ` Mark Rutland [this message]
2014-11-07 16:25 ` [PATCH 05/11] arm: perf: reject multi-pmu groups Mark Rutland
2014-11-07 16:25 ` [PATCH 06/11] arm: perf: probe number of counters on affine CPUs Mark Rutland
2014-11-07 16:25 ` [PATCH 07/11] arm: perf: document PMU affinity binding Mark Rutland
2014-11-17 11:14 ` Will Deacon
2014-11-17 14:32 ` Rob Herring
2014-11-17 15:01 ` Mark Rutland
2014-11-07 16:25 ` [PATCH 08/11] arm: perf: add functions to parse affinity from dt Mark Rutland
2014-11-17 11:16 ` Will Deacon
2014-11-17 15:02 ` Mark Rutland
2014-11-07 16:25 ` [PATCH 09/11] arm: perf: parse cpu " Mark Rutland
2014-11-17 11:20 ` Will Deacon
2014-11-17 15:08 ` Mark Rutland
2014-11-18 10:40 ` Will Deacon
2014-11-07 16:25 ` [PATCH 10/11] arm: perf: remove singleton PMU restriction Mark Rutland
2014-11-07 16:25 ` [PATCH 11/11] arm: dts: vexpress: describe all PMUs in TC2 dts Mark Rutland
2014-11-17 11:24 ` [PATCH 00/11] arm: perf: add support for heterogeneous PMUs 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=1415377536-12841-5-git-send-email-mark.rutland@arm.com \
--to=mark.rutland@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
/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).