[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230620141002.23914-1-bvanassche@acm.org>
Date: Tue, 20 Jun 2023 07:10:02 -0700
From: Bart Van Assche <bvanassche@....org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...hat.com>,
Vincent Guittot <vincent.guittot@...aro.org>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>,
Bart Van Assche <bvanassche@....org>,
Juri Lelli <juri.lelli@...hat.com>
Subject: [PATCH v2] sched/cputime: Make IRQ time accounting configurable at boot time
IRQ time accounting reduces performance by 40% for some block storage
workloads on Android. Despite this some producers of Android devices
want to keep IRQ time accounting enabled.
Android devices share a single kernel binary across device vendors.
Make IRQ time accounting configurable at boot time such that Android
vendors can decide individually whether or not to enable IRQ time
accounting.
Cc: Ingo Molnar <mingo@...hat.com>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Juri Lelli <juri.lelli@...hat.com>
Cc: Vincent Guittot <vincent.guittot@...aro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@....com>
Cc: Steven Rostedt <rostedt@...dmis.org>
Signed-off-by: Bart Van Assche <bvanassche@....org>
---
Changes between v1 and v2: improved patch description.
Documentation/admin-guide/kernel-parameters.txt | 4 ++++
kernel/sched/cputime.c | 13 +++++++++++++
2 files changed, 17 insertions(+)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 9e5bab29685f..67a2ad3af833 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5611,6 +5611,10 @@
non-zero "wait" parameter. See weight_single
and weight_many.
+ sched_clock_irqtime= [KNL]
+ Can be used to disable IRQ time accounting if
+ CONFIG_IRQ_TIME_ACCOUNTING=y.
+
skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
xtime_lock contention on larger systems, and/or RCU lock
contention on all systems with CONFIG_MAXSMP set.
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index af7952f12e6c..d9c65017024d 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -24,6 +24,19 @@ DEFINE_PER_CPU(struct irqtime, cpu_irqtime);
static int sched_clock_irqtime;
+static int __init sched_clock_irqtime_setup(char *arg)
+{
+ bool enabled;
+
+ if (kstrtobool(arg, &enabled) < 0)
+ pr_err("Invalid sched_clock_irqtime value\n");
+ else
+ sched_clock_irqtime = enabled;
+ return 1;
+}
+
+__setup("sched_clock_irqtime=", sched_clock_irqtime_setup);
+
void enable_sched_clock_irqtime(void)
{
sched_clock_irqtime = 1;
Powered by blists - more mailing lists