summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorIngo Molnar <mingo@kernel.org>2013-04-26 10:05:59 +0200
committerIngo Molnar <mingo@kernel.org>2013-04-26 10:13:42 +0200
commit47aa8b6cbcb839efe2edaa5b50fee21df115d37b (patch)
treeb3c54811c75f40c11ba4344b8089156724d5e074
parent65e709dc0c25dbd563861924815e9a3a93878b75 (diff)
downloadop-kernel-dev-47aa8b6cbcb839efe2edaa5b50fee21df115d37b.zip
op-kernel-dev-47aa8b6cbcb839efe2edaa5b50fee21df115d37b.tar.gz
nohz: Reduce overhead under high-freq idling patterns
One testbox of mine (Intel Nehalem, 16-way) uses MWAIT for its idle routine, which apparently can break out of its idle loop rather frequently, with high frequency. In that case NO_HZ_FULL=y kernels show high ksoftirqd overhead and constant context switching, because tick_nohz_stop_sched_tick() will, if delta_jiffies == 0, mis-identify this as a timer event - activating the TIMER_SOFTIRQ, which wakes up ksoftirqd. Fix this by treating delta_jiffies == 0 the same way we treat other short wakeups, delta_jiffies == 1. Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Chris Metcalf <cmetcalf@tilera.com> Cc: Christoph Lameter <cl@linux.com> Cc: Geoff Levand <geoff@infradead.org> Cc: Gilad Ben Yossef <gilad@benyossef.com> Cc: Hakan Akkan <hakanakkan@gmail.com> Cc: Kevin Hilman <khilman@linaro.org> Cc: Li Zhong <zhong@linux.vnet.ibm.com> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Ingo Molnar <mingo@kernel.org>
-rw-r--r--kernel/time/tick-sched.c7
1 files changed, 4 insertions, 3 deletions
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 85e05ab..da53c8f 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -565,11 +565,12 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
delta_jiffies = rcu_delta_jiffies;
}
}
+
/*
- * Do not stop the tick, if we are only one off
- * or if the cpu is required for rcu
+ * Do not stop the tick, if we are only one off (or less)
+ * or if the cpu is required for RCU:
*/
- if (!ts->tick_stopped && delta_jiffies == 1)
+ if (!ts->tick_stopped && delta_jiffies <= 1)
goto out;
/* Schedule the tick, if we are at least one jiffie off */
OpenPOWER on IntegriCloud