aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorIngo Molnar <mingo@elte.hu>2010-09-12 02:14:52 -0400
committerIngo Molnar <mingo@elte.hu>2010-09-13 14:17:11 -0400
commit0bf377bbb0bea6130f35613491887cc622e42a8b (patch)
tree1ecb56fb6195dcdb673ef3a08390b98e54f4f726
parent49553c2ef88749dd502687f4eb9c258bb10a4f44 (diff)
sched: Improve latencies under load by decreasing minimum scheduling granularity
Mathieu reported bad latencies with make -j10 kind of kbuild workloads - which is mostly caused by us scheduling with a too coarse granularity. Reduce the minimum granularity some more, to make sure we can meet the latency target. I got the following results (make -j10 kbuild load, average of 3 runs): vanilla: maximum latency: 38278.9 µs average latency: 7730.1 µs patched: maximum latency: 22702.1 µs average latency: 6684.8 µs Mathieu also measured it: | | * wakeup-latency.c (SIGEV_THREAD) with make -j10 | | - Mainline 2.6.35.2 kernel | | maximum latency: 45762.1 µs | average latency: 7348.6 µs | | - With only Peter's smaller min_gran (shown below): | | maximum latency: 29100.6 µs | average latency: 6684.1 µs | Reported-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Suggested-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <AANLkTi=8m4g01wZPacySoF7U0PevTNVgJoZZrHiUD-pN@mail.gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
-rw-r--r--kernel/sched_fair.c6
1 files changed, 3 insertions, 3 deletions
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 9b5b4f86b742..a171138a9402 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -54,13 +54,13 @@ enum sched_tunable_scaling sysctl_sched_tunable_scaling
54 * Minimal preemption granularity for CPU-bound tasks: 54 * Minimal preemption granularity for CPU-bound tasks:
55 * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds) 55 * (default: 2 msec * (1 + ilog(ncpus)), units: nanoseconds)
56 */ 56 */
57unsigned int sysctl_sched_min_granularity = 2000000ULL; 57unsigned int sysctl_sched_min_granularity = 750000ULL;
58unsigned int normalized_sysctl_sched_min_granularity = 2000000ULL; 58unsigned int normalized_sysctl_sched_min_granularity = 750000ULL;
59 59
60/* 60/*
61 * is kept at sysctl_sched_latency / sysctl_sched_min_granularity 61 * is kept at sysctl_sched_latency / sysctl_sched_min_granularity
62 */ 62 */
63static unsigned int sched_nr_latency = 3; 63static unsigned int sched_nr_latency = 8;
64 64
65/* 65/*
66 * After fork, child runs first. If set to 0 (default) then 66 * After fork, child runs first. If set to 0 (default) then