diff options
author | Ingo Molnar <mingo@elte.hu> | 2007-11-09 22:39:38 +0100 |
---|---|---|
committer | Ingo Molnar <mingo@elte.hu> | 2007-11-09 22:39:38 +0100 |
commit | 19978ca610946ed57c071bad63f8f6642ca1298b (patch) | |
tree | 3bdbe89a249b6091fbd71653f106f486337ca3d6 /kernel/sched.c | |
parent | sched: restore deterministic CPU accounting on powerpc (diff) | |
download | linux-19978ca610946ed57c071bad63f8f6642ca1298b.tar.xz linux-19978ca610946ed57c071bad63f8f6642ca1298b.zip |
sched: reintroduce SMP tunings again
Yanmin Zhang reported an aim7 regression and bisected it down to:
| commit 38ad464d410dadceda1563f36bdb0be7fe4c8938
| Author: Ingo Molnar <mingo@elte.hu>
| Date: Mon Oct 15 17:00:02 2007 +0200
|
| sched: uniform tunings
|
| use the same defaults on both UP and SMP.
fix this by reintroducing similar SMP tunings again. This resolves
the regression.
(also update the comments to match the ilog2(nr_cpus) tuning effect)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'kernel/sched.c')
-rw-r--r-- | kernel/sched.c | 28 |
1 files changed, 28 insertions, 0 deletions
diff --git a/kernel/sched.c b/kernel/sched.c index 3f6bd1112900..69cae271c63b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4992,6 +4992,32 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) */ cpumask_t nohz_cpu_mask = CPU_MASK_NONE; +/* + * Increase the granularity value when there are more CPUs, + * because with more CPUs the 'effective latency' as visible + * to users decreases. But the relationship is not linear, + * so pick a second-best guess by going with the log2 of the + * number of CPUs. + * + * This idea comes from the SD scheduler of Con Kolivas: + */ +static inline void sched_init_granularity(void) +{ + unsigned int factor = 1 + ilog2(num_online_cpus()); + const unsigned long limit = 200000000; + + sysctl_sched_min_granularity *= factor; + if (sysctl_sched_min_granularity > limit) + sysctl_sched_min_granularity = limit; + + sysctl_sched_latency *= factor; + if (sysctl_sched_latency > limit) + sysctl_sched_latency = limit; + + sysctl_sched_wakeup_granularity *= factor; + sysctl_sched_batch_wakeup_granularity *= factor; +} + #ifdef CONFIG_SMP /* * This is how migration works: @@ -6688,10 +6714,12 @@ void __init sched_init_smp(void) /* Move init over to a non-isolated CPU */ if (set_cpus_allowed(current, non_isolated_cpus) < 0) BUG(); + sched_init_granularity(); } #else void __init sched_init_smp(void) { + sched_init_granularity(); } #endif /* CONFIG_SMP */ |