* [patch] sched: improve tick time missed wakeup preempt protection
@ 2009-11-22 12:08 Mike Galbraith
2009-11-22 13:16 ` Peter Zijlstra
0 siblings, 1 reply; 3+ messages in thread
From: Mike Galbraith @ 2009-11-22 12:08 UTC (permalink / raw)
To: Peter Zijlstra, Ingo Molnar; +Cc: LKML
sched: improve tick time missed wakeup preempt protection
f685ceac provides protection from tasks just missing wakeup preemption, and then
having to wait a full slice. However, it offers this protection to tasks which
have no business receiving the benefit, namely SCHED_BATCH and SCHED_IDLE. It
also treats all tasks equally, which obviously isn't true. Exclude tasks of
other than SCHED_NORMAL class, and scale minimum runtime before a tick time
preemption by the difference in task weights, after which, we can just use the
standard wakeup preempt vruntime delta test, sysctl_sched_wakeup_granularity.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
---
kernel/sched_fair.c | 20 +++++++++++++-------
1 file changed, 13 insertions(+), 7 deletions(-)
Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -830,17 +830,23 @@ check_preempt_tick(struct cfs_rq *cfs_rq
* narrow margin doesn't have to wait for a full slice.
* This also mitigates buddy induced latencies under load.
*/
- if (!sched_feat(WAKEUP_PREEMPT))
+ if (!sched_feat(WAKEUP_PREEMPT) || cfs_rq->nr_running < 2)
return;
-
- if (delta_exec < sysctl_sched_min_granularity)
- return;
-
- if (cfs_rq->nr_running > 1) {
+ else {
struct sched_entity *se = __pick_next_entity(cfs_rq);
+ unsigned long min = sysctl_sched_min_granularity;
s64 delta = curr->vruntime - se->vruntime;
- if (delta > ideal_runtime)
+ if (task_of(se)->policy != SCHED_NORMAL)
+ return;
+ if (delta < 0)
+ return;
+ if (curr->load.weight != se->load.weight)
+ min = calc_delta_mine(min, curr->load.weight, &se->load);
+ if (delta_exec < min)
+ return;
+
+ if (delta > sysctl_sched_wakeup_granularity)
resched_task(rq_of(cfs_rq)->curr);
}
}
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [patch] sched: improve tick time missed wakeup preempt protection
2009-11-22 12:08 [patch] sched: improve tick time missed wakeup preempt protection Mike Galbraith
@ 2009-11-22 13:16 ` Peter Zijlstra
2009-11-22 16:50 ` Mike Galbraith
0 siblings, 1 reply; 3+ messages in thread
From: Peter Zijlstra @ 2009-11-22 13:16 UTC (permalink / raw)
To: Mike Galbraith; +Cc: Ingo Molnar, LKML
On Sun, 2009-11-22 at 13:08 +0100, Mike Galbraith wrote:
> sched: improve tick time missed wakeup preempt protection
>
> f685ceac provides protection from tasks just missing wakeup preemption, and then
> having to wait a full slice. However, it offers this protection to tasks which
> have no business receiving the benefit, namely SCHED_BATCH and SCHED_IDLE. It
> also treats all tasks equally, which obviously isn't true. Exclude tasks of
> other than SCHED_NORMAL class, and scale minimum runtime before a tick time
> preemption by the difference in task weights, after which, we can just use the
> standard wakeup preempt vruntime delta test, sysctl_sched_wakeup_granularity.
>
> Signed-off-by: Mike Galbraith <efault@gmx.de>
> Cc: Ingo Molnar <mingo@elte.hu>
> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
> LKML-Reference: <new-submission>
>
> ---
> kernel/sched_fair.c | 20 +++++++++++++-------
> 1 file changed, 13 insertions(+), 7 deletions(-)
>
> Index: linux-2.6/kernel/sched_fair.c
> ===================================================================
> --- linux-2.6.orig/kernel/sched_fair.c
> +++ linux-2.6/kernel/sched_fair.c
> @@ -830,17 +830,23 @@ check_preempt_tick(struct cfs_rq *cfs_rq
> * narrow margin doesn't have to wait for a full slice.
> * This also mitigates buddy induced latencies under load.
> */
> - if (!sched_feat(WAKEUP_PREEMPT))
> + if (!sched_feat(WAKEUP_PREEMPT) || cfs_rq->nr_running < 2)
> return;
> -
> - if (delta_exec < sysctl_sched_min_granularity)
> - return;
> -
> - if (cfs_rq->nr_running > 1) {
> + else {
> struct sched_entity *se = __pick_next_entity(cfs_rq);
> + unsigned long min = sysctl_sched_min_granularity;
> s64 delta = curr->vruntime - se->vruntime;
>
> - if (delta > ideal_runtime)
> + if (task_of(se)->policy != SCHED_NORMAL)
> + return;
> + if (delta < 0)
> + return;
> + if (curr->load.weight != se->load.weight)
> + min = calc_delta_mine(min, curr->load.weight, &se->load);
> + if (delta_exec < min)
> + return;
> +
> + if (delta > sysctl_sched_wakeup_granularity)
> resched_task(rq_of(cfs_rq)->curr);
> }
> }
You can loose the else, the if branch does an unconditional return,
there's no other way to get below there than 'else' ;-)
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [patch] sched: improve tick time missed wakeup preempt protection
2009-11-22 13:16 ` Peter Zijlstra
@ 2009-11-22 16:50 ` Mike Galbraith
0 siblings, 0 replies; 3+ messages in thread
From: Mike Galbraith @ 2009-11-22 16:50 UTC (permalink / raw)
To: Peter Zijlstra; +Cc: Ingo Molnar, LKML
On Sun, 2009-11-22 at 14:16 +0100, Peter Zijlstra wrote:
> You can loose the else, the if branch does an unconditional return,
> there's no other way to get below there than 'else' ;-)
Ok. Can't plug tail into a function, doesn't fit on a line, so..
sched: improve tick time missed wakeup preempt protection
f685ceac provides protection from tasks just missing wakeup preemption, and then
having to wait a full slice. However, it offers this protection to tasks which
have no business receiving the benefit, namely SCHED_BATCH and SCHED_IDLE. It
also treats all tasks equally, which obviously isn't true. Exclude tasks of
other than SCHED_NORMAL class, and scale minimum runtime before a tick time
preemption by the difference in task weights, after which, we can just use the
standard wakeup preempt vruntime delta test, sysctl_sched_wakeup_granularity.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
---
kernel/sched_fair.c | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)
Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -811,7 +811,10 @@ dequeue_entity(struct cfs_rq *cfs_rq, st
static void
check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
{
+ struct sched_entity *next;
unsigned long ideal_runtime, delta_exec;
+ unsigned long min = sysctl_sched_min_granularity;
+ s64 delta;
ideal_runtime = sched_slice(cfs_rq, curr);
delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime;
@@ -825,24 +828,28 @@ check_preempt_tick(struct cfs_rq *cfs_rq
return;
}
+ if (!sched_feat(WAKEUP_PREEMPT) || cfs_rq->nr_running < 2)
+ return;
+
/*
* Ensure that a task that missed wakeup preemption by a
* narrow margin doesn't have to wait for a full slice.
* This also mitigates buddy induced latencies under load.
*/
- if (!sched_feat(WAKEUP_PREEMPT))
- return;
+ next = __pick_next_entity(cfs_rq);
+ delta = curr->vruntime - next->vruntime;
- if (delta_exec < sysctl_sched_min_granularity)
+ if (task_of(next)->policy != SCHED_NORMAL)
+ return;
+ if (delta < 0)
+ return;
+ if (curr->load.weight != next->load.weight)
+ min = calc_delta_mine(min, curr->load.weight, &next->load);
+ if (delta_exec < min)
return;
- if (cfs_rq->nr_running > 1) {
- struct sched_entity *se = __pick_next_entity(cfs_rq);
- s64 delta = curr->vruntime - se->vruntime;
-
- if (delta > ideal_runtime)
- resched_task(rq_of(cfs_rq)->curr);
- }
+ if (delta > sysctl_sched_wakeup_granularity)
+ resched_task(rq_of(cfs_rq)->curr);
}
static void
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2009-11-22 16:50 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-11-22 12:08 [patch] sched: improve tick time missed wakeup preempt protection Mike Galbraith
2009-11-22 13:16 ` Peter Zijlstra
2009-11-22 16:50 ` Mike Galbraith
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.