Re: [RFC patch 1/2] sched: dynamically adapt granularity with nr_running

[Peter Zijlstra <peterz@infradead.org>]

On Mon, 2010-09-13 at 15:56 +0200, Mike Galbraith wrote:
> > One option is to simply get rid of that stuff in check_preempt_tick()
> > and instead do a wakeup-preempt check on the leftmost task instead.
> 
> That's what I wanted to boil it down to instead of putting the extra
> preempt check in, but it kills the longish slices of low load.  IIRC,
> when I tried that, it demolished throughput. 

Hrm.. yes it would..

So the reason for all this:

        /*
         * 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.
         */

Is to avoid tasks getting too far ahead in virtual time due to buddies,
right?

Would something like the below work? Don't actually use delta_exec to
filter, but use wakeup_gran + min_gran on virtual time, (much like Steve
suggested) and then verify using __sched_gran().

Or have I now totally confused myself backwards?

 - delta_exec is walltime, and should thus we compared against a
   weighted unit like slice,
 - delta is a vruntime unit, and is thus weight free, hence we can use
   granularity/unweighted units.


---
 kernel/sched_fair.c |   20 ++++++++++++++++----
 1 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 9b5b4f8..7f418de 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -485,6 +485,16 @@ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 	return slice;
 }
 
+static u64 __sched_gran(unsigned long nr_running)
+{
+	unsigned long latency = sysctl_sched_latency;
+
+	if (nr_running >= nr_latency)
+		return sysctl_sched_min_granularity;
+
+	return latency / nr_running;
+}
+
 /*
  * We calculate the vruntime slice of a to be inserted task
  *
@@ -865,14 +875,16 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
 	if (!sched_feat(WAKEUP_PREEMPT))
 		return;
 
-	if (delta_exec < sysctl_sched_min_granularity)
-		return;
-
 	if (cfs_rq->nr_running > 1) {
 		struct sched_entity *se = __pick_next_entity(cfs_rq);
 		s64 delta = curr->vruntime - se->vruntime;
+		u64 wakeup_gran = sysctl_sched_wakeup_granularity;
+		u64 min_gran = sysctl_sched_min_granularity;
+
+		if (delta < wakeup_gran + min_gran)
+			return;
 
-		if (delta > ideal_runtime)
+		if (delta > wakeup_gran + __sched_gran(cfs_rq->nr_running))
 			resched_task(rq_of(cfs_rq)->curr);
 	}
 }