Messages in this thread Patch in this message | | | Date | Tue, 02 Nov 2004 23:35:57 +1100 | From | Nick Piggin <> | Subject | Re: [PATCH] add requeue task |
| |
Con Kolivas wrote: > add requeue task > > > > ------------------------------------------------------------------------ > > We can requeue tasks for cheaper then doing a complete dequeue followed by > an enqueue. Add the requeue_task function and perform it where possible. > > Change the granularity code to requeue tasks at their best priority > instead of changing priority while they're running. This keeps tasks at > their top interactive level during their whole timeslice. >
I wonder... these things are all in sufficiently rarely used places, that the icache miss might be more costly than the operations saved.
But....
> Signed-off-by: Con Kolivas <kernel@kolivas.org> > > Index: linux-2.6.10-rc1-mm2/kernel/sched.c > =================================================================== > --- linux-2.6.10-rc1-mm2.orig/kernel/sched.c 2004-11-02 14:48:54.686316718 +1100 > +++ linux-2.6.10-rc1-mm2/kernel/sched.c 2004-11-02 14:52:51.805763544 +1100 > @@ -579,6 +579,16 @@ static void enqueue_task(struct task_str > } > > /* > + * Put task to the end of the run list without the overhead of dequeue > + * followed by enqueue. > + */ > +static void requeue_task(struct task_struct *p, prio_array_t *array) > +{ > + list_del(&p->run_list); > + list_add_tail(&p->run_list, array->queue + p->prio); > +} > + > +/* > * Used by the migration code - we pull tasks from the head of the > * remote queue so we want these tasks to show up at the head of the > * local queue: > @@ -2425,8 +2435,7 @@ void scheduler_tick(void) > set_tsk_need_resched(p); > > /* put it at the end of the queue: */ > - dequeue_task(p, rq->active); > - enqueue_task(p, rq->active); > + requeue_task(p, rq->active); > } > goto out_unlock; > } > @@ -2467,10 +2476,8 @@ void scheduler_tick(void) > (p->time_slice >= TIMESLICE_GRANULARITY(p)) && > (p->array == rq->active)) { > > - dequeue_task(p, rq->active); > + requeue_task(p, rq->active); > set_tsk_need_resched(p); > - p->prio = effective_prio(p); > - enqueue_task(p, rq->active); > } > } > out_unlock:
This isn't a 1:1 transformation. Looks like the effective_prio there might be superfluous, but if so that should be a different patch.
> @@ -3569,8 +3576,14 @@ asmlinkage long sys_sched_yield(void) > } else if (!rq->expired->nr_active) > schedstat_inc(rq, yld_exp_empty); > > - dequeue_task(current, array); > - enqueue_task(current, target); > + if (array != target) { > + dequeue_task(current, array); > + enqueue_task(current, target); > + } else > + /* > + * requeue_task is cheaper so perform that if possible. > + */ > + requeue_task(current, array); > > /* > * Since we are going to call schedule() anyway, there's >
Hmm if you have to go to this trouble I'd say its not worth it. Ingo may want to weigh in though. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |