mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Rusty Russell <rusty@rustcorp.com.au>
To: mingo@elte.hu
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] minor sched-E1 tweaks and questions
Date: Thu, 10 Jan 2002 14:48:42 +1100	[thread overview]
Message-ID: <E16OWCo-0000YO-00@wagner.rustcorp.com.au> (raw)
In-Reply-To: Your message of "Wed, 09 Jan 2002 11:41:19 BST." <Pine.LNX.4.33.0201091126520.2276-100000@localhost.localdomain>

In message <Pine.LNX.4.33.0201091126520.2276-100000@localhost.localdomain> you 
write:
> > Q. How can this happen in expire_task():
> > 	if (p->array != rq->active) {
> > 		p->need_resched = 1;
> > 		return;
> > 	}
> 
> if a task gets delayed by some really heavy IRQ load and the timer
> interrupt hits the task twice.

Hmm... still don't see it.  update_process_times() surely doesn't
re-enter?  And another CPU cannot load_balance() p (== current) away
from us.

Another question:

	if (likely(prev != next)) {
		rq->nr_switches++;
		rq->curr = next;
		next->cpu = prev->cpu;
		context_switch(prev, next);
		/*
		 * The runqueue pointer might be from another CPU
		 * if the new task was last running on a different
		 * CPU - thus re-load it.
		 */
		barrier();
		rq = this_rq();
	}
	spin_unlock_irq(&rq->lock);

I do not understand this comment.  How can rq (ie. smp_processor_id())
change?  Nothing sleeps here, and if it DID change, the
spin_unlock_irq() would be wrong...

> i've taken your fixes, please double-check the next patch whether all of
> them are correctly applied.

Um, missed one:

 static struct runqueue {
-	int cpu;
 	spinlock_t lock;
-	unsigned long nr_running, nr_switches, last_rt_event;
+	unsigned long nr_running, nr_switches;
 	task_t *curr, *idle;
 	prio_array_t *active, *expired, arrays[2];
-	char __pad [SMP_CACHE_BYTES];
+	int prev_nr_running[NR_CPUS];
 } runqueues [NR_CPUS] __cacheline_aligned;

You want each entry in the array to be aligned, not the whole array!
You need to define the struct runqueue to be the cacheline aligned
(using ____cacheline_aligned since it's a type), THEN put the
__cacheline_aligned after the array declaration so it gets put in the
aligned section:

 struct runqueue {
 	spinlock_t lock;
	unsigned long nr_running, nr_switches;
 	task_t *curr, *idle;
 	prio_array_t *active, *expired, arrays[2];
	int prev_nr_running[NR_CPUS];
 } ____cacheline_aligned;

 static struct runqueue runqueues [NR_CPUS] __cacheline_aligned;

This is why my __per_cpu patch was invented 8)
Cheers,
Rusty.
--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.

       reply	other threads:[~2002-01-10  5:50 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Pine.LNX.4.33.0201091126520.2276-100000@localhost.localdomain>
2002-01-10  3:48 ` Rusty Russell [this message]
2002-01-10  6:06   ` Davide Libenzi
2002-01-10 13:07   ` Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=E16OWCo-0000YO-00@wagner.rustcorp.com.au \
    --to=rusty@rustcorp.com.au \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®