All of lore.kernel.org
 help / color / mirror / Atom feed
From: Antti Salmela <asalmela@iki.fi>
To: Christian Ehrhardt <ehrhardt@mathematik.uni-ulm.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>, linux-kernel@vger.kernel.org
Subject: Re: [OOPS] 2.4.20-pre1-ac3, SMP (Dual PIII)
Date: Fri, 16 Aug 2002 19:50:57 +0300	[thread overview]
Message-ID: <20020816195057.A26010@wasala.fi> (raw)
In-Reply-To: <20020816141718.4766.qmail@thales.mathematik.uni-ulm.de>; from ehrhardt@mathematik.uni-ulm.de on Fri, Aug 16, 2002 at 04:17:18PM +0200

On Fri, Aug 16, 2002 at 04:17:18PM +0200, Christian Ehrhardt wrote:
> On Thu, Aug 15, 2002 at 02:33:02AM +0100, Alan Cox wrote:
> > Thanks - your analysis is informative to say the least. It looks like
> > the PIV load balancing code is the problem. 
> 
> The (untested) patch below should correct this problem along with
> a locking oddity (last hunk) that IMHO either needs fixing or a BIG
> comment. Be prepared for a few (up to 4) lines of fuzz due to additional
> BUG_ONs in both versions of the file.

With this patch I could boot 2.4.20-pre2-ac3 and it has now run nearly an
hour without any problems.

>      regards   Christian Ehrhardt
> 
> [1] http://www.atnf.csiro.au/people/rgooch/benchmarks/linux-scheduler.html
> 
> 
> --- /usr/src/linux-2.4.20-pre1-ac3/kernel/sched.c	Thu Aug 15 20:03:01 2002
> +++ sched.c	Fri Aug 16 16:15:57 2002
> @@ -769,7 +772,7 @@
>  			set_tsk_need_resched(p);
>  
>  			/* put it at the end of the queue: */
> -			dequeue_task(p, rq->active);
> +			dequeue_task(p, p->array);
>  			enqueue_task(p, rq->active);
>  		}
>  		goto out;
> @@ -785,7 +788,7 @@
>  	if (p->sleep_avg)
>  		p->sleep_avg--;
>  	if (!--p->time_slice) {
> -		dequeue_task(p, rq->active);
> +		dequeue_task(p, p->array);
>  		set_tsk_need_resched(p);
>  		p->prio = effective_prio(p);
>  		p->time_slice = TASK_TIMESLICE(p);
> @@ -1396,7 +1399,7 @@
>  	 */
>  	if (likely(current->prio == MAX_PRIO-1)) {
>  		if (current->time_slice <= 1) {
> -			dequeue_task(current, rq->active);
> +			dequeue_task(current, array);
>  			enqueue_task(current, rq->expired);
>  		} else
>  			current->time_slice--;
> @@ -1411,7 +1414,7 @@
>  		list_add_tail(&current->run_list, array->queue + current->prio);
>  		__set_bit(current->prio, array->bitmap);
>  	}
> -	spin_unlock(&rq->lock);
> +	rq_unlock (rq);
>  
>  	schedule();
>  

-- 
Antti Salmela

  reply	other threads:[~2002-08-16 16:47 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-08-14 11:54 [OOPS] 2.4.20-pre1-ac3, SMP (Dual PIII) Antti Salmela
2002-08-14 12:37 ` Alan Cox
2002-08-14 13:10   ` Antti Salmela
2002-08-14 13:27     ` Alan Cox
2002-08-14 15:55       ` Antti Salmela
2002-08-14 17:30         ` Christian Ehrhardt
2002-08-15  1:33           ` Alan Cox
2002-08-16 14:17             ` Christian Ehrhardt
2002-08-16 16:50               ` Antti Salmela [this message]
2002-08-14 13:16   ` pci-dma bug in pci_alloc_consistent on i386 ? Steffen Persvold

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=20020816195057.A26010@wasala.fi \
    --to=asalmela@iki.fi \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=ehrhardt@mathematik.uni-ulm.de \
    --cc=linux-kernel@vger.kernel.org \
    /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 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.