[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 8 Sep 2006 17:03:52 -0700
From: "Siddha, Suresh B" <suresh.b.siddha@...el.com>
To: Christoph Lameter <clameter@....com>
Cc: "Siddha, Suresh B" <suresh.b.siddha@...el.com>, akpm@...l.org,
linux-kernel@...r.kernel.org, npiggin@...e.de, mingo@...e.hu
Subject: Re: [PATCH] Fix longstanding load balancing bug in the scheduler.
On Fri, Sep 08, 2006 at 01:19:28PM -0700, Christoph Lameter wrote:
> On Fri, 8 Sep 2006, Siddha, Suresh B wrote:
>
> > On Fri, Sep 08, 2006 at 11:40:51AM -0700, Christoph Lameter wrote:
> > > The balancing operation is not that frequent and having to treat a special
> > > case in the callers would make code more complicated and likely offset the
> > > gains in this function.
> >
> > This solution as such is not accurate and clean :) and my suggestion is
> > not making it any more ugly.
> >
> > With increase in NR_CPUS, cost of cpumask operations will increase and
> > we shouldn't penalize the other logical threads or cores sharing the caches by
> > bringing in unnecessary cache lines.
>
> One cacheline sized 128bytes will support all 1024 cpus that IA64 allows.
> cacheline align the cpumask?
one or more, it is unnecessary for the common case.
thanks,
suresh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists