[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20071108194735.GW19691@waste.org>
Date: Thu, 8 Nov 2007 13:47:35 -0600
From: Matt Mackall <mpm@...enic.com>
To: Christoph Lameter <clameter@....com>
Cc: Mel Gorman <mel@...net.ie>, akpm@...ux-foundatin.org,
linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [patch 02/23] SLUB: Rename NUMA defrag_ratio to remote_node_defrag_ratio
On Thu, Nov 08, 2007 at 11:16:33AM -0800, Christoph Lameter wrote:
> On Thu, 8 Nov 2007, Matt Mackall wrote:
>
> > > I cannot figure out what the number of cycles currently showing on the TSC
> > > have to do with a ratio :(. I could semi-understand if we were counting up
> > > how many cycles were being spent trying to pack objects but that does not
> > > appear to be the case. The comment didn't help a whole lot either. It felt
> > > like a cost for packing, not a ratio
> >
> > It's just a random number generator. And a bad one: lots of arches
> > return 0. And I believe at least one of them has some NUMA support.
>
> Do we have a better one? Something with minimal processing overhead? I'd
> be glad to switch it.
Not really. drivers/char/random.c does:
__get_cpu_var(trickle_count)++ & 0xfff
for a similar purpose.
--
Mathematics is the supreme nostalgia of our time.
-
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