[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <521AE884.6090605@redhat.com>
Date: Mon, 26 Aug 2013 01:32:52 -0400
From: Rik van Riel <riel@...hat.com>
To: Peter Zijlstra <peterz@...radead.org>
CC: Mel Gorman <mgorman@...e.de>,
Andrew Morton <akpm@...ux-foundation.org>,
Miao Xie <miaox@...fujitsu.com>,
David Rientjes <rientjes@...gle.com>,
Christoph Lameter <cl@...ux.com>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] cpuset: mm: Reduce large amounts of memory barrier related
damage v3
On 08/23/2013 02:15 PM, Peter Zijlstra wrote:
> So I guess the quick and ugly solution is something like the below.
This still crashes :)
> --- a/mm/mempolicy.c
> +++ b/mm/mempolicy.c
> @@ -1762,19 +1762,21 @@ unsigned slab_node(void)
> static unsigned offset_il_node(struct mempolicy *pol,
> struct vm_area_struct *vma, unsigned long off)
> {
> - unsigned nnodes = nodes_weight(pol->v.nodes);
> - unsigned target;
> - int c;
> - int nid = -1;
> + unsigned nnodes, target;
> + int c, nid;
>
> +again:
> + nnodes = nodes_weight(pol->v.nodes);
> if (!nnodes)
> return numa_node_id();
> +
> target = (unsigned int)off % nnodes;
> - c = 0;
> - do {
> + for (c = 0, nid = -1; c <= target; c++)
> nid = next_node(nid, pol->v.nodes);
> - c++;
> - } while (c <= target);
> +
> + if (unlikely((unsigned)nid >= MAX_NUMNODES))
> + goto again;
I'll go kick off a compile that replaces the conditional above with:
if (unlikely(!node_online(nid)))
goto again;
> return nid;
> }
--
All rights reversed
--
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