Re: [PATCH] vmscan: limit concurrent reclaimers in shrink_zone

[Rik van Riel]

On 12/10/2009 09:03 PM, Minchan Kim wrote:
> On Fri, Dec 11, 2009 at 8:56 AM, Rik van Riel<riel@xxxxxxxxxx>  wrote:
> > Under very heavy multi-process workloads, like AIM7, the VM can
> > get into trouble in a variety of ways.  The trouble start when
> > there are hundreds, or even thousands of processes active in the
> > page reclaim code.

> Otherwise looks good to me.
>
> Reviewed-by: Minchan Kim<minchan.kim@xxxxxxxxx>

OK, we found three issues with my patch :)

1) there is a typo in sysctl.c

2) there is another typo in Documentation/vm/sysctl.c

3) the code in vmscan.c has a bug, where tasks without
    __GFP_IO or __GFP_FS can end up waiting for tasks
    with __GFP_IO or __GFP_FS, leading to a deadlock

I will fix these issues and send out a new patch.

--
All rights reversed.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/