2008-01-17 22:38:07

by Mike Travis

Subject: [PATCH 0/3] x86: Reduce memory and intra-node effects with large count NR_CPUs fixup

Fixup change NR_CPUS patchset by rebasing on 2.6.24-rc8-mm1 (from 2.6.24-rc6-mm1)
and adding last minute changes suggested by reviews.

Based on 2.6.24-rc8-mm1

Signed-off-by: Mike Travis <[email protected]>
Reviewed-by: Christoph Lameter <[email protected]>
---

--

2008-01-17 22:53:18

by Mike Travis

Subject: Re: [PATCH 0/3] x86: Reduce memory and intra-node effects with large count NR_CPUs fixup

Hi Andrew,

My automatic scripts accidentally sent this mail prematurely. Please hold
off applying yet.

Thanks,
Mike

[email protected] wrote:
> Fixup change NR_CPUS patchset by rebasing on 2.6.24-rc8-mm1 (from 2.6.24-rc6-mm1)
> and adding last minute changes suggested by reviews.
>
> Based on 2.6.24-rc8-mm1
>
> Signed-off-by: Mike Travis <[email protected]>
> Reviewed-by: Christoph Lameter <[email protected]>
> ---
>

2008-01-18 09:24:20

by Ingo Molnar

Subject: Re: [PATCH 0/3] x86: Reduce memory and intra-node effects with large count NR_CPUs fixup

* Mike Travis <[email protected]> wrote:

> Hi Andrew,
>
> My automatic scripts accidentally sent this mail prematurely. Please
> hold off applying yet.

I've picked it up for x86.git and i'll keep testing it (the patches seem
straightforward) and will report any problems with the bite-head-off
option unset.

[ The 32-bit NUMA compile issue is orthogonal to these patches - it's
due to the lack of 32-bit NUMA support in your changes :) That needs
fixing before this could go into v2.6.25. ]

Ingo

2008-01-18 12:59:20

by Mike Travis

Subject: Re: [PATCH 0/3] x86: Reduce memory and intra-node effects with large count NR_CPUs fixup

Ingo Molnar wrote:
> * Mike Travis <[email protected]> wrote:
>
>> Hi Andrew,
>>
>> My automatic scripts accidentally sent this mail prematurely. Please
>> hold off applying yet.
>
> I've picked it up for x86.git and i'll keep testing it (the patches seem
> straightforward) and will report any problems with the bite-head-off
> option unset.
>
> [ The 32-bit NUMA compile issue is orthogonal to these patches - it's
> due to the lack of 32-bit NUMA support in your changes :) That needs
> fixing before this could go into v2.6.25. ]
>
> Ingo

I hadn't considered doing 32-bit NUMA changes as I didn't know if the
NR_CPUS count would really be increased for the 32-bit architecture.
I have been trying though not to break it. ;-)

Thanks,
Mike

2008-01-18 18:55:13

by Christoph Lameter

Subject: Re: [PATCH 0/3] x86: Reduce memory and intra-node effects with large count NR_CPUs fixup

On Fri, 18 Jan 2008, Mike Travis wrote:

> I hadn't considered doing 32-bit NUMA changes as I didn't know if the
> NR_CPUS count would really be increased for the 32-bit architecture.
> I have been trying though not to break it. ;-)

32bit NUMA is tricky because ZONE_NORMAL memory is only available on node
0. There have been thorny difficult to debug issues in the past...