Re: [this_cpu_xx V3 00/19] Introduce per cpu atomic operations and avoid per cpu address arithmetic

Re: [this_cpu_xx V3 00/19] Introduce per cpu atomic operations and avoid per cpu address arithmetic

[Ingo Molnar]

* cl@linux-foundation.org <cl@linux-foundation.org> wrote:

> V2->V3:
> - Available via git tree against latest upstream from
> 	 git://git.kernel.org/pub/scm/linux/kernel/git/christoph/percpu.git linus
> - Rework SLUB per cpu operations. Get rid of dynamic DMA slab creation
>   for CONFIG_ZONE_DMA
> - Create fallback framework so that 64 bit ops on 32 bit platforms
>   can fallback to the use of preempt or interrupt disable. 64 bit
>   platforms can use 64 bit atomic per cpu ops.

I'm going to ask you (again...) to post future versions of this patchset 
to lkml.

linux-mm is a limited forum and a lot of people who might be interested 
in percpu matters simply wont know about your patch-set. per-cpu is not 
just a VM matter, obviously - it affects architectures, core kernel 
code, etc. etc.

I happened to see your patch-set and have a couple of comments about it 
but i will wait with discussing the issues until you submit these 
patches properly.

Thanks,

	Ingo

Re: [this_cpu_xx V3 00/19] Introduce per cpu atomic operations and avoid per cpu address arithmetic

[Ingo Molnar]

* Ingo Molnar <mingo@elte.hu> wrote:

> I'm going to ask you (again...) to post future versions of this 
> patchset to lkml.

... then i see the next version of the series, resent to lkml ;-)

Thanks for doing that - checking it now.

	Ingo