Re: [PATCH v2 1/2] kmem_cache: include allocators code directly intoslab_common

[Glauber Costa]

On 10/30/2012 07:31 PM, Christoph Lameter wrote:
> On Fri, 26 Oct 2012, JoonSoo Kim wrote:
> 
>> 2012/10/25 Christoph Lameter <cl@xxxxxxxxx>:
>>> On Wed, 24 Oct 2012, Pekka Enberg wrote:
>>>
>>>> So I hate this patch with a passion. We don't have any fastpaths in
>>>> mm/slab_common.c nor should we. Those should be allocator specific.
>>>
>>> I have similar thoughts on the issue. Lets keep the fast paths allocator
>>> specific until we find a better way to handle this issue.
>>
>> Okay. I see.
>> How about applying LTO not to the whole kernel code, but just to
>> slab_common.o + sl[aou]b.o?
>> I think that it may be possible, isn't it?
> 
> Well.... Andi: Is that possible?
> 

FYI: In the next version of my series, there is a patch that puts
the common code in an inline function in mm/slab.h. Then the allocators
just call that function.

I think it is the best we can do for now, given all the constraints.


--
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/