Re: [Patch] numa:x86_64: Cacheline aliasing makes for_each_populated_zoneextremely expensive -V2.

[H. Peter Anvin]

On 08/25/2010 04:08 AM, Robin Holt wrote:
>> So yes, this means adding an interface to the e820 allocator, even
>> though it's scheduled to be replaced.  Because the new implementation
>> will see the new interface and know they have to implement it, and the
>> interface will make it clear just what exactly is expected of the
>> implementation.
> 
> What is the new allocator called?  Who is developing it?  When do you
> expect it to be introduced into the kernel?
> 

memblock -- the core is already in the kernel and used by several
non-x86 architectures, and so we want to switch x86 to use it, as well.

Ben Herrenschmidt and Yinghai Lu are driving this one.

As with anything in Linux it will get integrated when it is ready.  It
would be nice if it was ready for .37, but it's not certain.

	-hpa
--
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/