Asymmetric Multiprocessing

[Peter Robinson]

On Mon, Apr 30, 2012 at 2:46 PM, John Watlington <wad at laptop.org> wrote:
>
> On Apr 30, 2012, at 9:28 AM, Peter Robinson wrote:
>
>> On Mon, Apr 30, 2012 at 2:23 PM, John Watlington <wad at laptop.org> wrote:
>>>
>>> Can someone please enlighten me as to the current state
>>> of Linux and asymmetric multiprocessing ?   A number of
>>> ARM SoCs on the market include both high performance
>>> and low power cores.
>>>
>>> Does Linux have a strategy for scheduling to these
>>> asymmetric processing units yet ?
>>
>> Presumable you're referring to the general cores like the Tegra3 and
>> not some of the supplemental cores like on the OMAP devices for media
>> processing?
>
> I'm referring to cores running the same instruction set, but with
> widely varying processing speeds (OMAP 5, Marvell PXA2128,
> i.e. ARM big.LITTLE).

Various links below, in fact yesterday Linaro I believe released a
qemu image that allows big.LITTLE to be emulated for dev and testing.

https://lwn.net/Articles/481055/
http://www.linaro.org/linaro-blog/2011/12/15/big-little-technology-two-usage-models/
https://wiki.linaro.org/WorkingGroups/Kernel/Big.Little.Switcher
http://lists.linaro.org/mailman/listinfo/linaro-big-little
http://git.linaro.org/gitweb?p=arm/big.LITTLE/switcher.git;a=summary
http://www.linaro.org/linaro-blog/2012/04/26/linaro-12-04-released/