On 10/07/2016 07:11 AM, David Gibson wrote:
> On Fri, Oct 07, 2016 at 04:01:52PM +1100, Benjamin Herrenschmidt wrote:
>> On Fri, 2016-10-07 at 15:32 +1100, David Gibson wrote:
>>> On Mon, Oct 03, 2016 at 09:24:39AM +0200, Cédric Le Goater wrote:
>>>> This will be used to build real HW ids for the cores and enforce
>>> some
>>>> limits on the available cores per chip.
>>>
>>> Is there actually a practical reason to allow the user (or machine
>>> type) to override the default core mask?
>>
>> None other than mimmicing real HW ... some cores can be disabled
>> on some chips and we *might* want to mimmic that for some test
>> scenarios.
>
> Ok, sounds like a good enough reason to me.
There are really different layouts on the field. Here is a S824,
16 cores, 4 sockets, with these HW ids :
0x4 0x4 0x5 0x4
0x5 0x5 0xc 0x5
0x6 0x6 0xd 0x6
0xe 0xe 0xe 0xd
Long term, we will want to support unplug I suppose.
C.
