lkml.org 
[lkml]   [2023]   [Sep]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [patch V4 24/41] x86/cpu: Provide cpu_init/parse_topology()
From
On 9/15/2023 4:54 AM, Peter Zijlstra wrote:
> On Tue, Aug 29, 2023 at 08:46:14AM +0530, K Prateek Nayak wrote:
>> Hello Arjan,
>>
>> On 8/28/2023 8:04 PM, Arjan van de Ven wrote:
>>> On 8/28/2023 7:28 AM, K Prateek Nayak wrote:
>>>>>     - Are these really different between AMD and Intel or is this some
>>>>>       naming convention issue which needs to be resolved?
>>>>     They do have different characteristics since, on Sapphire
>>>>     Rapids, the LLC is at a socket boundary despite having multiple
>>>>     tiles. (Please correct me if I'm wrong, I'm going off of
>>>>     llc_id shared in this report by Qiuxu Zhuo -
>>>>     https://lore.kernel.org/all/20230809161219.83084-1-qiuxu.zhuo@intel.com/)
>>>>
>>>
>>> Sapphire reports itself as 1 tile though (since logically it is) as far as I know
>>>
>>
>> I believe there are some variants with multiple tiles, at least the
>> following press-release suggests that:
>>
>> https://www.intc.com/news-events/press-releases/detail/1598/intel-launches-4th-gen-xeon-scalable-processors-max-series
>>
>> specifically "... combining up to four Intel 7-built tiles on a single
>> package, connected using Intel EMIB ...". Perhaps the one from Qiuxu
>> Zhuo's report does not contain multiple tiles.
>
> I think what Arjan was saying that despite them being build using
> multipe physical tiles, they describe themselves, in the topology leave,
> as being a single tile.

and more than that -- from a software perspective, they truely act as if they are 1 tile
(you can do SNC to break that sort of but that's not default and has its own list of downsides)


\
 
 \ /
  Last update: 2023-09-15 16:06    [W:0.108 / U:0.756 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site