On Mon, Sep 11, 2023 at 08:23:50PM -0700, Ricardo Neri wrote:
> Hi Andreas,
>
> Agreed. Testing is important. For the specific case of these patches, I
> booted CONFIG_PREEMPT_RT and !CONFIG_PREEMPT_RT kernels. Then I
> a) Ensured that the splat reported in commit 5944ce092b97
> ("arch_topology: Build cacheinfo from primary CPU") was not observed.
>
> b) Ensured that /sys/devices/system/cpu/cpuX/cache is present.
>
> c) Ensured that the contents /sys/devices/system/cpu/cpuX/cache is the
> same before and after my patches.
>
> I tested on the following systems: Intel Alder Lake, Intel Meteor
> Lake, 2-socket Intel Icelake server, 2-socket Intel Cascade Lake server,
> 2-socket Intel Skylake server, 4-socket Intel Broadwell server, 2-socket
> Intel Haswell server, 2-socket AMD Rome server, and 2-socket AMD Milan
> server.
>
> Thanks and BR,
> Ricardo
Thanks for all the tests and info.
--
Regards,
Andreas
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nürnberg, Germany
GF: Ivo Totev, Andrew McDonald, Werner Knoblich
(HRB 36809, AG Nürnberg)
