[PATCH v2 2/8] arch_topology: Set thread sibling cpumask only within the cluster
Sudeep Holla
sudeep.holla at arm.com
Fri May 20 06:20:25 PDT 2022
On Fri, May 20, 2022 at 02:32:19PM +0200, Dietmar Eggemann wrote:
> On 18/05/2022 11:33, Sudeep Holla wrote:
> > Currently the cluster identifier is not set on the DT based platforms.
> > The reset or default value is -1 for all the CPUs. Once we assign the
> > cluster identifier values correctly that imay result in getting the thread
> > siblings wrongs as the core identifiers can be same for 2 different CPUs
> > belonging to 2 different cluster.
> >
> > So, in order to get the thread sibling cpumasks correct, we need to
> > update them only if the cores they belong are in the same cluster within
> > the socket. Let us skip updation of the thread sibling cpumaks if the
> > cluster identifier doesn't match.
>
> So this issue should be there on ACPI systems as well then. Kunpeng920
> and Ampere Altra don't have SMT, so very likely nobody has noticed this
> so far.
No, I think it is handled correctly. May be the wordings of my commit log
needs changing. Previously the core ids were generated and unique across the
platform. But with this change I am assigned the id as read from the DT
node using core@<id>. That results in the problem and needs to be taken care.
This doesn't change the end result(the actual mask) but the way we arrive at
that. In short this is not fixing any issue that was broken before.
--
Regards,
Sudeep
More information about the linux-arm-kernel
mailing list