Re: [PATCH v2 0/4] powerpc/topology: Introduce llc_override kernel parameter

[Shrikanth Hegde <sshegde@linux.ibm.com>]

Hi Tim, Thanks for reviewing the changes.

On 10/23/26 11:28 PM, Tim Chen wrote:
> On Fri, 2026-10-23 at 18:46 +0530, Shrikanth Hegde wrote:
>> Provide an option for users to enable cache aware scheduling on powerpc
>> via a command line parameter.
>>
>> More details of the problem and context is in cover-letter of v1.
>> v1: https://lore.kernel.org/all/20261012102037.2477378-1-sshegde@linux.ibm.com/
>>
>> v1->v2:
>> - Move to kernel parameters from debugfs. This helps to keep
>>    llc_id consistent which doesn't change across sched domain
>>    rebuilds. (Tim Chen)
>> - Update arch_llc_mask to reflect the override (Tim Chen)
>> - Limit only to mc, pkg (Srikar Dronamraju)
>>
>> Notes:
>>
>> - A selftest for the functional validation of cache aware scheduling
>>    is a good idea as we discussed in v1. It is a significant change on
>>    its own and likely takes time. I am thinking it might be good idea to
>>    revisit selftests for scheduler in general. (We have very few).
>>    Maybe worth a discussion in coming OSPM.
>>
>> - I am on parental leave. This version is pretty much what I had last week
>>    and I haven't gotten any chance to update since. just sending it out.
>>    Sorry if I have missed to address any comments.
>>    Please expect delays in my response.
> 
> Thanks for sending this while on leave, and congratulations!
> 
> The updated approach with boot parameters looks good to me.
> Some minor points:
> 
> 1. llc_bytes. As Chen Yu noted, get_effective_llc_bytes() returns 0
>     on powerpc because the generic cacheinfo is not populated. With
>     NUMA balancing on and the default llc_aggr_tolerance, any process
>     with a footprint then fails exceed_llc_capacity(), so cache aware
>     scheduling still does not engage.
> 

Let me first see if powerpc can migrate to generic cacheinfo.

> 2. Bisection. Patch 2 moves SD_SHARE_LLC to MC/PKG, but
>     arch_llc_mask() only follows in patch 3, so a kernel at patch 2
>     booted with llc_override=mc has sd_llc at MC with per-L2 llc ids,
>     causing mismatch. It may be better to squash 2 and 3.

Alright.

> 
> 3. llc_override is only written at boot. Making it
>     
>          static int llc_override __ro_after_init;
> 
>     and then you don't have to worry about READ_ONCE and WRITE_ONCE
>     to it as Chen Yu pointed out.
> 

Sure.

> Tim
> 
>>
>> Shrikanth Hegde (4):
>>    powerpc/topology: Introduce llc_override kernel parameter
>>    powerpc/topology: Update mc/pkg flags based on llc_override
>>    powerpc/topology: update arch_llc_mask based llc_override
>>    Documentation/powerpc: Add details about llc_override
>>
>>   .../admin-guide/kernel-parameters.txt         | 15 ++++
>>   arch/powerpc/include/asm/topology.h           |  8 +--
>>   arch/powerpc/kernel/smp.c                     | 69 +++++++++++++++++--
>>   3 files changed, 81 insertions(+), 11 deletions(-)