Re: [PATCH v9 0/9] arm_mpam: Introduce Narrow-PARTID feature

[Zeng Heng]

Hi Jinjie, Shaopeng,

   Thanks for continuing the series.

在 10/22/26 上午8:02, Shaopeng Tan (Fujitsu) 写道:
> Hello Jinjie,
> 
> I would like to confirm the following points.
> 
>   - Is the number of reqPARTIDs per intPARTID dependent on the total number of intPARTIDs and reqPARTIDs?
>   - Given that requirements for the number of control and monitoring groups vary by user,
>     would dynamic configuration of the mapping ratio between intPARTID and reqPARTID be a more suitable approach?
>   - Regarding the mapping logic, some reqPARTIDs may remain unused. Is this the expected behavior?

Support the static-first scope of v9, consistent with James's
feedback on v8. Separating the driver-side grouping from the resctrl
translation, avoiding a new configuration ABI.

Dynamic allocation could improve utilisation when monitoring demand is
uneven across control groups, but can be addressed separately.

>   - Considering that Arm has previously implemented this,
>     are there any gaps between their implementation and Huawei’s requirements?
>     https://lore.kernel.org/linux-arm-kernel/20241212154000.330467-1-Dave.Martin@xxxxxxx/
> 

Dave's RFC explicitly describes itself as preparatory work,
rather than a complete narrowing implementation. In particular, it does
not program the request-to-internal PARTID mappings, retains the
internal PARTID limit, and documents unresolved control semantics when a
CLOSID spans multiple PARTIDs.

I would also like to thank Dave Martin for reviewing the earlier
revisions of the series and providing so much constructive feedback.


Best regards,
Zeng Heng