Re: [PATCH 06/18] x86/resctrl: Allow the allocator to check if a CLOSID can allocate clean RMID

[James Morse <james.morse@arm.com>]

Hi Shawn,

On 08/11/2022 15:57, Shawn Wang wrote:
> On 10/21/2022 9:11 PM, James Morse wrote:
>> MPAM's PMG bits extend its PARTID space, meaning the same PMG value can be
>> used for different control groups.
>>
>> This means once a CLOSID is allocated, all its monitoring ids may still be
>> dirty, and held in limbo.
>>
>> Add a helper to allow the CLOSID allocator to check if a CLOSID has dirty
>> RMID values. This behaviour is enabled by a kconfig option selected by
>> the architecture, which avoids a pointless search for x86.

>> +/**
>> + * resctrl_closid_is_dirty - Determine if clean RMID can be allocate for this
>> + *                           CLOSID.
>> + * @closid: The CLOSID that is being queried.
>> + *
>> + * MPAM's equivalent of RMID are per-CLOSID, meaning a freshly allocate CLOSID
>> + * may not be able to allocate clean RMID. To avoid this the allocator will
>> + * only return clean CLOSID.
>> + */
>> +bool resctrl_closid_is_dirty(u32 closid)
>> +{
>> +    struct rmid_entry *entry;
>> +    int i;
>> +
>> +    lockdep_assert_held(&rdtgroup_mutex);
>> +
>> +    if (IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))
>> +        return false;
> 
> Since dirty closid occurs when the kconfig option for MPAM is enabled, it seems
> that the condition of the if statement here should take the opposite value:
>     if (!IS_ENABLED(CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID))

Yup. Bother.

Thanks for spotting that! It was intended to avoid this work on x86 as its pointless, and
the number of RMID could be large.


Thanks!

James