Re: [RFC v2] arm,x86,fs/resctrl: Generic schema description Proof of Concept
From: Reinette Chatre
Date: Thu Aug 13 2026 - 11:21:18 EST
Hi Chenyu,
On 8/13/26 6:59 AM, Chen Yu wrote:
> On Tue, Aug 11, 2026 at 11:49:12AM +0800, Chen, Yu C wrote:
>>> I find "region-aware controllers at the same level as legacy controller" together with
>>> "per controller enable/disable" conflicting since it creates impression that user space can
>>> enable/disable the controllers independently while there is a required dependency among the
>>> different controllers based on whether hardware has region-aware MBA enabled or not. Even
>>> here there would be a requirement to communicate to userspace which controllers could be
>>> enabled/disabled together ... which is the same as the emulated list, no?
>>
>> I see. So we actually need a single option (control_mode) to switch between
>> modes (legacy vs.
>> native, mutually exclusive), rather than giving each controller its own
>> independent enable/disable
>> toggle. The emulated list is one way to implement control_mode, ensuring
>> "legacy" and "native"
>> are mutually exclusive.
>>
>
> While adjusting the region aware code on this emulation part, a minor issue is found
> in commit ("fs/resctrl: Support emulated controls")
>
> diff --git a/fs/resctrl/rdtgroup.c b/fs/resctrl/rdtgroup.c
> index cac4b3293cef..ab7d3fa3f337 100644
> --- a/fs/resctrl/rdtgroup.c
> +++ b/fs/resctrl/rdtgroup.c
> @@ -124,7 +124,7 @@ void rdt_staged_configs_clear(void)
> if (list_empty(&ctrl->emulated_by))
> continue;
> list_for_each_entry(em_ctrl, &ctrl->emulated_by, entry) {
> - list_for_each_entry_rcu(dom, &ctrl->domains, hdr.list, lockdep_is_cpus_held())
> + list_for_each_entry_rcu(dom, &em_ctrl->domains, hdr.list, lockdep_is_cpus_held())
> memset(dom->staged_config, 0, sizeof(dom->staged_config));
> }
> }
Thank you very much. Squashed and available in branch resctrl/controls_rfc_v2.2
Reinette