Re: [PATCH v4 11/19] x86/resctrl: Introduce mbm_total_cfg and mbm_local_cfg

["Moger, Babu" <bmoger@amd.com>]

Hi Reinette,

On 6/27/2024 3:56 PM, Reinette Chatre wrote:
> Hi Babu,
> 
> On 6/27/24 11:51 AM, Moger, Babu wrote:
>> Hi Reinette,
>>
>> On 6/13/24 20:43, Reinette Chatre wrote:
>>> Hi Babu,
>>>
>>> On 5/24/24 5:23 AM, Babu Moger wrote:
>>>> If the BMEC (Bandwidth Monitoring Event Configuration) feature is
>>>> supported, the bandwidth events can be configured to track specific
>>>> events. The event configuration is domain specific. ABMC (Assignable
>>>> Bandwidth Monitoring Counters) feature needs event configuration
>>>> information to assign hardware counter to an RMID. Event configurations
>>>> are not stored in resctrl but instead always read from or written to
>>>> hardware directly when prompted by user space.
>>>>
>>>> Read the event configuration from the hardware during the domain
>>>> initialization. Save the configuration information in the 
>>>> rdt_hw_domain,
>>>> so it can be used for counter assignment.
>>>>
>>>> Signed-off-by: Babu Moger <babu.moger@amd.com>
>>>> ---
>>>> v4: Read the configuration information from the hardware to initialize.
>>>>       Added few commit messages.
>>>>       Fixed the tab spaces.
>>>>
>>>> v3: Minor changes related to rebase in mbm_config_write_domain.
>>>>
>>>> v2: No changes.
>>>> ---
>>>>    arch/x86/kernel/cpu/resctrl/core.c     |  2 ++
>>>>    arch/x86/kernel/cpu/resctrl/internal.h |  5 +++++
>>>>    arch/x86/kernel/cpu/resctrl/monitor.c  | 21 +++++++++++++++++++++
>>>>    3 files changed, 28 insertions(+)
>>>>
>>>> diff --git a/arch/x86/kernel/cpu/resctrl/core.c
>>>> b/arch/x86/kernel/cpu/resctrl/core.c
>>>> index ec93f6a50308..856c46d12177 100644
>>>> --- a/arch/x86/kernel/cpu/resctrl/core.c
>>>> +++ b/arch/x86/kernel/cpu/resctrl/core.c
>>>> @@ -542,6 +542,8 @@ static void domain_add_cpu(int cpu, struct
>>>> rdt_resource *r)
>>>>            return;
>>>>        }
>>>>    +    arch_domain_mbm_evt_config(hw_dom);
>>>> +
>>>>        list_add_tail_rcu(&d->list, add_pos);
>>>>          err = resctrl_online_domain(r, d);
>>>> diff --git a/arch/x86/kernel/cpu/resctrl/internal.h
>>>> b/arch/x86/kernel/cpu/resctrl/internal.h
>>>> index 5e7e76cd512f..60a1ca0a11a7 100644
>>>> --- a/arch/x86/kernel/cpu/resctrl/internal.h
>>>> +++ b/arch/x86/kernel/cpu/resctrl/internal.h
>>>> @@ -373,6 +373,8 @@ struct arch_mbm_state {
>>>>     * @ctrl_val:    array of cache or mem ctrl values (indexed by 
>>>> CLOSID)
>>>>     * @arch_mbm_total:    arch private state for MBM total bandwidth
>>>>     * @arch_mbm_local:    arch private state for MBM local bandwidth
>>>> + * @mbm_total_cfg:    MBM total bandwidth configuration
>>>> + * @mbm_local_cfg:    MBM local bandwidth configuration
>>>>     *
>>>>     * Members of this structure are accessed via helpers that provide
>>>> abstraction.
>>>>     */
>>>> @@ -381,6 +383,8 @@ struct rdt_hw_domain {
>>>>        u32                *ctrl_val;
>>>>        struct arch_mbm_state        *arch_mbm_total;
>>>>        struct arch_mbm_state        *arch_mbm_local;
>>>> +    u32                mbm_total_cfg;
>>>> +    u32                mbm_local_cfg;
>>>>    };
>>>
>>> Similar to the abmc_enabled member of rdt_hw_resource, these new
>>> members of rdt_hw_domain are architecture specific and should never be
>>> touched directly by resctrl fs code, for example, from 
>>> mbm_config_show().
>>
>> Need some clarification here.
>>
>> I am thinking you want to introduce architecture specific routines to get
>> and set mbm_total_config/mbm_local_config for the domain.
>> Something like this.
>>
>> +int arch_get_mbm_evt_cfg(struct rdt_domain *d, enum resctrl_event_id 
>> eventid)
> 
> The prefix for arch specific calls converged to "resctrl_arch_".
> 
>> +{
>> +       struct rdt_hw_domain *hw_dom = resctrl_to_arch_dom(d);
>> +
>> +       switch (eventid) {
>> +       case QOS_L3_OCCUP_EVENT_ID:
>> +               break;
>> +       case QOS_L3_MBM_TOTAL_EVENT_ID:
>> +               return hw_dom->mbm_total_cfg;
>> +       case QOS_L3_MBM_LOCAL_EVENT_ID:
>> +               return hw_dom->mbm_local_cfg;
>> +       }
>> +
>> +       /* Never expect to get here */
>> +       WARN_ON_ONCE(1);
>> +
>> +       return -1;
>> +}
>> +
>> +void arch_set_mbm_evt_cfg(struct rdt_domain *d,
>> +                         enum resctrl_event_id eventid, u32 val)
>> +{
>> +       struct rdt_hw_domain *hw_dom = resctrl_to_arch_dom(d);
>> +
>> +       switch (eventid) {
>> +       case QOS_L3_OCCUP_EVENT_ID:
>> +               break;
>> +       case QOS_L3_MBM_TOTAL_EVENT_ID:
>> +               hw_dom->mbm_total_cfg = val;
>> +               break;
>> +       case QOS_L3_MBM_LOCAL_EVENT_ID:
>> +               hw_dom->mbm_local_cfg = val;
>> +       }
>> +
>> +       return;
>> +}
>> +
> 
> I expected that a call to set the event configuration on an architecture to
> also interact with the hardware. Essentially what mon_event_config_write()
> does today, but in addition also sets the cached 
> mbm_total_cfg/mbm_local_cfg.

Yea. Understood.
> 
> Part of this is done by the new MPAM portion [1]. With that the post-ABMC
> implementation of resctrl_arch_mon_event_config_read() may be what you
> have as arch_get_mbm_evt_cfg() above and 
> resctrl_arch_mon_event_config_write()
> does the same as in [1] but with addition of updating the cached
> mbm_local_cfg/mbm_total_cfg within struct rdt_hw_domain. Would this work
> for ABMC?


Yes. It does work. I can keep the same name as well.
Thanks for the clarifications.

> 
> Reinette
> 
> [1] 
> https://lore.kernel.org/lkml/20240614150033.10454-21-james.morse@arm.com/
> 
> 

-- 
- Babu Moger