Re: [PATCH v7 07/14] x86/resctrl: Maintain a count of enabled monitor features

[Reinette Chatre]

Hi Tony,

On 6/1/26 12:56 PM, Tony Luck wrote:
> AET (Application Energy Telemetry) may be enabled/disabled from one mount
> to the next depending on whether the pmt_telemetry module is loaded. If
> AET is the only monitoring feature supported on a system and it is enabled
> in one mount, but disabled in a subsequent mount this will result in empty
> mon_data directories.
> 
> Change from a boolean to a count of enabled monitor features inside
> architecture code. File system code only needs to know if any monitor

First sentence seems to be missing what is being changed.

> features are enabled so resctrl_arch_mon_capable() can still return
> boolean.
> 
> Signed-off-by: Tony Luck <tony.luck@xxxxxxxxx>
> ---
>  arch/x86/include/asm/resctrl.h         |  4 ++--
>  arch/x86/kernel/cpu/resctrl/internal.h |  2 +-
>  arch/x86/kernel/cpu/resctrl/core.c     | 24 +++++++++++++-----------
>  arch/x86/kernel/cpu/resctrl/monitor.c  | 11 +++--------
>  4 files changed, 19 insertions(+), 22 deletions(-)
> 
> diff --git a/arch/x86/include/asm/resctrl.h b/arch/x86/include/asm/resctrl.h
> index 575f8408a9e7..1e50c7dc3fe3 100644
> --- a/arch/x86/include/asm/resctrl.h
> +++ b/arch/x86/include/asm/resctrl.h
> @@ -43,7 +43,7 @@ struct resctrl_pqr_state {
>  DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);
>  
>  extern bool rdt_alloc_capable;
> -extern bool rdt_mon_capable;
> +extern int rdt_mon_feature_count;
>  
>  DECLARE_STATIC_KEY_FALSE(rdt_enable_key);
>  DECLARE_STATIC_KEY_FALSE(rdt_alloc_enable_key);
> @@ -68,7 +68,7 @@ static inline void resctrl_arch_disable_alloc(void)
>  
>  static inline bool resctrl_arch_mon_capable(void)
>  {
> -	return rdt_mon_capable;
> +	return !!rdt_mon_feature_count;
>  }
>  

This seem unnecessarily complicated to me. Can global "rdt_mon_capable" instead be dropped
and let resctrl_arch_mon_capable() just return "true" if any of the resources are
"mon_capable"?

Reinette