Re: [PATCH v2 22/35] x86/resctrl: Make resctrl_arch_pseudo_lock_fn() take a plr

[James Morse]

Hi Dave,

On 26/04/2024 16:08, Dave Martin wrote:
> From: James Morse <james.morse@xxxxxxx>
> 
> resctrl_arch_pseudo_lock_fn() has architecture specific behaviour,
> and takes a struct rdtgroup as an argument.
> 
> After the filesystem code moves to /fs/, the definition of struct
> rdtgroup will not be available to the architecture code.
> 
> The only reason resctrl_arch_pseudo_lock_fn() wants the rdtgroup is
> for the CLOSID. Embed that in the pseudo_lock_region as a closid,
> and move the definition of struct pseudo_lock_region to resctrl.h.

> FYI: James Morse:
> 
>  * Review the v2 change below, in case there was any functional reason
>    for distinguishing "hw" closids here.
> 
> Changes in v2:
> 
>  * [Commit message only] Typo fix:
>    s/hw_closid/closid/g
> 
>    (The name "closid" is used in the code.  If there is a distinction
>    between "hardware" closids and resctrl closids, it doesn't seem to
>    be explicit in the naming elsewhere.)

Heh, this is a blast from the past!

In 2018 I was trying to get resctrl to convert 'closid' to 'hw_closid' using some
architecture provided helpers. This was an alternative way of folding the CDP code to hide
the differences between x86/arm64. I decided the fs/resctrl code was cleaner without it -
and arm64 can just juggle the numbers instead.


Thanks,

James