Re: [PATCH v5 03/13] riscv: sbi: add FWFT extension interface

[Andrew Jones <ajones@ventanamicro.com>]

On Thu, Apr 24, 2025 at 02:32:01PM +0200, Clément Léger wrote:
> 
> 
> On 24/04/2025 13:00, Andrew Jones wrote:
> > On Thu, Apr 17, 2025 at 02:19:50PM +0200, Clément Léger wrote:
> >> This SBI extensions enables supervisor mode to control feature that are
> >> under M-mode control (For instance, Svadu menvcfg ADUE bit, Ssdbltrp
> >> DTE, etc). Add an interface to set local features for a specific cpu
> >> mask as well as for the online cpu mask.
> >>
> >> Signed-off-by: Clément Léger <cleger@rivosinc.com>
> >> Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
> >> ---
> >>  arch/riscv/include/asm/sbi.h | 17 +++++++++++
> >>  arch/riscv/kernel/sbi.c      | 57 ++++++++++++++++++++++++++++++++++++
> >>  2 files changed, 74 insertions(+)
> >>
> >> diff --git a/arch/riscv/include/asm/sbi.h b/arch/riscv/include/asm/sbi.h
> >> index 7ec249fea880..c8eab315c80e 100644
> >> --- a/arch/riscv/include/asm/sbi.h
> >> +++ b/arch/riscv/include/asm/sbi.h
> >> @@ -503,6 +503,23 @@ int sbi_remote_hfence_vvma_asid(const struct cpumask *cpu_mask,
> >>  				unsigned long asid);
> >>  long sbi_probe_extension(int ext);
> >>  
> >> +int sbi_fwft_set(u32 feature, unsigned long value, unsigned long flags);
> >> +int sbi_fwft_local_set_cpumask(const cpumask_t *mask, u32 feature,
> >> +			       unsigned long value, unsigned long flags);
> > 
> > I'm confused by the naming that includes 'local' and 'cpumask' together
> > and...
> > 
> >> +/**
> >> + * sbi_fwft_local_set() - Set a feature on all online cpus
> >> + * @feature: The feature to be set
> >> + * @value: The feature value to be set
> >> + * @flags: FWFT feature set flags
> >> + *
> >> + * Return: 0 on success, appropriate linux error code otherwise.
> >> + */
> >> +static inline int sbi_fwft_local_set(u32 feature, unsigned long value,
> >> +				     unsigned long flags)
> >> +{
> >> +	return sbi_fwft_local_set_cpumask(cpu_online_mask, feature, value, flags);
> > 
> > ...that something named with just 'local' is applied to all online cpus.
> > I've always considered 'local' functions to only affect the calling cpu.
> 
> Yeah I thought of that as well, local here refers to the fact that this
> function applies for a local feature (as described in the SBI
> documentation) but agreed that it's really missleading. Any idea for a
> better naming ?

I'd drop 'local' from these names and when we deal with global features we
can add 'global' to their function names. If we really must have something
now, then maybe 'per_cpu' instead of 'local'.

Thanks,
drew