Re: [PATCH v8 01/13] ACPI: CPPC: Add AMD pstate energy performance preference cppc control

[Huang Rui]

On Fri, Dec 23, 2022 at 11:19:57AM +0800, Yuan, Perry wrote:
> [AMD Official Use Only - General]
> 
> 
> 
> > -----Original Message-----
> > From: Huang, Ray <Ray.Huang@xxxxxxx>
> > Sent: Friday, December 23, 2022 10:16 AM
> > To: Yuan, Perry <Perry.Yuan@xxxxxxx>
> > Cc: rafael.j.wysocki@xxxxxxxxx; Limonciello, Mario
> > <Mario.Limonciello@xxxxxxx>; viresh.kumar@xxxxxxxxxx; Sharma, Deepak
> > <Deepak.Sharma@xxxxxxx>; Fontenot, Nathan
> > <Nathan.Fontenot@xxxxxxx>; Deucher, Alexander
> > <Alexander.Deucher@xxxxxxx>; Huang, Shimmer
> > <Shimmer.Huang@xxxxxxx>; Du, Xiaojian <Xiaojian.Du@xxxxxxx>; Meng,
> > Li (Jassmine) <Li.Meng@xxxxxxx>; Karny, Wyes <Wyes.Karny@xxxxxxx>;
> > linux-pm@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> > Subject: Re: [PATCH v8 01/13] ACPI: CPPC: Add AMD pstate energy
> > performance preference cppc control
> > 
> > On Mon, Dec 19, 2022 at 02:40:30PM +0800, Yuan, Perry wrote:
> > > From: Perry Yuan <Perry.Yuan@xxxxxxx>
> > >
> > > Add support for setting and querying EPP preferences to the generic
> > > CPPC driver.  This enables downstream drivers such as amd-pstate to
> > > discover and use these values
> > >
> > > Downstream drivers that want to use the new symbols cppc_get_epp_caps
> > > and cppc_set_epp_perf for querying and setting EPP preferences will
> > > need to call cppc_set_epp_perf to enable the EPP function firstly.
> > >
> > > Signed-off-by: Perry Yuan <Perry.Yuan@xxxxxxx>
> > > ---
> > >  drivers/acpi/cppc_acpi.c | 76
> > > +++++++++++++++++++++++++++++++++++++---
> > >  include/acpi/cppc_acpi.h | 12 +++++++
> > >  2 files changed, 83 insertions(+), 5 deletions(-)
> > >
> > > diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c index
> > > 093675b1a1ff..81081eb899ea 100644
> > > --- a/drivers/acpi/cppc_acpi.c
> > > +++ b/drivers/acpi/cppc_acpi.c
> > > @@ -1093,6 +1093,9 @@ static int cppc_get_perf(int cpunum, enum
> > > cppc_regs reg_idx, u64 *perf)  {
> > >  	struct cpc_desc *cpc_desc = per_cpu(cpc_desc_ptr, cpunum);
> > >  	struct cpc_register_resource *reg;
> > > +	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpunum);
> > > +	struct cppc_pcc_data *pcc_ss_data = NULL;
> > > +	int ret = -EINVAL;
> > >
> > >  	if (!cpc_desc) {
> > >  		pr_debug("No CPC descriptor for CPU:%d\n", cpunum); @@
> > -1102,10
> > > +1105,6 @@ static int cppc_get_perf(int cpunum, enum cppc_regs reg_idx,
> > u64 *perf)
> > >  	reg = &cpc_desc->cpc_regs[reg_idx];
> > >
> > >  	if (CPC_IN_PCC(reg)) {
> > > -		int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpunum);
> > > -		struct cppc_pcc_data *pcc_ss_data = NULL;
> > > -		int ret = 0;
> > > -
> > 
> > Do you have any specific reason to move this piece out of if-condition?
> 
> Move the declaration ahead of the If conditions like other functions did.
> It looks more reasonable and no functions impact. 
>

If one platform doesn't have any CPC registers, it even won't need define
these variables like pcc_ss_id, pcc_ss_data, and it will save more time.

Thanks,
Ray