linux-kernel - Re: [PATCH v1 3/4] ACPI: CPPC: Replace CPUFREQ

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAJZ5v0gxqNoX1CAAAuVtmYM5Hm-5nOGwLXo9hF=z3zPdVVNrCg@mail.gmail.com>
Date: Thu, 25 Sep 2025 18:57:36 +0200
From: "Rafael J. Wysocki" <rafael@...nel.org>
To: Mario Limonciello <mario.limonciello@....com>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>, Linux PM <linux-pm@...r.kernel.org>, 
	Shawn Guo <shawnguo@...nel.org>, Qais Yousef <qyousef@...alina.io>, 
	LKML <linux-kernel@...r.kernel.org>, Viresh Kumar <viresh.kumar@...aro.org>, 
	Prashanth Prakash <pprakash@...eaurora.org>, Pierre Gondois <pierre.gondois@....com>, 
	Linux ACPI <linux-acpi@...r.kernel.org>
Subject: Re: [PATCH v1 3/4] ACPI: CPPC: Replace CPUFREQ_ETERNAL with
 CPPC-specific symbol

On Thu, Sep 25, 2025 at 6:35 PM Mario Limonciello
<mario.limonciello@....com> wrote:
>
>
>
> On 9/25/2025 10:46 AM, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
> >
> > Instead of using CPUFREQ_ETERNAL for signaling error conditions in
> > cppc_get_transition_latency(), introduce CPPC_NO_DATA specifically
> > for this purpose and update all of the callers of this function to
> > use it.
> >
> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@...el.com>
> > ---
> >   drivers/acpi/cppc_acpi.c       |    6 +++---
> >   drivers/cpufreq/amd-pstate.c   |    4 ++--
> >   drivers/cpufreq/cppc_cpufreq.c |    2 +-
> >   include/acpi/cppc_acpi.h       |    4 +++-
> >   include/linux/cpufreq.h        |    3 ---
> >   5 files changed, 9 insertions(+), 10 deletions(-)
> >
> > --- a/drivers/acpi/cppc_acpi.c
> > +++ b/drivers/acpi/cppc_acpi.c
> > @@ -1897,16 +1897,16 @@ unsigned int cppc_get_transition_latency
> >
> >       cpc_desc = per_cpu(cpc_desc_ptr, cpu_num);
> >       if (!cpc_desc)
> > -             return CPUFREQ_ETERNAL;
> > +             return CPPC_NO_DATA;
> >
> >       desired_reg = &cpc_desc->cpc_regs[DESIRED_PERF];
> >       if (CPC_IN_SYSTEM_MEMORY(desired_reg) || CPC_IN_SYSTEM_IO(desired_reg))
> >               return 0;
> >       else if (!CPC_IN_PCC(desired_reg))
> > -             return CPUFREQ_ETERNAL;
> > +             return CPPC_NO_DATA;
> >
> >       if (pcc_ss_id < 0)
> > -             return CPUFREQ_ETERNAL;
> > +             return CPPC_NO_DATA;
> >
> >       pcc_ss_data = pcc_data[pcc_ss_id];
> >       if (pcc_ss_data->pcc_mpar)
> > --- a/drivers/cpufreq/amd-pstate.c
> > +++ b/drivers/cpufreq/amd-pstate.c
> > @@ -875,7 +875,7 @@ static u32 amd_pstate_get_transition_del
> >       u32 transition_delay_ns;
> >
> >       transition_delay_ns = cppc_get_transition_latency(cpu);
> > -     if (transition_delay_ns == CPUFREQ_ETERNAL) {
> > +     if (transition_delay_ns == CPPC_NO_DATA) {
> >               if (cpu_feature_enabled(X86_FEATURE_AMD_FAST_CPPC))
> >                       return AMD_PSTATE_FAST_CPPC_TRANSITION_DELAY;
> >               else
> > @@ -894,7 +894,7 @@ static u32 amd_pstate_get_transition_lat
> >       u32 transition_latency;
> >
> >       transition_latency = cppc_get_transition_latency(cpu);
> > -     if (transition_latency  == CPUFREQ_ETERNAL)
> > +     if (transition_latency  == CPPC_NO_DATA)
> >               return AMD_PSTATE_TRANSITION_LATENCY;
> >
> >       return transition_latency;
> > --- a/drivers/cpufreq/cppc_cpufreq.c
> > +++ b/drivers/cpufreq/cppc_cpufreq.c
> > @@ -312,7 +312,7 @@ static unsigned int get_transition_laten
> >   {
> >       unsigned int transition_latency_ns = cppc_get_transition_latency(cpu);
> >
> > -     if (transition_latency_ns == CPUFREQ_ETERNAL)
> > +     if (transition_latency_ns == CPPC_NO_DATA)
> >               return CPUFREQ_DEFAULT_TRANSITION_LATENCY_NS / NSEC_PER_USEC;
> >
> >       return transition_latency_ns / NSEC_PER_USEC;
> > --- a/include/acpi/cppc_acpi.h
> > +++ b/include/acpi/cppc_acpi.h
> > @@ -41,6 +41,8 @@
> >
> >   #define CPPC_ENERGY_PERF_MAX        (0xFF)
> >
> > +#define CPPC_NO_DATA (unsigned int)(-1)
> > +
>
> Would it be cleaner to just change all the functions that can return
> CPPC_NO_DATA to int instead of unsigned int and instead use -ENODATA?

Yes, that'll work too.

I'll send an update of this patch shortly.

> >   /* Each register has the folowing format. */
> >   struct cpc_reg {
> >       u8 descriptor;
> > @@ -218,7 +220,7 @@ static inline bool cppc_allow_fast_switc
> >   }
> >   static inline unsigned int cppc_get_transition_latency(int cpu)
> >   {
> > -     return CPUFREQ_ETERNAL;
> > +     return CPPC_NO_DATA;
> >   }
> >   static inline bool cpc_ffh_supported(void)
> >   {