| Message ID | 20230113052141.2874296-4-wyes.karny@amd.com (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Headers | show |
| Series | amd_pstate: Add guided autonomous mode support | expand |
On 1/12/23 23:21, Wyes Karny wrote: > From ACPI spec below 3 modes for CPPC can be defined: > 1. Non autonomous: OS scaling governor specifies operating frequency/ > performance level through `Desired Performance` register and platform > follows that. > 2. Guided autonomous: OS scaling governor specifies min and max > frequencies/ performance levels through `Minimum Performance` and > `Maximum Performance` register, and platform can autonomously select an > operating frequency in this range. > 3. Fully autonomous: OS only hints (via EPP) to platform for the required > energy performance preference for the workload and platform autonomously > scales the frequency. > > Currently (1) is supported by amd_pstate as passive mode, and (3) is > implemented by EPP support. This change is to support (2). > > In guided autonomous mode the min_perf is based on the input from the > scaling governor. For example, in case of schedutil this value depends > on the current utilization. And max_perf is set to max capacity. > > To activate guided auto mode ``amd_pstate=guided`` command line > parameter has to be passed in the kernel. > > Signed-off-by: Wyes Karny <wyes.karny@amd.com> > --- > .../admin-guide/kernel-parameters.txt | 15 ++++++---- > drivers/cpufreq/amd-pstate.c | 29 ++++++++++++++++--- > include/linux/amd-pstate.h | 2 ++ > 3 files changed, 37 insertions(+), 9 deletions(-) > > diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt > index e3618dfdb36a..0d8486325c9a 100644 > --- a/Documentation/admin-guide/kernel-parameters.txt > +++ b/Documentation/admin-guide/kernel-parameters.txt > @@ -7015,11 +7015,11 @@ > Do not enable amd_pstate as the default > scaling driver for the supported processors > passive > - Use amd_pstate as a scaling driver, driver requests a > - desired performance on this abstract scale and the power > - management firmware translates the requests into actual > - hardware states (core frequency, data fabric and memory > - clocks etc.) > + Use amd_pstate with passive mode as a scaling driver. > + In this mode autonomous selection is disabled. > + Driver requests a desired performance level and platform > + tires to match the same performance level (if it is > + satisfied by guaranteed performance level). > active > Use amd_pstate_epp driver instance as the scaling driver, > driver provides a hint to the hardware if software wants > @@ -7027,3 +7027,8 @@ > to the CPPC firmware. then CPPC power algorithm will > calculate the runtime workload and adjust the realtime cores > frequency. > + guided > + Activate guided autonomous mode. Driver requests minimum and > + maximum performance level and the platform autonomously > + selects a performance level in this range and appropriate > + to the current workload. > diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c > index 87450413cf45..20d78dad712d 100644 > --- a/drivers/cpufreq/amd-pstate.c > +++ b/drivers/cpufreq/amd-pstate.c > @@ -267,6 +267,20 @@ static int cppc_init_perf(struct amd_cpudata *cpudata) > cppc_perf.lowest_nonlinear_perf); > WRITE_ONCE(cpudata->lowest_perf, cppc_perf.lowest_perf); > > + ret = cppc_get_auto_sel_caps(cpudata->cpu, &cppc_perf); > + if (ret) { > + pr_warn("failed to get auto_sel\n"); To make debugging easier in the future perhaps: pr_warn("failed to get auto sel: %d\n", ret); > + return 0; > + } > + > + if (cppc_state == AMD_PSTATE_PASSIVE) > + ret = cppc_set_auto_sel(cpudata->cpu, 0); > + else if (cppc_state == AMD_PSTATE_GUIDED) > + ret = cppc_set_auto_sel(cpudata->cpu, 1); > + as a simplification maybe: ret = cppc_set_auto_sel(cpudata->cpu, cppc_state == AMD_PSTATE_PASSIVE ? 0 : 1); > + if (ret) > + pr_warn("failed to set auto_sel\n"); Again, probably good to emit the error code in this message; > + > return 0; As this could fail now shouldn't you be returning ret? > } > > @@ -344,12 +358,18 @@ static inline bool amd_pstate_sample(struct amd_cpudata *cpudata) > } > > static void amd_pstate_update(struct amd_cpudata *cpudata, u32 min_perf, > - u32 des_perf, u32 max_perf, bool fast_switch) > + u32 des_perf, u32 max_perf, bool fast_switch, int guv_flags) > { > u64 prev = READ_ONCE(cpudata->cppc_req_cached); > u64 value = prev; > > des_perf = clamp_t(unsigned long, des_perf, min_perf, max_perf); > + > + if ((cppc_state == AMD_PSTATE_GUIDED) && (guv_flags & CPUFREQ_GOV_DYNAMIC_SWITCHING)) { > + min_perf = des_perf; > + des_perf = 0; > + } > + > value &= ~AMD_CPPC_MIN_PERF(~0L); > value |= AMD_CPPC_MIN_PERF(min_perf); > > @@ -404,7 +424,7 @@ static int amd_pstate_target(struct cpufreq_policy *policy, > > cpufreq_freq_transition_begin(policy, &freqs); > amd_pstate_update(cpudata, min_perf, des_perf, > - max_perf, false); > + max_perf, false, policy->governor->flags); > cpufreq_freq_transition_end(policy, &freqs, false); > > return 0; > @@ -438,7 +458,8 @@ static void amd_pstate_adjust_perf(unsigned int cpu, > if (max_perf < min_perf) > max_perf = min_perf; > > - amd_pstate_update(cpudata, min_perf, des_perf, max_perf, true); > + amd_pstate_update(cpudata, min_perf, des_perf, max_perf, true, > + policy->governor->flags); > cpufreq_cpu_put(policy); > } > > @@ -1233,7 +1254,7 @@ static int __init amd_pstate_init(void) > /* capability check */ > if (boot_cpu_has(X86_FEATURE_CPPC)) { > pr_debug("AMD CPPC MSR based functionality is supported\n"); > - if (cppc_state == AMD_PSTATE_PASSIVE) > + if (cppc_state != AMD_PSTATE_ACTIVE) > default_pstate_driver->adjust_perf = amd_pstate_adjust_perf; > } else { > pr_debug("AMD CPPC shared memory based functionality is supported\n"); > diff --git a/include/linux/amd-pstate.h b/include/linux/amd-pstate.h > index 15761a581e82..e07cfbd63560 100644 > --- a/include/linux/amd-pstate.h > +++ b/include/linux/amd-pstate.h > @@ -92,6 +92,7 @@ enum amd_pstate_mode { > AMD_PSTATE_DISABLE = 0, > AMD_PSTATE_PASSIVE, > AMD_PSTATE_ACTIVE, > + AMD_PSTATE_GUIDED, > AMD_PSTATE_MAX, > }; > > @@ -99,6 +100,7 @@ static const char * const amd_pstate_mode_string[] = { > [AMD_PSTATE_DISABLE] = "disable", > [AMD_PSTATE_PASSIVE] = "passive", > [AMD_PSTATE_ACTIVE] = "active", > + [AMD_PSTATE_GUIDED] = "guided", > NULL, > }; >
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index e3618dfdb36a..0d8486325c9a 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -7015,11 +7015,11 @@ Do not enable amd_pstate as the default scaling driver for the supported processors passive - Use amd_pstate as a scaling driver, driver requests a - desired performance on this abstract scale and the power - management firmware translates the requests into actual - hardware states (core frequency, data fabric and memory - clocks etc.) + Use amd_pstate with passive mode as a scaling driver. + In this mode autonomous selection is disabled. + Driver requests a desired performance level and platform + tires to match the same performance level (if it is + satisfied by guaranteed performance level). active Use amd_pstate_epp driver instance as the scaling driver, driver provides a hint to the hardware if software wants @@ -7027,3 +7027,8 @@ to the CPPC firmware. then CPPC power algorithm will calculate the runtime workload and adjust the realtime cores frequency. + guided + Activate guided autonomous mode. Driver requests minimum and + maximum performance level and the platform autonomously + selects a performance level in this range and appropriate + to the current workload. diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 87450413cf45..20d78dad712d 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -267,6 +267,20 @@ static int cppc_init_perf(struct amd_cpudata *cpudata) cppc_perf.lowest_nonlinear_perf); WRITE_ONCE(cpudata->lowest_perf, cppc_perf.lowest_perf); + ret = cppc_get_auto_sel_caps(cpudata->cpu, &cppc_perf); + if (ret) { + pr_warn("failed to get auto_sel\n"); + return 0; + } + + if (cppc_state == AMD_PSTATE_PASSIVE) + ret = cppc_set_auto_sel(cpudata->cpu, 0); + else if (cppc_state == AMD_PSTATE_GUIDED) + ret = cppc_set_auto_sel(cpudata->cpu, 1); + + if (ret) + pr_warn("failed to set auto_sel\n"); + return 0; } @@ -344,12 +358,18 @@ static inline bool amd_pstate_sample(struct amd_cpudata *cpudata) } static void amd_pstate_update(struct amd_cpudata *cpudata, u32 min_perf, - u32 des_perf, u32 max_perf, bool fast_switch) + u32 des_perf, u32 max_perf, bool fast_switch, int guv_flags) { u64 prev = READ_ONCE(cpudata->cppc_req_cached); u64 value = prev; des_perf = clamp_t(unsigned long, des_perf, min_perf, max_perf); + + if ((cppc_state == AMD_PSTATE_GUIDED) && (guv_flags & CPUFREQ_GOV_DYNAMIC_SWITCHING)) { + min_perf = des_perf; + des_perf = 0; + } + value &= ~AMD_CPPC_MIN_PERF(~0L); value |= AMD_CPPC_MIN_PERF(min_perf); @@ -404,7 +424,7 @@ static int amd_pstate_target(struct cpufreq_policy *policy, cpufreq_freq_transition_begin(policy, &freqs); amd_pstate_update(cpudata, min_perf, des_perf, - max_perf, false); + max_perf, false, policy->governor->flags); cpufreq_freq_transition_end(policy, &freqs, false); return 0; @@ -438,7 +458,8 @@ static void amd_pstate_adjust_perf(unsigned int cpu, if (max_perf < min_perf) max_perf = min_perf; - amd_pstate_update(cpudata, min_perf, des_perf, max_perf, true); + amd_pstate_update(cpudata, min_perf, des_perf, max_perf, true, + policy->governor->flags); cpufreq_cpu_put(policy); } @@ -1233,7 +1254,7 @@ static int __init amd_pstate_init(void) /* capability check */ if (boot_cpu_has(X86_FEATURE_CPPC)) { pr_debug("AMD CPPC MSR based functionality is supported\n"); - if (cppc_state == AMD_PSTATE_PASSIVE) + if (cppc_state != AMD_PSTATE_ACTIVE) default_pstate_driver->adjust_perf = amd_pstate_adjust_perf; } else { pr_debug("AMD CPPC shared memory based functionality is supported\n"); diff --git a/include/linux/amd-pstate.h b/include/linux/amd-pstate.h index 15761a581e82..e07cfbd63560 100644 --- a/include/linux/amd-pstate.h +++ b/include/linux/amd-pstate.h @@ -92,6 +92,7 @@ enum amd_pstate_mode { AMD_PSTATE_DISABLE = 0, AMD_PSTATE_PASSIVE, AMD_PSTATE_ACTIVE, + AMD_PSTATE_GUIDED, AMD_PSTATE_MAX, }; @@ -99,6 +100,7 @@ static const char * const amd_pstate_mode_string[] = { [AMD_PSTATE_DISABLE] = "disable", [AMD_PSTATE_PASSIVE] = "passive", [AMD_PSTATE_ACTIVE] = "active", + [AMD_PSTATE_GUIDED] = "guided", NULL, };
From ACPI spec below 3 modes for CPPC can be defined: 1. Non autonomous: OS scaling governor specifies operating frequency/ performance level through `Desired Performance` register and platform follows that. 2. Guided autonomous: OS scaling governor specifies min and max frequencies/ performance levels through `Minimum Performance` and `Maximum Performance` register, and platform can autonomously select an operating frequency in this range. 3. Fully autonomous: OS only hints (via EPP) to platform for the required energy performance preference for the workload and platform autonomously scales the frequency. Currently (1) is supported by amd_pstate as passive mode, and (3) is implemented by EPP support. This change is to support (2). In guided autonomous mode the min_perf is based on the input from the scaling governor. For example, in case of schedutil this value depends on the current utilization. And max_perf is set to max capacity. To activate guided auto mode ``amd_pstate=guided`` command line parameter has to be passed in the kernel. Signed-off-by: Wyes Karny <wyes.karny@amd.com> --- .../admin-guide/kernel-parameters.txt | 15 ++++++---- drivers/cpufreq/amd-pstate.c | 29 ++++++++++++++++--- include/linux/amd-pstate.h | 2 ++ 3 files changed, 37 insertions(+), 9 deletions(-)