From ed7bde7a6dab521ee9e28fe2264018f83d83c7fa Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Fri, 26 Jun 2020 11:34:00 -0700 Subject: cpufreq: intel_pstate: Allow enable/disable energy efficiency By default intel_pstate the driver disables energy efficiency by setting MSR_IA32_POWER_CTL bit 19 for Kaby Lake desktop CPU model in HWP mode. This CPU model is also shared by Coffee Lake desktop CPUs. This allows these systems to reach maximum possible frequency. But this adds power penalty, which some customers don't want. They want some way to enable/ disable dynamically. So, add an additional attribute "energy_efficiency" under /sys/devices/system/cpu/intel_pstate/ for these CPU models. This allows to read and write bit 19 ("Disable Energy Efficiency Optimization") in the MSR IA32_POWER_CTL. This attribute is present in both HWP and non-HWP mode as this has an effect in both modes. Refer to Intel Software Developer's manual for details. The scope of this bit is package wide. Also these systems are single package systems. So read/write MSR on the current CPU is enough. The energy efficiency (EE) bit setting needs to be preserved during suspend/resume and CPU offline/online operation. To do this: - Restoring the EE setting from the cpufreq resume() callback, if there is change from the system default. - By default, don't disable EE from cpufreq init() callback for matching CPU models. Since the scope is package wide and is a single package system, move the disable EE calls from init() callback to intel_pstate_init() function, which is called only once. Suggested-by: Len Brown Signed-off-by: Srinivas Pandruvada Signed-off-by: Rafael J. Wysocki --- Documentation/admin-guide/pm/intel_pstate.rst | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/pm/intel_pstate.rst b/Documentation/admin-guide/pm/intel_pstate.rst index 39d80bc29ccd..25e1097fc332 100644 --- a/Documentation/admin-guide/pm/intel_pstate.rst +++ b/Documentation/admin-guide/pm/intel_pstate.rst @@ -431,6 +431,17 @@ argument is passed to the kernel in the command line. supported in the current configuration, writes to this attribute will fail with an appropriate error. +``energy_efficiency`` + This attribute is only present on platforms, which have CPUs matching + Kaby Lake or Coffee Lake desktop CPU model. By default + energy efficiency optimizations are disabled on these CPU models in HWP + mode by this driver. Enabling energy efficiency may limit maximum + operating frequency in both HWP and non HWP mode. In non HWP mode, + optimizations are done only in the turbo frequency range. In HWP mode, + optimizations are done in the entire frequency range. Setting this + attribute to "1" enables energy efficiency optimizations and setting + to "0" disables energy efficiency optimizations. + Interpretation of Policy Attributes ----------------------------------- -- cgit v1.2.3 From f473bf398bf1ed42b1bdbc5206a804d8d3140a2d Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Fri, 26 Jun 2020 11:34:01 -0700 Subject: cpufreq: intel_pstate: Allow raw energy performance preference value Currently using attribute "energy_performance_preference", user space can write one of the four per-defined preference string. These preference strings gets mapped to a hard-coded Energy-Performance Preference (EPP) or Energy-Performance Bias (EPB) knob. These four values are supposed to cover broad spectrum of use cases, but are not uniformly distributed in the range. There are number of cases, where this is not enough. For example: Suppose user wants more performance when connected to AC. Instead of using default "balance performance", the "performance" setting can be used. This changes EPP value from 0x80 to 0x00. But setting EPP to 0, results in electrical and thermal issues on some platforms. This results in aggressive throttling, which causes a drop in performance. But some value between 0x80 and 0x00 results in better performance. But that value can't be fixed as the power curve is not linear. In some cases just changing EPP from 0x80 to 0x75 is enough to get significant performance gain. Similarly on battery the default "balance_performance" mode can be aggressive in power consumption. But picking up the next choice "balance power" results in too much loss of performance, which results in bad user experience in use cases like "Google Hangout". It was observed that some value between these two EPP is optimal. This change allows fine grain EPP tuning for platform like Chromebook or for users who wants to fine tune power and performance. Here based on the product and use cases, different EPP values can be set. This change is similar to the change done for: /sys/devices/system/cpu/cpu*/power/energy_perf_bias where user has choice to write a predefined string or raw value. The change itself is trivial. When user preference doesn't match predefined string preferences and value is an unsigned integer and in range, use that value for EPP. When the EPP feature is not present writing raw value is not supported. Suggested-by: Len Brown Signed-off-by: Srinivas Pandruvada Signed-off-by: Rafael J. Wysocki --- Documentation/admin-guide/pm/intel_pstate.rst | 6 +++- drivers/cpufreq/intel_pstate.c | 50 +++++++++++++++++++++------ 2 files changed, 45 insertions(+), 11 deletions(-) (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/pm/intel_pstate.rst b/Documentation/admin-guide/pm/intel_pstate.rst index 25e1097fc332..40d481cca368 100644 --- a/Documentation/admin-guide/pm/intel_pstate.rst +++ b/Documentation/admin-guide/pm/intel_pstate.rst @@ -565,7 +565,11 @@ somewhere between the two extremes: Strings written to the ``energy_performance_preference`` attribute are internally translated to integer values written to the processor's Energy-Performance Preference (EPP) knob (if supported) or its -Energy-Performance Bias (EPB) knob. +Energy-Performance Bias (EPB) knob. It is also possible to write a positive +integer value between 0 to 255, if the EPP feature is present. If the EPP +feature is not present, writing integer value to this attribute is not +supported. In this case, user can use + "/sys/devices/system/cpu/cpu*/power/energy_perf_bias" interface. [Note that tasks may by migrated from one CPU to another by the scheduler's load-balancing algorithm and if different energy vs performance hints are diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index fd1bef0b9bb4..44c7b4677675 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -602,11 +602,12 @@ static const unsigned int epp_values[] = { HWP_EPP_POWERSAVE }; -static int intel_pstate_get_energy_pref_index(struct cpudata *cpu_data) +static int intel_pstate_get_energy_pref_index(struct cpudata *cpu_data, int *raw_epp) { s16 epp; int index = -EINVAL; + *raw_epp = 0; epp = intel_pstate_get_epp(cpu_data, 0); if (epp < 0) return epp; @@ -614,12 +615,14 @@ static int intel_pstate_get_energy_pref_index(struct cpudata *cpu_data) if (boot_cpu_has(X86_FEATURE_HWP_EPP)) { if (epp == HWP_EPP_PERFORMANCE) return 1; - if (epp <= HWP_EPP_BALANCE_PERFORMANCE) + if (epp == HWP_EPP_BALANCE_PERFORMANCE) return 2; - if (epp <= HWP_EPP_BALANCE_POWERSAVE) + if (epp == HWP_EPP_BALANCE_POWERSAVE) return 3; - else + if (epp == HWP_EPP_POWERSAVE) return 4; + *raw_epp = epp; + return 0; } else if (boot_cpu_has(X86_FEATURE_EPB)) { /* * Range: @@ -638,7 +641,8 @@ static int intel_pstate_get_energy_pref_index(struct cpudata *cpu_data) } static int intel_pstate_set_energy_pref_index(struct cpudata *cpu_data, - int pref_index) + int pref_index, bool use_raw, + u32 raw_epp) { int epp = -EINVAL; int ret; @@ -657,6 +661,16 @@ static int intel_pstate_set_energy_pref_index(struct cpudata *cpu_data, value &= ~GENMASK_ULL(31, 24); + if (use_raw) { + if (raw_epp > 255) { + ret = -EINVAL; + goto return_pref; + } + value |= (u64)raw_epp << 24; + ret = wrmsrl_on_cpu(cpu_data->cpu, MSR_HWP_REQUEST, value); + goto return_pref; + } + if (epp == -EINVAL) epp = epp_values[pref_index - 1]; @@ -694,6 +708,8 @@ static ssize_t store_energy_performance_preference( { struct cpudata *cpu_data = all_cpu_data[policy->cpu]; char str_preference[21]; + bool raw = false; + u32 epp; int ret; ret = sscanf(buf, "%20s", str_preference); @@ -701,10 +717,21 @@ static ssize_t store_energy_performance_preference( return -EINVAL; ret = match_string(energy_perf_strings, -1, str_preference); - if (ret < 0) + if (ret < 0) { + if (!boot_cpu_has(X86_FEATURE_HWP_EPP)) + return ret; + + ret = kstrtouint(buf, 10, &epp); + if (ret) + return ret; + + raw = true; + } + + ret = intel_pstate_set_energy_pref_index(cpu_data, ret, raw, epp); + if (ret) return ret; - intel_pstate_set_energy_pref_index(cpu_data, ret); return count; } @@ -712,13 +739,16 @@ static ssize_t show_energy_performance_preference( struct cpufreq_policy *policy, char *buf) { struct cpudata *cpu_data = all_cpu_data[policy->cpu]; - int preference; + int preference, raw_epp; - preference = intel_pstate_get_energy_pref_index(cpu_data); + preference = intel_pstate_get_energy_pref_index(cpu_data, &raw_epp); if (preference < 0) return preference; - return sprintf(buf, "%s\n", energy_perf_strings[preference]); + if (raw_epp) + return sprintf(buf, "%d\n", raw_epp); + else + return sprintf(buf, "%s\n", energy_perf_strings[preference]); } cpufreq_freq_attr_rw(energy_performance_preference); -- cgit v1.2.3 From 8412b4563e5910485c7bcd4fdcd8bcc3e728284c Mon Sep 17 00:00:00 2001 From: Quentin Perret Date: Mon, 29 Jun 2020 13:55:00 +0530 Subject: cpufreq: Specify default governor on command line Currently, the only way to specify the default CPUfreq governor is via Kconfig options, which suits users who can build the kernel themselves perfectly. However, for those who use a distro-like kernel (such as Android, with the Generic Kernel Image project), the only way to use a non-default governor is to boot to userspace, and to then switch using the sysfs interface. Being able to specify the default governor on the command line, like is the case for cpuidle, would allow those users to specify their governor of choice earlier on, and to simplify the userspace boot procedure slighlty. To support this use-case, add a kernel command line parameter allowing the default governor for CPUfreq to be specified, which takes precedence over the built-in default. This implementation has one notable limitation: the default governor must be registered before the driver. This is solved for builtin governors and drivers using appropriate *_initcall() functions. And in the modular case, this must be reflected as a constraint on the module loading order. Signed-off-by: Quentin Perret [ Viresh: Converted 'default_governor' to a string and parsing it only at initcall level, and several updates to cpufreq_init_policy(). ] Signed-off-by: Viresh Kumar [ rjw: Changelog ] Signed-off-by: Rafael J. Wysocki --- Documentation/admin-guide/kernel-parameters.txt | 5 ++++ Documentation/admin-guide/pm/cpufreq.rst | 6 ++--- drivers/cpufreq/cpufreq.c | 31 ++++++++++++++++++------- 3 files changed, 30 insertions(+), 12 deletions(-) (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index fb95fad81c79..8deb5a89328a 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -703,6 +703,11 @@ cpufreq.off=1 [CPU_FREQ] disable the cpufreq sub-system + cpufreq.default_governor= + [CPU_FREQ] Name of the default cpufreq governor or + policy to use. This governor must be registered in the + kernel before the cpufreq driver probes. + cpu_init_udelay=N [X86] Delay for N microsec between assert and de-assert of APIC INIT to start processors. This delay occurs diff --git a/Documentation/admin-guide/pm/cpufreq.rst b/Documentation/admin-guide/pm/cpufreq.rst index 0c74a7784964..368e612145d2 100644 --- a/Documentation/admin-guide/pm/cpufreq.rst +++ b/Documentation/admin-guide/pm/cpufreq.rst @@ -147,9 +147,9 @@ CPUs in it. The next major initialization step for a new policy object is to attach a scaling governor to it (to begin with, that is the default scaling governor -determined by the kernel configuration, but it may be changed later -via ``sysfs``). First, a pointer to the new policy object is passed to the -governor's ``->init()`` callback which is expected to initialize all of the +determined by the kernel command line or configuration, but it may be changed +later via ``sysfs``). First, a pointer to the new policy object is passed to +the governor's ``->init()`` callback which is expected to initialize all of the data structures necessary to handle the given policy and, possibly, to add a governor ``sysfs`` interface to it. Next, the governor is started by invoking its ``->start()`` callback. diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index e9e8200a0211..ad94b1d47ddb 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -50,6 +50,8 @@ static LIST_HEAD(cpufreq_governor_list); #define for_each_governor(__governor) \ list_for_each_entry(__governor, &cpufreq_governor_list, governor_list) +static char default_governor[CPUFREQ_NAME_LEN]; + /** * The "cpufreq driver" - the arch- or hardware-dependent low * level driver of CPUFreq support, and its spinlock. This lock @@ -1061,7 +1063,6 @@ __weak struct cpufreq_governor *cpufreq_default_governor(void) static int cpufreq_init_policy(struct cpufreq_policy *policy) { - struct cpufreq_governor *def_gov = cpufreq_default_governor(); struct cpufreq_governor *gov = NULL; unsigned int pol = CPUFREQ_POLICY_UNKNOWN; int ret; @@ -1071,21 +1072,27 @@ static int cpufreq_init_policy(struct cpufreq_policy *policy) gov = get_governor(policy->last_governor); if (gov) { pr_debug("Restoring governor %s for cpu %d\n", - policy->governor->name, policy->cpu); - } else if (def_gov) { - gov = def_gov; - __module_get(gov->owner); + gov->name, policy->cpu); } else { - return -ENODATA; + gov = get_governor(default_governor); + } + + if (!gov) { + gov = cpufreq_default_governor(); + if (!gov) + return -ENODATA; + __module_get(gov->owner); } + } else { + /* Use the default policy if there is no last_policy. */ if (policy->last_policy) { pol = policy->last_policy; - } else if (def_gov) { - pol = cpufreq_parse_policy(def_gov->name); + } else { + pol = cpufreq_parse_policy(default_governor); /* - * In case the default governor is neiter "performance" + * In case the default governor is neither "performance" * nor "powersave", fall back to the initial policy * value set by the driver. */ @@ -2795,13 +2802,19 @@ EXPORT_SYMBOL_GPL(cpufreq_unregister_driver); static int __init cpufreq_core_init(void) { + struct cpufreq_governor *gov = cpufreq_default_governor(); + if (cpufreq_disabled()) return -ENODEV; cpufreq_global_kobject = kobject_create_and_add("cpufreq", &cpu_subsys.dev_root->kobj); BUG_ON(!cpufreq_global_kobject); + if (!strlen(default_governor)) + strncpy(default_governor, gov->name, CPUFREQ_NAME_LEN); + return 0; } module_param(off, int, 0444); +module_param_string(default_governor, default_governor, CPUFREQ_NAME_LEN, 0444); core_initcall(cpufreq_core_init); -- cgit v1.2.3