diff options
| -rw-r--r-- | Documentation/ABI/testing/sysfs-devices-system-cpu | 54 | ||||
| -rw-r--r-- | Documentation/admin-guide/pm/intel_pstate.rst | 104 | ||||
| -rw-r--r-- | drivers/base/arch_topology.c | 52 | ||||
| -rw-r--r-- | drivers/base/topology.c | 52 | ||||
| -rw-r--r-- | drivers/cpufreq/amd-pstate-ut.c | 21 | ||||
| -rw-r--r-- | drivers/cpufreq/amd-pstate.c | 120 | ||||
| -rw-r--r-- | drivers/cpufreq/amd-pstate.h | 3 | ||||
| -rw-r--r-- | drivers/cpufreq/cppc_cpufreq.c | 109 | ||||
| -rw-r--r-- | drivers/cpufreq/cpufreq.c | 463 | ||||
| -rw-r--r-- | drivers/cpufreq/intel_pstate.c | 181 | ||||
| -rw-r--r-- | include/linux/arch_topology.h | 8 | ||||
| -rw-r--r-- | include/linux/cpufreq.h | 22 | ||||
| -rw-r--r-- | include/linux/topology.h | 9 | ||||
| -rw-r--r-- | kernel/sched/cpufreq_schedutil.c | 9 | ||||
| -rw-r--r-- | kernel/sched/sched.h | 2 | ||||
| -rw-r--r-- | kernel/sched/topology.c | 25 |
16 files changed, 826 insertions, 408 deletions
diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 6a1acabb29d8..03f43fb667a3 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -268,6 +268,60 @@ Description: Discover CPUs in the same CPU frequency coordination domain This file is only present if the acpi-cpufreq or the cppc-cpufreq drivers are in use. +What: /sys/devices/system/cpu/cpuX/cpufreq/auto_select +Date: May 2025 +Contact: linux-pm@vger.kernel.org +Description: Autonomous selection enable + + Read/write interface to control autonomous selection enable + Read returns autonomous selection status: + 0: autonomous selection is disabled + 1: autonomous selection is enabled + + Write 'y' or '1' or 'on' to enable autonomous selection. + Write 'n' or '0' or 'off' to disable autonomous selection. + + This file is only present if the cppc-cpufreq driver is in use. + +What: /sys/devices/system/cpu/cpuX/cpufreq/auto_act_window +Date: May 2025 +Contact: linux-pm@vger.kernel.org +Description: Autonomous activity window + + This file indicates a moving utilization sensitivity window to + the platform's autonomous selection policy. + + Read/write an integer represents autonomous activity window (in + microseconds) from/to this file. The max value to write is + 1270000000 but the max significand is 127. This means that if 128 + is written to this file, 127 will be stored. If the value is + greater than 130, only the first two digits will be saved as + significand. + + Writing a zero value to this file enable the platform to + determine an appropriate Activity Window depending on the workload. + + Writing to this file only has meaning when Autonomous Selection is + enabled. + + This file is only present if the cppc-cpufreq driver is in use. + +What: /sys/devices/system/cpu/cpuX/cpufreq/energy_performance_preference_val +Date: May 2025 +Contact: linux-pm@vger.kernel.org +Description: Energy performance preference + + Read/write an 8-bit integer from/to this file. This file + represents a range of values from 0 (performance preference) to + 0xFF (energy efficiency preference) that influences the rate of + performance increase/decrease and the result of the hardware's + energy efficiency and performance optimization policies. + + Writing to this file only has meaning when Autonomous Selection is + enabled. + + This file is only present if the cppc-cpufreq driver is in use. + What: /sys/devices/system/cpu/cpu*/cache/index3/cache_disable_{0,1} Date: August 2008 diff --git a/Documentation/admin-guide/pm/intel_pstate.rst b/Documentation/admin-guide/pm/intel_pstate.rst index 78fc83ed2a7e..26e702c7016e 100644 --- a/Documentation/admin-guide/pm/intel_pstate.rst +++ b/Documentation/admin-guide/pm/intel_pstate.rst @@ -329,6 +329,106 @@ information listed above is the same for all of the processors supporting the HWP feature, which is why ``intel_pstate`` works with all of them.] +Support for Hybrid Processors +============================= + +Some processors supported by ``intel_pstate`` contain two or more types of CPU +cores differing by the maximum turbo P-state, performance vs power characteristics, +cache sizes, and possibly other properties. They are commonly referred to as +hybrid processors. To support them, ``intel_pstate`` requires HWP to be enabled +and it assumes the HWP performance units to be the same for all CPUs in the +system, so a given HWP performance level always represents approximately the +same physical performance regardless of the core (CPU) type. + +Hybrid Processors with SMT +-------------------------- + +On systems where SMT (Simultaneous Multithreading), also referred to as +HyperThreading (HT) in the context of Intel processors, is enabled on at least +one core, ``intel_pstate`` assigns performance-based priorities to CPUs. Namely, +the priority of a given CPU reflects its highest HWP performance level which +causes the CPU scheduler to generally prefer more performant CPUs, so the less +performant CPUs are used when the other ones are fully loaded. However, SMT +siblings (that is, logical CPUs sharing one physical core) are treated in a +special way such that if one of them is in use, the effective priority of the +other ones is lowered below the priorities of the CPUs located in the other +physical cores. + +This approach maximizes performance in the majority of cases, but unfortunately +it also leads to excessive energy usage in some important scenarios, like video +playback, which is not generally desirable. While there is no other viable +choice with SMT enabled because the effective capacity and utilization of SMT +siblings are hard to determine, hybrid processors without SMT can be handled in +more energy-efficient ways. + +.. _CAS: + +Capacity-Aware Scheduling Support +--------------------------------- + +The capacity-aware scheduling (CAS) support in the CPU scheduler is enabled by +``intel_pstate`` by default on hybrid processors without SMT. CAS generally +causes the scheduler to put tasks on a CPU so long as there is a sufficient +amount of spare capacity on it, and if the utilization of a given task is too +high for it, the task will need to go somewhere else. + +Since CAS takes CPU capacities into account, it does not require CPU +prioritization and it allows tasks to be distributed more symmetrically among +the more performant and less performant CPUs. Once placed on a CPU with enough +capacity to accommodate it, a task may just continue to run there regardless of +whether or not the other CPUs are fully loaded, so on average CAS reduces the +utilization of the more performant CPUs which causes the energy usage to be more +balanced because the more performant CPUs are generally less energy-efficient +than the less performant ones. + +In order to use CAS, the scheduler needs to know the capacity of each CPU in +the system and it needs to be able to compute scale-invariant utilization of +CPUs, so ``intel_pstate`` provides it with the requisite information. + +First of all, the capacity of each CPU is represented by the ratio of its highest +HWP performance level, multiplied by 1024, to the highest HWP performance level +of the most performant CPU in the system, which works because the HWP performance +units are the same for all CPUs. Second, the frequency-invariance computations, +carried out by the scheduler to always express CPU utilization in the same units +regardless of the frequency it is currently running at, are adjusted to take the +CPU capacity into account. All of this happens when ``intel_pstate`` has +registered itself with the ``CPUFreq`` core and it has figured out that it is +running on a hybrid processor without SMT. + +Energy-Aware Scheduling Support +------------------------------- + +If ``CONFIG_ENERGY_MODEL`` has been set during kernel configuration and +``intel_pstate`` runs on a hybrid processor without SMT, in addition to enabling +`CAS <CAS_>`_ it registers an Energy Model for the processor. This allows the +Energy-Aware Scheduling (EAS) support to be enabled in the CPU scheduler if +``schedutil`` is used as the ``CPUFreq`` governor which requires ``intel_pstate`` +to operate in the `passive mode <Passive Mode_>`_. + +The Energy Model registered by ``intel_pstate`` is artificial (that is, it is +based on abstract cost values and it does not include any real power numbers) +and it is relatively simple to avoid unnecessary computations in the scheduler. +There is a performance domain in it for every CPU in the system and the cost +values for these performance domains have been chosen so that running a task on +a less performant (small) CPU appears to be always cheaper than running that +task on a more performant (big) CPU. However, for two CPUs of the same type, +the cost difference depends on their current utilization, and the CPU whose +current utilization is higher generally appears to be a more expensive +destination for a given task. This helps to balance the load among CPUs of the +same type. + +Since EAS works on top of CAS, high-utilization tasks are always migrated to +CPUs with enough capacity to accommodate them, but thanks to EAS, low-utilization +tasks tend to be placed on the CPUs that look less expensive to the scheduler. +Effectively, this causes the less performant and less loaded CPUs to be +preferred as long as they have enough spare capacity to run the given task +which generally leads to reduced energy usage. + +The Energy Model created by ``intel_pstate`` can be inspected by looking at +the ``energy_model`` directory in ``debugfs`` (typlically mounted on +``/sys/kernel/debug/``). + + User Space Interface in ``sysfs`` ================================= @@ -697,8 +797,8 @@ of them have to be prepended with the ``intel_pstate=`` prefix. Limits`_ for details). ``no_cas`` - Do not enable capacity-aware scheduling (CAS) which is enabled by - default on hybrid systems. + Do not enable `capacity-aware scheduling <CAS_>`_ which is enabled by + default on hybrid systems without SMT. Diagnostics and Tuning ====================== diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c index af0029d30dbe..1037169abb45 100644 --- a/drivers/base/arch_topology.c +++ b/drivers/base/arch_topology.c @@ -154,14 +154,6 @@ void topology_set_freq_scale(const struct cpumask *cpus, unsigned long cur_freq, per_cpu(arch_freq_scale, i) = scale; } -DEFINE_PER_CPU(unsigned long, cpu_scale) = SCHED_CAPACITY_SCALE; -EXPORT_PER_CPU_SYMBOL_GPL(cpu_scale); - -void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity) -{ - per_cpu(cpu_scale, cpu) = capacity; -} - DEFINE_PER_CPU(unsigned long, hw_pressure); /** @@ -207,53 +199,9 @@ void topology_update_hw_pressure(const struct cpumask *cpus, } EXPORT_SYMBOL_GPL(topology_update_hw_pressure); -static ssize_t cpu_capacity_show(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct cpu *cpu = container_of(dev, struct cpu, dev); - - return sysfs_emit(buf, "%lu\n", topology_get_cpu_scale(cpu->dev.id)); -} - static void update_topology_flags_workfn(struct work_struct *work); static DECLARE_WORK(update_topology_flags_work, update_topology_flags_workfn); -static DEVICE_ATTR_RO(cpu_capacity); - -static int cpu_capacity_sysctl_add(unsigned int cpu) -{ - struct device *cpu_dev = get_cpu_device(cpu); - - if (!cpu_dev) - return -ENOENT; - - device_create_file(cpu_dev, &dev_attr_cpu_capacity); - - return 0; -} - -static int cpu_capacity_sysctl_remove(unsigned int cpu) -{ - struct device *cpu_dev = get_cpu_device(cpu); - - if (!cpu_dev) - return -ENOENT; - - device_remove_file(cpu_dev, &dev_attr_cpu_capacity); - - return 0; -} - -static int register_cpu_capacity_sysctl(void) -{ - cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "topology/cpu-capacity", - cpu_capacity_sysctl_add, cpu_capacity_sysctl_remove); - - return 0; -} -subsys_initcall(register_cpu_capacity_sysctl); - static int update_topology; int topology_update_cpu_topology(void) diff --git a/drivers/base/topology.c b/drivers/base/topology.c index b962da263eee..8b42df05feff 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -208,3 +208,55 @@ static int __init topology_sysfs_init(void) } device_initcall(topology_sysfs_init); + +DEFINE_PER_CPU(unsigned long, cpu_scale) = SCHED_CAPACITY_SCALE; +EXPORT_PER_CPU_SYMBOL_GPL(cpu_scale); + +void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity) +{ + per_cpu(cpu_scale, cpu) = capacity; +} + +static ssize_t cpu_capacity_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct cpu *cpu = container_of(dev, struct cpu, dev); + + return sysfs_emit(buf, "%lu\n", topology_get_cpu_scale(cpu->dev.id)); +} + +static DEVICE_ATTR_RO(cpu_capacity); + +static int cpu_capacity_sysctl_add(unsigned int cpu) +{ + struct device *cpu_dev = get_cpu_device(cpu); + + if (!cpu_dev) + return -ENOENT; + + device_create_file(cpu_dev, &dev_attr_cpu_capacity); + + return 0; +} + +static int cpu_capacity_sysctl_remove(unsigned int cpu) +{ + struct device *cpu_dev = get_cpu_device(cpu); + + if (!cpu_dev) + return -ENOENT; + + device_remove_file(cpu_dev, &dev_attr_cpu_capacity); + + return 0; +} + +static int register_cpu_capacity_sysctl(void) +{ + cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "topology/cpu-capacity", + cpu_capacity_sysctl_add, cpu_capacity_sysctl_remove); + + return 0; +} +subsys_initcall(register_cpu_capacity_sysctl); diff --git a/drivers/cpufreq/amd-pstate-ut.c b/drivers/cpufreq/amd-pstate-ut.c index e671bc7d1550..65f9d2bae2d3 100644 --- a/drivers/cpufreq/amd-pstate-ut.c +++ b/drivers/cpufreq/amd-pstate-ut.c @@ -242,25 +242,30 @@ static int amd_pstate_set_mode(enum amd_pstate_mode mode) static int amd_pstate_ut_check_driver(u32 index) { enum amd_pstate_mode mode1, mode2 = AMD_PSTATE_DISABLE; + enum amd_pstate_mode orig_mode = amd_pstate_get_status(); + int ret; for (mode1 = AMD_PSTATE_DISABLE; mode1 < AMD_PSTATE_MAX; mode1++) { - int ret = amd_pstate_set_mode(mode1); + ret = amd_pstate_set_mode(mode1); if (ret) return ret; for (mode2 = AMD_PSTATE_DISABLE; mode2 < AMD_PSTATE_MAX; mode2++) { if (mode1 == mode2) continue; ret = amd_pstate_set_mode(mode2); - if (ret) { - pr_err("%s: failed to update status for %s->%s\n", __func__, - amd_pstate_get_mode_string(mode1), - amd_pstate_get_mode_string(mode2)); - return ret; - } + if (ret) + goto out; } } - return 0; +out: + if (ret) + pr_warn("%s: failed to update status for %s->%s: %d\n", __func__, + amd_pstate_get_mode_string(mode1), + amd_pstate_get_mode_string(mode2), ret); + + amd_pstate_set_mode(orig_mode); + return ret; } static int __init amd_pstate_ut_init(void) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index b961f3a3b580..7983563829b7 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -389,7 +389,8 @@ static inline int amd_pstate_cppc_enable(struct cpufreq_policy *policy) static int msr_init_perf(struct amd_cpudata *cpudata) { union perf_cached perf = READ_ONCE(cpudata->perf); - u64 cap1, numerator; + u64 cap1, numerator, cppc_req; + u8 min_perf; int ret = rdmsrl_safe_on_cpu(cpudata->cpu, MSR_AMD_CPPC_CAP1, &cap1); @@ -400,6 +401,22 @@ static int msr_init_perf(struct amd_cpudata *cpudata) if (ret) return ret; + ret = rdmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, &cppc_req); + if (ret) + return ret; + + WRITE_ONCE(cpudata->cppc_req_cached, cppc_req); + min_perf = FIELD_GET(AMD_CPPC_MIN_PERF_MASK, cppc_req); + + /* + * Clear out the min_perf part to check if the rest of the MSR is 0, if yes, this is an + * indication that the min_perf value is the one specified through the BIOS option + */ + cppc_req &= ~(AMD_CPPC_MIN_PERF_MASK); + + if (!cppc_req) + perf.bios_min_perf = min_perf; + perf.highest_perf = numerator; perf.max_limit_perf = numerator; perf.min_limit_perf = FIELD_GET(AMD_CPPC_LOWEST_PERF_MASK, cap1); @@ -554,6 +571,10 @@ static void amd_pstate_update(struct amd_cpudata *cpudata, u8 min_perf, if (!policy) return; + /* limit the max perf when core performance boost feature is disabled */ + if (!cpudata->boost_supported) + max_perf = min_t(u8, perf.nominal_perf, max_perf); + des_perf = clamp_t(u8, des_perf, min_perf, max_perf); policy->cur = perf_to_freq(perf, cpudata->nominal_freq, des_perf); @@ -563,10 +584,6 @@ static void amd_pstate_update(struct amd_cpudata *cpudata, u8 min_perf, des_perf = 0; } - /* limit the max perf when core performance boost feature is disabled */ - if (!cpudata->boost_supported) - max_perf = min_t(u8, perf.nominal_perf, max_perf); - if (trace_amd_pstate_perf_enabled() && amd_pstate_sample(cpudata)) { trace_amd_pstate_perf(min_perf, des_perf, max_perf, cpudata->freq, cpudata->cur.mperf, cpudata->cur.aperf, cpudata->cur.tsc, @@ -580,20 +597,26 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy_data) { /* * Initialize lower frequency limit (i.e.policy->min) with - * lowest_nonlinear_frequency which is the most energy efficient - * frequency. Override the initial value set by cpufreq core and - * amd-pstate qos_requests. + * lowest_nonlinear_frequency or the min frequency (if) specified in BIOS, + * Override the initial value set by cpufreq core and amd-pstate qos_requests. */ if (policy_data->min == FREQ_QOS_MIN_DEFAULT_VALUE) { struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(policy_data->cpu); struct amd_cpudata *cpudata; + union perf_cached perf; if (!policy) return -EINVAL; cpudata = policy->driver_data; - policy_data->min = cpudata->lowest_nonlinear_freq; + perf = READ_ONCE(cpudata->perf); + + if (perf.bios_min_perf) + policy_data->min = perf_to_freq(perf, cpudata->nominal_freq, + perf.bios_min_perf); + else + policy_data->min = cpudata->lowest_nonlinear_freq; } cpufreq_verify_within_cpu_limits(policy_data); @@ -808,19 +831,16 @@ static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata) sched_set_itmt_core_prio((int)READ_ONCE(cpudata->prefcore_ranking), cpudata->cpu); } -static void amd_pstate_update_limits(unsigned int cpu) +static void amd_pstate_update_limits(struct cpufreq_policy *policy) { - struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); struct amd_cpudata *cpudata; u32 prev_high = 0, cur_high = 0; bool highest_perf_changed = false; + unsigned int cpu = policy->cpu; if (!amd_pstate_prefcore) return; - if (!policy) - return; - if (amd_get_highest_perf(cpu, &cur_high)) return; @@ -1024,6 +1044,10 @@ free_cpudata1: static void amd_pstate_cpu_exit(struct cpufreq_policy *policy) { struct amd_cpudata *cpudata = policy->driver_data; + union perf_cached perf = READ_ONCE(cpudata->perf); + + /* Reset CPPC_REQ MSR to the BIOS value */ + amd_pstate_update_perf(policy, perf.bios_min_perf, 0U, 0U, 0U, false); freq_qos_remove_request(&cpudata->req[1]); freq_qos_remove_request(&cpudata->req[0]); @@ -1305,6 +1329,12 @@ static ssize_t amd_pstate_show_status(char *buf) return sysfs_emit(buf, "%s\n", amd_pstate_mode_string[cppc_state]); } +int amd_pstate_get_status(void) +{ + return cppc_state; +} +EXPORT_SYMBOL_GPL(amd_pstate_get_status); + int amd_pstate_update_status(const char *buf, size_t size) { int mode_idx; @@ -1419,7 +1449,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy) struct amd_cpudata *cpudata; union perf_cached perf; struct device *dev; - u64 value; int ret; /* @@ -1484,12 +1513,6 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy) cpudata->epp_default = AMD_CPPC_EPP_BALANCE_PERFORMANCE; } - if (cpu_feature_enabled(X86_FEATURE_CPPC)) { - ret = rdmsrl_on_cpu(cpudata->cpu, MSR_AMD_CPPC_REQ, &value); - if (ret) - return ret; - WRITE_ONCE(cpudata->cppc_req_cached, value); - } ret = amd_pstate_set_epp(policy, cpudata->epp_default); if (ret) return ret; @@ -1509,6 +1532,11 @@ static void amd_pstate_epp_cpu_exit(struct cpufreq_policy *policy) struct amd_cpudata *cpudata = policy->driver_data; if (cpudata) { + union perf_cached perf = READ_ONCE(cpudata->perf); + + /* Reset CPPC_REQ MSR to the BIOS value */ + amd_pstate_update_perf(policy, perf.bios_min_perf, 0U, 0U, 0U, false); + kfree(cpudata); policy->driver_data = NULL; } @@ -1559,21 +1587,38 @@ static int amd_pstate_epp_set_policy(struct cpufreq_policy *policy) return 0; } -static int amd_pstate_epp_cpu_online(struct cpufreq_policy *policy) +static int amd_pstate_cpu_online(struct cpufreq_policy *policy) { - pr_debug("AMD CPU Core %d going online\n", policy->cpu); - return amd_pstate_cppc_enable(policy); } -static int amd_pstate_epp_cpu_offline(struct cpufreq_policy *policy) +static int amd_pstate_cpu_offline(struct cpufreq_policy *policy) { - return 0; + struct amd_cpudata *cpudata = policy->driver_data; + union perf_cached perf = READ_ONCE(cpudata->perf); + + /* + * Reset CPPC_REQ MSR to the BIOS value, this will allow us to retain the BIOS specified + * min_perf value across kexec reboots. If this CPU is just onlined normally after this, the + * limits, epp and desired perf will get reset to the cached values in cpudata struct + */ + return amd_pstate_update_perf(policy, perf.bios_min_perf, 0U, 0U, 0U, false); } -static int amd_pstate_epp_suspend(struct cpufreq_policy *policy) +static int amd_pstate_suspend(struct cpufreq_policy *policy) { struct amd_cpudata *cpudata = policy->driver_data; + union perf_cached perf = READ_ONCE(cpudata->perf); + int ret; + + /* + * Reset CPPC_REQ MSR to the BIOS value, this will allow us to retain the BIOS specified + * min_perf value across kexec reboots. If this CPU is just resumed back without kexec, + * the limits, epp and desired perf will get reset to the cached values in cpudata struct + */ + ret = amd_pstate_update_perf(policy, perf.bios_min_perf, 0U, 0U, 0U, false); + if (ret) + return ret; /* invalidate to ensure it's rewritten during resume */ cpudata->cppc_req_cached = 0; @@ -1584,6 +1629,17 @@ static int amd_pstate_epp_suspend(struct cpufreq_policy *policy) return 0; } +static int amd_pstate_resume(struct cpufreq_policy *policy) +{ + struct amd_cpudata *cpudata = policy->driver_data; + union perf_cached perf = READ_ONCE(cpudata->perf); + int cur_perf = freq_to_perf(perf, cpudata->nominal_freq, policy->cur); + + /* Set CPPC_REQ to last sane value until the governor updates it */ + return amd_pstate_update_perf(policy, perf.min_limit_perf, cur_perf, perf.max_limit_perf, + 0U, false); +} + static int amd_pstate_epp_resume(struct cpufreq_policy *policy) { struct amd_cpudata *cpudata = policy->driver_data; @@ -1609,6 +1665,10 @@ static struct cpufreq_driver amd_pstate_driver = { .fast_switch = amd_pstate_fast_switch, .init = amd_pstate_cpu_init, .exit = amd_pstate_cpu_exit, + .online = amd_pstate_cpu_online, + .offline = amd_pstate_cpu_offline, + .suspend = amd_pstate_suspend, + .resume = amd_pstate_resume, .set_boost = amd_pstate_set_boost, .update_limits = amd_pstate_update_limits, .name = "amd-pstate", @@ -1621,9 +1681,9 @@ static struct cpufreq_driver amd_pstate_epp_driver = { .setpolicy = amd_pstate_epp_set_policy, .init = amd_pstate_epp_cpu_init, .exit = amd_pstate_epp_cpu_exit, - .offline = amd_pstate_epp_cpu_offline, - .online = amd_pstate_epp_cpu_online, - .suspend = amd_pstate_epp_suspend, + .offline = amd_pstate_cpu_offline, + .online = amd_pstate_cpu_online, + .suspend = amd_pstate_suspend, .resume = amd_pstate_epp_resume, .update_limits = amd_pstate_update_limits, .set_boost = amd_pstate_set_boost, diff --git a/drivers/cpufreq/amd-pstate.h b/drivers/cpufreq/amd-pstate.h index fbe1c08d3f06..cb45fdca27a6 100644 --- a/drivers/cpufreq/amd-pstate.h +++ b/drivers/cpufreq/amd-pstate.h @@ -30,6 +30,7 @@ * @lowest_perf: the absolute lowest performance level of the processor * @min_limit_perf: Cached value of the performance corresponding to policy->min * @max_limit_perf: Cached value of the performance corresponding to policy->max + * @bios_min_perf: Cached perf value corresponding to the "Requested CPU Min Frequency" BIOS option */ union perf_cached { struct { @@ -39,6 +40,7 @@ union perf_cached { u8 lowest_perf; u8 min_limit_perf; u8 max_limit_perf; + u8 bios_min_perf; }; u64 val; }; @@ -119,6 +121,7 @@ enum amd_pstate_mode { AMD_PSTATE_MAX, }; const char *amd_pstate_get_mode_string(enum amd_pstate_mode mode); +int amd_pstate_get_status(void); int amd_pstate_update_status(const char *buf, size_t size); #endif /* _LINUX_AMD_PSTATE_H */ diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c index cb93f00bafdb..b7c688a5659c 100644 --- a/drivers/cpufreq/cppc_cpufreq.c +++ b/drivers/cpufreq/cppc_cpufreq.c @@ -808,10 +808,119 @@ static ssize_t show_freqdomain_cpus(struct cpufreq_policy *policy, char *buf) return cpufreq_show_cpus(cpu_data->shared_cpu_map, buf); } + +static ssize_t show_auto_select(struct cpufreq_policy *policy, char *buf) +{ + bool val; + int ret; + + ret = cppc_get_auto_sel(policy->cpu, &val); + + /* show "<unsupported>" when this register is not supported by cpc */ + if (ret == -EOPNOTSUPP) + return sysfs_emit(buf, "<unsupported>\n"); + + if (ret) + return ret; + + return sysfs_emit(buf, "%d\n", val); +} + +static ssize_t store_auto_select(struct cpufreq_policy *policy, + const char *buf, size_t count) +{ + bool val; + int ret; + + ret = kstrtobool(buf, &val); + if (ret) + return ret; + + ret = cppc_set_auto_sel(policy->cpu, val); + if (ret) + return ret; + + return count; +} + +static ssize_t show_auto_act_window(struct cpufreq_policy *policy, char *buf) +{ + u64 val; + int ret; + + ret = cppc_get_auto_act_window(policy->cpu, &val); + + /* show "<unsupported>" when this register is not supported by cpc */ + if (ret == -EOPNOTSUPP) + return sysfs_emit(buf, "<unsupported>\n"); + + if (ret) + return ret; + + return sysfs_emit(buf, "%llu\n", val); +} + +static ssize_t store_auto_act_window(struct cpufreq_policy *policy, + const char *buf, size_t count) +{ + u64 usec; + int ret; + + ret = kstrtou64(buf, 0, &usec); + if (ret) + return ret; + + ret = cppc_set_auto_act_window(policy->cpu, usec); + if (ret) + return ret; + + return count; +} + +static ssize_t show_energy_performance_preference_val(struct cpufreq_policy *policy, char *buf) +{ + u64 val; + int ret; + + ret = cppc_get_epp_perf(policy->cpu, &val); + + /* show "<unsupported>" when this register is not supported by cpc */ + if (ret == -EOPNOTSUPP) + return sysfs_emit(buf, "<unsupported>\n"); + + if (ret) + return ret; + + return sysfs_emit(buf, "%llu\n", val); +} + +static ssize_t store_energy_performance_preference_val(struct cpufreq_policy *policy, + const char *buf, size_t count) +{ + u64 val; + int ret; + + ret = kstrtou64(buf, 0, &val); + if (ret) + return ret; + + ret = cppc_set_epp(policy->cpu, val); + if (ret) + return ret; + + return count; +} + cpufreq_freq_attr_ro(freqdomain_cpus); +cpufreq_freq_attr_rw(auto_select); +cpufreq_freq_attr_rw(auto_act_window); +cpufreq_freq_attr_rw(energy_performance_preference_val); static struct freq_attr *cppc_cpufreq_attr[] = { &freqdomain_cpus, + &auto_select, + &auto_act_window, + &energy_performance_preference_val, NULL, }; diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index f45ded62b0e0..d7426e1d8bdd 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -255,51 +255,6 @@ void cpufreq_cpu_put(struct cpufreq_policy *policy) } EXPORT_SYMBOL_GPL(cpufreq_cpu_put); -/** - * cpufreq_cpu_release - Unlock a policy and decrement its usage counter. - * @policy: cpufreq policy returned by cpufreq_cpu_acquire(). - */ -void cpufreq_cpu_release(struct cpufreq_policy *policy) -{ - if (WARN_ON(!policy)) - return; - - lockdep_assert_held(&policy->rwsem); - - up_write(&policy->rwsem); - - cpufreq_cpu_put(policy); -} - -/** - * cpufreq_cpu_acquire - Find policy for a CPU, mark it as busy and lock it. - * @cpu: CPU to find the policy for. - * - * Call cpufreq_cpu_get() to get a reference on the cpufreq policy for @cpu and - * if the policy returned by it is not NULL, acquire its rwsem for writing. - * Return the policy if it is active or release it and return NULL otherwise. - * - * The policy returned by this function has to be released with the help of - * cpufreq_cpu_release() in order to release its rwsem and balance its usage - * counter properly. - */ -struct cpufreq_policy *cpufreq_cpu_acquire(unsigned int cpu) -{ - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - - if (!policy) - return NULL; - - down_write(&policy->rwsem); - - if (policy_is_inactive(policy)) { - cpufreq_cpu_release(policy); - return NULL; - } - - return policy; -} - /********************************************************************* * EXTERNALLY AFFECTING FREQUENCY CHANGES * *********************************************************************/ @@ -636,6 +591,22 @@ static ssize_t show_local_boost(struct cpufreq_policy *policy, char *buf) return sysfs_emit(buf, "%d\n", policy->boost_enabled); } +static int policy_set_boost(struct cpufreq_policy *policy, bool enable) +{ + int ret; + + if (policy->boost_enabled == enable) + return 0; + + policy->boost_enabled = enable; + + ret = cpufreq_driver->set_boost(policy, enable); + if (ret) + policy->boost_enabled = !policy->boost_enabled; + + return ret; +} + static ssize_t store_local_boost(struct cpufreq_policy *policy, const char *buf, size_t count) { @@ -651,21 +622,11 @@ static ssize_t store_local_boost(struct cpufreq_policy *policy, if (!policy->boost_supported) return -EINVAL; - if (policy->boost_enabled == enable) + ret = policy_set_boost(policy, enable); + if (!ret) return count; - policy->boost_enabled = enable; - - cpus_read_lock(); - ret = cpufreq_driver->set_boost(policy, enable); - cpus_read_unlock(); - - if (ret) { - policy->boost_enabled = !policy->boost_enabled; - return ret; - } - - return count; + return ret; } static struct freq_attr local_boost = __ATTR(boost, 0644, show_local_boost, store_local_boost); @@ -845,7 +806,7 @@ static ssize_t show_scaling_governor(struct cpufreq_policy *policy, char *buf) static ssize_t store_scaling_governor(struct cpufreq_policy *policy, const char *buf, size_t count) { - char str_governor[16]; + char str_governor[CPUFREQ_NAME_LEN]; int ret; ret = sscanf(buf, "%15s", str_governor); @@ -956,9 +917,9 @@ static ssize_t store_scaling_setspeed(struct cpufreq_policy *policy, if (!policy->governor || !policy->governor->store_setspeed) return -EINVAL; - ret = sscanf(buf, "%u", &freq); - if (ret != 1) - return -EINVAL; + ret = kstrtouint(buf, 0, &freq); + if (ret) + return ret; policy->governor->store_setspeed(policy, freq); @@ -1025,17 +986,16 @@ static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf) { struct cpufreq_policy *policy = to_policy(kobj); struct freq_attr *fattr = to_attr(attr); - ssize_t ret = -EBUSY; if (!fattr->show) return -EIO; - down_read(&policy->rwsem); + guard(cpufreq_policy_read)(policy); + if (likely(!policy_is_inactive(policy))) - ret = fattr->show(policy, buf); - up_read(&policy->rwsem); + return fattr->show(policy, buf); - return ret; + return -EBUSY; } static ssize_t store(struct kobject *kobj, struct attribute *attr, @@ -1043,17 +1003,16 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr, { struct cpufreq_policy *policy = to_policy(kobj); struct freq_attr *fattr = to_attr(attr); - ssize_t ret = -EBUSY; if (!fattr->store) return -EIO; - down_write(&policy->rwsem); + guard(cpufreq_policy_write)(policy); + if (likely(!policy_is_inactive(policy))) - ret = fattr->store(policy, buf, count); - up_write(&policy->rwsem); + return fattr->store(policy, buf, count); - return ret; + return -EBUSY; } static void cpufreq_sysfs_release(struct kobject *kobj) @@ -1211,7 +1170,8 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy, unsigned int cp if (cpumask_test_cpu(cpu, policy->cpus)) return 0; - down_write(&policy->rwsem); + guard(cpufreq_policy_write)(policy); + if (has_target()) cpufreq_stop_governor(policy); @@ -1222,7 +1182,7 @@ static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy, unsigned int cp if (ret) pr_err("%s: Failed to start governor\n", __func__); } - up_write(&policy->rwsem); + return ret; } @@ -1242,9 +1202,10 @@ static void handle_update(struct work_struct *work) container_of(work, struct cpufreq_policy, update); pr_debug("handle_update for cpu %u called\n", policy->cpu); - down_write(&policy->rwsem); + + guard(cpufreq_policy_write)(policy); + refresh_frequency_limits(policy); - up_write(&policy->rwsem); } static int cpufreq_notifier_min(struct notifier_block *nb, unsigned long freq, @@ -1270,11 +1231,11 @@ static void cpufreq_policy_put_kobj(struct cpufreq_policy *policy) struct kobject *kobj; struct completion *cmp; - down_write(&policy->rwsem); - cpufreq_stats_free_table(policy); - kobj = &policy->kobj; - cmp = &policy->kobj_unregister; - up_write(&policy->rwsem); + scoped_guard(cpufreq_policy_write, policy) { + cpufreq_stats_free_table(policy); + kobj = &policy->kobj; + cmp = &policy->kobj_unregister; + } kobject_put(kobj); /* @@ -1350,7 +1311,6 @@ static struct cpufreq_policy *cpufreq_policy_alloc(unsigned int cpu) init_waitqueue_head(&policy->transition_wait); INIT_WORK(&policy->update, handle_update); - policy->cpu = cpu; return policy; err_min_qos_notifier: @@ -1419,35 +1379,17 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy) kfree(policy); } -static int cpufreq_online(unsigned int cpu) +static int cpufreq_policy_online(struct cpufreq_policy *policy, + unsigned int cpu, bool new_policy) { - struct cpufreq_policy *policy; - bool new_policy; unsigned long flags; unsigned int j; int ret; - pr_debug("%s: bringing CPU%u online\n", __func__, cpu); - - /* Check if this CPU already has a policy to manage it */ - policy = per_cpu(cpufreq_cpu_data, cpu); - if (policy) { - WARN_ON(!cpumask_test_cpu(cpu, policy->related_cpus)); - if (!policy_is_inactive(policy)) - return cpufreq_add_policy_cpu(policy, cpu); + guard(cpufreq_policy_write)(policy); - /* This is the only online CPU for the policy. Start over. */ - new_policy = false; - down_write(&policy->rwsem); - policy->cpu = cpu; - policy->governor = NULL; - } else { - new_policy = true; - policy = cpufreq_policy_alloc(cpu); - if (!policy) - return -ENOMEM; - down_write(&policy->rwsem); - } + policy->cpu = cpu; + policy->governor = NULL; if (!new_policy && cpufreq_driver->online) { /* Recover policy->cpus using related_cpus */ @@ -1470,7 +1412,7 @@ static int cpufreq_online(unsigned int cpu) if (ret) { pr_debug("%s: %d: initialization failed\n", __func__, __LINE__); - goto out_free_policy; + goto out_clear_policy; } /* @@ -1621,7 +1563,55 @@ static int cpufreq_online(unsigned int cpu) goto out_destroy_policy; } - up_write(&policy->rwsem); + return 0; + +out_destroy_policy: + for_each_cpu(j, policy->real_cpus) + remove_cpu_dev_symlink(policy, j, get_cpu_device(j)); + +out_offline_policy: + if (cpufreq_driver->offline) + cpufreq_driver->offline(policy); + +out_exit_policy: + if (cpufreq_driver->exit) + cpufreq_driver->exit(policy); + +out_clear_policy: + cpumask_clear(policy->cpus); + + return ret; +} + +static int cpufreq_online(unsigned int cpu) +{ + struct cpufreq_policy *policy; + bool new_policy; + int ret; + + pr_debug("%s: bringing CPU%u online\n", __func__, cpu); + + /* Check if this CPU already has a policy to manage it */ + policy = per_cpu(cpufreq_cpu_data, cpu); + if (policy) { + WARN_ON(!cpumask_test_cpu(cpu, policy->related_cpus)); + if (!policy_is_inactive(policy)) + return cpufreq_add_policy_cpu(policy, cpu); + + /* This is the only online CPU for the policy. Start over. */ + new_policy = false; + } else { + new_policy = true; + policy = cpufreq_policy_alloc(cpu); + if (!policy) + return -ENOMEM; + } + + ret = cpufreq_policy_online(policy, cpu, new_policy); + if (ret) { + cpufreq_policy_free(policy); + return ret; + } kobject_uevent(&policy->kobj, KOBJ_ADD); @@ -1633,41 +1623,24 @@ static int cpufreq_online(unsigned int cpu) if (new_policy && cpufreq_thermal_control_enabled(cpufreq_driver)) policy->cdev = of_cpufreq_cooling_register(policy); - /* Let the per-policy boost flag mirror the cpufreq_driver boost during init */ + /* + * Let the per-policy boost flag mirror the cpufreq_driver boost during + * initialization for a new policy. For an existing policy, maintain the + * previous boost value unless global boost is disabled. + */ if (cpufreq_driver->set_boost && policy->boost_supported && - policy->boost_enabled != cpufreq_boost_enabled()) { - policy->boost_enabled = cpufreq_boost_enabled(); - ret = cpufreq_driver->set_boost(policy, policy->boost_enabled); + (new_policy || !cpufreq_boost_enabled())) { + ret = policy_set_boost(policy, cpufreq_boost_enabled()); if (ret) { /* If the set_boost fails, the online operation is not affected */ pr_info("%s: CPU%d: Cannot %s BOOST\n", __func__, policy->cpu, - str_enable_disable(policy->boost_enabled)); - policy->boost_enabled = !policy->boost_enabled; + str_enable_disable(cpufreq_boost_enabled())); } } pr_debug("initialization complete\n"); return 0; - -out_destroy_policy: - for_each_cpu(j, policy->real_cpus) - remove_cpu_dev_symlink(policy, j, get_cpu_device(j)); - -out_offline_policy: - if (cpufreq_driver->offline) - cpufreq_driver->offline(policy); - -out_exit_policy: - if (cpufreq_driver->exit) - cpufreq_driver->exit(policy); - -out_free_policy: - cpumask_clear(policy->cpus); - up_write(&policy->rwsem); - - cpufreq_policy_free(policy); - return ret; } /** @@ -1757,11 +1730,10 @@ static int cpufreq_offline(unsigned int cpu) return 0; } - down_write(&policy->rwsem); + guard(cpufreq_policy_write)(policy); __cpufreq_offline(cpu, policy); - up_write(&policy->rwsem); return 0; } @@ -1778,33 +1750,29 @@ static void cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) if (!policy) return; - down_write(&policy->rwsem); + scoped_guard(cpufreq_policy_write, policy) { + if (cpu_online(cpu)) + __cpufreq_offline(cpu, policy); - if (cpu_online(cpu)) - __cpufreq_offline(cpu, policy); + remove_cpu_dev_symlink(policy, cpu, dev); - remove_cpu_dev_symlink(policy, cpu, dev); + if (!cpumask_empty(policy->real_cpus)) + return; - if (!cpumask_empty(policy->real_cpus)) { - up_write(&policy->rwsem); - return; - } + /* + * Unregister cpufreq cooling once all the CPUs of the policy + * are removed. + */ + if (cpufreq_thermal_control_enabled(cpufreq_driver)) { + cpufreq_cooling_unregister(policy->cdev); + policy->cdev = NULL; + } - /* - * Unregister cpufreq cooling once all the CPUs of the policy are - * removed. - */ - if (cpufreq_thermal_control_enabled(cpufreq_driver)) { - cpufreq_cooling_unregister(policy->cdev); - policy->cdev = NULL; + /* We did light-weight exit earlier, do full tear down now */ + if (cpufreq_driver->offline && cpufreq_driver->exit) + cpufreq_driver->exit(policy); } - /* We did light-weight exit earlier, do full tear down now */ - if (cpufreq_driver->offline && cpufreq_driver->exit) - cpufreq_driver->exit(policy); - - up_write(&policy->rwsem); - cpufreq_policy_free(policy); } @@ -1874,27 +1842,26 @@ static unsigned int cpufreq_verify_current_freq(struct cpufreq_policy *policy, b */ unsigned int cpufreq_quick_get(unsigned int cpu) { - struct cpufreq_policy *policy; - unsigned int ret_freq = 0; + struct cpufreq_policy *policy __free(put_cpufreq_policy) = NULL; unsigned long flags; read_lock_irqsave(&cpufreq_driver_lock, flags); if (cpufreq_driver && cpufreq_driver->setpolicy && cpufreq_driver->get) { - ret_freq = cpufreq_driver->get(cpu); + unsigned int ret_freq = cpufreq_driver->get(cpu); + read_unlock_irqrestore(&cpufreq_driver_lock, flags); + return ret_freq; } read_unlock_irqrestore(&cpufreq_driver_lock, flags); policy = cpufreq_cpu_get(cpu); - if (policy) { - ret_freq = policy->cur; - cpufreq_cpu_put(policy); - } + if (policy) + return policy->cur; - return ret_freq; + return 0; } EXPORT_SYMBOL(cpufreq_quick_get); @@ -1906,15 +1873,13 @@ EXPORT_SYMBOL(cpufreq_quick_get); */ unsigned int cpufreq_quick_get_max(unsigned int cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - unsigned int ret_freq = 0; + struct cpufreq_policy *policy __free(put_cpufreq_policy); - if (policy) { - ret_freq = policy->max; - cpufreq_cpu_put(policy); - } + policy = cpufreq_cpu_get(cpu); + if (policy) + return policy->max; - return ret_freq; + return 0; } EXPORT_SYMBOL(cpufreq_quick_get_max); @@ -1926,15 +1891,13 @@ EXPORT_SYMBOL(cpufreq_quick_get_max); */ __weak unsigned int cpufreq_get_hw_max_freq(unsigned int cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - unsigned int ret_freq = 0; + struct cpufreq_policy *policy __free(put_cpufreq_policy); - if (policy) { - ret_freq = policy->cpuinfo.max_freq; - cpufreq_cpu_put(policy); - } + policy = cpufreq_cpu_get(cpu); + if (policy) + return policy->cpuinfo.max_freq; - return ret_freq; + return 0; } EXPORT_SYMBOL(cpufreq_get_hw_max_freq); @@ -1954,19 +1917,18 @@ static unsigned int __cpufreq_get(struct cpufreq_policy *policy) */ unsigned int cpufreq_get(unsigned int cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - unsigned int ret_freq = 0; + struct cpufreq_policy *policy __free(put_cpufreq_policy); - if (policy) { - down_read(&policy->rwsem); - if (cpufreq_driver->get) - ret_freq = __cpufreq_get(policy); - up_read(&policy->rwsem); + policy = cpufreq_cpu_get(cpu); + if (!policy) + return 0; - cpufreq_cpu_put(policy); - } + guard(cpufreq_policy_read)(policy); + + if (cpufreq_driver->get) + return __cpufreq_get(policy); - return ret_freq; + return 0; } EXPORT_SYMBOL(cpufreq_get); @@ -2025,9 +1987,9 @@ void cpufreq_suspend(void) for_each_active_policy(policy) { if (has_target()) { - down_write(&policy->rwsem); - cpufreq_stop_governor(policy); - up_write(&policy->rwsem); + scoped_guard(cpufreq_policy_write, policy) { + cpufreq_stop_governor(policy); + } } if (cpufreq_driver->suspend && cpufreq_driver->suspend(policy)) @@ -2068,9 +2030,9 @@ void cpufreq_resume(void) pr_err("%s: Failed to resume driver: %s\n", __func__, cpufreq_driver->name); } else if (has_target()) { - down_write(&policy->rwsem); - ret = cpufreq_start_governor(policy); - up_write(&policy->rwsem); + scoped_guard(cpufreq_policy_write, policy) { + ret = cpufreq_start_governor(policy); + } if (ret) pr_err("%s: Failed to start governor for CPU%u's policy\n", @@ -2438,15 +2400,9 @@ int cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - int ret; - - down_write(&policy->rwsem); + guard(cpufreq_policy_write)(policy); - ret = __cpufreq_driver_target(policy, target_freq, relation); - - up_write(&policy->rwsem); - - return ret; + return __cpufreq_driver_target(policy, target_freq, relation); } EXPORT_SYMBOL_GPL(cpufreq_driver_target); @@ -2618,31 +2574,6 @@ EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); * POLICY INTERFACE * *********************************************************************/ -/** - * cpufreq_get_policy - get the current cpufreq_policy - * @policy: struct cpufreq_policy into which the current cpufreq_policy - * is written - * @cpu: CPU to find the policy for - * - * Reads the current cpufreq policy. - */ -int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu) -{ - struct cpufreq_policy *cpu_policy; - if (!policy) - return -EINVAL; - - cpu_policy = cpufreq_cpu_get(cpu); - if (!cpu_policy) - return -EINVAL; - - memcpy(policy, cpu_policy, sizeof(*policy)); - - cpufreq_cpu_put(cpu_policy); - return 0; -} -EXPORT_SYMBOL(cpufreq_get_policy); - DEFINE_PER_CPU(unsigned long, cpufreq_pressure); /** @@ -2793,6 +2724,21 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy, return ret; } +static void cpufreq_policy_refresh(struct cpufreq_policy *policy) +{ + guard(cpufreq_policy_write)(policy); + + /* + * BIOS might change freq behind our back + * -> ask driver for current freq and notify governors about a change + */ + if (cpufreq_driver->get && has_target() && + (cpufreq_suspended || WARN_ON(!cpufreq_verify_current_freq(policy, false)))) + return; + + refresh_frequency_limits(policy); +} + /** * cpufreq_update_policy - Re-evaluate an existing cpufreq policy. * @cpu: CPU to re-evaluate the policy for. @@ -2804,23 +2750,13 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy, */ void cpufreq_update_policy(unsigned int cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpu); + struct cpufreq_policy *policy __free(put_cpufreq_policy); + policy = cpufreq_cpu_get(cpu); if (!policy) return; - /* - * BIOS might change freq behind our back - * -> ask driver for current freq and notify governors about a change - */ - if (cpufreq_driver->get && has_target() && - (cpufreq_suspended || WARN_ON(!cpufreq_verify_current_freq(policy, false)))) - goto unlock; - - refresh_frequency_limits(policy); - -unlock: - cpufreq_cpu_release(policy); + cpufreq_policy_refresh(policy); } EXPORT_SYMBOL(cpufreq_update_policy); @@ -2829,7 +2765,7 @@ EXPORT_SYMBOL(cpufreq_update_policy); * @cpu: CPU to update the policy limits for. * * Invoke the driver's ->update_limits callback if present or call - * cpufreq_update_policy() for @cpu. + * cpufreq_policy_refresh() for @cpu. */ void cpufreq_update_limits(unsigned int cpu) { @@ -2840,9 +2776,9 @@ void cpufreq_update_limits(unsigned int cpu) return; if (cpufreq_driver->update_limits) - cpufreq_driver->update_limits(cpu); + cpufreq_driver->update_limits(policy); else - cpufreq_update_policy(cpu); + cpufreq_policy_refresh(policy); } EXPORT_SYMBOL_GPL(cpufreq_update_limits); @@ -2876,8 +2812,10 @@ static int cpufreq_boost_trigger_state(int state) unsigned long flags; int ret = 0; - if (cpufreq_driver->boost_enabled == state) - return 0; + /* + * Don't compare 'cpufreq_driver->boost_enabled' with 'state' here to + * make sure all policies are in sync with global boost flag. + */ write_lock_irqsave(&cpufreq_driver_lock, flags); cpufreq_driver->boost_enabled = state; @@ -2888,12 +2826,9 @@ static int cpufreq_boost_trigger_state(int state) if (!policy->boost_supported) continue; - policy->boost_enabled = state; - ret = cpufreq_driver->set_boost(policy, state); - if (ret) { - policy->boost_enabled = !policy->boost_enabled; + ret = policy_set_boost(policy, state); + if (ret) goto err_reset_state; - } } cpus_read_unlock(); @@ -3118,6 +3053,36 @@ static int __init cpufreq_core_init(void) return 0; } + +static bool cpufreq_policy_is_good_for_eas(unsigned int cpu) +{ + struct cpufreq_policy *policy __free(put_cpufreq_policy); + + policy = cpufreq_cpu_get(cpu); + if (!policy) { + pr_debug("cpufreq policy not set for CPU: %d\n", cpu); + return false; + } + + return sugov_is_governor(policy); +} + +bool cpufreq_ready_for_eas(const struct cpumask *cpu_mask) +{ + unsigned int cpu; + + /* Do not attempt EAS if schedutil is not being used. */ + for_each_cpu(cpu, cpu_mask) { + if (!cpufreq_policy_is_good_for_eas(cpu)) { + pr_debug("rd %*pbl: schedutil is mandatory for EAS\n", + cpumask_pr_args(cpu_mask)); + return false; + } + } + + return true; +} + module_param(off, int, 0444); module_param_string(default_governor, default_governor, CPUFREQ_NAME_LEN, 0444); core_initcall(cpufreq_core_init); diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index ba9bf06f1c77..0102c3e1441f 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -221,6 +221,7 @@ struct global_params { * @sched_flags: Store scheduler flags for possible cross CPU update * @hwp_boost_min: Last HWP boosted min performance * @suspended: Whether or not the driver has been suspended. + * @pd_registered: Set when a perf domain is registered for this CPU. * @hwp_notify_work: workqueue for HWP notifications. * * This structure stores per CPU instance data for all CPUs. @@ -260,6 +261,9 @@ struct cpudata { unsigned int sched_flags; u32 hwp_boost_min; bool suspended; +#ifdef CONFIG_ENERGY_MODEL + bool pd_registered; +#endif struct delayed_work hwp_notify_work; }; @@ -303,6 +307,7 @@ static bool hwp_is_hybrid; static struct cpufreq_driver *intel_pstate_driver __read_mostly; +#define INTEL_PSTATE_CORE_SCALING 100000 #define HYBRID_SCALING_FACTOR_ADL 78741 #define HYBRID_SCALING_FACTOR_MTL 80000 #define HYBRID_SCALING_FACTOR_LNL 86957 @@ -311,7 +316,7 @@ static int hybrid_scaling_factor; static inline int core_get_scaling(void) { - return 100000; + return INTEL_PSTATE_CORE_SCALING; } #ifdef CONFIG_ACPI @@ -948,12 +953,124 @@ static struct cpudata *hybrid_max_perf_cpu __read_mostly; */ static DEFINE_MUTEX(hybrid_capacity_lock); +#ifdef CONFIG_ENERGY_MODEL +#define HYBRID_EM_STATE_COUNT 4 + +static int hybrid_active_power(struct device *dev, unsigned long *power, + unsigned long *freq) +{ + /* + * Create "utilization bins" of 0-40%, 40%-60%, 60%-80%, and 80%-100% + * of the maximum capacity such that two CPUs of the same type will be + * regarded as equally attractive if the utilization of each of them + * falls into the same bin, which should prevent tasks from being + * migrated between them too often. + * + * For this purpose, return the "frequency" of 2 for the first + * performance level and otherwise leave the value set by the caller. + */ + if (!*freq) + *freq = 2; + + /* No power information. */ + *power = EM_MAX_POWER; + + return 0; +} + +static int hybrid_get_cost(struct device *dev, unsigned long freq, + unsigned long *cost) +{ + struct pstate_data *pstate = &all_cpu_data[dev->id]->pstate; + struct cpu_cacheinfo *cacheinfo = get_cpu_cacheinfo(dev->id); + + /* + * The smaller the perf-to-frequency scaling factor, the larger the IPC + * ratio between the given CPU and the least capable CPU in the system. + * Regard that IPC ratio as the primary cost component and assume that + * the scaling factors for different CPU types will differ by at least + * 5% and they will not be above INTEL_PSTATE_CORE_SCALING. + * + * Add the freq value to the cost, so that the cost of running on CPUs + * of the same type in different "utilization bins" is different. + */ + *cost = div_u64(100ULL * INTEL_PSTATE_CORE_SCALING, pstate->scaling) + freq; + /* + * Increase the cost slightly for CPUs able to access L3 to avoid + * touching it in case some other CPUs of the same type can do the work + * without it. + */ + if (cacheinfo) { + unsigned int i; + + /* Check if L3 cache is there. */ + for (i = 0; i < cacheinfo->num_leaves; i++) { + if (cacheinfo->info_list[i].level == 3) { + *cost += 2; + break; + } + } + } + + return 0; +} + +static bool hybrid_register_perf_domain(unsigned int cpu) +{ + static const struct em_data_callback cb + = EM_ADV_DATA_CB(hybrid_active_power, hybrid_get_cost); + struct cpudata *cpudata = all_cpu_data[cpu]; + struct device *cpu_dev; + + /* + * Registering EM perf domains without enabling asymmetric CPU capacity + * support is not really useful and one domain should not be registered + * more than once. + */ + if (!hybrid_max_perf_cpu || cpudata->pd_registered) + return false; + + cpu_dev = get_cpu_device(cpu); + if (!cpu_dev) + return false; + + if (em_dev_register_perf_domain(cpu_dev, HYBRID_EM_STATE_COUNT, &cb, + cpumask_of(cpu), false)) + return false; + + cpudata->pd_registered = true; + + return true; +} + +static void hybrid_register_all_perf_domains(void) +{ + unsigned int cpu; + + for_each_online_cpu(cpu) + hybrid_register_perf_domain(cpu); +} + +static void hybrid_update_perf_domain(struct cpudata *cpu) +{ + if (cpu->pd_registered) + em_adjust_cpu_capacity(cpu->cpu); +} +#else /* !CONFIG_ENERGY_MODEL */ +static inline bool hybrid_register_perf_domain(unsigned int cpu) { return false; } +static inline void hybrid_register_all_perf_domains(void) {} +static inline void hybrid_update_perf_domain(struct cpudata *cpu) {} +#endif /* CONFIG_ENERGY_MODEL */ + static void hybrid_set_cpu_capacity(struct cpudata *cpu) { arch_set_cpu_capacity(cpu->cpu, cpu->capacity_perf, hybrid_max_perf_cpu->capacity_perf, cpu->capacity_perf, cpu->pstate.max_pstate_physical); + hybrid_update_perf_domain(cpu); + + topology_set_cpu_scale(cpu->cpu, arch_scale_cpu_capacity(cpu->cpu)); pr_debug("CPU%d: perf = %u, max. perf = %u, base perf = %d\n", cpu->cpu, cpu->capacity_perf, hybrid_max_perf_cpu->capacity_perf, @@ -1042,6 +1159,11 @@ static void hybrid_refresh_cpu_capacity_scaling(void) guard(mutex)(&hybrid_capacity_lock); __hybrid_refresh_cpu_capacity_scaling(); + /* + * Perf domains are not registered before setting hybrid_max_perf_cpu, + * so register them all after setting up CPU capacity scaling. + */ + hybrid_register_all_perf_domains(); } static void hybrid_init_cpu_capacity_scaling(bool refresh) @@ -1069,7 +1191,7 @@ static void hybrid_init_cpu_capacity_scaling(bool refresh) hybrid_refresh_cpu_capacity_scaling(); /* * Disabling ITMT causes sched domains to be rebuilt to disable asym - * packing and enable asym capacity. + * packing and enable asym capacity and EAS. */ sched_clear_itmt_support(); } @@ -1147,6 +1269,14 @@ static void hybrid_update_capacity(struct cpudata *cpu) } hybrid_set_cpu_capacity(cpu); + /* + * If the CPU was offline to start with and it is going online for the + * first time, a perf domain needs to be registered for it if hybrid + * capacity scaling has been enabled already. In that case, sched + * domains need to be rebuilt to take the new perf domain into account. + */ + if (hybrid_register_perf_domain(cpu->cpu)) + em_rebuild_sched_domains(); unlock: mutex_unlock(&hybrid_capacity_lock); @@ -1356,9 +1486,11 @@ static void intel_pstate_update_policies(void) cpufreq_update_policy(cpu); } -static void __intel_pstate_update_max_freq(struct cpudata *cpudata, - struct cpufreq_policy *policy) +static void __intel_pstate_update_max_freq(struct cpufreq_policy *policy, + struct cpudata *cpudata) { + guard(cpufreq_policy_write)(policy); + if (hwp_active) intel_pstate_get_hwp_cap(cpudata); @@ -1368,42 +1500,34 @@ static void __intel_pstate_update_max_freq(struct cpudata *cpudata, refresh_frequency_limits(policy); } -static void intel_pstate_update_limits(unsigned int cpu) +static bool intel_pstate_update_max_freq(struct cpudata *cpudata) { - struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpu); - struct cpudata *cpudata; + struct cpufreq_policy *policy __free(put_cpufreq_policy); + policy = cpufreq_cpu_get(cpudata->cpu); if (!policy) - return; + return false; - cpudata = all_cpu_data[cpu]; + __intel_pstate_update_max_freq(policy, cpudata); - __intel_pstate_update_max_freq(cpudata, policy); + return true; +} - /* Prevent the driver from being unregistered now. */ - mutex_lock(&intel_pstate_driver_lock); +static void intel_pstate_update_limits(struct cpufreq_policy *policy) +{ + struct cpudata *cpudata = all_cpu_data[policy->cpu]; - cpufreq_cpu_release(policy); + __intel_pstate_update_max_freq(policy, cpudata); hybrid_update_capacity(cpudata); - - mutex_unlock(&intel_pstate_driver_lock); } static void intel_pstate_update_limits_for_all(void) { int cpu; - for_each_possible_cpu(cpu) { - struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpu); - - if (!policy) - continue; - - __intel_pstate_update_max_freq(all_cpu_data[cpu], policy); - - cpufreq_cpu_release(policy); - } + for_each_possible_cpu(cpu) + intel_pstate_update_max_freq(all_cpu_data[cpu]); mutex_lock(&hybrid_capacity_lock); @@ -1843,13 +1967,8 @@ static void intel_pstate_notify_work(struct work_struct *work) { struct cpudata *cpudata = container_of(to_delayed_work(work), struct cpudata, hwp_notify_work); - struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpudata->cpu); - - if (policy) { - __intel_pstate_update_max_freq(cpudata, policy); - - cpufreq_cpu_release(policy); + if (intel_pstate_update_max_freq(cpudata)) { /* * The driver will not be unregistered while this function is * running, so update the capacity without acquiring the driver diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h index 2222e8b03ff4..d72d6e5aa200 100644 --- a/include/linux/arch_topology.h +++ b/include/linux/arch_topology.h @@ -14,14 +14,6 @@ int topology_update_cpu_topology(void); struct device_node; bool topology_parse_cpu_capacity(struct device_node *cpu_node, int cpu); -DECLARE_PER_CPU(unsigned long, cpu_scale); - -static inline unsigned long topology_get_cpu_scale(int cpu) -{ - return per_cpu(cpu_scale, cpu); -} - -void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity); DECLARE_PER_CPU(unsigned long, capacity_freq_ref); diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 7a5b391dcc01..95f3807c8c55 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -170,6 +170,12 @@ struct cpufreq_policy { struct notifier_block nb_max; }; +DEFINE_GUARD(cpufreq_policy_write, struct cpufreq_policy *, + down_write(&_T->rwsem), up_write(&_T->rwsem)) + +DEFINE_GUARD(cpufreq_policy_read, struct cpufreq_policy *, + down_read(&_T->rwsem), up_read(&_T->rwsem)) + /* * Used for passing new cpufreq policy data to the cpufreq driver's ->verify() * callback for sanitization. That callback is only expected to modify the min @@ -235,9 +241,6 @@ void disable_cpufreq(void); u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy); -struct cpufreq_policy *cpufreq_cpu_acquire(unsigned int cpu); -void cpufreq_cpu_release(struct cpufreq_policy *policy); -int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu); void refresh_frequency_limits(struct cpufreq_policy *policy); void cpufreq_update_policy(unsigned int cpu); void cpufreq_update_limits(unsigned int cpu); @@ -395,7 +398,7 @@ struct cpufreq_driver { unsigned int (*get)(unsigned int cpu); /* Called to update policy limits on firmware notifications. */ - void (*update_limits)(unsigned int cpu); + void (*update_limits)(struct cpufreq_policy *policy); /* optional */ int (*bios_limit)(int cpu, unsigned int *limit); @@ -647,6 +650,15 @@ module_exit(__governor##_exit) struct cpufreq_governor *cpufreq_default_governor(void); struct cpufreq_governor *cpufreq_fallback_governor(void); +#ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL +bool sugov_is_governor(struct cpufreq_policy *policy); +#else +static inline bool sugov_is_governor(struct cpufreq_policy *policy) +{ + return false; +} +#endif + static inline void cpufreq_policy_apply_limits(struct cpufreq_policy *policy) { if (policy->max < policy->cur) @@ -1225,6 +1237,8 @@ void cpufreq_generic_init(struct cpufreq_policy *policy, struct cpufreq_frequency_table *table, unsigned int transition_latency); +bool cpufreq_ready_for_eas(const struct cpumask *cpu_mask); + static inline void cpufreq_register_em_with_opp(struct cpufreq_policy *policy) { dev_pm_opp_of_register_em(get_cpu_device(policy->cpu), diff --git a/include/linux/topology.h b/include/linux/topology.h index 24e715f0f6d2..cd6b4bdc9cfd 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -332,4 +332,13 @@ sched_numa_hop_mask(unsigned int node, unsigned int hops) !IS_ERR_OR_NULL(mask); \ __hops++) +DECLARE_PER_CPU(unsigned long, cpu_scale); + +static inline unsigned long topology_get_cpu_scale(int cpu) +{ + return per_cpu(cpu_scale, cpu); +} + +void topology_set_cpu_scale(unsigned int cpu, unsigned long capacity); + #endif /* _LINUX_TOPOLOGY_H */ diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 816f07f9d30f..461242ec958a 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -630,7 +630,7 @@ static const struct kobj_type sugov_tunables_ktype = { /********************** cpufreq governor interface *********************/ -struct cpufreq_governor schedutil_gov; +static struct cpufreq_governor schedutil_gov; static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy) { @@ -909,7 +909,7 @@ static void sugov_limits(struct cpufreq_policy *policy) WRITE_ONCE(sg_policy->limits_changed, true); } -struct cpufreq_governor schedutil_gov = { +static struct cpufreq_governor schedutil_gov = { .name = "schedutil", .owner = THIS_MODULE, .flags = CPUFREQ_GOV_DYNAMIC_SWITCHING, @@ -927,4 +927,9 @@ struct cpufreq_governor *cpufreq_default_governor(void) } #endif +bool sugov_is_governor(struct cpufreq_policy *policy) +{ + return policy->governor == &schedutil_gov; +} + cpufreq_governor_init(schedutil_gov); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 47972f34ea70..5cbe3fd93fda 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -3509,8 +3509,6 @@ static inline bool sched_energy_enabled(void) return static_branch_unlikely(&sched_energy_present); } -extern struct cpufreq_governor schedutil_gov; - #else /* ! (CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */ #define perf_domain_span(pd) NULL diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index f1ebc60d967f..580c28fed539 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -212,8 +212,6 @@ static bool sched_energy_update; static bool sched_is_eas_possible(const struct cpumask *cpu_mask) { bool any_asym_capacity = false; - struct cpufreq_policy *policy; - struct cpufreq_governor *gov; int i; /* EAS is enabled for asymmetric CPU capacity topologies. */ @@ -248,25 +246,12 @@ static bool sched_is_eas_possible(const struct cpumask *cpu_mask) return false; } - /* Do not attempt EAS if schedutil is not being used. */ - for_each_cpu(i, cpu_mask) { - policy = cpufreq_cpu_get(i); - if (!policy) { - if (sched_debug()) { - pr_info("rd %*pbl: Checking EAS, cpufreq policy not set for CPU: %d", - cpumask_pr_args(cpu_mask), i); - } - return false; - } - gov = policy->governor; - cpufreq_cpu_put(policy); - if (gov != &schedutil_gov) { - if (sched_debug()) { - pr_info("rd %*pbl: Checking EAS, schedutil is mandatory\n", - cpumask_pr_args(cpu_mask)); - } - return false; + if (!cpufreq_ready_for_eas(cpu_mask)) { + if (sched_debug()) { + pr_info("rd %*pbl: Checking EAS: cpufreq is not ready\n", + cpumask_pr_args(cpu_mask)); } + return false; } return true; |