diff options
Diffstat (limited to 'drivers/cpufreq/cpufreq.c')
| -rw-r--r-- | drivers/cpufreq/cpufreq.c | 3246 |
1 files changed, 2120 insertions, 1126 deletions
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 0937b8d6c2a4..4472bb1ec83c 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * linux/drivers/cpufreq/cpufreq.c * @@ -9,35 +10,48 @@ * Added handling for CPU hotplug * Feb 2006 - Jacob Shin <jacob.shin@amd.com> * Fix handling for CPU hotplug -- affected CPUs - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#include <asm/cputime.h> -#include <linux/kernel.h> -#include <linux/kernel_stat.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/notifier.h> +#include <linux/cpu.h> #include <linux/cpufreq.h> +#include <linux/cpu_cooling.h> #include <linux/delay.h> -#include <linux/interrupt.h> -#include <linux/spinlock.h> -#include <linux/tick.h> #include <linux/device.h> -#include <linux/slab.h> -#include <linux/cpu.h> -#include <linux/completion.h> +#include <linux/init.h> +#include <linux/kernel_stat.h> +#include <linux/module.h> #include <linux/mutex.h> +#include <linux/pm_qos.h> +#include <linux/slab.h> +#include <linux/string_choices.h> +#include <linux/suspend.h> #include <linux/syscore_ops.h> - +#include <linux/tick.h> +#include <linux/units.h> #include <trace/events/power.h> -/** +static LIST_HEAD(cpufreq_policy_list); + +/* Macros to iterate over CPU policies */ +#define for_each_suitable_policy(__policy, __active) \ + list_for_each_entry(__policy, &cpufreq_policy_list, policy_list) \ + if ((__active) == !policy_is_inactive(__policy)) + +#define for_each_active_policy(__policy) \ + for_each_suitable_policy(__policy, true) +#define for_each_inactive_policy(__policy) \ + for_each_suitable_policy(__policy, false) + +/* Iterate over governors */ +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 * also protects the cpufreq_cpu_data array. @@ -45,64 +59,38 @@ static struct cpufreq_driver *cpufreq_driver; static DEFINE_PER_CPU(struct cpufreq_policy *, cpufreq_cpu_data); static DEFINE_RWLOCK(cpufreq_driver_lock); -static DEFINE_MUTEX(cpufreq_governor_lock); -#ifdef CONFIG_HOTPLUG_CPU -/* This one keeps track of the previously set governor of a removed CPU */ -static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor); -#endif - -/* - * cpu_policy_rwsem is a per CPU reader-writer semaphore designed to cure - * all cpufreq/hotplug/workqueue/etc related lock issues. - * - * The rules for this semaphore: - * - Any routine that wants to read from the policy structure will - * do a down_read on this semaphore. - * - Any routine that will write to the policy structure and/or may take away - * the policy altogether (eg. CPU hotplug), will hold this lock in write - * mode before doing so. - * - * Additional rules: - * - Governor routines that can be called in cpufreq hotplug path should not - * take this sem as top level hotplug notifier handler takes this. - * - Lock should not be held across - * __cpufreq_governor(data, CPUFREQ_GOV_STOP); - */ -static DEFINE_PER_CPU(int, cpufreq_policy_cpu); -static DEFINE_PER_CPU(struct rw_semaphore, cpu_policy_rwsem); - -#define lock_policy_rwsem(mode, cpu) \ -static int lock_policy_rwsem_##mode(int cpu) \ -{ \ - int policy_cpu = per_cpu(cpufreq_policy_cpu, cpu); \ - BUG_ON(policy_cpu == -1); \ - down_##mode(&per_cpu(cpu_policy_rwsem, policy_cpu)); \ - \ - return 0; \ +static DEFINE_STATIC_KEY_FALSE(cpufreq_freq_invariance); +bool cpufreq_supports_freq_invariance(void) +{ + return static_branch_likely(&cpufreq_freq_invariance); } -lock_policy_rwsem(read, cpu); -lock_policy_rwsem(write, cpu); +/* Flag to suspend/resume CPUFreq governors */ +static bool cpufreq_suspended; -#define unlock_policy_rwsem(mode, cpu) \ -static void unlock_policy_rwsem_##mode(int cpu) \ -{ \ - int policy_cpu = per_cpu(cpufreq_policy_cpu, cpu); \ - BUG_ON(policy_cpu == -1); \ - up_##mode(&per_cpu(cpu_policy_rwsem, policy_cpu)); \ +static inline bool has_target(void) +{ + return cpufreq_driver->target_index || cpufreq_driver->target; } -unlock_policy_rwsem(read, cpu); -unlock_policy_rwsem(write, cpu); +bool has_target_index(void) +{ + return !!cpufreq_driver->target_index; +} /* internal prototypes */ -static int __cpufreq_governor(struct cpufreq_policy *policy, - unsigned int event); -static unsigned int __cpufreq_get(unsigned int cpu); -static void handle_update(struct work_struct *work); +static unsigned int __cpufreq_get(struct cpufreq_policy *policy); +static int cpufreq_init_governor(struct cpufreq_policy *policy); +static void cpufreq_exit_governor(struct cpufreq_policy *policy); +static void cpufreq_governor_limits(struct cpufreq_policy *policy); +static int cpufreq_set_policy(struct cpufreq_policy *policy, + struct cpufreq_governor *new_gov, + unsigned int new_pol); +static bool cpufreq_boost_supported(void); +static int cpufreq_boost_trigger_state(int state); -/** +/* * Two notifier lists: the "policy" list is involved in the * validation process for a new CPU frequency policy; the * "transition" list for kernel code that needs to handle @@ -110,16 +98,7 @@ static void handle_update(struct work_struct *work); * The mutex locks both lists. */ static BLOCKING_NOTIFIER_HEAD(cpufreq_policy_notifier_list); -static struct srcu_notifier_head cpufreq_transition_notifier_list; - -static bool init_cpufreq_transition_notifier_list_called; -static int __init init_cpufreq_transition_notifier_list(void) -{ - srcu_init_notifier_head(&cpufreq_transition_notifier_list); - init_cpufreq_transition_notifier_list_called = true; - return 0; -} -pure_initcall(init_cpufreq_transition_notifier_list); +SRCU_NOTIFIER_HEAD_STATIC(cpufreq_transition_notifier_list); static int off __read_mostly; static int cpufreq_disabled(void) @@ -130,15 +109,18 @@ void disable_cpufreq(void) { off = 1; } -static LIST_HEAD(cpufreq_governor_list); +EXPORT_SYMBOL_GPL(disable_cpufreq); + static DEFINE_MUTEX(cpufreq_governor_mutex); bool have_governor_per_policy(void) { - return cpufreq_driver->have_governor_per_policy; + return !!(cpufreq_driver->flags & CPUFREQ_HAVE_GOVERNOR_PER_POLICY); } EXPORT_SYMBOL_GPL(have_governor_per_policy); +static struct kobject *cpufreq_global_kobject; + struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy) { if (have_governor_per_policy()) @@ -150,24 +132,27 @@ EXPORT_SYMBOL_GPL(get_governor_parent_kobj); static inline u64 get_cpu_idle_time_jiffy(unsigned int cpu, u64 *wall) { - u64 idle_time; + struct kernel_cpustat kcpustat; u64 cur_wall_time; + u64 idle_time; u64 busy_time; - cur_wall_time = jiffies64_to_cputime64(get_jiffies_64()); + cur_wall_time = jiffies64_to_nsecs(get_jiffies_64()); - busy_time = kcpustat_cpu(cpu).cpustat[CPUTIME_USER]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SYSTEM]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_IRQ]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SOFTIRQ]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_STEAL]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_NICE]; + kcpustat_cpu_fetch(&kcpustat, cpu); + + busy_time = kcpustat.cpustat[CPUTIME_USER]; + busy_time += kcpustat.cpustat[CPUTIME_SYSTEM]; + busy_time += kcpustat.cpustat[CPUTIME_IRQ]; + busy_time += kcpustat.cpustat[CPUTIME_SOFTIRQ]; + busy_time += kcpustat.cpustat[CPUTIME_STEAL]; + busy_time += kcpustat.cpustat[CPUTIME_NICE]; idle_time = cur_wall_time - busy_time; if (wall) - *wall = cputime_to_usecs(cur_wall_time); + *wall = div_u64(cur_wall_time, NSEC_PER_USEC); - return cputime_to_usecs(idle_time); + return div_u64(idle_time, NSEC_PER_USEC); } u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy) @@ -183,261 +168,530 @@ u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy) } EXPORT_SYMBOL_GPL(get_cpu_idle_time); -static struct cpufreq_policy *__cpufreq_cpu_get(unsigned int cpu, bool sysfs) +/* + * This is a generic cpufreq init() routine which can be used by cpufreq + * drivers of SMP systems. It will do following: + * - validate & show freq table passed + * - set policies transition latency + * - policy->cpus with all possible CPUs + */ +void cpufreq_generic_init(struct cpufreq_policy *policy, + struct cpufreq_frequency_table *table, + unsigned int transition_latency) { - struct cpufreq_policy *data; - unsigned long flags; + policy->freq_table = table; + policy->cpuinfo.transition_latency = transition_latency; - if (cpu >= nr_cpu_ids) - goto err_out; - - /* get the cpufreq driver */ - read_lock_irqsave(&cpufreq_driver_lock, flags); - - if (!cpufreq_driver) - goto err_out_unlock; - - if (!try_module_get(cpufreq_driver->owner)) - goto err_out_unlock; + /* + * The driver only supports the SMP configuration where all processors + * share the clock and voltage and clock. + */ + cpumask_setall(policy->cpus); +} +EXPORT_SYMBOL_GPL(cpufreq_generic_init); - /* get the CPU */ - data = per_cpu(cpufreq_cpu_data, cpu); +struct cpufreq_policy *cpufreq_cpu_get_raw(unsigned int cpu) +{ + struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); - if (!data) - goto err_out_put_module; + return policy && cpumask_test_cpu(cpu, policy->cpus) ? policy : NULL; +} +EXPORT_SYMBOL_GPL(cpufreq_cpu_get_raw); - if (!sysfs && !kobject_get(&data->kobj)) - goto err_out_put_module; +unsigned int cpufreq_generic_get(unsigned int cpu) +{ + struct cpufreq_policy *policy = cpufreq_cpu_get_raw(cpu); - read_unlock_irqrestore(&cpufreq_driver_lock, flags); - return data; + if (!policy || IS_ERR(policy->clk)) { + pr_err("%s: No %s associated to cpu: %d\n", + __func__, policy ? "clk" : "policy", cpu); + return 0; + } -err_out_put_module: - module_put(cpufreq_driver->owner); -err_out_unlock: - read_unlock_irqrestore(&cpufreq_driver_lock, flags); -err_out: - return NULL; + return clk_get_rate(policy->clk) / 1000; } +EXPORT_SYMBOL_GPL(cpufreq_generic_get); +/** + * cpufreq_cpu_get - Return policy for a CPU and mark it as busy. + * @cpu: CPU to find the policy for. + * + * Call cpufreq_cpu_get_raw() to obtain a cpufreq policy for @cpu and increment + * the kobject reference counter of that policy. Return a valid policy on + * success or NULL on failure. + * + * The policy returned by this function has to be released with the help of + * cpufreq_cpu_put() to balance its kobject reference counter properly. + */ struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu) { - if (cpufreq_disabled()) - return NULL; + struct cpufreq_policy *policy = NULL; + unsigned long flags; - return __cpufreq_cpu_get(cpu, false); -} -EXPORT_SYMBOL_GPL(cpufreq_cpu_get); + if (WARN_ON(cpu >= nr_cpu_ids)) + return NULL; -static struct cpufreq_policy *cpufreq_cpu_get_sysfs(unsigned int cpu) -{ - return __cpufreq_cpu_get(cpu, true); -} + /* get the cpufreq driver */ + read_lock_irqsave(&cpufreq_driver_lock, flags); -static void __cpufreq_cpu_put(struct cpufreq_policy *data, bool sysfs) -{ - if (!sysfs) - kobject_put(&data->kobj); - module_put(cpufreq_driver->owner); -} + if (cpufreq_driver) { + /* get the CPU */ + policy = cpufreq_cpu_get_raw(cpu); + if (policy) + kobject_get(&policy->kobj); + } -void cpufreq_cpu_put(struct cpufreq_policy *data) -{ - if (cpufreq_disabled()) - return; + read_unlock_irqrestore(&cpufreq_driver_lock, flags); - __cpufreq_cpu_put(data, false); + return policy; } -EXPORT_SYMBOL_GPL(cpufreq_cpu_put); +EXPORT_SYMBOL_GPL(cpufreq_cpu_get); -static void cpufreq_cpu_put_sysfs(struct cpufreq_policy *data) +/** + * cpufreq_cpu_put - Decrement kobject usage counter for cpufreq policy. + * @policy: cpufreq policy returned by cpufreq_cpu_get(). + */ +void cpufreq_cpu_put(struct cpufreq_policy *policy) { - __cpufreq_cpu_put(data, true); + kobject_put(&policy->kobj); } +EXPORT_SYMBOL_GPL(cpufreq_cpu_put); /********************************************************************* * EXTERNALLY AFFECTING FREQUENCY CHANGES * *********************************************************************/ /** - * adjust_jiffies - adjust the system "loops_per_jiffy" + * adjust_jiffies - Adjust the system "loops_per_jiffy". + * @val: CPUFREQ_PRECHANGE or CPUFREQ_POSTCHANGE. + * @ci: Frequency change information. * * This function alters the system "loops_per_jiffy" for the clock * speed change. Note that loops_per_jiffy cannot be updated on SMP * systems as each CPU might be scaled differently. So, use the arch * per-CPU loops_per_jiffy value wherever possible. */ -#ifndef CONFIG_SMP -static unsigned long l_p_j_ref; -static unsigned int l_p_j_ref_freq; - static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) { +#ifndef CONFIG_SMP + static unsigned long l_p_j_ref; + static unsigned int l_p_j_ref_freq; + if (ci->flags & CPUFREQ_CONST_LOOPS) return; if (!l_p_j_ref_freq) { l_p_j_ref = loops_per_jiffy; l_p_j_ref_freq = ci->old; - pr_debug("saving %lu as reference value for loops_per_jiffy; " - "freq is %u kHz\n", l_p_j_ref, l_p_j_ref_freq); + pr_debug("saving %lu as reference value for loops_per_jiffy; freq is %u kHz\n", + l_p_j_ref, l_p_j_ref_freq); } - if ((val == CPUFREQ_POSTCHANGE && ci->old != ci->new) || - (val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) { + if (val == CPUFREQ_POSTCHANGE && ci->old != ci->new) { loops_per_jiffy = cpufreq_scale(l_p_j_ref, l_p_j_ref_freq, ci->new); - pr_debug("scaling loops_per_jiffy to %lu " - "for frequency %u kHz\n", loops_per_jiffy, ci->new); + pr_debug("scaling loops_per_jiffy to %lu for frequency %u kHz\n", + loops_per_jiffy, ci->new); } -} -#else -static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) -{ - return; -} #endif +} -static void __cpufreq_notify_transition(struct cpufreq_policy *policy, - struct cpufreq_freqs *freqs, unsigned int state) +/** + * cpufreq_notify_transition - Notify frequency transition and adjust jiffies. + * @policy: cpufreq policy to enable fast frequency switching for. + * @freqs: contain details of the frequency update. + * @state: set to CPUFREQ_PRECHANGE or CPUFREQ_POSTCHANGE. + * + * This function calls the transition notifiers and adjust_jiffies(). + * + * It is called twice on all CPU frequency changes that have external effects. + */ +static void cpufreq_notify_transition(struct cpufreq_policy *policy, + struct cpufreq_freqs *freqs, + unsigned int state) { + int cpu; + BUG_ON(irqs_disabled()); if (cpufreq_disabled()) return; + freqs->policy = policy; freqs->flags = cpufreq_driver->flags; pr_debug("notification %u of frequency transition to %u kHz\n", - state, freqs->new); + state, freqs->new); switch (state) { - case CPUFREQ_PRECHANGE: - if (WARN(policy->transition_ongoing == - cpumask_weight(policy->cpus), - "In middle of another frequency transition\n")) - return; - - policy->transition_ongoing++; - - /* detect if the driver reported a value as "old frequency" + /* + * Detect if the driver reported a value as "old frequency" * which is not equal to what the cpufreq core thinks is * "old frequency". */ - if (!(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) { - if ((policy) && (policy->cpu == freqs->cpu) && - (policy->cur) && (policy->cur != freqs->old)) { - pr_debug("Warning: CPU frequency is" - " %u, cpufreq assumed %u kHz.\n", - freqs->old, policy->cur); - freqs->old = policy->cur; - } + if (policy->cur && policy->cur != freqs->old) { + pr_debug("Warning: CPU frequency is %u, cpufreq assumed %u kHz\n", + freqs->old, policy->cur); + freqs->old = policy->cur; } + srcu_notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_PRECHANGE, freqs); + CPUFREQ_PRECHANGE, freqs); + adjust_jiffies(CPUFREQ_PRECHANGE, freqs); break; case CPUFREQ_POSTCHANGE: - if (WARN(!policy->transition_ongoing, - "No frequency transition in progress\n")) - return; + adjust_jiffies(CPUFREQ_POSTCHANGE, freqs); + pr_debug("FREQ: %u - CPUs: %*pbl\n", freqs->new, + cpumask_pr_args(policy->cpus)); - policy->transition_ongoing--; + for_each_cpu(cpu, policy->cpus) + trace_cpu_frequency(freqs->new, cpu); - adjust_jiffies(CPUFREQ_POSTCHANGE, freqs); - pr_debug("FREQ: %lu - CPU: %lu", (unsigned long)freqs->new, - (unsigned long)freqs->cpu); - trace_cpu_frequency(freqs->new, freqs->cpu); srcu_notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_POSTCHANGE, freqs); - if (likely(policy) && likely(policy->cpu == freqs->cpu)) - policy->cur = freqs->new; - break; + CPUFREQ_POSTCHANGE, freqs); + + cpufreq_stats_record_transition(policy, freqs->new); + policy->cur = freqs->new; + } +} + +/* Do post notifications when there are chances that transition has failed */ +static void cpufreq_notify_post_transition(struct cpufreq_policy *policy, + struct cpufreq_freqs *freqs, int transition_failed) +{ + cpufreq_notify_transition(policy, freqs, CPUFREQ_POSTCHANGE); + if (!transition_failed) + return; + + swap(freqs->old, freqs->new); + cpufreq_notify_transition(policy, freqs, CPUFREQ_PRECHANGE); + cpufreq_notify_transition(policy, freqs, CPUFREQ_POSTCHANGE); +} + +void cpufreq_freq_transition_begin(struct cpufreq_policy *policy, + struct cpufreq_freqs *freqs) +{ + + /* + * Catch double invocations of _begin() which lead to self-deadlock. + * ASYNC_NOTIFICATION drivers are left out because the cpufreq core + * doesn't invoke _begin() on their behalf, and hence the chances of + * double invocations are very low. Moreover, there are scenarios + * where these checks can emit false-positive warnings in these + * drivers; so we avoid that by skipping them altogether. + */ + WARN_ON(!(cpufreq_driver->flags & CPUFREQ_ASYNC_NOTIFICATION) + && current == policy->transition_task); + +wait: + wait_event(policy->transition_wait, !policy->transition_ongoing); + + spin_lock(&policy->transition_lock); + + if (unlikely(policy->transition_ongoing)) { + spin_unlock(&policy->transition_lock); + goto wait; + } + + policy->transition_ongoing = true; + policy->transition_task = current; + + spin_unlock(&policy->transition_lock); + + cpufreq_notify_transition(policy, freqs, CPUFREQ_PRECHANGE); +} +EXPORT_SYMBOL_GPL(cpufreq_freq_transition_begin); + +void cpufreq_freq_transition_end(struct cpufreq_policy *policy, + struct cpufreq_freqs *freqs, int transition_failed) +{ + if (WARN_ON(!policy->transition_ongoing)) + return; + + cpufreq_notify_post_transition(policy, freqs, transition_failed); + + arch_set_freq_scale(policy->related_cpus, + policy->cur, + arch_scale_freq_ref(policy->cpu)); + + spin_lock(&policy->transition_lock); + policy->transition_ongoing = false; + policy->transition_task = NULL; + spin_unlock(&policy->transition_lock); + + wake_up(&policy->transition_wait); +} +EXPORT_SYMBOL_GPL(cpufreq_freq_transition_end); + +/* + * Fast frequency switching status count. Positive means "enabled", negative + * means "disabled" and 0 means "not decided yet". + */ +static int cpufreq_fast_switch_count; +static DEFINE_MUTEX(cpufreq_fast_switch_lock); + +static void cpufreq_list_transition_notifiers(void) +{ + struct notifier_block *nb; + + pr_info("Registered transition notifiers:\n"); + + mutex_lock(&cpufreq_transition_notifier_list.mutex); + + for (nb = cpufreq_transition_notifier_list.head; nb; nb = nb->next) + pr_info("%pS\n", nb->notifier_call); + + mutex_unlock(&cpufreq_transition_notifier_list.mutex); +} + +/** + * cpufreq_enable_fast_switch - Enable fast frequency switching for policy. + * @policy: cpufreq policy to enable fast frequency switching for. + * + * Try to enable fast frequency switching for @policy. + * + * The attempt will fail if there is at least one transition notifier registered + * at this point, as fast frequency switching is quite fundamentally at odds + * with transition notifiers. Thus if successful, it will make registration of + * transition notifiers fail going forward. + */ +void cpufreq_enable_fast_switch(struct cpufreq_policy *policy) +{ + lockdep_assert_held(&policy->rwsem); + + if (!policy->fast_switch_possible) + return; + + mutex_lock(&cpufreq_fast_switch_lock); + if (cpufreq_fast_switch_count >= 0) { + cpufreq_fast_switch_count++; + policy->fast_switch_enabled = true; + } else { + pr_warn("CPU%u: Fast frequency switching not enabled\n", + policy->cpu); + cpufreq_list_transition_notifiers(); + } + mutex_unlock(&cpufreq_fast_switch_lock); +} +EXPORT_SYMBOL_GPL(cpufreq_enable_fast_switch); + +/** + * cpufreq_disable_fast_switch - Disable fast frequency switching for policy. + * @policy: cpufreq policy to disable fast frequency switching for. + */ +void cpufreq_disable_fast_switch(struct cpufreq_policy *policy) +{ + mutex_lock(&cpufreq_fast_switch_lock); + if (policy->fast_switch_enabled) { + policy->fast_switch_enabled = false; + if (!WARN_ON(cpufreq_fast_switch_count <= 0)) + cpufreq_fast_switch_count--; } + mutex_unlock(&cpufreq_fast_switch_lock); +} +EXPORT_SYMBOL_GPL(cpufreq_disable_fast_switch); + +static unsigned int __resolve_freq(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int min, unsigned int max, + unsigned int relation) +{ + unsigned int idx; + + target_freq = clamp_val(target_freq, min, max); + + if (!policy->freq_table) + return target_freq; + + idx = cpufreq_frequency_table_target(policy, target_freq, min, max, relation); + policy->cached_resolved_idx = idx; + policy->cached_target_freq = target_freq; + return policy->freq_table[idx].frequency; } /** - * cpufreq_notify_transition - call notifier chain and adjust_jiffies - * on frequency transition. + * cpufreq_driver_resolve_freq - Map a target frequency to a driver-supported + * one. + * @policy: associated policy to interrogate + * @target_freq: target frequency to resolve. + * + * The target to driver frequency mapping is cached in the policy. * - * This function calls the transition notifiers and the "adjust_jiffies" - * function. It is called twice on all CPU frequency changes that have - * external effects. + * Return: Lowest driver-supported frequency greater than or equal to the + * given target_freq, subject to policy (min/max) and driver limitations. */ -void cpufreq_notify_transition(struct cpufreq_policy *policy, - struct cpufreq_freqs *freqs, unsigned int state) +unsigned int cpufreq_driver_resolve_freq(struct cpufreq_policy *policy, + unsigned int target_freq) { - for_each_cpu(freqs->cpu, policy->cpus) - __cpufreq_notify_transition(policy, freqs, state); + unsigned int min = READ_ONCE(policy->min); + unsigned int max = READ_ONCE(policy->max); + + /* + * If this function runs in parallel with cpufreq_set_policy(), it may + * read policy->min before the update and policy->max after the update + * or the other way around, so there is no ordering guarantee. + * + * Resolve this by always honoring the max (in case it comes from + * thermal throttling or similar). + */ + if (unlikely(min > max)) + min = max; + + return __resolve_freq(policy, target_freq, min, max, CPUFREQ_RELATION_LE); } -EXPORT_SYMBOL_GPL(cpufreq_notify_transition); +EXPORT_SYMBOL_GPL(cpufreq_driver_resolve_freq); +unsigned int cpufreq_policy_transition_delay_us(struct cpufreq_policy *policy) +{ + unsigned int latency; + + if (policy->transition_delay_us) + return policy->transition_delay_us; + + latency = policy->cpuinfo.transition_latency / NSEC_PER_USEC; + if (latency) + /* Give a 50% breathing room between updates */ + return latency + (latency >> 1); + + return USEC_PER_MSEC; +} +EXPORT_SYMBOL_GPL(cpufreq_policy_transition_delay_us); /********************************************************************* * SYSFS INTERFACE * *********************************************************************/ +static ssize_t show_boost(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sysfs_emit(buf, "%d\n", cpufreq_driver->boost_enabled); +} + +static ssize_t store_boost(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + bool enable; + + if (kstrtobool(buf, &enable)) + return -EINVAL; + + if (cpufreq_boost_trigger_state(enable)) { + pr_err("%s: Cannot %s BOOST!\n", + __func__, str_enable_disable(enable)); + return -EINVAL; + } + + pr_debug("%s: cpufreq BOOST %s\n", + __func__, str_enabled_disabled(enable)); + + return count; +} +define_one_global_rw(boost); + +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 struct cpufreq_governor *__find_governor(const char *str_governor) +static ssize_t store_local_boost(struct cpufreq_policy *policy, + const char *buf, size_t count) +{ + int ret; + bool enable; + + if (kstrtobool(buf, &enable)) + return -EINVAL; + + if (!cpufreq_driver->boost_enabled) + return -EINVAL; + + if (!policy->boost_supported) + return -EINVAL; + + ret = policy_set_boost(policy, enable); + if (!ret) + return count; + + return ret; +} + +static struct freq_attr local_boost = __ATTR(boost, 0644, show_local_boost, store_local_boost); + +static struct cpufreq_governor *find_governor(const char *str_governor) { struct cpufreq_governor *t; - list_for_each_entry(t, &cpufreq_governor_list, governor_list) - if (!strnicmp(str_governor, t->name, CPUFREQ_NAME_LEN)) + for_each_governor(t) + if (!strncasecmp(str_governor, t->name, CPUFREQ_NAME_LEN)) return t; return NULL; } -/** - * cpufreq_parse_governor - parse a governor string - */ -static int cpufreq_parse_governor(char *str_governor, unsigned int *policy, - struct cpufreq_governor **governor) +static struct cpufreq_governor *get_governor(const char *str_governor) { - int err = -EINVAL; + struct cpufreq_governor *t; - if (!cpufreq_driver) - goto out; + mutex_lock(&cpufreq_governor_mutex); + t = find_governor(str_governor); + if (!t) + goto unlock; - if (cpufreq_driver->setpolicy) { - if (!strnicmp(str_governor, "performance", CPUFREQ_NAME_LEN)) { - *policy = CPUFREQ_POLICY_PERFORMANCE; - err = 0; - } else if (!strnicmp(str_governor, "powersave", - CPUFREQ_NAME_LEN)) { - *policy = CPUFREQ_POLICY_POWERSAVE; - err = 0; - } - } else if (cpufreq_driver->target) { - struct cpufreq_governor *t; + if (!try_module_get(t->owner)) + t = NULL; + +unlock: + mutex_unlock(&cpufreq_governor_mutex); - mutex_lock(&cpufreq_governor_mutex); + return t; +} - t = __find_governor(str_governor); +static unsigned int cpufreq_parse_policy(char *str_governor) +{ + if (!strncasecmp(str_governor, "performance", strlen("performance"))) + return CPUFREQ_POLICY_PERFORMANCE; - if (t == NULL) { - int ret; + if (!strncasecmp(str_governor, "powersave", strlen("powersave"))) + return CPUFREQ_POLICY_POWERSAVE; - mutex_unlock(&cpufreq_governor_mutex); - ret = request_module("cpufreq_%s", str_governor); - mutex_lock(&cpufreq_governor_mutex); + return CPUFREQ_POLICY_UNKNOWN; +} - if (ret == 0) - t = __find_governor(str_governor); - } +/** + * cpufreq_parse_governor - parse a governor string only for has_target() + * @str_governor: Governor name. + */ +static struct cpufreq_governor *cpufreq_parse_governor(char *str_governor) +{ + struct cpufreq_governor *t; - if (t != NULL) { - *governor = t; - err = 0; - } + t = get_governor(str_governor); + if (t) + return t; - mutex_unlock(&cpufreq_governor_mutex); - } -out: - return err; + if (request_module("cpufreq_%s", str_governor)) + return NULL; + + return get_governor(str_governor); } -/** +/* * cpufreq_per_cpu_attr_read() / show_##file_name() - * print out cpufreq information * @@ -449,7 +703,7 @@ out: static ssize_t show_##file_name \ (struct cpufreq_policy *policy, char *buf) \ { \ - return sprintf(buf, "%u\n", policy->object); \ + return sysfs_emit(buf, "%u\n", policy->object); \ } show_one(cpuinfo_min_freq, cpuinfo.min_freq); @@ -457,103 +711,135 @@ show_one(cpuinfo_max_freq, cpuinfo.max_freq); show_one(cpuinfo_transition_latency, cpuinfo.transition_latency); show_one(scaling_min_freq, min); show_one(scaling_max_freq, max); -show_one(scaling_cur_freq, cur); -static int __cpufreq_set_policy(struct cpufreq_policy *data, - struct cpufreq_policy *policy); +__weak int arch_freq_get_on_cpu(int cpu) +{ + return -EOPNOTSUPP; +} -/** +static inline bool cpufreq_avg_freq_supported(struct cpufreq_policy *policy) +{ + return arch_freq_get_on_cpu(policy->cpu) != -EOPNOTSUPP; +} + +static ssize_t show_scaling_cur_freq(struct cpufreq_policy *policy, char *buf) +{ + ssize_t ret; + int freq; + + freq = IS_ENABLED(CONFIG_CPUFREQ_ARCH_CUR_FREQ) + ? arch_freq_get_on_cpu(policy->cpu) + : 0; + + if (freq > 0) + ret = sysfs_emit(buf, "%u\n", freq); + else if (cpufreq_driver->setpolicy && cpufreq_driver->get) + ret = sysfs_emit(buf, "%u\n", cpufreq_driver->get(policy->cpu)); + else + ret = sysfs_emit(buf, "%u\n", policy->cur); + return ret; +} + +/* * cpufreq_per_cpu_attr_write() / store_##file_name() - sysfs write access */ #define store_one(file_name, object) \ static ssize_t store_##file_name \ (struct cpufreq_policy *policy, const char *buf, size_t count) \ { \ - unsigned int ret; \ - struct cpufreq_policy new_policy; \ + unsigned long val; \ + int ret; \ \ - ret = cpufreq_get_policy(&new_policy, policy->cpu); \ + ret = kstrtoul(buf, 0, &val); \ if (ret) \ - return -EINVAL; \ - \ - ret = sscanf(buf, "%u", &new_policy.object); \ - if (ret != 1) \ - return -EINVAL; \ + return ret; \ \ - ret = __cpufreq_set_policy(policy, &new_policy); \ - policy->user_policy.object = policy->object; \ - \ - return ret ? ret : count; \ + ret = freq_qos_update_request(policy->object##_freq_req, val);\ + return ret >= 0 ? count : ret; \ } store_one(scaling_min_freq, min); store_one(scaling_max_freq, max); -/** +/* * show_cpuinfo_cur_freq - current CPU frequency as detected by hardware */ static ssize_t show_cpuinfo_cur_freq(struct cpufreq_policy *policy, char *buf) { - unsigned int cur_freq = __cpufreq_get(policy->cpu); - if (!cur_freq) - return sprintf(buf, "<unknown>"); - return sprintf(buf, "%u\n", cur_freq); + unsigned int cur_freq = __cpufreq_get(policy); + + if (cur_freq) + return sysfs_emit(buf, "%u\n", cur_freq); + + return sysfs_emit(buf, "<unknown>\n"); +} + +/* + * show_cpuinfo_avg_freq - average CPU frequency as detected by hardware + */ +static ssize_t show_cpuinfo_avg_freq(struct cpufreq_policy *policy, + char *buf) +{ + int avg_freq = arch_freq_get_on_cpu(policy->cpu); + + if (avg_freq > 0) + return sysfs_emit(buf, "%u\n", avg_freq); + return avg_freq != 0 ? avg_freq : -EINVAL; } -/** +/* * show_scaling_governor - show the current policy for the specified CPU */ static ssize_t show_scaling_governor(struct cpufreq_policy *policy, char *buf) { if (policy->policy == CPUFREQ_POLICY_POWERSAVE) - return sprintf(buf, "powersave\n"); + return sysfs_emit(buf, "powersave\n"); else if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) - return sprintf(buf, "performance\n"); + return sysfs_emit(buf, "performance\n"); else if (policy->governor) - return scnprintf(buf, CPUFREQ_NAME_PLEN, "%s\n", - policy->governor->name); + return sysfs_emit(buf, "%s\n", policy->governor->name); return -EINVAL; } -/** +/* * store_scaling_governor - store policy for the specified CPU */ static ssize_t store_scaling_governor(struct cpufreq_policy *policy, const char *buf, size_t count) { - unsigned int ret; - char str_governor[16]; - struct cpufreq_policy new_policy; - - ret = cpufreq_get_policy(&new_policy, policy->cpu); - if (ret) - return ret; + char str_governor[CPUFREQ_NAME_LEN]; + int ret; ret = sscanf(buf, "%15s", str_governor); if (ret != 1) return -EINVAL; - if (cpufreq_parse_governor(str_governor, &new_policy.policy, - &new_policy.governor)) - return -EINVAL; + if (cpufreq_driver->setpolicy) { + unsigned int new_pol; - /* - * Do not use cpufreq_set_policy here or the user_policy.max - * will be wrongly overridden - */ - ret = __cpufreq_set_policy(policy, &new_policy); + new_pol = cpufreq_parse_policy(str_governor); + if (!new_pol) + return -EINVAL; - policy->user_policy.policy = policy->policy; - policy->user_policy.governor = policy->governor; + ret = cpufreq_set_policy(policy, NULL, new_pol); + } else { + struct cpufreq_governor *new_gov; - if (ret) - return ret; - else - return count; + new_gov = cpufreq_parse_governor(str_governor); + if (!new_gov) + return -EINVAL; + + ret = cpufreq_set_policy(policy, new_gov, + CPUFREQ_POLICY_UNKNOWN); + + module_put(new_gov->owner); + } + + return ret ? ret : count; } -/** +/* * show_scaling_driver - show the cpufreq driver currently loaded */ static ssize_t show_scaling_driver(struct cpufreq_policy *policy, char *buf) @@ -561,7 +847,7 @@ static ssize_t show_scaling_driver(struct cpufreq_policy *policy, char *buf) return scnprintf(buf, CPUFREQ_NAME_PLEN, "%s\n", cpufreq_driver->name); } -/** +/* * show_scaling_available_governors - show the available CPUfreq governors */ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy, @@ -570,19 +856,21 @@ static ssize_t show_scaling_available_governors(struct cpufreq_policy *policy, ssize_t i = 0; struct cpufreq_governor *t; - if (!cpufreq_driver->target) { - i += sprintf(buf, "performance powersave"); + if (!has_target()) { + i += sysfs_emit(buf, "performance powersave"); goto out; } - list_for_each_entry(t, &cpufreq_governor_list, governor_list) { + mutex_lock(&cpufreq_governor_mutex); + for_each_governor(t) { if (i >= (ssize_t) ((PAGE_SIZE / sizeof(char)) - (CPUFREQ_NAME_LEN + 2))) - goto out; - i += scnprintf(&buf[i], CPUFREQ_NAME_PLEN, "%s ", t->name); + break; + i += sysfs_emit_at(buf, i, "%s ", t->name); } + mutex_unlock(&cpufreq_governor_mutex); out: - i += sprintf(&buf[i], "\n"); + i += sysfs_emit_at(buf, i, "\n"); return i; } @@ -592,18 +880,20 @@ ssize_t cpufreq_show_cpus(const struct cpumask *mask, char *buf) unsigned int cpu; for_each_cpu(cpu, mask) { - if (i) - i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " "); - i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu); + i += sysfs_emit_at(buf, i, "%u ", cpu); if (i >= (PAGE_SIZE - 5)) break; } - i += sprintf(&buf[i], "\n"); + + /* Remove the extra space at the end */ + i--; + + i += sysfs_emit_at(buf, i, "\n"); return i; } EXPORT_SYMBOL_GPL(cpufreq_show_cpus); -/** +/* * show_related_cpus - show the CPUs affected by each transition even if * hw coordination is in use */ @@ -612,7 +902,7 @@ static ssize_t show_related_cpus(struct cpufreq_policy *policy, char *buf) return cpufreq_show_cpus(policy->related_cpus, buf); } -/** +/* * show_affected_cpus - show the CPUs affected by each transition */ static ssize_t show_affected_cpus(struct cpufreq_policy *policy, char *buf) @@ -624,14 +914,14 @@ static ssize_t store_scaling_setspeed(struct cpufreq_policy *policy, const char *buf, size_t count) { unsigned int freq = 0; - unsigned int ret; + int ret; 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); @@ -641,27 +931,26 @@ static ssize_t store_scaling_setspeed(struct cpufreq_policy *policy, static ssize_t show_scaling_setspeed(struct cpufreq_policy *policy, char *buf) { if (!policy->governor || !policy->governor->show_setspeed) - return sprintf(buf, "<unsupported>\n"); + return sysfs_emit(buf, "<unsupported>\n"); return policy->governor->show_setspeed(policy, buf); } -/** +/* * show_bios_limit - show the current cpufreq HW/BIOS limitation */ static ssize_t show_bios_limit(struct cpufreq_policy *policy, char *buf) { unsigned int limit; int ret; - if (cpufreq_driver->bios_limit) { - ret = cpufreq_driver->bios_limit(policy->cpu, &limit); - if (!ret) - return sprintf(buf, "%u\n", limit); - } - return sprintf(buf, "%u\n", policy->cpuinfo.max_freq); + ret = cpufreq_driver->bios_limit(policy->cpu, &limit); + if (!ret) + return sysfs_emit(buf, "%u\n", limit); + return sysfs_emit(buf, "%u\n", policy->cpuinfo.max_freq); } cpufreq_freq_attr_ro_perm(cpuinfo_cur_freq, 0400); +cpufreq_freq_attr_ro(cpuinfo_avg_freq); cpufreq_freq_attr_ro(cpuinfo_min_freq); cpufreq_freq_attr_ro(cpuinfo_max_freq); cpufreq_freq_attr_ro(cpuinfo_transition_latency); @@ -676,10 +965,11 @@ cpufreq_freq_attr_rw(scaling_max_freq); cpufreq_freq_attr_rw(scaling_governor); cpufreq_freq_attr_rw(scaling_setspeed); -static struct attribute *default_attrs[] = { +static struct attribute *cpufreq_attrs[] = { &cpuinfo_min_freq.attr, &cpuinfo_max_freq.attr, &cpuinfo_transition_latency.attr, + &scaling_cur_freq.attr, &scaling_min_freq.attr, &scaling_max_freq.attr, &affected_cpus.attr, @@ -690,6 +980,7 @@ static struct attribute *default_attrs[] = { &scaling_setspeed.attr, NULL }; +ATTRIBUTE_GROUPS(cpufreq); #define to_policy(k) container_of(k, struct cpufreq_policy, kobj) #define to_attr(a) container_of(a, struct freq_attr, attr) @@ -698,24 +989,16 @@ 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 = -EINVAL; - policy = cpufreq_cpu_get_sysfs(policy->cpu); - if (!policy) - goto no_policy; - if (lock_policy_rwsem_read(policy->cpu) < 0) - goto fail; + if (!fattr->show) + return -EIO; - if (fattr->show) - ret = fattr->show(policy, buf); - else - ret = -EIO; + guard(cpufreq_policy_read)(policy); - unlock_policy_rwsem_read(policy->cpu); -fail: - cpufreq_cpu_put_sysfs(policy); -no_policy: - return ret; + if (likely(!policy_is_inactive(policy))) + return fattr->show(policy, buf); + + return -EBUSY; } static ssize_t store(struct kobject *kobj, struct attribute *attr, @@ -723,24 +1006,16 @@ 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 = -EINVAL; - policy = cpufreq_cpu_get_sysfs(policy->cpu); - if (!policy) - goto no_policy; - if (lock_policy_rwsem_write(policy->cpu) < 0) - goto fail; + if (!fattr->store) + return -EIO; - if (fattr->store) - ret = fattr->store(policy, buf, count); - else - ret = -EIO; + guard(cpufreq_policy_write)(policy); - unlock_policy_rwsem_write(policy->cpu); -fail: - cpufreq_cpu_put_sysfs(policy); -no_policy: - return ret; + if (likely(!policy_is_inactive(policy))) + return fattr->store(policy, buf, count); + + return -EBUSY; } static void cpufreq_sysfs_release(struct kobject *kobj) @@ -755,251 +1030,236 @@ static const struct sysfs_ops sysfs_ops = { .store = store, }; -static struct kobj_type ktype_cpufreq = { +static const struct kobj_type ktype_cpufreq = { .sysfs_ops = &sysfs_ops, - .default_attrs = default_attrs, + .default_groups = cpufreq_groups, .release = cpufreq_sysfs_release, }; -struct kobject *cpufreq_global_kobject; -EXPORT_SYMBOL(cpufreq_global_kobject); - -static int cpufreq_global_kobject_usage; - -int cpufreq_get_global_kobject(void) -{ - if (!cpufreq_global_kobject_usage++) - return kobject_add(cpufreq_global_kobject, - &cpu_subsys.dev_root->kobj, "%s", "cpufreq"); - - return 0; -} -EXPORT_SYMBOL(cpufreq_get_global_kobject); - -void cpufreq_put_global_kobject(void) -{ - if (!--cpufreq_global_kobject_usage) - kobject_del(cpufreq_global_kobject); -} -EXPORT_SYMBOL(cpufreq_put_global_kobject); - -int cpufreq_sysfs_create_file(const struct attribute *attr) +static void add_cpu_dev_symlink(struct cpufreq_policy *policy, unsigned int cpu, + struct device *dev) { - int ret = cpufreq_get_global_kobject(); + if (unlikely(!dev)) + return; - if (!ret) { - ret = sysfs_create_file(cpufreq_global_kobject, attr); - if (ret) - cpufreq_put_global_kobject(); - } + if (cpumask_test_and_set_cpu(cpu, policy->real_cpus)) + return; - return ret; + dev_dbg(dev, "%s: Adding symlink\n", __func__); + if (sysfs_create_link(&dev->kobj, &policy->kobj, "cpufreq")) + dev_err(dev, "cpufreq symlink creation failed\n"); } -EXPORT_SYMBOL(cpufreq_sysfs_create_file); -void cpufreq_sysfs_remove_file(const struct attribute *attr) +static void remove_cpu_dev_symlink(struct cpufreq_policy *policy, int cpu, + struct device *dev) { - sysfs_remove_file(cpufreq_global_kobject, attr); - cpufreq_put_global_kobject(); + dev_dbg(dev, "%s: Removing symlink\n", __func__); + sysfs_remove_link(&dev->kobj, "cpufreq"); + cpumask_clear_cpu(cpu, policy->real_cpus); } -EXPORT_SYMBOL(cpufreq_sysfs_remove_file); -/* symlink affected CPUs */ -static int cpufreq_add_dev_symlink(unsigned int cpu, - struct cpufreq_policy *policy) +static int cpufreq_add_dev_interface(struct cpufreq_policy *policy) { - unsigned int j; + struct freq_attr **drv_attr; int ret = 0; - for_each_cpu(j, policy->cpus) { - struct cpufreq_policy *managed_policy; - struct device *cpu_dev; - - if (j == cpu) - continue; - - pr_debug("CPU %u already managed, adding link\n", j); - managed_policy = cpufreq_cpu_get(cpu); - cpu_dev = get_cpu_device(j); - ret = sysfs_create_link(&cpu_dev->kobj, &policy->kobj, - "cpufreq"); - if (ret) { - cpufreq_cpu_put(managed_policy); + /* Attributes that need freq_table */ + if (policy->freq_table) { + ret = sysfs_create_file(&policy->kobj, + &cpufreq_freq_attr_scaling_available_freqs.attr); + if (ret) return ret; + + if (cpufreq_boost_supported()) { + ret = sysfs_create_file(&policy->kobj, + &cpufreq_freq_attr_scaling_boost_freqs.attr); + if (ret) + return ret; } } - return ret; -} - -static int cpufreq_add_dev_interface(unsigned int cpu, - struct cpufreq_policy *policy, - struct device *dev) -{ - struct cpufreq_policy new_policy; - struct freq_attr **drv_attr; - unsigned long flags; - int ret = 0; - unsigned int j; - - /* prepare interface data */ - ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, - &dev->kobj, "cpufreq"); - if (ret) - return ret; /* set up files for this cpu device */ drv_attr = cpufreq_driver->attr; - while ((drv_attr) && (*drv_attr)) { + while (drv_attr && *drv_attr) { ret = sysfs_create_file(&policy->kobj, &((*drv_attr)->attr)); if (ret) - goto err_out_kobj_put; + return ret; drv_attr++; } if (cpufreq_driver->get) { ret = sysfs_create_file(&policy->kobj, &cpuinfo_cur_freq.attr); if (ret) - goto err_out_kobj_put; + return ret; } - if (cpufreq_driver->target) { - ret = sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr); + + if (cpufreq_avg_freq_supported(policy)) { + ret = sysfs_create_file(&policy->kobj, &cpuinfo_avg_freq.attr); if (ret) - goto err_out_kobj_put; + return ret; } + if (cpufreq_driver->bios_limit) { ret = sysfs_create_file(&policy->kobj, &bios_limit.attr); if (ret) - goto err_out_kobj_put; + return ret; } - write_lock_irqsave(&cpufreq_driver_lock, flags); - for_each_cpu(j, policy->cpus) { - per_cpu(cpufreq_cpu_data, j) = policy; - per_cpu(cpufreq_policy_cpu, j) = policy->cpu; + if (cpufreq_boost_supported()) { + ret = sysfs_create_file(&policy->kobj, &local_boost.attr); + if (ret) + return ret; } - write_unlock_irqrestore(&cpufreq_driver_lock, flags); - ret = cpufreq_add_dev_symlink(cpu, policy); - if (ret) - goto err_out_kobj_put; + return 0; +} - memcpy(&new_policy, policy, sizeof(struct cpufreq_policy)); - /* assure that the starting sequence is run in __cpufreq_set_policy */ - policy->governor = NULL; +static int cpufreq_init_policy(struct cpufreq_policy *policy) +{ + struct cpufreq_governor *gov = NULL; + unsigned int pol = CPUFREQ_POLICY_UNKNOWN; + int ret; - /* set default policy */ - ret = __cpufreq_set_policy(policy, &new_policy); - policy->user_policy.policy = policy->policy; - policy->user_policy.governor = policy->governor; + if (has_target()) { + /* Update policy governor to the one used before hotplug. */ + if (policy->last_governor[0] != '\0') + gov = get_governor(policy->last_governor); + if (gov) { + pr_debug("Restoring governor %s for cpu %d\n", + gov->name, policy->cpu); + } else { + gov = get_governor(default_governor); + } - if (ret) { - pr_debug("setting policy failed\n"); - if (cpufreq_driver->exit) - cpufreq_driver->exit(policy); + if (!gov) { + gov = cpufreq_default_governor(); + __module_get(gov->owner); + } + + } else { + + /* Use the default policy if there is no last_policy. */ + if (policy->last_policy) { + pol = policy->last_policy; + } else { + pol = cpufreq_parse_policy(default_governor); + /* + * In case the default governor is neither "performance" + * nor "powersave", fall back to the initial policy + * value set by the driver. + */ + if (pol == CPUFREQ_POLICY_UNKNOWN) + pol = policy->policy; + } + if (pol != CPUFREQ_POLICY_PERFORMANCE && + pol != CPUFREQ_POLICY_POWERSAVE) + return -ENODATA; } - return ret; -err_out_kobj_put: - kobject_put(&policy->kobj); - wait_for_completion(&policy->kobj_unregister); + ret = cpufreq_set_policy(policy, gov, pol); + if (gov) + module_put(gov->owner); + return ret; } -#ifdef CONFIG_HOTPLUG_CPU -static int cpufreq_add_policy_cpu(unsigned int cpu, unsigned int sibling, - struct device *dev) +static int cpufreq_add_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu) { - struct cpufreq_policy *policy; - int ret = 0, has_target = !!cpufreq_driver->target; - unsigned long flags; - - policy = cpufreq_cpu_get(sibling); - WARN_ON(!policy); + int ret = 0; - if (has_target) - __cpufreq_governor(policy, CPUFREQ_GOV_STOP); + /* Has this CPU been taken care of already? */ + if (cpumask_test_cpu(cpu, policy->cpus)) + return 0; - lock_policy_rwsem_write(sibling); + guard(cpufreq_policy_write)(policy); - write_lock_irqsave(&cpufreq_driver_lock, flags); + if (has_target()) + cpufreq_stop_governor(policy); cpumask_set_cpu(cpu, policy->cpus); - per_cpu(cpufreq_policy_cpu, cpu) = policy->cpu; - per_cpu(cpufreq_cpu_data, cpu) = policy; - write_unlock_irqrestore(&cpufreq_driver_lock, flags); - - unlock_policy_rwsem_write(sibling); - if (has_target) { - __cpufreq_governor(policy, CPUFREQ_GOV_START); - __cpufreq_governor(policy, CPUFREQ_GOV_LIMITS); + if (has_target()) { + ret = cpufreq_start_governor(policy); + if (ret) + pr_err("%s: Failed to start governor\n", __func__); } - ret = sysfs_create_link(&dev->kobj, &policy->kobj, "cpufreq"); - if (ret) { - cpufreq_cpu_put(policy); - return ret; + return ret; +} + +void refresh_frequency_limits(struct cpufreq_policy *policy) +{ + if (!policy_is_inactive(policy)) { + pr_debug("updating policy for CPU %u\n", policy->cpu); + + cpufreq_set_policy(policy, policy->governor, policy->policy); } +} +EXPORT_SYMBOL(refresh_frequency_limits); + +static void handle_update(struct work_struct *work) +{ + struct cpufreq_policy *policy = + container_of(work, struct cpufreq_policy, update); + + pr_debug("handle_update for cpu %u called\n", policy->cpu); + + guard(cpufreq_policy_write)(policy); + + refresh_frequency_limits(policy); +} +static int cpufreq_notifier_min(struct notifier_block *nb, unsigned long freq, + void *data) +{ + struct cpufreq_policy *policy = container_of(nb, struct cpufreq_policy, nb_min); + + schedule_work(&policy->update); return 0; } -#endif -/** - * cpufreq_add_dev - add a CPU device - * - * Adds the cpufreq interface for a CPU device. - * - * The Oracle says: try running cpufreq registration/unregistration concurrently - * with with cpu hotplugging and all hell will break loose. Tried to clean this - * mess up, but more thorough testing is needed. - Mathieu - */ -static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif) +static int cpufreq_notifier_max(struct notifier_block *nb, unsigned long freq, + void *data) { - unsigned int j, cpu = dev->id; - int ret = -ENOMEM; - struct cpufreq_policy *policy; - unsigned long flags; -#ifdef CONFIG_HOTPLUG_CPU - struct cpufreq_governor *gov; - int sibling; -#endif + struct cpufreq_policy *policy = container_of(nb, struct cpufreq_policy, nb_max); - if (cpu_is_offline(cpu)) - return 0; + schedule_work(&policy->update); + return 0; +} - pr_debug("adding CPU %u\n", cpu); +static void cpufreq_policy_put_kobj(struct cpufreq_policy *policy) +{ + struct kobject *kobj; + struct completion *cmp; -#ifdef CONFIG_SMP - /* check whether a different CPU already registered this - * CPU because it is in the same boat. */ - policy = cpufreq_cpu_get(cpu); - if (unlikely(policy)) { - cpufreq_cpu_put(policy); - return 0; + scoped_guard(cpufreq_policy_write, policy) { + cpufreq_stats_free_table(policy); + kobj = &policy->kobj; + cmp = &policy->kobj_unregister; } + kobject_put(kobj); -#ifdef CONFIG_HOTPLUG_CPU - /* Check if this cpu was hot-unplugged earlier and has siblings */ - read_lock_irqsave(&cpufreq_driver_lock, flags); - for_each_online_cpu(sibling) { - struct cpufreq_policy *cp = per_cpu(cpufreq_cpu_data, sibling); - if (cp && cpumask_test_cpu(cpu, cp->related_cpus)) { - read_unlock_irqrestore(&cpufreq_driver_lock, flags); - return cpufreq_add_policy_cpu(cpu, sibling, dev); - } - } - read_unlock_irqrestore(&cpufreq_driver_lock, flags); -#endif -#endif + /* + * We need to make sure that the underlying kobj is + * actually not referenced anymore by anybody before we + * proceed with unloading. + */ + pr_debug("waiting for dropping of refcount\n"); + wait_for_completion(cmp); + pr_debug("wait complete\n"); +} - if (!try_module_get(cpufreq_driver->owner)) { - ret = -EINVAL; - goto module_out; - } +static struct cpufreq_policy *cpufreq_policy_alloc(unsigned int cpu) +{ + struct cpufreq_policy *policy; + struct device *dev = get_cpu_device(cpu); + int ret; - policy = kzalloc(sizeof(struct cpufreq_policy), GFP_KERNEL); + if (!dev) + return NULL; + + policy = kzalloc(sizeof(*policy), GFP_KERNEL); if (!policy) - goto nomem_out; + return NULL; if (!alloc_cpumask_var(&policy->cpus, GFP_KERNEL)) goto err_free_policy; @@ -1007,263 +1267,576 @@ static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif) if (!zalloc_cpumask_var(&policy->related_cpus, GFP_KERNEL)) goto err_free_cpumask; - policy->cpu = cpu; - policy->governor = CPUFREQ_DEFAULT_GOVERNOR; - cpumask_copy(policy->cpus, cpumask_of(cpu)); - - /* Initially set CPU itself as the policy_cpu */ - per_cpu(cpufreq_policy_cpu, cpu) = cpu; + if (!zalloc_cpumask_var(&policy->real_cpus, GFP_KERNEL)) + goto err_free_rcpumask; init_completion(&policy->kobj_unregister); - INIT_WORK(&policy->update, handle_update); - - /* call driver. From then on the cpufreq must be able - * to accept all calls to ->verify and ->setpolicy for this CPU - */ - ret = cpufreq_driver->init(policy); + ret = kobject_init_and_add(&policy->kobj, &ktype_cpufreq, + cpufreq_global_kobject, "policy%u", cpu); if (ret) { - pr_debug("initialization failed\n"); - goto err_set_policy_cpu; + dev_err(dev, "%s: failed to init policy->kobj: %d\n", __func__, ret); + /* + * The entire policy object will be freed below, but the extra + * memory allocated for the kobject name needs to be freed by + * releasing the kobject. + */ + kobject_put(&policy->kobj); + goto err_free_real_cpus; } - /* related cpus should atleast have policy->cpus */ - cpumask_or(policy->related_cpus, policy->related_cpus, policy->cpus); + init_rwsem(&policy->rwsem); - /* - * affected cpus must always be the one, which are online. We aren't - * managing offline cpus here. - */ - cpumask_and(policy->cpus, policy->cpus, cpu_online_mask); + freq_constraints_init(&policy->constraints); - policy->user_policy.min = policy->min; - policy->user_policy.max = policy->max; + policy->nb_min.notifier_call = cpufreq_notifier_min; + policy->nb_max.notifier_call = cpufreq_notifier_max; - blocking_notifier_call_chain(&cpufreq_policy_notifier_list, - CPUFREQ_START, policy); + ret = freq_qos_add_notifier(&policy->constraints, FREQ_QOS_MIN, + &policy->nb_min); + if (ret) { + dev_err(dev, "Failed to register MIN QoS notifier: %d (CPU%u)\n", + ret, cpu); + goto err_kobj_remove; + } -#ifdef CONFIG_HOTPLUG_CPU - gov = __find_governor(per_cpu(cpufreq_cpu_governor, cpu)); - if (gov) { - policy->governor = gov; - pr_debug("Restoring governor %s for cpu %d\n", - policy->governor->name, cpu); + ret = freq_qos_add_notifier(&policy->constraints, FREQ_QOS_MAX, + &policy->nb_max); + if (ret) { + dev_err(dev, "Failed to register MAX QoS notifier: %d (CPU%u)\n", + ret, cpu); + goto err_min_qos_notifier; } -#endif - ret = cpufreq_add_dev_interface(cpu, policy, dev); - if (ret) - goto err_out_unregister; + INIT_LIST_HEAD(&policy->policy_list); + spin_lock_init(&policy->transition_lock); + init_waitqueue_head(&policy->transition_wait); + INIT_WORK(&policy->update, handle_update); - kobject_uevent(&policy->kobj, KOBJ_ADD); - module_put(cpufreq_driver->owner); - pr_debug("initialization complete\n"); + return policy; - return 0; +err_min_qos_notifier: + freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MIN, + &policy->nb_min); +err_kobj_remove: + cpufreq_policy_put_kobj(policy); +err_free_real_cpus: + free_cpumask_var(policy->real_cpus); +err_free_rcpumask: + free_cpumask_var(policy->related_cpus); +err_free_cpumask: + free_cpumask_var(policy->cpus); +err_free_policy: + kfree(policy); + + return NULL; +} + +static void cpufreq_policy_free(struct cpufreq_policy *policy) +{ + unsigned long flags; + int cpu; -err_out_unregister: + /* + * The callers must ensure the policy is inactive by now, to avoid any + * races with show()/store() callbacks. + */ + if (unlikely(!policy_is_inactive(policy))) + pr_warn("%s: Freeing active policy\n", __func__); + + /* Remove policy from list */ write_lock_irqsave(&cpufreq_driver_lock, flags); - for_each_cpu(j, policy->cpus) - per_cpu(cpufreq_cpu_data, j) = NULL; + list_del(&policy->policy_list); + + for_each_cpu(cpu, policy->related_cpus) + per_cpu(cpufreq_cpu_data, cpu) = NULL; write_unlock_irqrestore(&cpufreq_driver_lock, flags); - kobject_put(&policy->kobj); - wait_for_completion(&policy->kobj_unregister); + freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MAX, + &policy->nb_max); + freq_qos_remove_notifier(&policy->constraints, FREQ_QOS_MIN, + &policy->nb_min); + + /* Cancel any pending policy->update work before freeing the policy. */ + cancel_work_sync(&policy->update); -err_set_policy_cpu: - per_cpu(cpufreq_policy_cpu, cpu) = -1; + if (policy->max_freq_req) { + /* + * Remove max_freq_req after sending CPUFREQ_REMOVE_POLICY + * notification, since CPUFREQ_CREATE_POLICY notification was + * sent after adding max_freq_req earlier. + */ + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_REMOVE_POLICY, policy); + freq_qos_remove_request(policy->max_freq_req); + } + + freq_qos_remove_request(policy->min_freq_req); + kfree(policy->min_freq_req); + + cpufreq_policy_put_kobj(policy); + free_cpumask_var(policy->real_cpus); free_cpumask_var(policy->related_cpus); -err_free_cpumask: free_cpumask_var(policy->cpus); -err_free_policy: kfree(policy); -nomem_out: - module_put(cpufreq_driver->owner); -module_out: - return ret; } -static void update_policy_cpu(struct cpufreq_policy *policy, unsigned int cpu) +static int cpufreq_policy_online(struct cpufreq_policy *policy, + unsigned int cpu, bool new_policy) { - int j; + unsigned long flags; + unsigned int j; + int ret; + + guard(cpufreq_policy_write)(policy); - policy->last_cpu = policy->cpu; policy->cpu = cpu; + policy->governor = NULL; - for_each_cpu(j, policy->cpus) - per_cpu(cpufreq_policy_cpu, j) = cpu; + if (!new_policy && cpufreq_driver->online) { + /* Recover policy->cpus using related_cpus */ + cpumask_copy(policy->cpus, policy->related_cpus); -#ifdef CONFIG_CPU_FREQ_TABLE - cpufreq_frequency_table_update_policy_cpu(policy); -#endif - blocking_notifier_call_chain(&cpufreq_policy_notifier_list, - CPUFREQ_UPDATE_POLICY_CPU, policy); + ret = cpufreq_driver->online(policy); + if (ret) { + pr_debug("%s: %d: initialization failed\n", __func__, + __LINE__); + goto out_exit_policy; + } + } else { + cpumask_copy(policy->cpus, cpumask_of(cpu)); + + /* + * Call driver. From then on the cpufreq must be able + * to accept all calls to ->verify and ->setpolicy for this CPU. + */ + ret = cpufreq_driver->init(policy); + if (ret) { + pr_debug("%s: %d: initialization failed\n", __func__, + __LINE__); + goto out_clear_policy; + } + + /* + * The initialization has succeeded and the policy is online. + * If there is a problem with its frequency table, take it + * offline and drop it. + */ + if (policy->freq_table_sorted != CPUFREQ_TABLE_SORTED_ASCENDING && + policy->freq_table_sorted != CPUFREQ_TABLE_SORTED_DESCENDING) { + ret = cpufreq_table_validate_and_sort(policy); + if (ret) + goto out_offline_policy; + } + + /* related_cpus should at least include policy->cpus. */ + cpumask_copy(policy->related_cpus, policy->cpus); + } + + /* + * affected cpus must always be the one, which are online. We aren't + * managing offline cpus here. + */ + cpumask_and(policy->cpus, policy->cpus, cpu_online_mask); + + if (new_policy) { + for_each_cpu(j, policy->related_cpus) { + per_cpu(cpufreq_cpu_data, j) = policy; + add_cpu_dev_symlink(policy, j, get_cpu_device(j)); + } + + policy->min_freq_req = kzalloc(2 * sizeof(*policy->min_freq_req), + GFP_KERNEL); + if (!policy->min_freq_req) { + ret = -ENOMEM; + goto out_destroy_policy; + } + + ret = freq_qos_add_request(&policy->constraints, + policy->min_freq_req, FREQ_QOS_MIN, + FREQ_QOS_MIN_DEFAULT_VALUE); + if (ret < 0) { + /* + * So we don't call freq_qos_remove_request() for an + * uninitialized request. + */ + kfree(policy->min_freq_req); + policy->min_freq_req = NULL; + goto out_destroy_policy; + } + + /* + * This must be initialized right here to avoid calling + * freq_qos_remove_request() on uninitialized request in case + * of errors. + */ + policy->max_freq_req = policy->min_freq_req + 1; + + ret = freq_qos_add_request(&policy->constraints, + policy->max_freq_req, FREQ_QOS_MAX, + FREQ_QOS_MAX_DEFAULT_VALUE); + if (ret < 0) { + policy->max_freq_req = NULL; + goto out_destroy_policy; + } + + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_CREATE_POLICY, policy); + } else { + ret = freq_qos_update_request(policy->max_freq_req, policy->max); + if (ret < 0) + goto out_destroy_policy; + } + + if (cpufreq_driver->get && has_target()) { + policy->cur = cpufreq_driver->get(policy->cpu); + if (!policy->cur) { + ret = -EIO; + pr_err("%s: ->get() failed\n", __func__); + goto out_destroy_policy; + } + } + + /* + * Sometimes boot loaders set CPU frequency to a value outside of + * frequency table present with cpufreq core. In such cases CPU might be + * unstable if it has to run on that frequency for long duration of time + * and so its better to set it to a frequency which is specified in + * freq-table. This also makes cpufreq stats inconsistent as + * cpufreq-stats would fail to register because current frequency of CPU + * isn't found in freq-table. + * + * Because we don't want this change to effect boot process badly, we go + * for the next freq which is >= policy->cur ('cur' must be set by now, + * otherwise we will end up setting freq to lowest of the table as 'cur' + * is initialized to zero). + * + * We are passing target-freq as "policy->cur - 1" otherwise + * __cpufreq_driver_target() would simply fail, as policy->cur will be + * equal to target-freq. + */ + if ((cpufreq_driver->flags & CPUFREQ_NEED_INITIAL_FREQ_CHECK) + && has_target()) { + unsigned int old_freq = policy->cur; + + /* Are we running at unknown frequency ? */ + ret = cpufreq_frequency_table_get_index(policy, old_freq); + if (ret == -EINVAL) { + ret = __cpufreq_driver_target(policy, old_freq - 1, + CPUFREQ_RELATION_L); + + /* + * Reaching here after boot in a few seconds may not + * mean that system will remain stable at "unknown" + * frequency for longer duration. Hence, a BUG_ON(). + */ + BUG_ON(ret); + pr_info("%s: CPU%d: Running at unlisted initial frequency: %u kHz, changing to: %u kHz\n", + __func__, policy->cpu, old_freq, policy->cur); + } + } + + if (new_policy) { + ret = cpufreq_add_dev_interface(policy); + if (ret) + goto out_destroy_policy; + + cpufreq_stats_create_table(policy); + + write_lock_irqsave(&cpufreq_driver_lock, flags); + list_add(&policy->policy_list, &cpufreq_policy_list); + write_unlock_irqrestore(&cpufreq_driver_lock, flags); + + /* + * Register with the energy model before + * em_rebuild_sched_domains() is called, which will result + * in rebuilding of the sched domains, which should only be done + * once the energy model is properly initialized for the policy + * first. + * + * Also, this should be called before the policy is registered + * with cooling framework. + */ + if (cpufreq_driver->register_em) + cpufreq_driver->register_em(policy); + } + + ret = cpufreq_init_policy(policy); + if (ret) { + pr_err("%s: Failed to initialize policy for cpu: %d (%d)\n", + __func__, cpu, ret); + goto out_destroy_policy; + } + + 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; } -/** - * __cpufreq_remove_dev - remove a CPU device - * - * Removes the cpufreq interface for a CPU device. - * Caller should already have policy_rwsem in write mode for this CPU. - * This routine frees the rwsem before returning. - */ -static int __cpufreq_remove_dev(struct device *dev, - struct subsys_interface *sif) +static int cpufreq_online(unsigned int cpu) { - unsigned int cpu = dev->id, ret, cpus; - unsigned long flags; - struct cpufreq_policy *data; - struct kobject *kobj; - struct completion *cmp; - struct device *cpu_dev; + struct cpufreq_policy *policy; + bool new_policy; + int ret; - pr_debug("%s: unregistering CPU %u\n", __func__, cpu); + pr_debug("%s: bringing CPU%u online\n", __func__, cpu); - write_lock_irqsave(&cpufreq_driver_lock, flags); + /* 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); - data = per_cpu(cpufreq_cpu_data, cpu); - per_cpu(cpufreq_cpu_data, cpu) = NULL; + /* 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; + } - write_unlock_irqrestore(&cpufreq_driver_lock, flags); + ret = cpufreq_policy_online(policy, cpu, new_policy); + if (ret) { + cpufreq_policy_free(policy); + return ret; + } - if (!data) { - pr_debug("%s: No cpu_data found\n", __func__); - return -EINVAL; + kobject_uevent(&policy->kobj, KOBJ_ADD); + + /* Callback for handling stuff after policy is ready */ + if (cpufreq_driver->ready) + cpufreq_driver->ready(policy); + + /* Register cpufreq cooling only for a new policy */ + 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 + * 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 && + (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(cpufreq_boost_enabled())); + } } - if (cpufreq_driver->target) - __cpufreq_governor(data, CPUFREQ_GOV_STOP); + pr_debug("initialization complete\n"); -#ifdef CONFIG_HOTPLUG_CPU - if (!cpufreq_driver->setpolicy) - strncpy(per_cpu(cpufreq_cpu_governor, cpu), - data->governor->name, CPUFREQ_NAME_LEN); -#endif + return 0; +} - WARN_ON(lock_policy_rwsem_write(cpu)); - cpus = cpumask_weight(data->cpus); +/** + * cpufreq_add_dev - the cpufreq interface for a CPU device. + * @dev: CPU device. + * @sif: Subsystem interface structure pointer (not used) + */ +static int cpufreq_add_dev(struct device *dev, struct subsys_interface *sif) +{ + struct cpufreq_policy *policy; + unsigned cpu = dev->id; + int ret; - if (cpus > 1) - cpumask_clear_cpu(cpu, data->cpus); - unlock_policy_rwsem_write(cpu); + dev_dbg(dev, "%s: adding CPU%u\n", __func__, cpu); - if (cpu != data->cpu) { - sysfs_remove_link(&dev->kobj, "cpufreq"); - } else if (cpus > 1) { - /* first sibling now owns the new sysfs dir */ - cpu_dev = get_cpu_device(cpumask_first(data->cpus)); - sysfs_remove_link(&cpu_dev->kobj, "cpufreq"); - ret = kobject_move(&data->kobj, &cpu_dev->kobj); - if (ret) { - pr_err("%s: Failed to move kobj: %d", __func__, ret); + if (cpu_online(cpu)) { + ret = cpufreq_online(cpu); + if (ret) + return ret; + } - WARN_ON(lock_policy_rwsem_write(cpu)); - cpumask_set_cpu(cpu, data->cpus); + /* Create sysfs link on CPU registration */ + policy = per_cpu(cpufreq_cpu_data, cpu); + if (policy) + add_cpu_dev_symlink(policy, cpu, dev); + + return 0; +} - write_lock_irqsave(&cpufreq_driver_lock, flags); - per_cpu(cpufreq_cpu_data, cpu) = data; - write_unlock_irqrestore(&cpufreq_driver_lock, flags); +static void __cpufreq_offline(unsigned int cpu, struct cpufreq_policy *policy) +{ + int ret; - unlock_policy_rwsem_write(cpu); + if (has_target()) + cpufreq_stop_governor(policy); - ret = sysfs_create_link(&cpu_dev->kobj, &data->kobj, - "cpufreq"); - return -EINVAL; + cpumask_clear_cpu(cpu, policy->cpus); + + if (!policy_is_inactive(policy)) { + /* Nominate a new CPU if necessary. */ + if (cpu == policy->cpu) + policy->cpu = cpumask_any(policy->cpus); + + /* Start the governor again for the active policy. */ + if (has_target()) { + ret = cpufreq_start_governor(policy); + if (ret) + pr_err("%s: Failed to start governor\n", __func__); } - WARN_ON(lock_policy_rwsem_write(cpu)); - update_policy_cpu(data, cpu_dev->id); - unlock_policy_rwsem_write(cpu); - pr_debug("%s: policy Kobject moved to cpu: %d from: %d\n", - __func__, cpu_dev->id, cpu); + return; } - if ((cpus == 1) && (cpufreq_driver->target)) - __cpufreq_governor(data, CPUFREQ_GOV_POLICY_EXIT); + if (has_target()) { + strscpy(policy->last_governor, policy->governor->name, + CPUFREQ_NAME_LEN); + cpufreq_exit_governor(policy); + } else { + policy->last_policy = policy->policy; + } - pr_debug("%s: removing link, cpu: %d\n", __func__, cpu); - cpufreq_cpu_put(data); + /* + * Perform the ->offline() during light-weight tear-down, as + * that allows fast recovery when the CPU comes back. + */ + if (cpufreq_driver->offline) { + cpufreq_driver->offline(policy); + return; + } - /* If cpu is last user of policy, free policy */ - if (cpus == 1) { - lock_policy_rwsem_read(cpu); - kobj = &data->kobj; - cmp = &data->kobj_unregister; - unlock_policy_rwsem_read(cpu); - kobject_put(kobj); + if (cpufreq_driver->exit) + cpufreq_driver->exit(policy); - /* we need to make sure that the underlying kobj is actually - * not referenced anymore by anybody before we proceed with - * unloading. - */ - pr_debug("waiting for dropping of refcount\n"); - wait_for_completion(cmp); - pr_debug("wait complete\n"); + policy->freq_table = NULL; +} - if (cpufreq_driver->exit) - cpufreq_driver->exit(data); +static int cpufreq_offline(unsigned int cpu) +{ + struct cpufreq_policy *policy; + + pr_debug("%s: unregistering CPU %u\n", __func__, cpu); - free_cpumask_var(data->related_cpus); - free_cpumask_var(data->cpus); - kfree(data); - } else if (cpufreq_driver->target) { - __cpufreq_governor(data, CPUFREQ_GOV_START); - __cpufreq_governor(data, CPUFREQ_GOV_LIMITS); + policy = cpufreq_cpu_get_raw(cpu); + if (!policy) { + pr_debug("%s: No cpu_data found\n", __func__); + return 0; } - per_cpu(cpufreq_policy_cpu, cpu) = -1; + guard(cpufreq_policy_write)(policy); + + __cpufreq_offline(cpu, policy); + return 0; } -static int cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) +/* + * cpufreq_remove_dev - remove a CPU device + * + * Removes the cpufreq interface for a CPU device. + */ +static void cpufreq_remove_dev(struct device *dev, struct subsys_interface *sif) { unsigned int cpu = dev->id; - int retval; + struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); - if (cpu_is_offline(cpu)) - return 0; + if (!policy) + return; - retval = __cpufreq_remove_dev(dev, sif); - return retval; -} + scoped_guard(cpufreq_policy_write, policy) { + if (cpu_online(cpu)) + __cpufreq_offline(cpu, policy); -static void handle_update(struct work_struct *work) -{ - struct cpufreq_policy *policy = - container_of(work, struct cpufreq_policy, update); - unsigned int cpu = policy->cpu; - pr_debug("handle_update for cpu %u called\n", cpu); - cpufreq_update_policy(cpu); + remove_cpu_dev_symlink(policy, cpu, dev); + + if (!cpumask_empty(policy->real_cpus)) + 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; + } + + /* We did light-weight exit earlier, do full tear down now */ + if (cpufreq_driver->offline && cpufreq_driver->exit) + cpufreq_driver->exit(policy); + } + + cpufreq_policy_free(policy); } /** - * cpufreq_out_of_sync - If actual and saved CPU frequency differs, we're - * in deep trouble. - * @cpu: cpu number - * @old_freq: CPU frequency the kernel thinks the CPU runs at - * @new_freq: CPU frequency the CPU actually runs at + * cpufreq_out_of_sync - Fix up actual and saved CPU frequency difference. + * @policy: Policy managing CPUs. + * @new_freq: New CPU frequency. * - * We adjust to current frequency first, and need to clean up later. - * So either call to cpufreq_update_policy() or schedule handle_update()). + * Adjust to the current frequency first and clean up later by either calling + * cpufreq_update_policy(), or scheduling handle_update(). */ -static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, +static void cpufreq_out_of_sync(struct cpufreq_policy *policy, unsigned int new_freq) { - struct cpufreq_policy *policy; struct cpufreq_freqs freqs; - unsigned long flags; - pr_debug("Warning: CPU frequency out of sync: cpufreq and timing " - "core thinks of %u, is %u kHz.\n", old_freq, new_freq); + pr_debug("Warning: CPU frequency out of sync: cpufreq and timing core thinks of %u, is %u kHz\n", + policy->cur, new_freq); - freqs.old = old_freq; + freqs.old = policy->cur; freqs.new = new_freq; - read_lock_irqsave(&cpufreq_driver_lock, flags); - policy = per_cpu(cpufreq_cpu_data, cpu); - read_unlock_irqrestore(&cpufreq_driver_lock, flags); + cpufreq_freq_transition_begin(policy, &freqs); + cpufreq_freq_transition_end(policy, &freqs, 0); +} + +static unsigned int cpufreq_verify_current_freq(struct cpufreq_policy *policy, bool update) +{ + unsigned int new_freq; + + if (!cpufreq_driver->get) + return 0; - cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE); - cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE); + new_freq = cpufreq_driver->get(policy->cpu); + if (!new_freq) + return 0; + + /* + * If fast frequency switching is used with the given policy, the check + * against policy->cur is pointless, so skip it in that case. + */ + if (policy->fast_switch_enabled || !has_target()) + return new_freq; + + if (policy->cur != new_freq) { + /* + * For some platforms, the frequency returned by hardware may be + * slightly different from what is provided in the frequency + * table, for example hardware may return 499 MHz instead of 500 + * MHz. In such cases it is better to avoid getting into + * unnecessary frequency updates. + */ + if (abs(policy->cur - new_freq) < KHZ_PER_MHZ) + return policy->cur; + + cpufreq_out_of_sync(policy, new_freq); + if (update) + schedule_work(&policy->update); + } + + return new_freq; } /** @@ -1275,19 +1848,25 @@ static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, */ unsigned int cpufreq_quick_get(unsigned int cpu) { - struct cpufreq_policy *policy; - unsigned int ret_freq = 0; + unsigned long flags; - if (cpufreq_driver && cpufreq_driver->setpolicy && cpufreq_driver->get) - return cpufreq_driver->get(cpu); + read_lock_irqsave(&cpufreq_driver_lock, flags); - policy = cpufreq_cpu_get(cpu); - if (policy) { - ret_freq = policy->cur; - cpufreq_cpu_put(policy); + if (cpufreq_driver && cpufreq_driver->setpolicy && cpufreq_driver->get) { + unsigned int ret_freq = cpufreq_driver->get(cpu); + + read_unlock_irqrestore(&cpufreq_driver_lock, flags); + + return ret_freq; } - return ret_freq; + read_unlock_irqrestore(&cpufreq_driver_lock, flags); + + struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); + if (policy) + return policy->cur; + + return 0; } EXPORT_SYMBOL(cpufreq_quick_get); @@ -1299,39 +1878,36 @@ 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) = cpufreq_cpu_get(cpu); + if (policy) + return policy->max; - if (policy) { - ret_freq = policy->max; - cpufreq_cpu_put(policy); - } - - return ret_freq; + return 0; } EXPORT_SYMBOL(cpufreq_quick_get_max); -static unsigned int __cpufreq_get(unsigned int cpu) +/** + * cpufreq_get_hw_max_freq - get the max hardware frequency of the CPU + * @cpu: CPU number + * + * The default return value is the max_freq field of cpuinfo. + */ +__weak unsigned int cpufreq_get_hw_max_freq(unsigned int cpu) { - struct cpufreq_policy *policy = per_cpu(cpufreq_cpu_data, cpu); - unsigned int ret_freq = 0; + struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); + if (policy) + return policy->cpuinfo.max_freq; - if (!cpufreq_driver->get) - return ret_freq; - - ret_freq = cpufreq_driver->get(cpu); + return 0; +} +EXPORT_SYMBOL(cpufreq_get_hw_max_freq); - if (ret_freq && policy->cur && - !(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) { - /* verify no discrepancy between actual and - saved value exists */ - if (unlikely(ret_freq != policy->cur)) { - cpufreq_out_of_sync(cpu, policy->cur, ret_freq); - schedule_work(&policy->update); - } - } +static unsigned int __cpufreq_get(struct cpufreq_policy *policy) +{ + if (unlikely(policy_is_inactive(policy))) + return 0; - return ret_freq; + return cpufreq_verify_current_freq(policy, true); } /** @@ -1342,23 +1918,13 @@ static unsigned int __cpufreq_get(unsigned int cpu) */ unsigned int cpufreq_get(unsigned int cpu) { - unsigned int ret_freq = 0; - struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - + struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); if (!policy) - goto out; - - if (unlikely(lock_policy_rwsem_read(cpu))) - goto out_policy; - - ret_freq = __cpufreq_get(cpu); + return 0; - unlock_policy_rwsem_read(cpu); + guard(cpufreq_policy_read)(policy); -out_policy: - cpufreq_cpu_put(policy); -out: - return ret_freq; + return __cpufreq_get(policy); } EXPORT_SYMBOL(cpufreq_get); @@ -1369,89 +1935,125 @@ static struct subsys_interface cpufreq_interface = { .remove_dev = cpufreq_remove_dev, }; +/* + * In case platform wants some specific frequency to be configured + * during suspend.. + */ +int cpufreq_generic_suspend(struct cpufreq_policy *policy) +{ + int ret; + + if (!policy->suspend_freq) { + pr_debug("%s: suspend_freq not defined\n", __func__); + return 0; + } + + pr_debug("%s: Setting suspend-freq: %u\n", __func__, + policy->suspend_freq); + + ret = __cpufreq_driver_target(policy, policy->suspend_freq, + CPUFREQ_RELATION_H); + if (ret) + pr_err("%s: unable to set suspend-freq: %u. err: %d\n", + __func__, policy->suspend_freq, ret); + + return ret; +} +EXPORT_SYMBOL(cpufreq_generic_suspend); + /** - * cpufreq_bp_suspend - Prepare the boot CPU for system suspend. + * cpufreq_suspend() - Suspend CPUFreq governors. * - * This function is only executed for the boot processor. The other CPUs - * have been put offline by means of CPU hotplug. + * Called during system wide Suspend/Hibernate cycles for suspending governors + * as some platforms can't change frequency after this point in suspend cycle. + * Because some of the devices (like: i2c, regulators, etc) they use for + * changing frequency are suspended quickly after this point. */ -static int cpufreq_bp_suspend(void) +void cpufreq_suspend(void) { - int ret = 0; + struct cpufreq_policy *policy; - int cpu = smp_processor_id(); - struct cpufreq_policy *cpu_policy; + if (!cpufreq_driver) + return; - pr_debug("suspending cpu %u\n", cpu); + if (!has_target() && !cpufreq_driver->suspend) + goto suspend; - /* If there's no policy for the boot CPU, we have nothing to do. */ - cpu_policy = cpufreq_cpu_get(cpu); - if (!cpu_policy) - return 0; + pr_debug("%s: Suspending Governors\n", __func__); - if (cpufreq_driver->suspend) { - ret = cpufreq_driver->suspend(cpu_policy); - if (ret) - printk(KERN_ERR "cpufreq: suspend failed in ->suspend " - "step on CPU %u\n", cpu_policy->cpu); + for_each_active_policy(policy) { + if (has_target()) { + scoped_guard(cpufreq_policy_write, policy) { + cpufreq_stop_governor(policy); + } + } + + if (cpufreq_driver->suspend && cpufreq_driver->suspend(policy)) + pr_err("%s: Failed to suspend driver: %s\n", __func__, + cpufreq_driver->name); } - cpufreq_cpu_put(cpu_policy); - return ret; +suspend: + cpufreq_suspended = true; } /** - * cpufreq_bp_resume - Restore proper frequency handling of the boot CPU. - * - * 1.) resume CPUfreq hardware support (cpufreq_driver->resume()) - * 2.) schedule call cpufreq_update_policy() ASAP as interrupts are - * restored. It will verify that the current freq is in sync with - * what we believe it to be. This is a bit later than when it - * should be, but nonethteless it's better than calling - * cpufreq_driver->get() here which might re-enable interrupts... + * cpufreq_resume() - Resume CPUFreq governors. * - * This function is only executed for the boot CPU. The other CPUs have not - * been turned on yet. + * Called during system wide Suspend/Hibernate cycle for resuming governors that + * are suspended with cpufreq_suspend(). */ -static void cpufreq_bp_resume(void) +void cpufreq_resume(void) { - int ret = 0; + struct cpufreq_policy *policy; + int ret; - int cpu = smp_processor_id(); - struct cpufreq_policy *cpu_policy; + if (!cpufreq_driver) + return; - pr_debug("resuming cpu %u\n", cpu); + if (unlikely(!cpufreq_suspended)) + return; + + cpufreq_suspended = false; - /* If there's no policy for the boot CPU, we have nothing to do. */ - cpu_policy = cpufreq_cpu_get(cpu); - if (!cpu_policy) + if (!has_target() && !cpufreq_driver->resume) return; - if (cpufreq_driver->resume) { - ret = cpufreq_driver->resume(cpu_policy); - if (ret) { - printk(KERN_ERR "cpufreq: resume failed in ->resume " - "step on CPU %u\n", cpu_policy->cpu); - goto fail; - } - } + pr_debug("%s: Resuming Governors\n", __func__); - schedule_work(&cpu_policy->update); + for_each_active_policy(policy) { + if (cpufreq_driver->resume && cpufreq_driver->resume(policy)) { + pr_err("%s: Failed to resume driver: %s\n", __func__, + cpufreq_driver->name); + } else if (has_target()) { + scoped_guard(cpufreq_policy_write, policy) { + ret = cpufreq_start_governor(policy); + } -fail: - cpufreq_cpu_put(cpu_policy); + if (ret) + pr_err("%s: Failed to start governor for CPU%u's policy\n", + __func__, policy->cpu); + } + } } -static struct syscore_ops cpufreq_syscore_ops = { - .suspend = cpufreq_bp_suspend, - .resume = cpufreq_bp_resume, -}; +/** + * cpufreq_driver_test_flags - Test cpufreq driver's flags against given ones. + * @flags: Flags to test against the current cpufreq driver's flags. + * + * Assumes that the driver is there, so callers must ensure that this is the + * case. + */ +bool cpufreq_driver_test_flags(u16 flags) +{ + return !!(cpufreq_driver->flags & flags); +} /** - * cpufreq_get_current_driver - return current driver's name + * cpufreq_get_current_driver - Return the current driver's name. * - * Return the name string of the currently loaded cpufreq driver - * or NULL, if none. + * Return the name string of the currently registered cpufreq driver or NULL if + * none. */ const char *cpufreq_get_current_driver(void) { @@ -1462,22 +2064,36 @@ const char *cpufreq_get_current_driver(void) } EXPORT_SYMBOL_GPL(cpufreq_get_current_driver); +/** + * cpufreq_get_driver_data - Return current driver data. + * + * Return the private data of the currently registered cpufreq driver, or NULL + * if no cpufreq driver has been registered. + */ +void *cpufreq_get_driver_data(void) +{ + if (cpufreq_driver) + return cpufreq_driver->driver_data; + + return NULL; +} +EXPORT_SYMBOL_GPL(cpufreq_get_driver_data); + /********************************************************************* * NOTIFIER LISTS INTERFACE * *********************************************************************/ /** - * cpufreq_register_notifier - register a driver with cpufreq - * @nb: notifier function to register - * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER + * cpufreq_register_notifier - Register a notifier with cpufreq. + * @nb: notifier function to register. + * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER. * - * Add a driver to one of two lists: either a list of drivers that - * are notified about clock rate changes (once before and once after - * the transition), or a list of drivers that are notified about - * changes in cpufreq policy. + * Add a notifier to one of two lists: either a list of notifiers that run on + * clock rate changes (once before and once after every transition), or a list + * of notifiers that ron on cpufreq policy changes. * - * This function may sleep, and has the same return conditions as - * blocking_notifier_chain_register. + * This function may sleep and it has the same return values as + * blocking_notifier_chain_register(). */ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) { @@ -1486,12 +2102,20 @@ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) if (cpufreq_disabled()) return -EINVAL; - WARN_ON(!init_cpufreq_transition_notifier_list_called); - switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: + mutex_lock(&cpufreq_fast_switch_lock); + + if (cpufreq_fast_switch_count > 0) { + mutex_unlock(&cpufreq_fast_switch_lock); + return -EBUSY; + } ret = srcu_notifier_chain_register( &cpufreq_transition_notifier_list, nb); + if (!ret) + cpufreq_fast_switch_count--; + + mutex_unlock(&cpufreq_fast_switch_lock); break; case CPUFREQ_POLICY_NOTIFIER: ret = blocking_notifier_chain_register( @@ -1506,14 +2130,14 @@ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) EXPORT_SYMBOL(cpufreq_register_notifier); /** - * cpufreq_unregister_notifier - unregister a driver with cpufreq - * @nb: notifier block to be unregistered - * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER + * cpufreq_unregister_notifier - Unregister a notifier from cpufreq. + * @nb: notifier block to be unregistered. + * @list: CPUFREQ_TRANSITION_NOTIFIER or CPUFREQ_POLICY_NOTIFIER. * - * Remove a driver from the CPU frequency notifier list. + * Remove a notifier from one of the cpufreq notifier lists. * - * This function may sleep, and has the same return conditions as - * blocking_notifier_chain_unregister. + * This function may sleep and it has the same return values as + * blocking_notifier_chain_unregister(). */ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list) { @@ -1524,8 +2148,14 @@ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list) switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: + mutex_lock(&cpufreq_fast_switch_lock); + ret = srcu_notifier_chain_unregister( &cpufreq_transition_notifier_list, nb); + if (!ret && !WARN_ON(cpufreq_fast_switch_count >= 0)) + cpufreq_fast_switch_count++; + + mutex_unlock(&cpufreq_fast_switch_lock); break; case CPUFREQ_POLICY_NOTIFIER: ret = blocking_notifier_chain_unregister( @@ -1544,34 +2174,221 @@ EXPORT_SYMBOL(cpufreq_unregister_notifier); * GOVERNORS * *********************************************************************/ +/** + * cpufreq_driver_fast_switch - Carry out a fast CPU frequency switch. + * @policy: cpufreq policy to switch the frequency for. + * @target_freq: New frequency to set (may be approximate). + * + * Carry out a fast frequency switch without sleeping. + * + * The driver's ->fast_switch() callback invoked by this function must be + * suitable for being called from within RCU-sched read-side critical sections + * and it is expected to select the minimum available frequency greater than or + * equal to @target_freq (CPUFREQ_RELATION_L). + * + * This function must not be called if policy->fast_switch_enabled is unset. + * + * Governors calling this function must guarantee that it will never be invoked + * twice in parallel for the same policy and that it will never be called in + * parallel with either ->target() or ->target_index() for the same policy. + * + * Returns the actual frequency set for the CPU. + * + * If 0 is returned by the driver's ->fast_switch() callback to indicate an + * error condition, the hardware configuration must be preserved. + */ +unsigned int cpufreq_driver_fast_switch(struct cpufreq_policy *policy, + unsigned int target_freq) +{ + unsigned int freq; + int cpu; + + target_freq = clamp_val(target_freq, policy->min, policy->max); + freq = cpufreq_driver->fast_switch(policy, target_freq); + + if (!freq) + return 0; + + policy->cur = freq; + arch_set_freq_scale(policy->related_cpus, freq, + arch_scale_freq_ref(policy->cpu)); + cpufreq_stats_record_transition(policy, freq); + + if (trace_cpu_frequency_enabled()) { + for_each_cpu(cpu, policy->cpus) + trace_cpu_frequency(freq, cpu); + } + + return freq; +} +EXPORT_SYMBOL_GPL(cpufreq_driver_fast_switch); + +/** + * cpufreq_driver_adjust_perf - Adjust CPU performance level in one go. + * @cpu: Target CPU. + * @min_perf: Minimum (required) performance level (units of @capacity). + * @target_perf: Target (desired) performance level (units of @capacity). + * @capacity: Capacity of the target CPU. + * + * Carry out a fast performance level switch of @cpu without sleeping. + * + * The driver's ->adjust_perf() callback invoked by this function must be + * suitable for being called from within RCU-sched read-side critical sections + * and it is expected to select a suitable performance level equal to or above + * @min_perf and preferably equal to or below @target_perf. + * + * This function must not be called if policy->fast_switch_enabled is unset. + * + * Governors calling this function must guarantee that it will never be invoked + * twice in parallel for the same CPU and that it will never be called in + * parallel with either ->target() or ->target_index() or ->fast_switch() for + * the same CPU. + */ +void cpufreq_driver_adjust_perf(unsigned int cpu, + unsigned long min_perf, + unsigned long target_perf, + unsigned long capacity) +{ + cpufreq_driver->adjust_perf(cpu, min_perf, target_perf, capacity); +} + +/** + * cpufreq_driver_has_adjust_perf - Check "direct fast switch" callback. + * + * Return 'true' if the ->adjust_perf callback is present for the + * current driver or 'false' otherwise. + */ +bool cpufreq_driver_has_adjust_perf(void) +{ + return !!cpufreq_driver->adjust_perf; +} + +/* Must set freqs->new to intermediate frequency */ +static int __target_intermediate(struct cpufreq_policy *policy, + struct cpufreq_freqs *freqs, int index) +{ + int ret; + + freqs->new = cpufreq_driver->get_intermediate(policy, index); + + /* We don't need to switch to intermediate freq */ + if (!freqs->new) + return 0; + + pr_debug("%s: cpu: %d, switching to intermediate freq: oldfreq: %u, intermediate freq: %u\n", + __func__, policy->cpu, freqs->old, freqs->new); + + cpufreq_freq_transition_begin(policy, freqs); + ret = cpufreq_driver->target_intermediate(policy, index); + cpufreq_freq_transition_end(policy, freqs, ret); + + if (ret) + pr_err("%s: Failed to change to intermediate frequency: %d\n", + __func__, ret); + + return ret; +} + +static int __target_index(struct cpufreq_policy *policy, int index) +{ + struct cpufreq_freqs freqs = {.old = policy->cur, .flags = 0}; + unsigned int restore_freq, intermediate_freq = 0; + unsigned int newfreq = policy->freq_table[index].frequency; + int retval = -EINVAL; + bool notify; + + if (newfreq == policy->cur) + return 0; + + /* Save last value to restore later on errors */ + restore_freq = policy->cur; + + notify = !(cpufreq_driver->flags & CPUFREQ_ASYNC_NOTIFICATION); + if (notify) { + /* Handle switching to intermediate frequency */ + if (cpufreq_driver->get_intermediate) { + retval = __target_intermediate(policy, &freqs, index); + if (retval) + return retval; + + intermediate_freq = freqs.new; + /* Set old freq to intermediate */ + if (intermediate_freq) + freqs.old = freqs.new; + } + + freqs.new = newfreq; + pr_debug("%s: cpu: %d, oldfreq: %u, new freq: %u\n", + __func__, policy->cpu, freqs.old, freqs.new); + + cpufreq_freq_transition_begin(policy, &freqs); + } + + retval = cpufreq_driver->target_index(policy, index); + if (retval) + pr_err("%s: Failed to change cpu frequency: %d\n", __func__, + retval); + + if (notify) { + cpufreq_freq_transition_end(policy, &freqs, retval); + + /* + * Failed after setting to intermediate freq? Driver should have + * reverted back to initial frequency and so should we. Check + * here for intermediate_freq instead of get_intermediate, in + * case we haven't switched to intermediate freq at all. + */ + if (unlikely(retval && intermediate_freq)) { + freqs.old = intermediate_freq; + freqs.new = restore_freq; + cpufreq_freq_transition_begin(policy, &freqs); + cpufreq_freq_transition_end(policy, &freqs, 0); + } + } + + return retval; +} + int __cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - int retval = -EINVAL; unsigned int old_target_freq = target_freq; if (cpufreq_disabled()) return -ENODEV; - if (policy->transition_ongoing) - return -EBUSY; - /* Make sure that target_freq is within supported range */ - if (target_freq > policy->max) - target_freq = policy->max; - if (target_freq < policy->min) - target_freq = policy->min; + target_freq = __resolve_freq(policy, target_freq, policy->min, + policy->max, relation); pr_debug("target for CPU %u: %u kHz, relation %u, requested %u kHz\n", - policy->cpu, target_freq, relation, old_target_freq); + policy->cpu, target_freq, relation, old_target_freq); - if (target_freq == policy->cur) + /* + * This might look like a redundant call as we are checking it again + * after finding index. But it is left intentionally for cases where + * exactly same freq is called again and so we can save on few function + * calls. + */ + if (target_freq == policy->cur && + !(cpufreq_driver->flags & CPUFREQ_NEED_UPDATE_LIMITS)) return 0; - if (cpufreq_driver->target) - retval = cpufreq_driver->target(policy, target_freq, relation); + if (cpufreq_driver->target) { + /* + * If the driver hasn't setup a single inefficient frequency, + * it's unlikely it knows how to decode CPUFREQ_RELATION_E. + */ + if (!policy->efficiencies_available) + relation &= ~CPUFREQ_RELATION_E; - return retval; + return cpufreq_driver->target(policy, target_freq, relation); + } + + if (!cpufreq_driver->target_index) + return -EINVAL; + + return __target_index(policy, policy->cached_resolved_idx); } EXPORT_SYMBOL_GPL(__cpufreq_driver_target); @@ -1579,111 +2396,122 @@ int cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - int ret = -EINVAL; - - if (unlikely(lock_policy_rwsem_write(policy->cpu))) - goto fail; - - ret = __cpufreq_driver_target(policy, target_freq, relation); - - unlock_policy_rwsem_write(policy->cpu); + guard(cpufreq_policy_write)(policy); -fail: - return ret; + return __cpufreq_driver_target(policy, target_freq, relation); } EXPORT_SYMBOL_GPL(cpufreq_driver_target); -int __cpufreq_driver_getavg(struct cpufreq_policy *policy, unsigned int cpu) +__weak struct cpufreq_governor *cpufreq_fallback_governor(void) { - if (cpufreq_disabled()) - return 0; - - if (!cpufreq_driver->getavg) - return 0; - - return cpufreq_driver->getavg(policy, cpu); + return NULL; } -EXPORT_SYMBOL_GPL(__cpufreq_driver_getavg); - -/* - * when "event" is CPUFREQ_GOV_LIMITS - */ -static int __cpufreq_governor(struct cpufreq_policy *policy, - unsigned int event) +static int cpufreq_init_governor(struct cpufreq_policy *policy) { int ret; - /* Only must be defined when default governor is known to have latency - restrictions, like e.g. conservative or ondemand. - That this is the case is already ensured in Kconfig - */ -#ifdef CONFIG_CPU_FREQ_GOV_PERFORMANCE - struct cpufreq_governor *gov = &cpufreq_gov_performance; -#else - struct cpufreq_governor *gov = NULL; -#endif + /* Don't start any governor operations if we are entering suspend */ + if (cpufreq_suspended) + return 0; + /* + * Governor might not be initiated here if ACPI _PPC changed + * notification happened, so check it. + */ + if (!policy->governor) + return -EINVAL; - if (policy->governor->max_transition_latency && - policy->cpuinfo.transition_latency > - policy->governor->max_transition_latency) { - if (!gov) - return -EINVAL; - else { - printk(KERN_WARNING "%s governor failed, too long" - " transition latency of HW, fallback" - " to %s governor\n", - policy->governor->name, - gov->name); + /* Platform doesn't want dynamic frequency switching ? */ + if (policy->governor->flags & CPUFREQ_GOV_DYNAMIC_SWITCHING && + cpufreq_driver->flags & CPUFREQ_NO_AUTO_DYNAMIC_SWITCHING) { + struct cpufreq_governor *gov = cpufreq_fallback_governor(); + + if (gov) { + pr_warn("Can't use %s governor as dynamic switching is disallowed. Fallback to %s governor\n", + policy->governor->name, gov->name); policy->governor = gov; + } else { + return -EINVAL; } } if (!try_module_get(policy->governor->owner)) return -EINVAL; - pr_debug("__cpufreq_governor for CPU %u, event %u\n", - policy->cpu, event); + pr_debug("%s: for CPU %u\n", __func__, policy->cpu); - mutex_lock(&cpufreq_governor_lock); - if ((!policy->governor_enabled && (event == CPUFREQ_GOV_STOP)) || - (policy->governor_enabled && (event == CPUFREQ_GOV_START))) { - mutex_unlock(&cpufreq_governor_lock); - return -EBUSY; + if (policy->governor->init) { + ret = policy->governor->init(policy); + if (ret) { + module_put(policy->governor->owner); + return ret; + } } - if (event == CPUFREQ_GOV_STOP) - policy->governor_enabled = false; - else if (event == CPUFREQ_GOV_START) - policy->governor_enabled = true; + policy->strict_target = !!(policy->governor->flags & CPUFREQ_GOV_STRICT_TARGET); - mutex_unlock(&cpufreq_governor_lock); + return 0; +} - ret = policy->governor->governor(policy, event); +static void cpufreq_exit_governor(struct cpufreq_policy *policy) +{ + if (cpufreq_suspended || !policy->governor) + return; - if (!ret) { - if (event == CPUFREQ_GOV_POLICY_INIT) - policy->governor->initialized++; - else if (event == CPUFREQ_GOV_POLICY_EXIT) - policy->governor->initialized--; - } else { - /* Restore original values */ - mutex_lock(&cpufreq_governor_lock); - if (event == CPUFREQ_GOV_STOP) - policy->governor_enabled = true; - else if (event == CPUFREQ_GOV_START) - policy->governor_enabled = false; - mutex_unlock(&cpufreq_governor_lock); - } - - /* we keep one module reference alive for - each CPU governed by this CPU */ - if ((event != CPUFREQ_GOV_START) || ret) - module_put(policy->governor->owner); - if ((event == CPUFREQ_GOV_STOP) && !ret) - module_put(policy->governor->owner); + pr_debug("%s: for CPU %u\n", __func__, policy->cpu); - return ret; + if (policy->governor->exit) + policy->governor->exit(policy); + + module_put(policy->governor->owner); +} + +int cpufreq_start_governor(struct cpufreq_policy *policy) +{ + int ret; + + if (cpufreq_suspended) + return 0; + + if (!policy->governor) + return -EINVAL; + + pr_debug("%s: for CPU %u\n", __func__, policy->cpu); + + cpufreq_verify_current_freq(policy, false); + + if (policy->governor->start) { + ret = policy->governor->start(policy); + if (ret) + return ret; + } + + if (policy->governor->limits) + policy->governor->limits(policy); + + return 0; +} + +void cpufreq_stop_governor(struct cpufreq_policy *policy) +{ + if (cpufreq_suspended || !policy->governor) + return; + + pr_debug("%s: for CPU %u\n", __func__, policy->cpu); + + if (policy->governor->stop) + policy->governor->stop(policy); +} + +static void cpufreq_governor_limits(struct cpufreq_policy *policy) +{ + if (cpufreq_suspended || !policy->governor) + return; + + pr_debug("%s: for CPU %u\n", __func__, policy->cpu); + + if (policy->governor->limits) + policy->governor->limits(policy); } int cpufreq_register_governor(struct cpufreq_governor *governor) @@ -1698,9 +2526,8 @@ int cpufreq_register_governor(struct cpufreq_governor *governor) mutex_lock(&cpufreq_governor_mutex); - governor->initialized = 0; err = -EBUSY; - if (__find_governor(governor->name) == NULL) { + if (!find_governor(governor->name)) { err = 0; list_add(&governor->governor_list, &cpufreq_governor_list); } @@ -1712,9 +2539,8 @@ EXPORT_SYMBOL_GPL(cpufreq_register_governor); void cpufreq_unregister_governor(struct cpufreq_governor *governor) { -#ifdef CONFIG_HOTPLUG_CPU - int cpu; -#endif + struct cpufreq_policy *policy; + unsigned long flags; if (!governor) return; @@ -1722,19 +2548,19 @@ void cpufreq_unregister_governor(struct cpufreq_governor *governor) if (cpufreq_disabled()) return; -#ifdef CONFIG_HOTPLUG_CPU - for_each_present_cpu(cpu) { - if (cpu_online(cpu)) - continue; - if (!strcmp(per_cpu(cpufreq_cpu_governor, cpu), governor->name)) - strcpy(per_cpu(cpufreq_cpu_governor, cpu), "\0"); + /* clear last_governor for all inactive policies */ + read_lock_irqsave(&cpufreq_driver_lock, flags); + for_each_inactive_policy(policy) { + if (!strcmp(policy->last_governor, governor->name)) { + policy->governor = NULL; + policy->last_governor[0] = '\0'; + } } -#endif + read_unlock_irqrestore(&cpufreq_driver_lock, flags); mutex_lock(&cpufreq_governor_mutex); list_del(&governor->governor_list); mutex_unlock(&cpufreq_governor_mutex); - return; } EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); @@ -1743,226 +2569,324 @@ EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); * POLICY INTERFACE * *********************************************************************/ +DEFINE_PER_CPU(unsigned long, cpufreq_pressure); + /** - * cpufreq_get_policy - get the current cpufreq_policy - * @policy: struct cpufreq_policy into which the current cpufreq_policy - * is written + * cpufreq_update_pressure() - Update cpufreq pressure for CPUs + * @policy: cpufreq policy of the CPUs. * - * Reads the current cpufreq policy. + * Update the value of cpufreq pressure for all @cpus in the policy. */ -int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu) +static void cpufreq_update_pressure(struct cpufreq_policy *policy) { - struct cpufreq_policy *cpu_policy; - if (!policy) - return -EINVAL; + unsigned long max_capacity, capped_freq, pressure; + u32 max_freq; + int cpu; - cpu_policy = cpufreq_cpu_get(cpu); - if (!cpu_policy) - return -EINVAL; + cpu = cpumask_first(policy->related_cpus); + max_freq = arch_scale_freq_ref(cpu); + capped_freq = policy->max; - memcpy(policy, cpu_policy, sizeof(struct cpufreq_policy)); + /* + * Handle properly the boost frequencies, which should simply clean + * the cpufreq pressure value. + */ + if (max_freq <= capped_freq) { + pressure = 0; + } else { + max_capacity = arch_scale_cpu_capacity(cpu); + pressure = max_capacity - + mult_frac(max_capacity, capped_freq, max_freq); + } - cpufreq_cpu_put(cpu_policy); - return 0; + for_each_cpu(cpu, policy->related_cpus) + WRITE_ONCE(per_cpu(cpufreq_pressure, cpu), pressure); } -EXPORT_SYMBOL(cpufreq_get_policy); -/* - * data : current policy. - * policy : policy to be set. +/** + * cpufreq_set_policy - Modify cpufreq policy parameters. + * @policy: Policy object to modify. + * @new_gov: Policy governor pointer. + * @new_pol: Policy value (for drivers with built-in governors). + * + * Invoke the cpufreq driver's ->verify() callback to sanity-check the frequency + * limits to be set for the policy, update @policy with the verified limits + * values and either invoke the driver's ->setpolicy() callback (if present) or + * carry out a governor update for @policy. That is, run the current governor's + * ->limits() callback (if @new_gov points to the same object as the one in + * @policy) or replace the governor for @policy with @new_gov. + * + * The cpuinfo part of @policy is not updated by this function. */ -static int __cpufreq_set_policy(struct cpufreq_policy *data, - struct cpufreq_policy *policy) +static int cpufreq_set_policy(struct cpufreq_policy *policy, + struct cpufreq_governor *new_gov, + unsigned int new_pol) { - int ret = 0, failed = 1; - - pr_debug("setting new policy for CPU %u: %u - %u kHz\n", policy->cpu, - policy->min, policy->max); + struct cpufreq_policy_data new_data; + struct cpufreq_governor *old_gov; + int ret; - memcpy(&policy->cpuinfo, &data->cpuinfo, - sizeof(struct cpufreq_cpuinfo)); + memcpy(&new_data.cpuinfo, &policy->cpuinfo, sizeof(policy->cpuinfo)); + new_data.freq_table = policy->freq_table; + new_data.cpu = policy->cpu; + /* + * PM QoS framework collects all the requests from users and provide us + * the final aggregated value here. + */ + new_data.min = freq_qos_read_value(&policy->constraints, FREQ_QOS_MIN); + new_data.max = freq_qos_read_value(&policy->constraints, FREQ_QOS_MAX); - if (policy->min > data->max || policy->max < data->min) { - ret = -EINVAL; - goto error_out; - } + pr_debug("setting new policy for CPU %u: %u - %u kHz\n", + new_data.cpu, new_data.min, new_data.max); - /* verify the cpu speed can be set within this limit */ - ret = cpufreq_driver->verify(policy); + /* + * Verify that the CPU speed can be set within these limits and make sure + * that min <= max. + */ + ret = cpufreq_driver->verify(&new_data); if (ret) - goto error_out; - - /* adjust if necessary - all reasons */ - blocking_notifier_call_chain(&cpufreq_policy_notifier_list, - CPUFREQ_ADJUST, policy); - - /* adjust if necessary - hardware incompatibility*/ - blocking_notifier_call_chain(&cpufreq_policy_notifier_list, - CPUFREQ_INCOMPATIBLE, policy); + return ret; /* - * verify the cpu speed can be set within this limit, which might be - * different to the first one + * Resolve policy min/max to available frequencies. It ensures + * no frequency resolution will neither overshoot the requested maximum + * nor undershoot the requested minimum. + * + * Avoid storing intermediate values in policy->max or policy->min and + * compiler optimizations around them because they may be accessed + * concurrently by cpufreq_driver_resolve_freq() during the update. */ - ret = cpufreq_driver->verify(policy); - if (ret) - goto error_out; + WRITE_ONCE(policy->max, __resolve_freq(policy, new_data.max, + new_data.min, new_data.max, + CPUFREQ_RELATION_H)); + new_data.min = __resolve_freq(policy, new_data.min, new_data.min, + new_data.max, CPUFREQ_RELATION_L); + WRITE_ONCE(policy->min, new_data.min > policy->max ? policy->max : new_data.min); - /* notification of the new policy */ - blocking_notifier_call_chain(&cpufreq_policy_notifier_list, - CPUFREQ_NOTIFY, policy); + trace_cpu_frequency_limits(policy); - data->min = policy->min; - data->max = policy->max; + cpufreq_update_pressure(policy); + + policy->cached_target_freq = UINT_MAX; pr_debug("new min and max freqs are %u - %u kHz\n", - data->min, data->max); + policy->min, policy->max); if (cpufreq_driver->setpolicy) { - data->policy = policy->policy; + policy->policy = new_pol; pr_debug("setting range\n"); - ret = cpufreq_driver->setpolicy(policy); - } else { - if (policy->governor != data->governor) { - /* save old, working values */ - struct cpufreq_governor *old_gov = data->governor; - - pr_debug("governor switch\n"); - - /* end old governor */ - if (data->governor) { - __cpufreq_governor(data, CPUFREQ_GOV_STOP); - unlock_policy_rwsem_write(policy->cpu); - __cpufreq_governor(data, - CPUFREQ_GOV_POLICY_EXIT); - lock_policy_rwsem_write(policy->cpu); - } + return cpufreq_driver->setpolicy(policy); + } - /* start new governor */ - data->governor = policy->governor; - if (!__cpufreq_governor(data, CPUFREQ_GOV_POLICY_INIT)) { - if (!__cpufreq_governor(data, CPUFREQ_GOV_START)) { - failed = 0; - } else { - unlock_policy_rwsem_write(policy->cpu); - __cpufreq_governor(data, - CPUFREQ_GOV_POLICY_EXIT); - lock_policy_rwsem_write(policy->cpu); - } - } + if (new_gov == policy->governor) { + pr_debug("governor limits update\n"); + cpufreq_governor_limits(policy); + return 0; + } - if (failed) { - /* new governor failed, so re-start old one */ - pr_debug("starting governor %s failed\n", - data->governor->name); - if (old_gov) { - data->governor = old_gov; - __cpufreq_governor(data, - CPUFREQ_GOV_POLICY_INIT); - __cpufreq_governor(data, - CPUFREQ_GOV_START); - } - ret = -EINVAL; - goto error_out; - } - /* might be a policy change, too, so fall through */ + pr_debug("governor switch\n"); + + /* save old, working values */ + old_gov = policy->governor; + /* end old governor */ + if (old_gov) { + cpufreq_stop_governor(policy); + cpufreq_exit_governor(policy); + } + + /* start new governor */ + policy->governor = new_gov; + ret = cpufreq_init_governor(policy); + if (!ret) { + ret = cpufreq_start_governor(policy); + if (!ret) { + pr_debug("governor change\n"); + return 0; + } + cpufreq_exit_governor(policy); + } + + /* new governor failed, so re-start old one */ + pr_debug("starting governor %s failed\n", policy->governor->name); + if (old_gov) { + policy->governor = old_gov; + if (cpufreq_init_governor(policy)) { + policy->governor = NULL; + } else if (cpufreq_start_governor(policy)) { + cpufreq_exit_governor(policy); + policy->governor = NULL; } - pr_debug("governor: change or update limits\n"); - __cpufreq_governor(data, CPUFREQ_GOV_LIMITS); } -error_out: 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. + * + * Update the current frequency for the cpufreq policy of @cpu and use + * cpufreq_set_policy() to re-apply the min and max limits, which triggers the + * evaluation of policy notifiers and the cpufreq driver's ->verify() callback + * for the policy in question, among other things. + */ +void cpufreq_update_policy(unsigned int cpu) +{ + struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); + if (!policy) + return; + + cpufreq_policy_refresh(policy); +} +EXPORT_SYMBOL(cpufreq_update_policy); + /** - * cpufreq_update_policy - re-evaluate an existing cpufreq policy - * @cpu: CPU which shall be re-evaluated + * cpufreq_update_limits - Update policy limits for a given CPU. + * @cpu: CPU to update the policy limits for. * - * Useful for policy notifiers which have different necessities - * at different times. + * Invoke the driver's ->update_limits callback if present or call + * cpufreq_policy_refresh() for @cpu. */ -int cpufreq_update_policy(unsigned int cpu) +void cpufreq_update_limits(unsigned int cpu) +{ + struct cpufreq_policy *policy __free(put_cpufreq_policy) = cpufreq_cpu_get(cpu); + if (!policy) + return; + + if (cpufreq_driver->update_limits) + cpufreq_driver->update_limits(policy); + else + cpufreq_policy_refresh(policy); +} +EXPORT_SYMBOL_GPL(cpufreq_update_limits); + +/********************************************************************* + * BOOST * + *********************************************************************/ +int cpufreq_boost_set_sw(struct cpufreq_policy *policy, int state) { - struct cpufreq_policy *data = cpufreq_cpu_get(cpu); - struct cpufreq_policy policy; int ret; - if (!data) { - ret = -ENODEV; - goto no_policy; - } + if (!policy->freq_table) + return -ENXIO; - if (unlikely(lock_policy_rwsem_write(cpu))) { - ret = -EINVAL; - goto fail; + ret = cpufreq_frequency_table_cpuinfo(policy); + if (ret) { + pr_err("%s: Policy frequency update failed\n", __func__); + return ret; } - pr_debug("updating policy for CPU %u\n", cpu); - memcpy(&policy, data, sizeof(struct cpufreq_policy)); - policy.min = data->user_policy.min; - policy.max = data->user_policy.max; - policy.policy = data->user_policy.policy; - policy.governor = data->user_policy.governor; + ret = freq_qos_update_request(policy->max_freq_req, policy->max); + if (ret < 0) + return ret; + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_boost_set_sw); + +static int cpufreq_boost_trigger_state(int state) +{ + struct cpufreq_policy *policy; + unsigned long flags; + int ret = 0; /* - * BIOS might change freq behind our back - * -> ask driver for current freq and notify governors about a change + * Don't compare 'cpufreq_driver->boost_enabled' with 'state' here to + * make sure all policies are in sync with global boost flag. */ - if (cpufreq_driver->get) { - policy.cur = cpufreq_driver->get(cpu); - if (!data->cur) { - pr_debug("Driver did not initialize current freq"); - data->cur = policy.cur; - } else { - if (data->cur != policy.cur && cpufreq_driver->target) - cpufreq_out_of_sync(cpu, data->cur, - policy.cur); - } + + write_lock_irqsave(&cpufreq_driver_lock, flags); + cpufreq_driver->boost_enabled = state; + write_unlock_irqrestore(&cpufreq_driver_lock, flags); + + cpus_read_lock(); + for_each_active_policy(policy) { + if (!policy->boost_supported) + continue; + + ret = policy_set_boost(policy, state); + if (ret) + goto err_reset_state; } + cpus_read_unlock(); - ret = __cpufreq_set_policy(data, &policy); + return 0; + +err_reset_state: + cpus_read_unlock(); + + write_lock_irqsave(&cpufreq_driver_lock, flags); + cpufreq_driver->boost_enabled = !state; + write_unlock_irqrestore(&cpufreq_driver_lock, flags); - unlock_policy_rwsem_write(cpu); + pr_err("%s: Cannot %s BOOST\n", + __func__, str_enable_disable(state)); -fail: - cpufreq_cpu_put(data); -no_policy: return ret; } -EXPORT_SYMBOL(cpufreq_update_policy); -static int __cpuinit cpufreq_cpu_callback(struct notifier_block *nfb, - unsigned long action, void *hcpu) +static bool cpufreq_boost_supported(void) { - unsigned int cpu = (unsigned long)hcpu; - struct device *dev; + return cpufreq_driver->set_boost; +} - dev = get_cpu_device(cpu); - if (dev) { - switch (action) { - case CPU_ONLINE: - cpufreq_add_dev(dev, NULL); - break; - case CPU_DOWN_PREPARE: - case CPU_UP_CANCELED_FROZEN: - __cpufreq_remove_dev(dev, NULL); - break; - case CPU_DOWN_FAILED: - cpufreq_add_dev(dev, NULL); - break; - } - } - return NOTIFY_OK; +static int create_boost_sysfs_file(void) +{ + int ret; + + ret = sysfs_create_file(cpufreq_global_kobject, &boost.attr); + if (ret) + pr_err("%s: cannot register global BOOST sysfs file\n", + __func__); + + return ret; } -static struct notifier_block __refdata cpufreq_cpu_notifier = { - .notifier_call = cpufreq_cpu_callback, -}; +static void remove_boost_sysfs_file(void) +{ + if (cpufreq_boost_supported()) + sysfs_remove_file(cpufreq_global_kobject, &boost.attr); +} + +bool cpufreq_boost_enabled(void) +{ + return cpufreq_driver->boost_enabled; +} +EXPORT_SYMBOL_GPL(cpufreq_boost_enabled); /********************************************************************* * REGISTER / UNREGISTER CPUFREQ DRIVER * *********************************************************************/ +static enum cpuhp_state hp_online; + +static int cpuhp_cpufreq_online(unsigned int cpu) +{ + cpufreq_online(cpu); + + return 0; +} + +static int cpuhp_cpufreq_offline(unsigned int cpu) +{ + cpufreq_offline(cpu); + + return 0; +} /** * cpufreq_register_driver - register a CPU Frequency driver @@ -1970,7 +2894,7 @@ static struct notifier_block __refdata cpufreq_cpu_notifier = { * submitted by the CPU Frequency driver. * * Registers a CPU Frequency driver to this core code. This code - * returns zero on success, -EBUSY when another driver got here first + * returns zero on success, -EEXIST when another driver got here first * (and isn't unregistered in the meantime). * */ @@ -1982,61 +2906,94 @@ int cpufreq_register_driver(struct cpufreq_driver *driver_data) if (cpufreq_disabled()) return -ENODEV; + /* + * The cpufreq core depends heavily on the availability of device + * structure, make sure they are available before proceeding further. + */ + if (!get_cpu_device(0)) + return -EPROBE_DEFER; + if (!driver_data || !driver_data->verify || !driver_data->init || - ((!driver_data->setpolicy) && (!driver_data->target))) + (driver_data->target_index && driver_data->target) || + (!!driver_data->setpolicy == (driver_data->target_index || driver_data->target)) || + (!driver_data->get_intermediate != !driver_data->target_intermediate) || + (!driver_data->online != !driver_data->offline) || + (driver_data->adjust_perf && !driver_data->fast_switch)) return -EINVAL; pr_debug("trying to register driver %s\n", driver_data->name); - if (driver_data->setpolicy) - driver_data->flags |= CPUFREQ_CONST_LOOPS; + /* Protect against concurrent CPU online/offline. */ + cpus_read_lock(); write_lock_irqsave(&cpufreq_driver_lock, flags); if (cpufreq_driver) { write_unlock_irqrestore(&cpufreq_driver_lock, flags); - return -EBUSY; + ret = -EEXIST; + goto out; } cpufreq_driver = driver_data; write_unlock_irqrestore(&cpufreq_driver_lock, flags); - ret = subsys_interface_register(&cpufreq_interface); - if (ret) - goto err_null_driver; + if (driver_data->setpolicy) + driver_data->flags |= CPUFREQ_CONST_LOOPS; - if (!(cpufreq_driver->flags & CPUFREQ_STICKY)) { - int i; - ret = -ENODEV; + if (cpufreq_boost_supported()) { + ret = create_boost_sysfs_file(); + if (ret) + goto err_null_driver; + } - /* check for at least one working CPU */ - for (i = 0; i < nr_cpu_ids; i++) - if (cpu_possible(i) && per_cpu(cpufreq_cpu_data, i)) { - ret = 0; - break; - } + /* + * Mark support for the scheduler's frequency invariance engine for + * drivers that implement target(), target_index() or fast_switch(). + */ + if (!cpufreq_driver->setpolicy) { + static_branch_enable_cpuslocked(&cpufreq_freq_invariance); + pr_debug("cpufreq: supports frequency invariance\n"); + } + ret = subsys_interface_register(&cpufreq_interface); + if (ret) + goto err_boost_unreg; + + if (unlikely(list_empty(&cpufreq_policy_list))) { /* if all ->init() calls failed, unregister */ - if (ret) { - pr_debug("no CPU initialized for driver %s\n", - driver_data->name); - goto err_if_unreg; - } + ret = -ENODEV; + pr_debug("%s: No CPU initialized for driver %s\n", __func__, + driver_data->name); + goto err_if_unreg; } - register_hotcpu_notifier(&cpufreq_cpu_notifier); + ret = cpuhp_setup_state_nocalls_cpuslocked(CPUHP_AP_ONLINE_DYN, + "cpufreq:online", + cpuhp_cpufreq_online, + cpuhp_cpufreq_offline); + if (ret < 0) + goto err_if_unreg; + hp_online = ret; + ret = 0; + pr_debug("driver %s up and running\n", driver_data->name); + goto out; - return 0; err_if_unreg: subsys_interface_unregister(&cpufreq_interface); +err_boost_unreg: + if (!cpufreq_driver->setpolicy) + static_branch_disable_cpuslocked(&cpufreq_freq_invariance); + remove_boost_sysfs_file(); err_null_driver: write_lock_irqsave(&cpufreq_driver_lock, flags); cpufreq_driver = NULL; write_unlock_irqrestore(&cpufreq_driver_lock, flags); +out: + cpus_read_unlock(); return ret; } EXPORT_SYMBOL_GPL(cpufreq_register_driver); -/** +/* * cpufreq_unregister_driver - unregister the current CPUFreq driver * * Unregister the current CPUFreq driver. Only call this if you have @@ -2044,42 +3001,79 @@ EXPORT_SYMBOL_GPL(cpufreq_register_driver); * Returns zero if successful, and -EINVAL if the cpufreq_driver is * currently not initialised. */ -int cpufreq_unregister_driver(struct cpufreq_driver *driver) +void cpufreq_unregister_driver(struct cpufreq_driver *driver) { unsigned long flags; - if (!cpufreq_driver || (driver != cpufreq_driver)) - return -EINVAL; + if (WARN_ON(!cpufreq_driver || (driver != cpufreq_driver))) + return; pr_debug("unregistering driver %s\n", driver->name); + /* Protect against concurrent cpu hotplug */ + cpus_read_lock(); subsys_interface_unregister(&cpufreq_interface); - unregister_hotcpu_notifier(&cpufreq_cpu_notifier); + remove_boost_sysfs_file(); + static_branch_disable_cpuslocked(&cpufreq_freq_invariance); + cpuhp_remove_state_nocalls_cpuslocked(hp_online); write_lock_irqsave(&cpufreq_driver_lock, flags); + cpufreq_driver = NULL; - write_unlock_irqrestore(&cpufreq_driver_lock, flags); - return 0; + write_unlock_irqrestore(&cpufreq_driver_lock, flags); + cpus_read_unlock(); } EXPORT_SYMBOL_GPL(cpufreq_unregister_driver); static int __init cpufreq_core_init(void) { - int cpu; + struct cpufreq_governor *gov = cpufreq_default_governor(); + struct device *dev_root; if (cpufreq_disabled()) return -ENODEV; - for_each_possible_cpu(cpu) { - per_cpu(cpufreq_policy_cpu, cpu) = -1; - init_rwsem(&per_cpu(cpu_policy_rwsem, cpu)); + dev_root = bus_get_dev_root(&cpu_subsys); + if (dev_root) { + cpufreq_global_kobject = kobject_create_and_add("cpufreq", &dev_root->kobj); + put_device(dev_root); } - - cpufreq_global_kobject = kobject_create(); BUG_ON(!cpufreq_global_kobject); - register_syscore_ops(&cpufreq_syscore_ops); + + if (!strlen(default_governor)) + strscpy(default_governor, gov->name, CPUFREQ_NAME_LEN); return 0; } + +static bool cpufreq_policy_is_good_for_eas(unsigned int cpu) +{ + struct cpufreq_policy *policy __free(put_cpufreq_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); |