diff options
author | Linus Torvalds <torvalds@ppc970.osdl.org> | 2005-04-16 15:20:36 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@ppc970.osdl.org> | 2005-04-16 15:20:36 -0700 |
commit | 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch) | |
tree | 0bba044c4ce775e45a88a51686b5d9f90697ea9d /drivers/cpufreq | |
download | lwn-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.tar.gz lwn-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.zip |
Linux-2.6.12-rc2v2.6.12-rc2
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Diffstat (limited to 'drivers/cpufreq')
-rw-r--r-- | drivers/cpufreq/Kconfig | 118 | ||||
-rw-r--r-- | drivers/cpufreq/Makefile | 14 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq.c | 1428 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_ondemand.c | 491 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_performance.c | 61 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_powersave.c | 59 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_stats.c | 334 | ||||
-rw-r--r-- | drivers/cpufreq/cpufreq_userspace.c | 207 | ||||
-rw-r--r-- | drivers/cpufreq/freq_table.c | 225 |
9 files changed, 2937 insertions, 0 deletions
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig new file mode 100644 index 000000000000..95882bb1950e --- /dev/null +++ b/drivers/cpufreq/Kconfig @@ -0,0 +1,118 @@ +config CPU_FREQ + bool "CPU Frequency scaling" + help + CPU Frequency scaling allows you to change the clock speed of + CPUs on the fly. This is a nice method to save power, because + the lower the CPU clock speed, the less power the CPU consumes. + + Note that this driver doesn't automatically change the CPU + clock speed, you need to either enable a dynamic cpufreq governor + (see below) after boot, or use a userspace tool. + + For details, take a look at <file:Documentation/cpu-freq>. + + If in doubt, say N. + +if CPU_FREQ + +config CPU_FREQ_TABLE + def_tristate m + +config CPU_FREQ_DEBUG + bool "Enable CPUfreq debugging" + help + Say Y here to enable CPUfreq subsystem (including drivers) + debugging. You will need to activate it via the kernel + command line by passing + cpufreq.debug=<value> + + To get <value>, add + 1 to activate CPUfreq core debugging, + 2 to activate CPUfreq drivers debugging, and + 4 to activate CPUfreq governor debugging + +config CPU_FREQ_STAT + tristate "CPU frequency translation statistics" + select CPU_FREQ_TABLE + default y + help + This driver exports CPU frequency statistics information through sysfs + file system + +config CPU_FREQ_STAT_DETAILS + bool "CPU frequency translation statistics details" + depends on CPU_FREQ_STAT + help + This will show detail CPU frequency translation table in sysfs file + system + +choice + prompt "Default CPUFreq governor" + default CPU_FREQ_DEFAULT_GOV_USERSPACE if CPU_FREQ_SA1100 || CPU_FREQ_SA1110 + default CPU_FREQ_DEFAULT_GOV_PERFORMANCE + help + This option sets which CPUFreq governor shall be loaded at + startup. If in doubt, select 'performance'. + +config CPU_FREQ_DEFAULT_GOV_PERFORMANCE + bool "performance" + select CPU_FREQ_GOV_PERFORMANCE + help + Use the CPUFreq governor 'performance' as default. This sets + the frequency statically to the highest frequency supported by + the CPU. + +config CPU_FREQ_DEFAULT_GOV_USERSPACE + bool "userspace" + select CPU_FREQ_GOV_USERSPACE + help + Use the CPUFreq governor 'userspace' as default. This allows + you to set the CPU frequency manually or when an userspace + program shall be able to set the CPU dynamically without having + to enable the userspace governor manually. + +endchoice + +config CPU_FREQ_GOV_PERFORMANCE + tristate "'performance' governor" + help + This cpufreq governor sets the frequency statically to the + highest available CPU frequency. + + If in doubt, say Y. + +config CPU_FREQ_GOV_POWERSAVE + tristate "'powersave' governor" + help + This cpufreq governor sets the frequency statically to the + lowest available CPU frequency. + + If in doubt, say Y. + +config CPU_FREQ_GOV_USERSPACE + tristate "'userspace' governor for userspace frequency scaling" + help + Enable this cpufreq governor when you either want to set the + CPU frequency manually or when an userspace program shall + be able to set the CPU dynamically, like on LART + <http://www.lart.tudelft.nl/> + + For details, take a look at <file:Documentation/cpu-freq/>. + + If in doubt, say Y. + +config CPU_FREQ_GOV_ONDEMAND + tristate "'ondemand' cpufreq policy governor" + help + 'ondemand' - This driver adds a dynamic cpufreq policy governor. + The governor does a periodic polling and + changes frequency based on the CPU utilization. + The support for this governor depends on CPU capability to + do fast frequency switching (i.e, very low latency frequency + transitions). + + For details, take a look at linux/Documentation/cpu-freq. + + If in doubt, say N. + +endif # CPU_FREQ diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile new file mode 100644 index 000000000000..67b16e5a41a7 --- /dev/null +++ b/drivers/cpufreq/Makefile @@ -0,0 +1,14 @@ +# CPUfreq core +obj-$(CONFIG_CPU_FREQ) += cpufreq.o +# CPUfreq stats +obj-$(CONFIG_CPU_FREQ_STAT) += cpufreq_stats.o + +# CPUfreq governors +obj-$(CONFIG_CPU_FREQ_GOV_PERFORMANCE) += cpufreq_performance.o +obj-$(CONFIG_CPU_FREQ_GOV_POWERSAVE) += cpufreq_powersave.o +obj-$(CONFIG_CPU_FREQ_GOV_USERSPACE) += cpufreq_userspace.o +obj-$(CONFIG_CPU_FREQ_GOV_ONDEMAND) += cpufreq_ondemand.o + +# CPUfreq cross-arch helpers +obj-$(CONFIG_CPU_FREQ_TABLE) += freq_table.o + diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c new file mode 100644 index 000000000000..b30001f31610 --- /dev/null +++ b/drivers/cpufreq/cpufreq.c @@ -0,0 +1,1428 @@ +/* + * linux/drivers/cpufreq/cpufreq.c + * + * Copyright (C) 2001 Russell King + * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> + * + * 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. + * + */ + +#include <linux/config.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/notifier.h> +#include <linux/cpufreq.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/spinlock.h> +#include <linux/device.h> +#include <linux/slab.h> +#include <linux/cpu.h> +#include <linux/completion.h> + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_CORE, "cpufreq-core", msg) + +/** + * The "cpufreq driver" - the arch- or hardware-dependend low + * level driver of CPUFreq support, and its spinlock. This lock + * also protects the cpufreq_cpu_data array. + */ +static struct cpufreq_driver *cpufreq_driver; +static struct cpufreq_policy *cpufreq_cpu_data[NR_CPUS]; +static DEFINE_SPINLOCK(cpufreq_driver_lock); + + +/* we keep a copy of all ->add'ed CPU's struct sys_device here; + * as it is only accessed in ->add and ->remove, no lock or reference + * count is necessary. + */ +static struct sys_device *cpu_sys_devices[NR_CPUS]; + + +/* internal prototypes */ +static int __cpufreq_governor(struct cpufreq_policy *policy, unsigned int event); +static void handle_update(void *data); +static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci); + +/** + * 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 + * changes to devices when the CPU clock speed changes. + * The mutex locks both lists. + */ +static struct notifier_block *cpufreq_policy_notifier_list; +static struct notifier_block *cpufreq_transition_notifier_list; +static DECLARE_RWSEM (cpufreq_notifier_rwsem); + + +static LIST_HEAD(cpufreq_governor_list); +static DECLARE_MUTEX (cpufreq_governor_sem); + +struct cpufreq_policy * cpufreq_cpu_get(unsigned int cpu) +{ + struct cpufreq_policy *data; + unsigned long flags; + + if (cpu >= NR_CPUS) + goto err_out; + + /* get the cpufreq driver */ + spin_lock_irqsave(&cpufreq_driver_lock, flags); + + if (!cpufreq_driver) + goto err_out_unlock; + + if (!try_module_get(cpufreq_driver->owner)) + goto err_out_unlock; + + + /* get the CPU */ + data = cpufreq_cpu_data[cpu]; + + if (!data) + goto err_out_put_module; + + if (!kobject_get(&data->kobj)) + goto err_out_put_module; + + + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + + return data; + + err_out_put_module: + module_put(cpufreq_driver->owner); + err_out_unlock: + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + err_out: + return NULL; +} +EXPORT_SYMBOL_GPL(cpufreq_cpu_get); + +void cpufreq_cpu_put(struct cpufreq_policy *data) +{ + kobject_put(&data->kobj); + module_put(cpufreq_driver->owner); +} +EXPORT_SYMBOL_GPL(cpufreq_cpu_put); + + +/********************************************************************* + * UNIFIED DEBUG HELPERS * + *********************************************************************/ +#ifdef CONFIG_CPU_FREQ_DEBUG + +/* what part(s) of the CPUfreq subsystem are debugged? */ +static unsigned int debug; + +/* is the debug output ratelimit'ed using printk_ratelimit? User can + * set or modify this value. + */ +static unsigned int debug_ratelimit = 1; + +/* is the printk_ratelimit'ing enabled? It's enabled after a successful + * loading of a cpufreq driver, temporarily disabled when a new policy + * is set, and disabled upon cpufreq driver removal + */ +static unsigned int disable_ratelimit = 1; +static DEFINE_SPINLOCK(disable_ratelimit_lock); + +static inline void cpufreq_debug_enable_ratelimit(void) +{ + unsigned long flags; + + spin_lock_irqsave(&disable_ratelimit_lock, flags); + if (disable_ratelimit) + disable_ratelimit--; + spin_unlock_irqrestore(&disable_ratelimit_lock, flags); +} + +static inline void cpufreq_debug_disable_ratelimit(void) +{ + unsigned long flags; + + spin_lock_irqsave(&disable_ratelimit_lock, flags); + disable_ratelimit++; + spin_unlock_irqrestore(&disable_ratelimit_lock, flags); +} + +void cpufreq_debug_printk(unsigned int type, const char *prefix, const char *fmt, ...) +{ + char s[256]; + va_list args; + unsigned int len; + unsigned long flags; + + WARN_ON(!prefix); + if (type & debug) { + spin_lock_irqsave(&disable_ratelimit_lock, flags); + if (!disable_ratelimit && debug_ratelimit && !printk_ratelimit()) { + spin_unlock_irqrestore(&disable_ratelimit_lock, flags); + return; + } + spin_unlock_irqrestore(&disable_ratelimit_lock, flags); + + len = snprintf(s, 256, KERN_DEBUG "%s: ", prefix); + + va_start(args, fmt); + len += vsnprintf(&s[len], (256 - len), fmt, args); + va_end(args); + + printk(s); + + WARN_ON(len < 5); + } +} +EXPORT_SYMBOL(cpufreq_debug_printk); + + +module_param(debug, uint, 0644); +MODULE_PARM_DESC(debug, "CPUfreq debugging: add 1 to debug core, 2 to debug drivers, and 4 to debug governors."); + +module_param(debug_ratelimit, uint, 0644); +MODULE_PARM_DESC(debug_ratelimit, "CPUfreq debugging: set to 0 to disable ratelimiting."); + +#else /* !CONFIG_CPU_FREQ_DEBUG */ + +static inline void cpufreq_debug_enable_ratelimit(void) { return; } +static inline void cpufreq_debug_disable_ratelimit(void) { return; } + +#endif /* CONFIG_CPU_FREQ_DEBUG */ + + +/********************************************************************* + * EXTERNALLY AFFECTING FREQUENCY CHANGES * + *********************************************************************/ + +/** + * adjust_jiffies - adjust the system "loops_per_jiffy" + * + * 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 inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) +{ + 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; + dprintk("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_PRECHANGE && ci->old < ci->new) || + (val == CPUFREQ_POSTCHANGE && ci->old > ci->new) || + (val == CPUFREQ_RESUMECHANGE)) { + loops_per_jiffy = cpufreq_scale(l_p_j_ref, l_p_j_ref_freq, ci->new); + dprintk("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 + + +/** + * cpufreq_notify_transition - call notifier chain and adjust_jiffies on frequency transition + * + * This function calls the transition notifiers and the "adjust_jiffies" function. It is called + * twice on all CPU frequency changes that have external effects. + */ +void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) +{ + BUG_ON(irqs_disabled()); + + freqs->flags = cpufreq_driver->flags; + dprintk("notification %u of frequency transition to %u kHz\n", state, freqs->new); + + down_read(&cpufreq_notifier_rwsem); + switch (state) { + case CPUFREQ_PRECHANGE: + /* 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 ((likely(cpufreq_cpu_data[freqs->cpu])) && + (likely(cpufreq_cpu_data[freqs->cpu]->cpu == freqs->cpu)) && + (likely(cpufreq_cpu_data[freqs->cpu]->cur)) && + (unlikely(freqs->old != cpufreq_cpu_data[freqs->cpu]->cur))) + { + printk(KERN_WARNING "Warning: CPU frequency is %u, " + "cpufreq assumed %u kHz.\n", freqs->old, cpufreq_cpu_data[freqs->cpu]->cur); + freqs->old = cpufreq_cpu_data[freqs->cpu]->cur; + } + } + notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_PRECHANGE, freqs); + adjust_jiffies(CPUFREQ_PRECHANGE, freqs); + break; + case CPUFREQ_POSTCHANGE: + adjust_jiffies(CPUFREQ_POSTCHANGE, freqs); + notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_POSTCHANGE, freqs); + if ((likely(cpufreq_cpu_data[freqs->cpu])) && + (likely(cpufreq_cpu_data[freqs->cpu]->cpu == freqs->cpu))) + cpufreq_cpu_data[freqs->cpu]->cur = freqs->new; + break; + } + up_read(&cpufreq_notifier_rwsem); +} +EXPORT_SYMBOL_GPL(cpufreq_notify_transition); + + + +/********************************************************************* + * SYSFS INTERFACE * + *********************************************************************/ + +/** + * cpufreq_parse_governor - parse a governor string + */ +static int cpufreq_parse_governor (char *str_governor, unsigned int *policy, + struct cpufreq_governor **governor) +{ + if (!cpufreq_driver) + return -EINVAL; + if (cpufreq_driver->setpolicy) { + if (!strnicmp(str_governor, "performance", CPUFREQ_NAME_LEN)) { + *policy = CPUFREQ_POLICY_PERFORMANCE; + return 0; + } else if (!strnicmp(str_governor, "powersave", CPUFREQ_NAME_LEN)) { + *policy = CPUFREQ_POLICY_POWERSAVE; + return 0; + } + return -EINVAL; + } else { + struct cpufreq_governor *t; + down(&cpufreq_governor_sem); + if (!cpufreq_driver || !cpufreq_driver->target) + goto out; + list_for_each_entry(t, &cpufreq_governor_list, governor_list) { + if (!strnicmp(str_governor,t->name,CPUFREQ_NAME_LEN)) { + *governor = t; + up(&cpufreq_governor_sem); + return 0; + } + } + out: + up(&cpufreq_governor_sem); + } + return -EINVAL; +} +EXPORT_SYMBOL_GPL(cpufreq_parse_governor); + + +/* drivers/base/cpu.c */ +extern struct sysdev_class cpu_sysdev_class; + + +/** + * cpufreq_per_cpu_attr_read() / show_##file_name() - print out cpufreq information + * + * Write out information from cpufreq_driver->policy[cpu]; object must be + * "unsigned int". + */ + +#define show_one(file_name, object) \ +static ssize_t show_##file_name \ +(struct cpufreq_policy * policy, char *buf) \ +{ \ + return sprintf (buf, "%u\n", policy->object); \ +} + +show_one(cpuinfo_min_freq, cpuinfo.min_freq); +show_one(cpuinfo_max_freq, cpuinfo.max_freq); +show_one(scaling_min_freq, min); +show_one(scaling_max_freq, max); +show_one(scaling_cur_freq, cur); + +/** + * 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 = -EINVAL; \ + struct cpufreq_policy new_policy; \ + \ + ret = cpufreq_get_policy(&new_policy, policy->cpu); \ + if (ret) \ + return -EINVAL; \ + \ + ret = sscanf (buf, "%u", &new_policy.object); \ + if (ret != 1) \ + return -EINVAL; \ + \ + ret = cpufreq_set_policy(&new_policy); \ + \ + return ret ? ret : count; \ +} + +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); +} + + +/** + * 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"); + else if (policy->policy == CPUFREQ_POLICY_PERFORMANCE) + return sprintf(buf, "performance\n"); + else if (policy->governor) + return scnprintf(buf, CPUFREQ_NAME_LEN, "%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 = -EINVAL; + char str_governor[16]; + struct cpufreq_policy new_policy; + + ret = cpufreq_get_policy(&new_policy, policy->cpu); + if (ret) + return 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; + + ret = cpufreq_set_policy(&new_policy); + + 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) +{ + return scnprintf(buf, CPUFREQ_NAME_LEN, "%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, + char *buf) +{ + ssize_t i = 0; + struct cpufreq_governor *t; + + if (!cpufreq_driver->target) { + i += sprintf(buf, "performance powersave"); + goto out; + } + + list_for_each_entry(t, &cpufreq_governor_list, governor_list) { + if (i >= (ssize_t) ((PAGE_SIZE / sizeof(char)) - (CPUFREQ_NAME_LEN + 2))) + goto out; + i += scnprintf(&buf[i], CPUFREQ_NAME_LEN, "%s ", t->name); + } + out: + i += sprintf(&buf[i], "\n"); + return i; +} +/** + * show_affected_cpus - show the CPUs affected by each transition + */ +static ssize_t show_affected_cpus (struct cpufreq_policy * policy, char *buf) +{ + ssize_t i = 0; + unsigned int cpu; + + for_each_cpu_mask(cpu, policy->cpus) { + if (i) + i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " "); + i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu); + if (i >= (PAGE_SIZE - 5)) + break; + } + i += sprintf(&buf[i], "\n"); + return i; +} + + +#define define_one_ro(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0444, show_##_name, NULL) + +#define define_one_ro0400(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0400, show_##_name, NULL) + +#define define_one_rw(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0644, show_##_name, store_##_name) + +define_one_ro0400(cpuinfo_cur_freq); +define_one_ro(cpuinfo_min_freq); +define_one_ro(cpuinfo_max_freq); +define_one_ro(scaling_available_governors); +define_one_ro(scaling_driver); +define_one_ro(scaling_cur_freq); +define_one_ro(affected_cpus); +define_one_rw(scaling_min_freq); +define_one_rw(scaling_max_freq); +define_one_rw(scaling_governor); + +static struct attribute * default_attrs[] = { + &cpuinfo_min_freq.attr, + &cpuinfo_max_freq.attr, + &scaling_min_freq.attr, + &scaling_max_freq.attr, + &affected_cpus.attr, + &scaling_governor.attr, + &scaling_driver.attr, + &scaling_available_governors.attr, + NULL +}; + +#define to_policy(k) container_of(k,struct cpufreq_policy,kobj) +#define to_attr(a) container_of(a,struct freq_attr,attr) + +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; + policy = cpufreq_cpu_get(policy->cpu); + if (!policy) + return -EINVAL; + ret = fattr->show ? fattr->show(policy,buf) : 0; + cpufreq_cpu_put(policy); + return ret; +} + +static ssize_t store(struct kobject * kobj, struct attribute * attr, + const char * buf, size_t count) +{ + struct cpufreq_policy * policy = to_policy(kobj); + struct freq_attr * fattr = to_attr(attr); + ssize_t ret; + policy = cpufreq_cpu_get(policy->cpu); + if (!policy) + return -EINVAL; + ret = fattr->store ? fattr->store(policy,buf,count) : 0; + cpufreq_cpu_put(policy); + return ret; +} + +static void cpufreq_sysfs_release(struct kobject * kobj) +{ + struct cpufreq_policy * policy = to_policy(kobj); + dprintk("last reference is dropped\n"); + complete(&policy->kobj_unregister); +} + +static struct sysfs_ops sysfs_ops = { + .show = show, + .store = store, +}; + +static struct kobj_type ktype_cpufreq = { + .sysfs_ops = &sysfs_ops, + .default_attrs = default_attrs, + .release = cpufreq_sysfs_release, +}; + + +/** + * cpufreq_add_dev - add a CPU device + * + * Adds the cpufreq interface for a CPU device. + */ +static int cpufreq_add_dev (struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + int ret = 0; + struct cpufreq_policy new_policy; + struct cpufreq_policy *policy; + struct freq_attr **drv_attr; + unsigned long flags; + unsigned int j; + + cpufreq_debug_disable_ratelimit(); + dprintk("adding CPU %u\n", cpu); + +#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)) { + cpu_sys_devices[cpu] = sys_dev; + dprintk("CPU already managed, adding link\n"); + sysfs_create_link(&sys_dev->kobj, &policy->kobj, "cpufreq"); + cpufreq_debug_enable_ratelimit(); + return 0; + } +#endif + + if (!try_module_get(cpufreq_driver->owner)) { + ret = -EINVAL; + goto module_out; + } + + policy = kmalloc(sizeof(struct cpufreq_policy), GFP_KERNEL); + if (!policy) { + ret = -ENOMEM; + goto nomem_out; + } + memset(policy, 0, sizeof(struct cpufreq_policy)); + + policy->cpu = cpu; + policy->cpus = cpumask_of_cpu(cpu); + + init_MUTEX_LOCKED(&policy->lock); + init_completion(&policy->kobj_unregister); + INIT_WORK(&policy->update, handle_update, (void *)(long)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) { + dprintk("initialization failed\n"); + goto err_out; + } + + memcpy(&new_policy, policy, sizeof(struct cpufreq_policy)); + + /* prepare interface data */ + policy->kobj.parent = &sys_dev->kobj; + policy->kobj.ktype = &ktype_cpufreq; + strlcpy(policy->kobj.name, "cpufreq", KOBJ_NAME_LEN); + + ret = kobject_register(&policy->kobj); + if (ret) + goto err_out; + + /* set up files for this cpu device */ + drv_attr = cpufreq_driver->attr; + while ((drv_attr) && (*drv_attr)) { + sysfs_create_file(&policy->kobj, &((*drv_attr)->attr)); + drv_attr++; + } + if (cpufreq_driver->get) + sysfs_create_file(&policy->kobj, &cpuinfo_cur_freq.attr); + if (cpufreq_driver->target) + sysfs_create_file(&policy->kobj, &scaling_cur_freq.attr); + + spin_lock_irqsave(&cpufreq_driver_lock, flags); + for_each_cpu_mask(j, policy->cpus) + cpufreq_cpu_data[j] = policy; + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + policy->governor = NULL; /* to assure that the starting sequence is + * run in cpufreq_set_policy */ + up(&policy->lock); + + /* set default policy */ + + ret = cpufreq_set_policy(&new_policy); + if (ret) { + dprintk("setting policy failed\n"); + goto err_out_unregister; + } + + module_put(cpufreq_driver->owner); + cpu_sys_devices[cpu] = sys_dev; + dprintk("initialization complete\n"); + cpufreq_debug_enable_ratelimit(); + + return 0; + + +err_out_unregister: + spin_lock_irqsave(&cpufreq_driver_lock, flags); + for_each_cpu_mask(j, policy->cpus) + cpufreq_cpu_data[j] = NULL; + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + + kobject_unregister(&policy->kobj); + wait_for_completion(&policy->kobj_unregister); + +err_out: + kfree(policy); + +nomem_out: + module_put(cpufreq_driver->owner); + module_out: + cpufreq_debug_enable_ratelimit(); + return ret; +} + + +/** + * cpufreq_remove_dev - remove a CPU device + * + * Removes the cpufreq interface for a CPU device. + */ +static int cpufreq_remove_dev (struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + unsigned long flags; + struct cpufreq_policy *data; +#ifdef CONFIG_SMP + unsigned int j; +#endif + + cpufreq_debug_disable_ratelimit(); + dprintk("unregistering CPU %u\n", cpu); + + spin_lock_irqsave(&cpufreq_driver_lock, flags); + data = cpufreq_cpu_data[cpu]; + + if (!data) { + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + cpu_sys_devices[cpu] = NULL; + cpufreq_debug_enable_ratelimit(); + return -EINVAL; + } + cpufreq_cpu_data[cpu] = NULL; + + +#ifdef CONFIG_SMP + /* if this isn't the CPU which is the parent of the kobj, we + * only need to unlink, put and exit + */ + if (unlikely(cpu != data->cpu)) { + dprintk("removing link\n"); + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + sysfs_remove_link(&sys_dev->kobj, "cpufreq"); + cpu_sys_devices[cpu] = NULL; + cpufreq_cpu_put(data); + cpufreq_debug_enable_ratelimit(); + return 0; + } +#endif + + cpu_sys_devices[cpu] = NULL; + + if (!kobject_get(&data->kobj)) { + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + cpufreq_debug_enable_ratelimit(); + return -EFAULT; + } + +#ifdef CONFIG_SMP + /* if we have other CPUs still registered, we need to unlink them, + * or else wait_for_completion below will lock up. Clean the + * cpufreq_cpu_data[] while holding the lock, and remove the sysfs + * links afterwards. + */ + if (unlikely(cpus_weight(data->cpus) > 1)) { + for_each_cpu_mask(j, data->cpus) { + if (j == cpu) + continue; + cpufreq_cpu_data[j] = NULL; + } + } + + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + + if (unlikely(cpus_weight(data->cpus) > 1)) { + for_each_cpu_mask(j, data->cpus) { + if (j == cpu) + continue; + dprintk("removing link for cpu %u\n", j); + sysfs_remove_link(&cpu_sys_devices[j]->kobj, "cpufreq"); + cpufreq_cpu_put(data); + } + } +#else + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); +#endif + + down(&data->lock); + if (cpufreq_driver->target) + __cpufreq_governor(data, CPUFREQ_GOV_STOP); + cpufreq_driver->target = NULL; + up(&data->lock); + + kobject_unregister(&data->kobj); + + kobject_put(&data->kobj); + + /* we need to make sure that the underlying kobj is actually + * not referenced anymore by anybody before we proceed with + * unloading. + */ + dprintk("waiting for dropping of refcount\n"); + wait_for_completion(&data->kobj_unregister); + dprintk("wait complete\n"); + + if (cpufreq_driver->exit) + cpufreq_driver->exit(data); + + kfree(data); + + cpufreq_debug_enable_ratelimit(); + + return 0; +} + + +static void handle_update(void *data) +{ + unsigned int cpu = (unsigned int)(long)data; + dprintk("handle_update for cpu %u called\n", cpu); + cpufreq_update_policy(cpu); +} + +/** + * 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 + * + * We adjust to current frequency first, and need to clean up later. So either call + * to cpufreq_update_policy() or schedule handle_update()). + */ +static void cpufreq_out_of_sync(unsigned int cpu, unsigned int old_freq, unsigned int new_freq) +{ + struct cpufreq_freqs freqs; + + printk(KERN_WARNING "Warning: CPU frequency out of sync: cpufreq and timing " + "core thinks of %u, is %u kHz.\n", old_freq, new_freq); + + freqs.cpu = cpu; + freqs.old = old_freq; + freqs.new = new_freq; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); +} + + +/** + * cpufreq_get - get the current CPU frequency (in kHz) + * @cpu: CPU number + * + * Get the CPU current (static) CPU frequency + */ +unsigned int cpufreq_get(unsigned int cpu) +{ + struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); + unsigned int ret = 0; + + if (!policy) + return 0; + + if (!cpufreq_driver->get) + goto out; + + down(&policy->lock); + + ret = cpufreq_driver->get(cpu); + + if (ret && policy->cur && !(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) + { + /* verify no discrepancy between actual and saved value exists */ + if (unlikely(ret != policy->cur)) { + cpufreq_out_of_sync(cpu, policy->cur, ret); + schedule_work(&policy->update); + } + } + + up(&policy->lock); + + out: + cpufreq_cpu_put(policy); + + return (ret); +} +EXPORT_SYMBOL(cpufreq_get); + + +/** + * cpufreq_resume - restore proper CPU frequency handling after resume + * + * 1.) resume CPUfreq hardware support (cpufreq_driver->resume()) + * 2.) if ->target and !CPUFREQ_CONST_LOOPS: verify we're in sync + * 3.) schedule call cpufreq_update_policy() ASAP as interrupts are restored. + */ +static int cpufreq_resume(struct sys_device * sysdev) +{ + int cpu = sysdev->id; + unsigned int ret = 0; + struct cpufreq_policy *cpu_policy; + + dprintk("resuming cpu %u\n", cpu); + + if (!cpu_online(cpu)) + return 0; + + /* we may be lax here as interrupts are off. Nonetheless + * we need to grab the correct cpu policy, as to check + * whether we really run on this CPU. + */ + + cpu_policy = cpufreq_cpu_get(cpu); + if (!cpu_policy) + return -EINVAL; + + /* only handle each CPU group once */ + if (unlikely(cpu_policy->cpu != cpu)) { + cpufreq_cpu_put(cpu_policy); + return 0; + } + + 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); + cpufreq_cpu_put(cpu_policy); + return ret; + } + } + + if (!(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) { + unsigned int cur_freq = 0; + + if (cpufreq_driver->get) + cur_freq = cpufreq_driver->get(cpu_policy->cpu); + + if (!cur_freq || !cpu_policy->cur) { + printk(KERN_ERR "cpufreq: resume failed to assert current frequency is what timing core thinks it is.\n"); + goto out; + } + + if (unlikely(cur_freq != cpu_policy->cur)) { + struct cpufreq_freqs freqs; + + printk(KERN_WARNING "Warning: CPU frequency is %u, " + "cpufreq assumed %u kHz.\n", cur_freq, cpu_policy->cur); + + freqs.cpu = cpu; + freqs.old = cpu_policy->cur; + freqs.new = cur_freq; + + notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_RESUMECHANGE, &freqs); + adjust_jiffies(CPUFREQ_RESUMECHANGE, &freqs); + + cpu_policy->cur = cur_freq; + } + } + +out: + schedule_work(&cpu_policy->update); + cpufreq_cpu_put(cpu_policy); + return ret; +} + +static struct sysdev_driver cpufreq_sysdev_driver = { + .add = cpufreq_add_dev, + .remove = cpufreq_remove_dev, + .resume = cpufreq_resume, +}; + + +/********************************************************************* + * NOTIFIER LISTS INTERFACE * + *********************************************************************/ + +/** + * cpufreq_register_notifier - register a driver 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. + * + * This function may sleep, and has the same return conditions as + * notifier_chain_register. + */ +int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) +{ + int ret; + + down_write(&cpufreq_notifier_rwsem); + switch (list) { + case CPUFREQ_TRANSITION_NOTIFIER: + ret = notifier_chain_register(&cpufreq_transition_notifier_list, nb); + break; + case CPUFREQ_POLICY_NOTIFIER: + ret = notifier_chain_register(&cpufreq_policy_notifier_list, nb); + break; + default: + ret = -EINVAL; + } + up_write(&cpufreq_notifier_rwsem); + + return ret; +} +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 + * + * Remove a driver from the CPU frequency notifier list. + * + * This function may sleep, and has the same return conditions as + * notifier_chain_unregister. + */ +int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list) +{ + int ret; + + down_write(&cpufreq_notifier_rwsem); + switch (list) { + case CPUFREQ_TRANSITION_NOTIFIER: + ret = notifier_chain_unregister(&cpufreq_transition_notifier_list, nb); + break; + case CPUFREQ_POLICY_NOTIFIER: + ret = notifier_chain_unregister(&cpufreq_policy_notifier_list, nb); + break; + default: + ret = -EINVAL; + } + up_write(&cpufreq_notifier_rwsem); + + return ret; +} +EXPORT_SYMBOL(cpufreq_unregister_notifier); + + +/********************************************************************* + * GOVERNORS * + *********************************************************************/ + + +int __cpufreq_driver_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + int retval = -EINVAL; + lock_cpu_hotplug(); + dprintk("target for CPU %u: %u kHz, relation %u\n", policy->cpu, + target_freq, relation); + if (cpu_online(policy->cpu) && cpufreq_driver->target) + retval = cpufreq_driver->target(policy, target_freq, relation); + unlock_cpu_hotplug(); + return retval; +} +EXPORT_SYMBOL_GPL(__cpufreq_driver_target); + + +int cpufreq_driver_target(struct cpufreq_policy *policy, + unsigned int target_freq, + unsigned int relation) +{ + unsigned int ret; + + policy = cpufreq_cpu_get(policy->cpu); + if (!policy) + return -EINVAL; + + down(&policy->lock); + + ret = __cpufreq_driver_target(policy, target_freq, relation); + + up(&policy->lock); + + cpufreq_cpu_put(policy); + + return ret; +} +EXPORT_SYMBOL_GPL(cpufreq_driver_target); + + +static int __cpufreq_governor(struct cpufreq_policy *policy, unsigned int event) +{ + int ret = -EINVAL; + + if (!try_module_get(policy->governor->owner)) + return -EINVAL; + + dprintk("__cpufreq_governor for CPU %u, event %u\n", policy->cpu, event); + ret = policy->governor->governor(policy, event); + + /* 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); + + return ret; +} + + +int cpufreq_governor(unsigned int cpu, unsigned int event) +{ + int ret = 0; + struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); + + if (!policy) + return -EINVAL; + + down(&policy->lock); + ret = __cpufreq_governor(policy, event); + up(&policy->lock); + + cpufreq_cpu_put(policy); + + return ret; +} +EXPORT_SYMBOL_GPL(cpufreq_governor); + + +int cpufreq_register_governor(struct cpufreq_governor *governor) +{ + struct cpufreq_governor *t; + + if (!governor) + return -EINVAL; + + down(&cpufreq_governor_sem); + + list_for_each_entry(t, &cpufreq_governor_list, governor_list) { + if (!strnicmp(governor->name,t->name,CPUFREQ_NAME_LEN)) { + up(&cpufreq_governor_sem); + return -EBUSY; + } + } + list_add(&governor->governor_list, &cpufreq_governor_list); + + up(&cpufreq_governor_sem); + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_register_governor); + + +void cpufreq_unregister_governor(struct cpufreq_governor *governor) +{ + if (!governor) + return; + + down(&cpufreq_governor_sem); + list_del(&governor->governor_list); + up(&cpufreq_governor_sem); + return; +} +EXPORT_SYMBOL_GPL(cpufreq_unregister_governor); + + + +/********************************************************************* + * POLICY INTERFACE * + *********************************************************************/ + +/** + * cpufreq_get_policy - get the current cpufreq_policy + * @policy: struct cpufreq_policy into which the current cpufreq_policy is written + * + * Reads the current cpufreq policy. + */ +int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu) +{ + struct cpufreq_policy *cpu_policy; + if (!policy) + return -EINVAL; + + cpu_policy = cpufreq_cpu_get(cpu); + if (!cpu_policy) + return -EINVAL; + + down(&cpu_policy->lock); + memcpy(policy, cpu_policy, sizeof(struct cpufreq_policy)); + up(&cpu_policy->lock); + + cpufreq_cpu_put(cpu_policy); + + return 0; +} +EXPORT_SYMBOL(cpufreq_get_policy); + + +static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_policy *policy) +{ + int ret = 0; + + cpufreq_debug_disable_ratelimit(); + dprintk("setting new policy for CPU %u: %u - %u kHz\n", policy->cpu, + policy->min, policy->max); + + memcpy(&policy->cpuinfo, + &data->cpuinfo, + sizeof(struct cpufreq_cpuinfo)); + + /* verify the cpu speed can be set within this limit */ + ret = cpufreq_driver->verify(policy); + if (ret) + goto error_out; + + down_read(&cpufreq_notifier_rwsem); + + /* adjust if necessary - all reasons */ + notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_ADJUST, + policy); + + /* adjust if necessary - hardware incompatibility*/ + notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_INCOMPATIBLE, + policy); + + /* verify the cpu speed can be set within this limit, + which might be different to the first one */ + ret = cpufreq_driver->verify(policy); + if (ret) { + up_read(&cpufreq_notifier_rwsem); + goto error_out; + } + + /* notification of the new policy */ + notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_NOTIFY, + policy); + + up_read(&cpufreq_notifier_rwsem); + + data->min = policy->min; + data->max = policy->max; + + dprintk("new min and max freqs are %u - %u kHz\n", data->min, data->max); + + if (cpufreq_driver->setpolicy) { + data->policy = policy->policy; + dprintk("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; + + dprintk("governor switch\n"); + + /* end old governor */ + if (data->governor) + __cpufreq_governor(data, CPUFREQ_GOV_STOP); + + /* start new governor */ + data->governor = policy->governor; + if (__cpufreq_governor(data, CPUFREQ_GOV_START)) { + /* new governor failed, so re-start old one */ + dprintk("starting governor %s failed\n", data->governor->name); + if (old_gov) { + data->governor = old_gov; + __cpufreq_governor(data, CPUFREQ_GOV_START); + } + ret = -EINVAL; + goto error_out; + } + /* might be a policy change, too, so fall through */ + } + dprintk("governor: change or update limits\n"); + __cpufreq_governor(data, CPUFREQ_GOV_LIMITS); + } + + error_out: + cpufreq_debug_enable_ratelimit(); + return ret; +} + +/** + * cpufreq_set_policy - set a new CPUFreq policy + * @policy: policy to be set. + * + * Sets a new CPU frequency and voltage scaling policy. + */ +int cpufreq_set_policy(struct cpufreq_policy *policy) +{ + int ret = 0; + struct cpufreq_policy *data; + + if (!policy) + return -EINVAL; + + data = cpufreq_cpu_get(policy->cpu); + if (!data) + return -EINVAL; + + /* lock this CPU */ + down(&data->lock); + + ret = __cpufreq_set_policy(data, policy); + data->user_policy.min = data->min; + data->user_policy.max = data->max; + data->user_policy.policy = data->policy; + data->user_policy.governor = data->governor; + + up(&data->lock); + cpufreq_cpu_put(data); + + return ret; +} +EXPORT_SYMBOL(cpufreq_set_policy); + + +/** + * cpufreq_update_policy - re-evaluate an existing cpufreq policy + * @cpu: CPU which shall be re-evaluated + * + * Usefull for policy notifiers which have different necessities + * at different times. + */ +int cpufreq_update_policy(unsigned int cpu) +{ + struct cpufreq_policy *data = cpufreq_cpu_get(cpu); + struct cpufreq_policy policy; + int ret = 0; + + if (!data) + return -ENODEV; + + down(&data->lock); + + dprintk("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 = __cpufreq_set_policy(data, &policy); + + up(&data->lock); + + cpufreq_cpu_put(data); + return ret; +} +EXPORT_SYMBOL(cpufreq_update_policy); + + +/********************************************************************* + * REGISTER / UNREGISTER CPUFREQ DRIVER * + *********************************************************************/ + +/** + * cpufreq_register_driver - register a CPU Frequency driver + * @driver_data: A struct cpufreq_driver containing the values# + * 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 + * (and isn't unregistered in the meantime). + * + */ +int cpufreq_register_driver(struct cpufreq_driver *driver_data) +{ + unsigned long flags; + int ret; + + if (!driver_data || !driver_data->verify || !driver_data->init || + ((!driver_data->setpolicy) && (!driver_data->target))) + return -EINVAL; + + dprintk("trying to register driver %s\n", driver_data->name); + + if (driver_data->setpolicy) + driver_data->flags |= CPUFREQ_CONST_LOOPS; + + spin_lock_irqsave(&cpufreq_driver_lock, flags); + if (cpufreq_driver) { + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + return -EBUSY; + } + cpufreq_driver = driver_data; + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + + ret = sysdev_driver_register(&cpu_sysdev_class,&cpufreq_sysdev_driver); + + if ((!ret) && !(cpufreq_driver->flags & CPUFREQ_STICKY)) { + int i; + ret = -ENODEV; + + /* check for at least one working CPU */ + for (i=0; i<NR_CPUS; i++) + if (cpufreq_cpu_data[i]) + ret = 0; + + /* if all ->init() calls failed, unregister */ + if (ret) { + dprintk("no CPU initialized for driver %s\n", driver_data->name); + sysdev_driver_unregister(&cpu_sysdev_class, &cpufreq_sysdev_driver); + + spin_lock_irqsave(&cpufreq_driver_lock, flags); + cpufreq_driver = NULL; + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + } + } + + if (!ret) { + dprintk("driver %s up and running\n", driver_data->name); + cpufreq_debug_enable_ratelimit(); + } + + 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 + * the right to do so, i.e. if you have succeeded in initialising before! + * Returns zero if successful, and -EINVAL if the cpufreq_driver is + * currently not initialised. + */ +int cpufreq_unregister_driver(struct cpufreq_driver *driver) +{ + unsigned long flags; + + cpufreq_debug_disable_ratelimit(); + + if (!cpufreq_driver || (driver != cpufreq_driver)) { + cpufreq_debug_enable_ratelimit(); + return -EINVAL; + } + + dprintk("unregistering driver %s\n", driver->name); + + sysdev_driver_unregister(&cpu_sysdev_class, &cpufreq_sysdev_driver); + + spin_lock_irqsave(&cpufreq_driver_lock, flags); + cpufreq_driver = NULL; + spin_unlock_irqrestore(&cpufreq_driver_lock, flags); + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_unregister_driver); diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c new file mode 100644 index 000000000000..8d83a21c6477 --- /dev/null +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -0,0 +1,491 @@ +/* + * drivers/cpufreq/cpufreq_ondemand.c + * + * Copyright (C) 2001 Russell King + * (C) 2003 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>. + * Jun Nakajima <jun.nakajima@intel.com> + * + * 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. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/ctype.h> +#include <linux/cpufreq.h> +#include <linux/sysctl.h> +#include <linux/types.h> +#include <linux/fs.h> +#include <linux/sysfs.h> +#include <linux/sched.h> +#include <linux/kmod.h> +#include <linux/workqueue.h> +#include <linux/jiffies.h> +#include <linux/kernel_stat.h> +#include <linux/percpu.h> + +/* + * dbs is used in this file as a shortform for demandbased switching + * It helps to keep variable names smaller, simpler + */ + +#define DEF_FREQUENCY_UP_THRESHOLD (80) +#define MIN_FREQUENCY_UP_THRESHOLD (0) +#define MAX_FREQUENCY_UP_THRESHOLD (100) + +#define DEF_FREQUENCY_DOWN_THRESHOLD (20) +#define MIN_FREQUENCY_DOWN_THRESHOLD (0) +#define MAX_FREQUENCY_DOWN_THRESHOLD (100) + +/* + * The polling frequency of this governor depends on the capability of + * the processor. Default polling frequency is 1000 times the transition + * latency of the processor. The governor will work on any processor with + * transition latency <= 10mS, using appropriate sampling + * rate. + * For CPUs with transition latency > 10mS (mostly drivers with CPUFREQ_ETERNAL) + * this governor will not work. + * All times here are in uS. + */ +static unsigned int def_sampling_rate; +#define MIN_SAMPLING_RATE (def_sampling_rate / 2) +#define MAX_SAMPLING_RATE (500 * def_sampling_rate) +#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER (1000) +#define DEF_SAMPLING_DOWN_FACTOR (10) +#define TRANSITION_LATENCY_LIMIT (10 * 1000) +#define sampling_rate_in_HZ(x) (((x * HZ) < (1000 * 1000))?1:((x * HZ) / (1000 * 1000))) + +static void do_dbs_timer(void *data); + +struct cpu_dbs_info_s { + struct cpufreq_policy *cur_policy; + unsigned int prev_cpu_idle_up; + unsigned int prev_cpu_idle_down; + unsigned int enable; +}; +static DEFINE_PER_CPU(struct cpu_dbs_info_s, cpu_dbs_info); + +static unsigned int dbs_enable; /* number of CPUs using this policy */ + +static DECLARE_MUTEX (dbs_sem); +static DECLARE_WORK (dbs_work, do_dbs_timer, NULL); + +struct dbs_tuners { + unsigned int sampling_rate; + unsigned int sampling_down_factor; + unsigned int up_threshold; + unsigned int down_threshold; +}; + +static struct dbs_tuners dbs_tuners_ins = { + .up_threshold = DEF_FREQUENCY_UP_THRESHOLD, + .down_threshold = DEF_FREQUENCY_DOWN_THRESHOLD, + .sampling_down_factor = DEF_SAMPLING_DOWN_FACTOR, +}; + +/************************** sysfs interface ************************/ +static ssize_t show_sampling_rate_max(struct cpufreq_policy *policy, char *buf) +{ + return sprintf (buf, "%u\n", MAX_SAMPLING_RATE); +} + +static ssize_t show_sampling_rate_min(struct cpufreq_policy *policy, char *buf) +{ + return sprintf (buf, "%u\n", MIN_SAMPLING_RATE); +} + +#define define_one_ro(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0444, show_##_name, NULL) + +define_one_ro(sampling_rate_max); +define_one_ro(sampling_rate_min); + +/* cpufreq_ondemand Governor Tunables */ +#define show_one(file_name, object) \ +static ssize_t show_##file_name \ +(struct cpufreq_policy *unused, char *buf) \ +{ \ + return sprintf(buf, "%u\n", dbs_tuners_ins.object); \ +} +show_one(sampling_rate, sampling_rate); +show_one(sampling_down_factor, sampling_down_factor); +show_one(up_threshold, up_threshold); +show_one(down_threshold, down_threshold); + +static ssize_t store_sampling_down_factor(struct cpufreq_policy *unused, + const char *buf, size_t count) +{ + unsigned int input; + int ret; + ret = sscanf (buf, "%u", &input); + if (ret != 1 ) + return -EINVAL; + + down(&dbs_sem); + dbs_tuners_ins.sampling_down_factor = input; + up(&dbs_sem); + + return count; +} + +static ssize_t store_sampling_rate(struct cpufreq_policy *unused, + const char *buf, size_t count) +{ + unsigned int input; + int ret; + ret = sscanf (buf, "%u", &input); + + down(&dbs_sem); + if (ret != 1 || input > MAX_SAMPLING_RATE || input < MIN_SAMPLING_RATE) { + up(&dbs_sem); + return -EINVAL; + } + + dbs_tuners_ins.sampling_rate = input; + up(&dbs_sem); + + return count; +} + +static ssize_t store_up_threshold(struct cpufreq_policy *unused, + const char *buf, size_t count) +{ + unsigned int input; + int ret; + ret = sscanf (buf, "%u", &input); + + down(&dbs_sem); + if (ret != 1 || input > MAX_FREQUENCY_UP_THRESHOLD || + input < MIN_FREQUENCY_UP_THRESHOLD || + input <= dbs_tuners_ins.down_threshold) { + up(&dbs_sem); + return -EINVAL; + } + + dbs_tuners_ins.up_threshold = input; + up(&dbs_sem); + + return count; +} + +static ssize_t store_down_threshold(struct cpufreq_policy *unused, + const char *buf, size_t count) +{ + unsigned int input; + int ret; + ret = sscanf (buf, "%u", &input); + + down(&dbs_sem); + if (ret != 1 || input > MAX_FREQUENCY_DOWN_THRESHOLD || + input < MIN_FREQUENCY_DOWN_THRESHOLD || + input >= dbs_tuners_ins.up_threshold) { + up(&dbs_sem); + return -EINVAL; + } + + dbs_tuners_ins.down_threshold = input; + up(&dbs_sem); + + return count; +} + +#define define_one_rw(_name) \ +static struct freq_attr _name = \ +__ATTR(_name, 0644, show_##_name, store_##_name) + +define_one_rw(sampling_rate); +define_one_rw(sampling_down_factor); +define_one_rw(up_threshold); +define_one_rw(down_threshold); + +static struct attribute * dbs_attributes[] = { + &sampling_rate_max.attr, + &sampling_rate_min.attr, + &sampling_rate.attr, + &sampling_down_factor.attr, + &up_threshold.attr, + &down_threshold.attr, + NULL +}; + +static struct attribute_group dbs_attr_group = { + .attrs = dbs_attributes, + .name = "ondemand", +}; + +/************************** sysfs end ************************/ + +static void dbs_check_cpu(int cpu) +{ + unsigned int idle_ticks, up_idle_ticks, down_idle_ticks; + unsigned int total_idle_ticks; + unsigned int freq_down_step; + unsigned int freq_down_sampling_rate; + static int down_skip[NR_CPUS]; + struct cpu_dbs_info_s *this_dbs_info; + + struct cpufreq_policy *policy; + unsigned int j; + + this_dbs_info = &per_cpu(cpu_dbs_info, cpu); + if (!this_dbs_info->enable) + return; + + policy = this_dbs_info->cur_policy; + /* + * The default safe range is 20% to 80% + * Every sampling_rate, we check + * - If current idle time is less than 20%, then we try to + * increase frequency + * Every sampling_rate*sampling_down_factor, we check + * - If current idle time is more than 80%, then we try to + * decrease frequency + * + * Any frequency increase takes it to the maximum frequency. + * Frequency reduction happens at minimum steps of + * 5% of max_frequency + */ + + /* Check for frequency increase */ + total_idle_ticks = kstat_cpu(cpu).cpustat.idle + + kstat_cpu(cpu).cpustat.iowait; + idle_ticks = total_idle_ticks - + this_dbs_info->prev_cpu_idle_up; + this_dbs_info->prev_cpu_idle_up = total_idle_ticks; + + + for_each_cpu_mask(j, policy->cpus) { + unsigned int tmp_idle_ticks; + struct cpu_dbs_info_s *j_dbs_info; + + if (j == cpu) + continue; + + j_dbs_info = &per_cpu(cpu_dbs_info, j); + /* Check for frequency increase */ + total_idle_ticks = kstat_cpu(j).cpustat.idle + + kstat_cpu(j).cpustat.iowait; + tmp_idle_ticks = total_idle_ticks - + j_dbs_info->prev_cpu_idle_up; + j_dbs_info->prev_cpu_idle_up = total_idle_ticks; + + if (tmp_idle_ticks < idle_ticks) + idle_ticks = tmp_idle_ticks; + } + + /* Scale idle ticks by 100 and compare with up and down ticks */ + idle_ticks *= 100; + up_idle_ticks = (100 - dbs_tuners_ins.up_threshold) * + sampling_rate_in_HZ(dbs_tuners_ins.sampling_rate); + + if (idle_ticks < up_idle_ticks) { + __cpufreq_driver_target(policy, policy->max, + CPUFREQ_RELATION_H); + down_skip[cpu] = 0; + this_dbs_info->prev_cpu_idle_down = total_idle_ticks; + return; + } + + /* Check for frequency decrease */ + down_skip[cpu]++; + if (down_skip[cpu] < dbs_tuners_ins.sampling_down_factor) + return; + + total_idle_ticks = kstat_cpu(cpu).cpustat.idle + + kstat_cpu(cpu).cpustat.iowait; + idle_ticks = total_idle_ticks - + this_dbs_info->prev_cpu_idle_down; + this_dbs_info->prev_cpu_idle_down = total_idle_ticks; + + for_each_cpu_mask(j, policy->cpus) { + unsigned int tmp_idle_ticks; + struct cpu_dbs_info_s *j_dbs_info; + + if (j == cpu) + continue; + + j_dbs_info = &per_cpu(cpu_dbs_info, j); + /* Check for frequency increase */ + total_idle_ticks = kstat_cpu(j).cpustat.idle + + kstat_cpu(j).cpustat.iowait; + tmp_idle_ticks = total_idle_ticks - + j_dbs_info->prev_cpu_idle_down; + j_dbs_info->prev_cpu_idle_down = total_idle_ticks; + + if (tmp_idle_ticks < idle_ticks) + idle_ticks = tmp_idle_ticks; + } + + /* Scale idle ticks by 100 and compare with up and down ticks */ + idle_ticks *= 100; + down_skip[cpu] = 0; + + freq_down_sampling_rate = dbs_tuners_ins.sampling_rate * + dbs_tuners_ins.sampling_down_factor; + down_idle_ticks = (100 - dbs_tuners_ins.down_threshold) * + sampling_rate_in_HZ(freq_down_sampling_rate); + + if (idle_ticks > down_idle_ticks ) { + freq_down_step = (5 * policy->max) / 100; + + /* max freq cannot be less than 100. But who knows.... */ + if (unlikely(freq_down_step == 0)) + freq_down_step = 5; + + __cpufreq_driver_target(policy, + policy->cur - freq_down_step, + CPUFREQ_RELATION_H); + return; + } +} + +static void do_dbs_timer(void *data) +{ + int i; + down(&dbs_sem); + for (i = 0; i < NR_CPUS; i++) + if (cpu_online(i)) + dbs_check_cpu(i); + schedule_delayed_work(&dbs_work, + sampling_rate_in_HZ(dbs_tuners_ins.sampling_rate)); + up(&dbs_sem); +} + +static inline void dbs_timer_init(void) +{ + INIT_WORK(&dbs_work, do_dbs_timer, NULL); + schedule_delayed_work(&dbs_work, + sampling_rate_in_HZ(dbs_tuners_ins.sampling_rate)); + return; +} + +static inline void dbs_timer_exit(void) +{ + cancel_delayed_work(&dbs_work); + return; +} + +static int cpufreq_governor_dbs(struct cpufreq_policy *policy, + unsigned int event) +{ + unsigned int cpu = policy->cpu; + struct cpu_dbs_info_s *this_dbs_info; + unsigned int j; + + this_dbs_info = &per_cpu(cpu_dbs_info, cpu); + + switch (event) { + case CPUFREQ_GOV_START: + if ((!cpu_online(cpu)) || + (!policy->cur)) + return -EINVAL; + + if (policy->cpuinfo.transition_latency > + (TRANSITION_LATENCY_LIMIT * 1000)) + return -EINVAL; + if (this_dbs_info->enable) /* Already enabled */ + break; + + down(&dbs_sem); + for_each_cpu_mask(j, policy->cpus) { + struct cpu_dbs_info_s *j_dbs_info; + j_dbs_info = &per_cpu(cpu_dbs_info, j); + j_dbs_info->cur_policy = policy; + + j_dbs_info->prev_cpu_idle_up = + kstat_cpu(j).cpustat.idle + + kstat_cpu(j).cpustat.iowait; + j_dbs_info->prev_cpu_idle_down = + kstat_cpu(j).cpustat.idle + + kstat_cpu(j).cpustat.iowait; + } + this_dbs_info->enable = 1; + sysfs_create_group(&policy->kobj, &dbs_attr_group); + dbs_enable++; + /* + * Start the timerschedule work, when this governor + * is used for first time + */ + if (dbs_enable == 1) { + unsigned int latency; + /* policy latency is in nS. Convert it to uS first */ + + latency = policy->cpuinfo.transition_latency; + if (latency < 1000) + latency = 1000; + + def_sampling_rate = (latency / 1000) * + DEF_SAMPLING_RATE_LATENCY_MULTIPLIER; + dbs_tuners_ins.sampling_rate = def_sampling_rate; + + dbs_timer_init(); + } + + up(&dbs_sem); + break; + + case CPUFREQ_GOV_STOP: + down(&dbs_sem); + this_dbs_info->enable = 0; + sysfs_remove_group(&policy->kobj, &dbs_attr_group); + dbs_enable--; + /* + * Stop the timerschedule work, when this governor + * is used for first time + */ + if (dbs_enable == 0) + dbs_timer_exit(); + + up(&dbs_sem); + + break; + + case CPUFREQ_GOV_LIMITS: + down(&dbs_sem); + if (policy->max < this_dbs_info->cur_policy->cur) + __cpufreq_driver_target( + this_dbs_info->cur_policy, + policy->max, CPUFREQ_RELATION_H); + else if (policy->min > this_dbs_info->cur_policy->cur) + __cpufreq_driver_target( + this_dbs_info->cur_policy, + policy->min, CPUFREQ_RELATION_L); + up(&dbs_sem); + break; + } + return 0; +} + +struct cpufreq_governor cpufreq_gov_dbs = { + .name = "ondemand", + .governor = cpufreq_governor_dbs, + .owner = THIS_MODULE, +}; +EXPORT_SYMBOL(cpufreq_gov_dbs); + +static int __init cpufreq_gov_dbs_init(void) +{ + return cpufreq_register_governor(&cpufreq_gov_dbs); +} + +static void __exit cpufreq_gov_dbs_exit(void) +{ + /* Make sure that the scheduled work is indeed not running */ + flush_scheduled_work(); + + cpufreq_unregister_governor(&cpufreq_gov_dbs); +} + + +MODULE_AUTHOR ("Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>"); +MODULE_DESCRIPTION ("'cpufreq_ondemand' - A dynamic cpufreq governor for " + "Low Latency Frequency Transition capable processors"); +MODULE_LICENSE ("GPL"); + +module_init(cpufreq_gov_dbs_init); +module_exit(cpufreq_gov_dbs_exit); diff --git a/drivers/cpufreq/cpufreq_performance.c b/drivers/cpufreq/cpufreq_performance.c new file mode 100644 index 000000000000..8d536b40deb8 --- /dev/null +++ b/drivers/cpufreq/cpufreq_performance.c @@ -0,0 +1,61 @@ +/* + * linux/drivers/cpufreq/cpufreq_performance.c + * + * Copyright (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> + * + * + * 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. + * + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/cpufreq.h> +#include <linux/init.h> + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_GOVERNOR, "performance", msg) + + +static int cpufreq_governor_performance(struct cpufreq_policy *policy, + unsigned int event) +{ + switch (event) { + case CPUFREQ_GOV_START: + case CPUFREQ_GOV_LIMITS: + dprintk("setting to %u kHz because of event %u\n", policy->max, event); + __cpufreq_driver_target(policy, policy->max, CPUFREQ_RELATION_H); + break; + default: + break; + } + return 0; +} + +struct cpufreq_governor cpufreq_gov_performance = { + .name = "performance", + .governor = cpufreq_governor_performance, + .owner = THIS_MODULE, +}; +EXPORT_SYMBOL(cpufreq_gov_performance); + + +static int __init cpufreq_gov_performance_init(void) +{ + return cpufreq_register_governor(&cpufreq_gov_performance); +} + + +static void __exit cpufreq_gov_performance_exit(void) +{ + cpufreq_unregister_governor(&cpufreq_gov_performance); +} + + +MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); +MODULE_DESCRIPTION("CPUfreq policy governor 'performance'"); +MODULE_LICENSE("GPL"); + +fs_initcall(cpufreq_gov_performance_init); +module_exit(cpufreq_gov_performance_exit); diff --git a/drivers/cpufreq/cpufreq_powersave.c b/drivers/cpufreq/cpufreq_powersave.c new file mode 100644 index 000000000000..c85edda7feb0 --- /dev/null +++ b/drivers/cpufreq/cpufreq_powersave.c @@ -0,0 +1,59 @@ +/* + * linux/drivers/cpufreq/cpufreq_powersave.c + * + * Copyright (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> + * + * + * 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. + * + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/cpufreq.h> +#include <linux/init.h> + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_GOVERNOR, "powersave", msg) + +static int cpufreq_governor_powersave(struct cpufreq_policy *policy, + unsigned int event) +{ + switch (event) { + case CPUFREQ_GOV_START: + case CPUFREQ_GOV_LIMITS: + dprintk("setting to %u kHz because of event %u\n", policy->min, event); + __cpufreq_driver_target(policy, policy->min, CPUFREQ_RELATION_L); + break; + default: + break; + } + return 0; +} + +static struct cpufreq_governor cpufreq_gov_powersave = { + .name = "powersave", + .governor = cpufreq_governor_powersave, + .owner = THIS_MODULE, +}; + + +static int __init cpufreq_gov_powersave_init(void) +{ + return cpufreq_register_governor(&cpufreq_gov_powersave); +} + + +static void __exit cpufreq_gov_powersave_exit(void) +{ + cpufreq_unregister_governor(&cpufreq_gov_powersave); +} + + +MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); +MODULE_DESCRIPTION("CPUfreq policy governor 'powersave'"); +MODULE_LICENSE("GPL"); + +module_init(cpufreq_gov_powersave_init); +module_exit(cpufreq_gov_powersave_exit); diff --git a/drivers/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c new file mode 100644 index 000000000000..2084593937c6 --- /dev/null +++ b/drivers/cpufreq/cpufreq_stats.c @@ -0,0 +1,334 @@ +/* + * drivers/cpufreq/cpufreq_stats.c + * + * Copyright (C) 2003-2004 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>. + * (C) 2004 Zou Nan hai <nanhai.zou@intel.com>. + * + * 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. + */ + +#include <linux/config.h> +#include <linux/kernel.h> +#include <linux/sysdev.h> +#include <linux/cpu.h> +#include <linux/sysfs.h> +#include <linux/cpufreq.h> +#include <linux/jiffies.h> +#include <linux/percpu.h> +#include <linux/kobject.h> +#include <linux/spinlock.h> + +static spinlock_t cpufreq_stats_lock; + +#define CPUFREQ_STATDEVICE_ATTR(_name,_mode,_show) \ +static struct freq_attr _attr_##_name = {\ + .attr = {.name = __stringify(_name), .owner = THIS_MODULE, \ + .mode = _mode, }, \ + .show = _show,\ +}; + +static unsigned long +delta_time(unsigned long old, unsigned long new) +{ + return (old > new) ? (old - new): (new + ~old + 1); +} + +struct cpufreq_stats { + unsigned int cpu; + unsigned int total_trans; + unsigned long long last_time; + unsigned int max_state; + unsigned int state_num; + unsigned int last_index; + unsigned long long *time_in_state; + unsigned int *freq_table; +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS + unsigned int *trans_table; +#endif +}; + +static struct cpufreq_stats *cpufreq_stats_table[NR_CPUS]; + +struct cpufreq_stats_attribute { + struct attribute attr; + ssize_t(*show) (struct cpufreq_stats *, char *); +}; + +static int +cpufreq_stats_update (unsigned int cpu) +{ + struct cpufreq_stats *stat; + spin_lock(&cpufreq_stats_lock); + stat = cpufreq_stats_table[cpu]; + if (stat->time_in_state) + stat->time_in_state[stat->last_index] += + delta_time(stat->last_time, jiffies); + stat->last_time = jiffies; + spin_unlock(&cpufreq_stats_lock); + return 0; +} + +static ssize_t +show_total_trans(struct cpufreq_policy *policy, char *buf) +{ + struct cpufreq_stats *stat = cpufreq_stats_table[policy->cpu]; + if(!stat) + return 0; + return sprintf(buf, "%d\n", + cpufreq_stats_table[stat->cpu]->total_trans); +} + +static ssize_t +show_time_in_state(struct cpufreq_policy *policy, char *buf) +{ + ssize_t len = 0; + int i; + struct cpufreq_stats *stat = cpufreq_stats_table[policy->cpu]; + if(!stat) + return 0; + cpufreq_stats_update(stat->cpu); + for (i = 0; i < stat->state_num; i++) { + len += sprintf(buf + len, "%u %llu\n", + stat->freq_table[i], stat->time_in_state[i]); + } + return len; +} + +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS +static ssize_t +show_trans_table(struct cpufreq_policy *policy, char *buf) +{ + ssize_t len = 0; + int i, j; + + struct cpufreq_stats *stat = cpufreq_stats_table[policy->cpu]; + if(!stat) + return 0; + cpufreq_stats_update(stat->cpu); + for (i = 0; i < stat->state_num; i++) { + if (len >= PAGE_SIZE) + break; + len += snprintf(buf + len, PAGE_SIZE - len, "%9u:\t", + stat->freq_table[i]); + + for (j = 0; j < stat->state_num; j++) { + if (len >= PAGE_SIZE) + break; + len += snprintf(buf + len, PAGE_SIZE - len, "%u\t", + stat->trans_table[i*stat->max_state+j]); + } + len += snprintf(buf + len, PAGE_SIZE - len, "\n"); + } + return len; +} +CPUFREQ_STATDEVICE_ATTR(trans_table,0444,show_trans_table); +#endif + +CPUFREQ_STATDEVICE_ATTR(total_trans,0444,show_total_trans); +CPUFREQ_STATDEVICE_ATTR(time_in_state,0444,show_time_in_state); + +static struct attribute *default_attrs[] = { + &_attr_total_trans.attr, + &_attr_time_in_state.attr, +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS + &_attr_trans_table.attr, +#endif + NULL +}; +static struct attribute_group stats_attr_group = { + .attrs = default_attrs, + .name = "stats" +}; + +static int +freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq) +{ + int index; + for (index = 0; index < stat->max_state; index++) + if (stat->freq_table[index] == freq) + return index; + return -1; +} + +static void +cpufreq_stats_free_table (unsigned int cpu) +{ + struct cpufreq_stats *stat = cpufreq_stats_table[cpu]; + struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); + if (policy && policy->cpu == cpu) + sysfs_remove_group(&policy->kobj, &stats_attr_group); + if (stat) { + kfree(stat->time_in_state); + kfree(stat); + } + cpufreq_stats_table[cpu] = NULL; + if (policy) + cpufreq_cpu_put(policy); +} + +static int +cpufreq_stats_create_table (struct cpufreq_policy *policy, + struct cpufreq_frequency_table *table) +{ + unsigned int i, j, count = 0, ret = 0; + struct cpufreq_stats *stat; + struct cpufreq_policy *data; + unsigned int alloc_size; + unsigned int cpu = policy->cpu; + if (cpufreq_stats_table[cpu]) + return -EBUSY; + if ((stat = kmalloc(sizeof(struct cpufreq_stats), GFP_KERNEL)) == NULL) + return -ENOMEM; + memset(stat, 0, sizeof (struct cpufreq_stats)); + + data = cpufreq_cpu_get(cpu); + if ((ret = sysfs_create_group(&data->kobj, &stats_attr_group))) + goto error_out; + + stat->cpu = cpu; + cpufreq_stats_table[cpu] = stat; + + for (i=0; table[i].frequency != CPUFREQ_TABLE_END; i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + count++; + } + + alloc_size = count * sizeof(int) + count * sizeof(long long); + +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS + alloc_size += count * count * sizeof(int); +#endif + stat->max_state = count; + stat->time_in_state = kmalloc(alloc_size, GFP_KERNEL); + if (!stat->time_in_state) { + ret = -ENOMEM; + goto error_out; + } + memset(stat->time_in_state, 0, alloc_size); + stat->freq_table = (unsigned int *)(stat->time_in_state + count); + +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS + stat->trans_table = stat->freq_table + count; +#endif + j = 0; + for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + if (freq_table_get_index(stat, freq) == -1) + stat->freq_table[j++] = freq; + } + stat->state_num = j; + spin_lock(&cpufreq_stats_lock); + stat->last_time = jiffies; + stat->last_index = freq_table_get_index(stat, policy->cur); + spin_unlock(&cpufreq_stats_lock); + cpufreq_cpu_put(data); + return 0; +error_out: + cpufreq_cpu_put(data); + kfree(stat); + cpufreq_stats_table[cpu] = NULL; + return ret; +} + +static int +cpufreq_stat_notifier_policy (struct notifier_block *nb, unsigned long val, + void *data) +{ + int ret; + struct cpufreq_policy *policy = data; + struct cpufreq_frequency_table *table; + unsigned int cpu = policy->cpu; + if (val != CPUFREQ_NOTIFY) + return 0; + table = cpufreq_frequency_get_table(cpu); + if (!table) + return 0; + if ((ret = cpufreq_stats_create_table(policy, table))) + return ret; + return 0; +} + +static int +cpufreq_stat_notifier_trans (struct notifier_block *nb, unsigned long val, + void *data) +{ + struct cpufreq_freqs *freq = data; + struct cpufreq_stats *stat; + int old_index, new_index; + + if (val != CPUFREQ_POSTCHANGE) + return 0; + + stat = cpufreq_stats_table[freq->cpu]; + if (!stat) + return 0; + old_index = freq_table_get_index(stat, freq->old); + new_index = freq_table_get_index(stat, freq->new); + + cpufreq_stats_update(freq->cpu); + if (old_index == new_index) + return 0; + + spin_lock(&cpufreq_stats_lock); + stat->last_index = new_index; +#ifdef CONFIG_CPU_FREQ_STAT_DETAILS + stat->trans_table[old_index * stat->max_state + new_index]++; +#endif + stat->total_trans++; + spin_unlock(&cpufreq_stats_lock); + return 0; +} + +static struct notifier_block notifier_policy_block = { + .notifier_call = cpufreq_stat_notifier_policy +}; + +static struct notifier_block notifier_trans_block = { + .notifier_call = cpufreq_stat_notifier_trans +}; + +static int +__init cpufreq_stats_init(void) +{ + int ret; + unsigned int cpu; + spin_lock_init(&cpufreq_stats_lock); + if ((ret = cpufreq_register_notifier(¬ifier_policy_block, + CPUFREQ_POLICY_NOTIFIER))) + return ret; + + if ((ret = cpufreq_register_notifier(¬ifier_trans_block, + CPUFREQ_TRANSITION_NOTIFIER))) { + cpufreq_unregister_notifier(¬ifier_policy_block, + CPUFREQ_POLICY_NOTIFIER); + return ret; + } + + for_each_cpu(cpu) + cpufreq_update_policy(cpu); + return 0; +} +static void +__exit cpufreq_stats_exit(void) +{ + unsigned int cpu; + cpufreq_unregister_notifier(¬ifier_policy_block, + CPUFREQ_POLICY_NOTIFIER); + cpufreq_unregister_notifier(¬ifier_trans_block, + CPUFREQ_TRANSITION_NOTIFIER); + for_each_cpu(cpu) + cpufreq_stats_free_table(cpu); +} + +MODULE_AUTHOR ("Zou Nan hai <nanhai.zou@intel.com>"); +MODULE_DESCRIPTION ("'cpufreq_stats' - A driver to export cpufreq stats through sysfs filesystem"); +MODULE_LICENSE ("GPL"); + +module_init(cpufreq_stats_init); +module_exit(cpufreq_stats_exit); diff --git a/drivers/cpufreq/cpufreq_userspace.c b/drivers/cpufreq/cpufreq_userspace.c new file mode 100644 index 000000000000..d32bf3593cd3 --- /dev/null +++ b/drivers/cpufreq/cpufreq_userspace.c @@ -0,0 +1,207 @@ +/* + * linux/drivers/cpufreq/cpufreq_userspace.c + * + * Copyright (C) 2001 Russell King + * (C) 2002 - 2004 Dominik Brodowski <linux@brodo.de> + * + * 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. + * + */ + +#include <linux/config.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/smp.h> +#include <linux/init.h> +#include <linux/spinlock.h> +#include <linux/interrupt.h> +#include <linux/cpufreq.h> +#include <linux/types.h> +#include <linux/fs.h> +#include <linux/sysfs.h> + +#include <asm/uaccess.h> + + +/** + * A few values needed by the userspace governor + */ +static unsigned int cpu_max_freq[NR_CPUS]; +static unsigned int cpu_min_freq[NR_CPUS]; +static unsigned int cpu_cur_freq[NR_CPUS]; /* current CPU freq */ +static unsigned int cpu_set_freq[NR_CPUS]; /* CPU freq desired by userspace */ +static unsigned int cpu_is_managed[NR_CPUS]; +static struct cpufreq_policy current_policy[NR_CPUS]; + +static DECLARE_MUTEX (userspace_sem); + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_GOVERNOR, "userspace", msg) + +/* keep track of frequency transitions */ +static int +userspace_cpufreq_notifier(struct notifier_block *nb, unsigned long val, + void *data) +{ + struct cpufreq_freqs *freq = data; + + dprintk("saving cpu_cur_freq of cpu %u to be %u kHz\n", freq->cpu, freq->new); + cpu_cur_freq[freq->cpu] = freq->new; + + return 0; +} + +static struct notifier_block userspace_cpufreq_notifier_block = { + .notifier_call = userspace_cpufreq_notifier +}; + + +/** + * cpufreq_set - set the CPU frequency + * @freq: target frequency in kHz + * @cpu: CPU for which the frequency is to be set + * + * Sets the CPU frequency to freq. + */ +static int cpufreq_set(unsigned int freq, unsigned int cpu) +{ + int ret = -EINVAL; + + dprintk("cpufreq_set for cpu %u, freq %u kHz\n", cpu, freq); + + down(&userspace_sem); + if (!cpu_is_managed[cpu]) + goto err; + + cpu_set_freq[cpu] = freq; + + if (freq < cpu_min_freq[cpu]) + freq = cpu_min_freq[cpu]; + if (freq > cpu_max_freq[cpu]) + freq = cpu_max_freq[cpu]; + + /* + * We're safe from concurrent calls to ->target() here + * as we hold the userspace_sem lock. If we were calling + * cpufreq_driver_target, a deadlock situation might occur: + * A: cpufreq_set (lock userspace_sem) -> cpufreq_driver_target(lock policy->lock) + * B: cpufreq_set_policy(lock policy->lock) -> __cpufreq_governor -> cpufreq_governor_userspace (lock userspace_sem) + */ + ret = __cpufreq_driver_target(¤t_policy[cpu], freq, + CPUFREQ_RELATION_L); + + err: + up(&userspace_sem); + return ret; +} + + +/************************** sysfs interface ************************/ +static ssize_t show_speed (struct cpufreq_policy *policy, char *buf) +{ + return sprintf (buf, "%u\n", cpu_cur_freq[policy->cpu]); +} + +static ssize_t +store_speed (struct cpufreq_policy *policy, const char *buf, size_t count) +{ + unsigned int freq = 0; + unsigned int ret; + + ret = sscanf (buf, "%u", &freq); + if (ret != 1) + return -EINVAL; + + cpufreq_set(freq, policy->cpu); + + return count; +} + +static struct freq_attr freq_attr_scaling_setspeed = +{ + .attr = { .name = "scaling_setspeed", .mode = 0644, .owner = THIS_MODULE }, + .show = show_speed, + .store = store_speed, +}; + +static int cpufreq_governor_userspace(struct cpufreq_policy *policy, + unsigned int event) +{ + unsigned int cpu = policy->cpu; + switch (event) { + case CPUFREQ_GOV_START: + if (!cpu_online(cpu)) + return -EINVAL; + BUG_ON(!policy->cur); + down(&userspace_sem); + cpu_is_managed[cpu] = 1; + cpu_min_freq[cpu] = policy->min; + cpu_max_freq[cpu] = policy->max; + cpu_cur_freq[cpu] = policy->cur; + cpu_set_freq[cpu] = policy->cur; + sysfs_create_file (&policy->kobj, &freq_attr_scaling_setspeed.attr); + memcpy (¤t_policy[cpu], policy, sizeof(struct cpufreq_policy)); + dprintk("managing cpu %u started (%u - %u kHz, currently %u kHz)\n", cpu, cpu_min_freq[cpu], cpu_max_freq[cpu], cpu_cur_freq[cpu]); + up(&userspace_sem); + break; + case CPUFREQ_GOV_STOP: + down(&userspace_sem); + cpu_is_managed[cpu] = 0; + cpu_min_freq[cpu] = 0; + cpu_max_freq[cpu] = 0; + cpu_set_freq[cpu] = 0; + sysfs_remove_file (&policy->kobj, &freq_attr_scaling_setspeed.attr); + dprintk("managing cpu %u stopped\n", cpu); + up(&userspace_sem); + break; + case CPUFREQ_GOV_LIMITS: + down(&userspace_sem); + cpu_min_freq[cpu] = policy->min; + cpu_max_freq[cpu] = policy->max; + dprintk("limit event for cpu %u: %u - %u kHz, currently %u kHz, last set to %u kHz\n", cpu, cpu_min_freq[cpu], cpu_max_freq[cpu], cpu_cur_freq[cpu], cpu_set_freq[cpu]); + if (policy->max < cpu_set_freq[cpu]) { + __cpufreq_driver_target(¤t_policy[cpu], policy->max, + CPUFREQ_RELATION_H); + } else if (policy->min > cpu_set_freq[cpu]) { + __cpufreq_driver_target(¤t_policy[cpu], policy->min, + CPUFREQ_RELATION_L); + } else { + __cpufreq_driver_target(¤t_policy[cpu], cpu_set_freq[cpu], + CPUFREQ_RELATION_L); + } + memcpy (¤t_policy[cpu], policy, sizeof(struct cpufreq_policy)); + up(&userspace_sem); + break; + } + return 0; +} + + +struct cpufreq_governor cpufreq_gov_userspace = { + .name = "userspace", + .governor = cpufreq_governor_userspace, + .owner = THIS_MODULE, +}; +EXPORT_SYMBOL(cpufreq_gov_userspace); + +static int __init cpufreq_gov_userspace_init(void) +{ + cpufreq_register_notifier(&userspace_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); + return cpufreq_register_governor(&cpufreq_gov_userspace); +} + + +static void __exit cpufreq_gov_userspace_exit(void) +{ + cpufreq_unregister_governor(&cpufreq_gov_userspace); + cpufreq_unregister_notifier(&userspace_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER); +} + + +MODULE_AUTHOR ("Dominik Brodowski <linux@brodo.de>, Russell King <rmk@arm.linux.org.uk>"); +MODULE_DESCRIPTION ("CPUfreq policy governor 'userspace'"); +MODULE_LICENSE ("GPL"); + +fs_initcall(cpufreq_gov_userspace_init); +module_exit(cpufreq_gov_userspace_exit); diff --git a/drivers/cpufreq/freq_table.c b/drivers/cpufreq/freq_table.c new file mode 100644 index 000000000000..ba460bdea60f --- /dev/null +++ b/drivers/cpufreq/freq_table.c @@ -0,0 +1,225 @@ +/* + * linux/drivers/cpufreq/freq_table.c + * + * Copyright (C) 2002 - 2003 Dominik Brodowski + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/cpufreq.h> + +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_CORE, "freq-table", msg) + +/********************************************************************* + * FREQUENCY TABLE HELPERS * + *********************************************************************/ + +int cpufreq_frequency_table_cpuinfo(struct cpufreq_policy *policy, + struct cpufreq_frequency_table *table) +{ + unsigned int min_freq = ~0; + unsigned int max_freq = 0; + unsigned int i = 0; + + for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) { + dprintk("table entry %u is invalid, skipping\n", i); + + continue; + } + dprintk("table entry %u: %u kHz, %u index\n", i, freq, table[i].index); + if (freq < min_freq) + min_freq = freq; + if (freq > max_freq) + max_freq = freq; + } + + policy->min = policy->cpuinfo.min_freq = min_freq; + policy->max = policy->cpuinfo.max_freq = max_freq; + + if (policy->min == ~0) + return -EINVAL; + else + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_cpuinfo); + + +int cpufreq_frequency_table_verify(struct cpufreq_policy *policy, + struct cpufreq_frequency_table *table) +{ + unsigned int next_larger = ~0; + unsigned int i = 0; + unsigned int count = 0; + + dprintk("request for verification of policy (%u - %u kHz) for cpu %u\n", policy->min, policy->max, policy->cpu); + + if (!cpu_online(policy->cpu)) + return -EINVAL; + + cpufreq_verify_within_limits(policy, + policy->cpuinfo.min_freq, + policy->cpuinfo.max_freq); + + for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + if ((freq >= policy->min) && (freq <= policy->max)) + count++; + else if ((next_larger > freq) && (freq > policy->max)) + next_larger = freq; + } + + if (!count) + policy->max = next_larger; + + cpufreq_verify_within_limits(policy, + policy->cpuinfo.min_freq, + policy->cpuinfo.max_freq); + + dprintk("verification lead to (%u - %u kHz) for cpu %u\n", policy->min, policy->max, policy->cpu); + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_verify); + + +int cpufreq_frequency_table_target(struct cpufreq_policy *policy, + struct cpufreq_frequency_table *table, + unsigned int target_freq, + unsigned int relation, + unsigned int *index) +{ + struct cpufreq_frequency_table optimal = { .index = ~0, }; + struct cpufreq_frequency_table suboptimal = { .index = ~0, }; + unsigned int i; + + dprintk("request for target %u kHz (relation: %u) for cpu %u\n", target_freq, relation, policy->cpu); + + switch (relation) { + case CPUFREQ_RELATION_H: + optimal.frequency = 0; + suboptimal.frequency = ~0; + break; + case CPUFREQ_RELATION_L: + optimal.frequency = ~0; + suboptimal.frequency = 0; + break; + } + + if (!cpu_online(policy->cpu)) + return -EINVAL; + + for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + if ((freq < policy->min) || (freq > policy->max)) + continue; + switch(relation) { + case CPUFREQ_RELATION_H: + if (freq <= target_freq) { + if (freq >= optimal.frequency) { + optimal.frequency = freq; + optimal.index = i; + } + } else { + if (freq <= suboptimal.frequency) { + suboptimal.frequency = freq; + suboptimal.index = i; + } + } + break; + case CPUFREQ_RELATION_L: + if (freq >= target_freq) { + if (freq <= optimal.frequency) { + optimal.frequency = freq; + optimal.index = i; + } + } else { + if (freq >= suboptimal.frequency) { + suboptimal.frequency = freq; + suboptimal.index = i; + } + } + break; + } + } + if (optimal.index > i) { + if (suboptimal.index > i) + return -EINVAL; + *index = suboptimal.index; + } else + *index = optimal.index; + + dprintk("target is %u (%u kHz, %u)\n", *index, table[*index].frequency, + table[*index].index); + + return 0; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_target); + +static struct cpufreq_frequency_table *show_table[NR_CPUS]; +/** + * show_scaling_governor - show the current policy for the specified CPU + */ +static ssize_t show_available_freqs (struct cpufreq_policy *policy, char *buf) +{ + unsigned int i = 0; + unsigned int cpu = policy->cpu; + ssize_t count = 0; + struct cpufreq_frequency_table *table; + + if (!show_table[cpu]) + return -ENODEV; + + table = show_table[cpu]; + + for (i=0; (table[i].frequency != CPUFREQ_TABLE_END); i++) { + if (table[i].frequency == CPUFREQ_ENTRY_INVALID) + continue; + count += sprintf(&buf[count], "%d ", table[i].frequency); + } + count += sprintf(&buf[count], "\n"); + + return count; + +} + +struct freq_attr cpufreq_freq_attr_scaling_available_freqs = { + .attr = { .name = "scaling_available_frequencies", .mode = 0444, .owner=THIS_MODULE }, + .show = show_available_freqs, +}; +EXPORT_SYMBOL_GPL(cpufreq_freq_attr_scaling_available_freqs); + +/* + * if you use these, you must assure that the frequency table is valid + * all the time between get_attr and put_attr! + */ +void cpufreq_frequency_table_get_attr(struct cpufreq_frequency_table *table, + unsigned int cpu) +{ + dprintk("setting show_table for cpu %u to %p\n", cpu, table); + show_table[cpu] = table; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_get_attr); + +void cpufreq_frequency_table_put_attr(unsigned int cpu) +{ + dprintk("clearing show_table for cpu %u\n", cpu); + show_table[cpu] = NULL; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_table_put_attr); + +struct cpufreq_frequency_table *cpufreq_frequency_get_table(unsigned int cpu) +{ + return show_table[cpu]; +} +EXPORT_SYMBOL_GPL(cpufreq_frequency_get_table); + +MODULE_AUTHOR ("Dominik Brodowski <linux@brodo.de>"); +MODULE_DESCRIPTION ("CPUfreq frequency table helpers"); +MODULE_LICENSE ("GPL"); |