/* * drivers/cpufreq/cpufreq_stats.c * * Copyright (C) 2003-2004 Venkatesh Pallipadi . * (C) 2004 Zou Nan hai . * * 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 #include #include #include #include #include #include static spinlock_t cpufreq_stats_lock; 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; u64 *time_in_state; int *tempMax; int *tempMin; u32 *tempCount; //maximum: 0xffffffff which #define TEMPCOUNT_MAX 0xffffffff //represents the overflow state u64 *tempSum; unsigned int *freq_table; #ifdef CONFIG_CPU_FREQ_STAT_DETAILS unsigned int *trans_table; #endif }; static DEFINE_PER_CPU(struct cpufreq_stats *, cpufreq_stats_table); struct cpufreq_stats_attribute { struct attribute attr; ssize_t(*show) (struct cpufreq_stats *, char *); }; static int cpufreq_stats_update(unsigned int cpu, int *temp) { struct cpufreq_stats *stat; unsigned long long cur_time; cur_time = get_jiffies_64(); spin_lock(&cpufreq_stats_lock); stat = per_cpu(cpufreq_stats_table, cpu); if (stat->time_in_state) stat->time_in_state[stat->last_index] += cur_time - stat->last_time; stat->last_time = cur_time; if(temp) { if(stat->tempCount[stat->last_index] == 0) { stat->tempMax[stat->last_index]= *temp; stat->tempMin[stat->last_index]= *temp; stat->tempCount[stat->last_index] = 1; } else { if(stat->tempMax[stat->last_index] < *temp) { stat->tempMax[stat->last_index] = *temp; } if(stat->tempMin[stat->last_index] > *temp) { stat->tempMin[stat->last_index] = *temp; } if( stat->tempCount[stat->last_index] < TEMPCOUNT_MAX ) { stat->tempCount[stat->last_index]++; } //thus TEMPCOUNT_MAX indicates overflow with high probability } if( *temp < 0 ) { *temp=0; } //CPU should never be able to reach negative temperatures. If it does we might consider overclocking for more performance. stat->tempSum[stat->last_index] += *temp; } spin_unlock(&cpufreq_stats_lock); return 0; } static ssize_t show_total_trans(struct cpufreq_policy *policy, char *buf) { struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) return 0; return sprintf(buf, "%d\n", per_cpu(cpufreq_stats_table, stat->cpu)->total_trans); } ssize_t store_time_in_state(struct cpufreq_policy *policy, const char *buf, size_t count) { struct cpufreq_stats *stat; stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) return -EIO; //clear the current statistics memset( stat->time_in_state, 0, stat->max_state * sizeof( *(stat->time_in_state) ) ); memset( stat->tempMax, 0, stat->max_state * sizeof( *(stat->tempMax) ) ); memset( stat->tempMin, 0, stat->max_state * sizeof( *(stat->tempMin) ) ); memset( stat->tempCount, 0, stat->max_state * sizeof( *(stat->tempCount) ) ); memset( stat->tempSum, 0, stat->max_state * sizeof( *(stat->tempSum) ) ); return count; } static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf) { ssize_t len = 0; int i; int temp=-1337; struct cpufreq_stats *stat; stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) return 0; if( TemperatureSensorGetCurrentTemp(0, NULL, &temp) != 0 ) { printk(KERN_ERR "Could not get CPU temperature."); cpufreq_stats_update(stat->cpu, 0); } else { cpufreq_stats_update(stat->cpu, &temp); } len += sprintf(buf + len, "{Frequency TimeInFrequencyState TemperatureMin(1/10 °) TemperatureSum(1/10 °) TemperatureMeasurements TemperatureMax(1/10 °)}[%d]:\n", stat->state_num); for (i = 0; i < stat->state_num; i++) { if( stat->tempCount[i] == 0 ) { len += sprintf(buf + len, "%u %llu none none none none ", stat->freq_table[i], (unsigned long long) cputime64_to_clock_t(stat->time_in_state[i]) ); } else if(stat->tempCount[i]==TEMPCOUNT_MAX) { len += sprintf(buf + len, "%u %llu %d overflow overflow %d ", stat->freq_table[i], (unsigned long long) cputime64_to_clock_t(stat->time_in_state[i]), stat->tempMin[i], stat->tempMax[i] ); } else { len += sprintf(buf + len, "%u %llu %d %llu %u %d ", stat->freq_table[i], (unsigned long long) cputime64_to_clock_t(stat->time_in_state[i]), stat->tempMin[i], stat->tempSum[i], stat->tempCount[i], stat->tempMax[i] ); } } len += sprintf(buf + len, "\n"); 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; int temp=-1337; struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) return 0; if( TemperatureSensorGetCurrentTemp(0, NULL, &temp) != 0 ) { printk(KERN_ERR "Could not get CPU temperature."); cpufreq_stats_update(stat->cpu, 0); } else { cpufreq_stats_update(stat->cpu, &temp); } len += snprintf(buf + len, PAGE_SIZE - len, " From : To\n"); len += snprintf(buf + len, PAGE_SIZE - len, " : "); for (i = 0; i < stat->state_num; i++) { if (len >= PAGE_SIZE) break; len += snprintf(buf + len, PAGE_SIZE - len, "%9u ", stat->freq_table[i]); } if (len >= PAGE_SIZE) return PAGE_SIZE; len += snprintf(buf + len, PAGE_SIZE - len, "\n"); for (i = 0; i < stat->state_num; i++) { if (len >= PAGE_SIZE) break; len += snprintf(buf + len, PAGE_SIZE - len, "%9u: ", stat->freq_table[i]); for (j = 0; j < stat->state_num; j++) { if (len >= PAGE_SIZE) break; len += snprintf(buf + len, PAGE_SIZE - len, "%9u ", stat->trans_table[i*stat->max_state+j]); } if (len >= PAGE_SIZE) break; len += snprintf(buf + len, PAGE_SIZE - len, "\n"); } if (len >= PAGE_SIZE) return PAGE_SIZE; return len; } cpufreq_freq_attr_ro(trans_table); #endif cpufreq_freq_attr_ro(total_trans); cpufreq_freq_attr_rw(time_in_state); static struct attribute *default_attrs[] = { &total_trans.attr, &time_in_state.attr, #ifdef CONFIG_CPU_FREQ_STAT_DETAILS &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(struct cpufreq_policy *policy) { struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->cpu); if (!stat) return; pr_debug("%s: Free stat table\n", __func__); sysfs_remove_group(&policy->kobj, &stats_attr_group); kfree(stat->time_in_state); kfree(stat); per_cpu(cpufreq_stats_table, policy->cpu) = NULL; } static void cpufreq_stats_free_table(unsigned int cpu) { struct cpufreq_policy *policy; policy = cpufreq_cpu_get(cpu); if (!policy) return; if (cpufreq_frequency_get_table(policy->cpu)) __cpufreq_stats_free_table(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 *current_policy; unsigned int alloc_size; unsigned int cpu = policy->cpu; if (per_cpu(cpufreq_stats_table, cpu)) return -EBUSY; stat = kzalloc(sizeof(*stat), GFP_KERNEL); if ((stat) == NULL) return -ENOMEM; current_policy = cpufreq_cpu_get(cpu); if (current_policy == NULL) { ret = -EINVAL; goto error_get_fail; } ret = sysfs_create_group(¤t_policy->kobj, &stats_attr_group); if (ret) goto error_out; stat->cpu = cpu; per_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(u64); #ifdef CONFIG_CPU_FREQ_STAT_DETAILS alloc_size += count * count * sizeof(int); #endif stat->max_state = count; //add space for the current temperature statistics alloc_size += count*sizeof( *(stat->tempMax) ); alloc_size += count*sizeof( *(stat->tempSum) ); alloc_size += count*sizeof( *(stat->tempMin) ); alloc_size += count*sizeof( *(stat->tempCount) ); stat->time_in_state = kzalloc(alloc_size, GFP_KERNEL); if (!stat->time_in_state) { ret = -ENOMEM; goto error_out; } //poke current temperature statistics out of the allocated chunk stat->tempMax = (int *)(stat->time_in_state + count); stat->tempSum = (u64 *)(stat->tempMax + count); stat->tempMin = (int *)(stat->tempSum + count); stat->tempCount = (u32 *)(stat->tempMin + count); stat->freq_table = (unsigned int *)(stat->tempCount + 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 = get_jiffies_64(); stat->last_index = freq_table_get_index(stat, policy->cur); spin_unlock(&cpufreq_stats_lock); cpufreq_cpu_put(current_policy); return 0; error_out: cpufreq_cpu_put(current_policy); error_get_fail: kfree(stat); per_cpu(cpufreq_stats_table, cpu) = NULL; return ret; } static void cpufreq_stats_create_table(unsigned int cpu) { struct cpufreq_policy *policy; struct cpufreq_frequency_table *table; /* * "likely(!policy)" because normally cpufreq_stats will be registered * before cpufreq driver */ policy = cpufreq_cpu_get(cpu); if (likely(!policy)) return; table = cpufreq_frequency_get_table(policy->cpu); if (likely(table)) __cpufreq_stats_create_table(policy, table); cpufreq_cpu_put(policy); } static void cpufreq_stats_update_policy_cpu(struct cpufreq_policy *policy) { struct cpufreq_stats *stat = per_cpu(cpufreq_stats_table, policy->last_cpu); pr_debug("Updating stats_table for new_cpu %u from last_cpu %u\n", policy->cpu, policy->last_cpu); per_cpu(cpufreq_stats_table, policy->cpu) = per_cpu(cpufreq_stats_table, policy->last_cpu); per_cpu(cpufreq_stats_table, policy->last_cpu) = NULL; stat->cpu = policy->cpu; } static int cpufreq_stat_notifier_policy(struct notifier_block *nb, unsigned long val, void *data) { int ret = 0; struct cpufreq_policy *policy = data; struct cpufreq_frequency_table *table; unsigned int cpu = policy->cpu; if (val == CPUFREQ_UPDATE_POLICY_CPU) { cpufreq_stats_update_policy_cpu(policy); return 0; } table = cpufreq_frequency_get_table(cpu); if (!table) return 0; if (val == CPUFREQ_CREATE_POLICY) ret = __cpufreq_stats_create_table(policy, table); else if (val == CPUFREQ_REMOVE_POLICY) __cpufreq_stats_free_table(policy); return ret; } 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; int temp=-1337; stat = per_cpu(cpufreq_stats_table, freq->cpu); if (!stat) return 0; if(val==CPUFREQ_NOCHANGE) { if( TemperatureSensorGetCurrentTemp(0, NULL, &temp) != 0 ) { cpufreq_stats_update(freq->cpu, 0); } else { cpufreq_stats_update(freq->cpu, &temp); } return 0; } if (val != CPUFREQ_POSTCHANGE) return 0; old_index = stat->last_index; new_index = freq_table_get_index(stat, freq->new); /* We can't do stat->time_in_state[-1]= .. */ if (old_index == -1 || new_index == -1) return 0; if( TemperatureSensorGetCurrentTemp(0, NULL, &temp) != 0 ) { cpufreq_stats_update(freq->cpu, 0); } else { cpufreq_stats_update(freq->cpu, &temp); } 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 }; /* Fake CPU Frequency Driver On vr9, the frequency is not adjusted for hardware reasons. However to get statistics on CPU frequency and temperature we employ a fake CPU driver which does nothing but lets the governor think he can govern something so the cpufreq_stats will get updated periodically. */ struct cpufreq_frequency_table freq_table[] = { {.driver_data = 42, .frequency = 0 /* to be set at runtime in fake_init, see below */ }, {.driver_data = 0, .frequency = CPUFREQ_TABLE_END,} }; int fake_target_index(struct cpufreq_policy *policy,unsigned int index) { return 0; } int fake_init(struct cpufreq_policy *policy) { int ret; freq_table[0].frequency = avm_get_clock(avm_clock_id_cpu)/1000; ret = cpufreq_table_validate_and_show(policy, freq_table); if (ret) { pr_err("%s: invalid frequency table: %d\n", __func__, ret); return ret; } /* * The driver only supports the SMP configuartion where all processors * share the clock and voltage and clock. */ cpumask_setall(policy->cpus); return 0; } unsigned int fake_get(unsigned int cpu) { return avm_get_clock(avm_clock_id_cpu)/1000; //current frequency in kHz } static struct cpufreq_driver fake_cpufreq_driver = { .flags = CPUFREQ_STICKY, .verify = cpufreq_generic_frequency_table_verify, .target_index = fake_target_index, .get = fake_get, .init = fake_init, .name = "fake_driver", }; static int __init cpufreq_stats_init(void) { int ret; unsigned int cpu; spin_lock_init(&cpufreq_stats_lock); ret = cpufreq_register_driver(&fake_cpufreq_driver); if( ret != 0) { printk(KERN_ERR "Could not register fake CPU frequency driver.\n"); return ret; } ret = cpufreq_register_notifier(¬ifier_policy_block, CPUFREQ_POLICY_NOTIFIER); if (ret) return ret; for_each_online_cpu(cpu) cpufreq_stats_create_table(cpu); ret = cpufreq_register_notifier(¬ifier_trans_block, CPUFREQ_TRANSITION_NOTIFIER); if (ret) { cpufreq_unregister_notifier(¬ifier_policy_block, CPUFREQ_POLICY_NOTIFIER); for_each_online_cpu(cpu) cpufreq_stats_free_table(cpu); return ret; } 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_online_cpu(cpu) cpufreq_stats_free_table(cpu); } MODULE_AUTHOR("Zou Nan hai "); 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);