From dfa5bb622555d9da0df21b50f46ebdeef390041b Mon Sep 17 00:00:00 2001 From: Stratos Karafotis Date: Wed, 5 Jun 2013 19:01:25 +0300 Subject: [PATCH 1/3] cpufreq: ondemand: Change the calculation of target frequency The ondemand governor calculates load in terms of frequency and increases it only if load_freq is greater than up_threshold multiplied by the current or average frequency. This appears to produce oscillations of frequency between min and max because, for example, a relatively small load can easily saturate minimum frequency and lead the CPU to the max. Then, it will decrease back to the min due to small load_freq. Change the calculation method of load and target frequency on the basis of the following two observations: - Load computation should not depend on the current or average measured frequency. For example, absolute load of 80% at 100MHz is not necessarily equivalent to 8% at 1000MHz in the next sampling interval. - It should be possible to increase the target frequency to any value present in the frequency table proportional to the absolute load, rather than to the max only, so that: Target frequency = C * load where we take C = policy->cpuinfo.max_freq / 100. Tested on Intel i7-3770 CPU @ 3.40GHz and on Quad core 1500MHz Krait. Phoronix benchmark of Linux Kernel Compilation 3.1 test shows an increase ~1.5% in performance. cpufreq_stats (time_in_state) shows that middle frequencies are used more, with this patch. Highest and lowest frequencies were used less by ~9%. [rjw: We have run multiple other tests on kernels with this change applied and in the vast majority of cases it turns out that the resulting performance improvement also leads to reduced consumption of energy. The change is additionally justified by the overall simplification of the code in question.] Signed-off-by: Stratos Karafotis Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/cpufreq_governor.c | 10 +------- drivers/cpufreq/cpufreq_governor.h | 1 - drivers/cpufreq/cpufreq_ondemand.c | 39 ++++++------------------------ 3 files changed, 8 insertions(+), 42 deletions(-) diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c index 7b839a8db2a7..7409dbd1d897 100644 --- a/drivers/cpufreq/cpufreq_governor.c +++ b/drivers/cpufreq/cpufreq_governor.c @@ -53,7 +53,7 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu) policy = cdbs->cur_policy; - /* Get Absolute Load (in terms of freq for ondemand gov) */ + /* Get Absolute Load */ for_each_cpu(j, policy->cpus) { struct cpu_dbs_common_info *j_cdbs; u64 cur_wall_time, cur_idle_time; @@ -104,14 +104,6 @@ void dbs_check_cpu(struct dbs_data *dbs_data, int cpu) load = 100 * (wall_time - idle_time) / wall_time; - if (dbs_data->cdata->governor == GOV_ONDEMAND) { - int freq_avg = __cpufreq_driver_getavg(policy, j); - if (freq_avg <= 0) - freq_avg = policy->cur; - - load *= freq_avg; - } - if (load > max_load) max_load = load; } diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h index 6663ec3b3056..0e0dd4c82020 100644 --- a/drivers/cpufreq/cpufreq_governor.h +++ b/drivers/cpufreq/cpufreq_governor.h @@ -169,7 +169,6 @@ struct od_dbs_tuners { unsigned int sampling_rate; unsigned int sampling_down_factor; unsigned int up_threshold; - unsigned int adj_up_threshold; unsigned int powersave_bias; unsigned int io_is_busy; }; diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index 93eb5cbcc1f6..a3c5574f9b3a 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -29,11 +29,9 @@ #include "cpufreq_governor.h" /* On-demand governor macros */ -#define DEF_FREQUENCY_DOWN_DIFFERENTIAL (10) #define DEF_FREQUENCY_UP_THRESHOLD (80) #define DEF_SAMPLING_DOWN_FACTOR (1) #define MAX_SAMPLING_DOWN_FACTOR (100000) -#define MICRO_FREQUENCY_DOWN_DIFFERENTIAL (3) #define MICRO_FREQUENCY_UP_THRESHOLD (95) #define MICRO_FREQUENCY_MIN_SAMPLE_RATE (10000) #define MIN_FREQUENCY_UP_THRESHOLD (11) @@ -161,14 +159,10 @@ static void dbs_freq_increase(struct cpufreq_policy *p, unsigned int freq) /* * Every sampling_rate, we check, if current idle time is less than 20% - * (default), then we try to increase frequency. Every sampling_rate, we look - * for the lowest frequency which can sustain the load while keeping idle time - * over 30%. If such a frequency exist, we try to decrease to this frequency. - * - * Any frequency increase takes it to the maximum frequency. Frequency reduction - * happens at minimum steps of 5% (default) of current frequency + * (default), then we try to increase frequency. Else, we adjust the frequency + * proportional to load. */ -static void od_check_cpu(int cpu, unsigned int load_freq) +static void od_check_cpu(int cpu, unsigned int load) { struct od_cpu_dbs_info_s *dbs_info = &per_cpu(od_cpu_dbs_info, cpu); struct cpufreq_policy *policy = dbs_info->cdbs.cur_policy; @@ -178,29 +172,17 @@ static void od_check_cpu(int cpu, unsigned int load_freq) dbs_info->freq_lo = 0; /* Check for frequency increase */ - if (load_freq > od_tuners->up_threshold * policy->cur) { + if (load > od_tuners->up_threshold) { /* If switching to max speed, apply sampling_down_factor */ if (policy->cur < policy->max) dbs_info->rate_mult = od_tuners->sampling_down_factor; dbs_freq_increase(policy, policy->max); return; - } - - /* Check for frequency decrease */ - /* if we cannot reduce the frequency anymore, break out early */ - if (policy->cur == policy->min) - return; - - /* - * The optimal frequency is the frequency that is the lowest that can - * support the current CPU usage without triggering the up policy. To be - * safe, we focus 10 points under the threshold. - */ - if (load_freq < od_tuners->adj_up_threshold - * policy->cur) { + } else { + /* Calculate the next frequency proportional to load */ unsigned int freq_next; - freq_next = load_freq / od_tuners->adj_up_threshold; + freq_next = load * policy->cpuinfo.max_freq / 100; /* No longer fully busy, reset rate_mult */ dbs_info->rate_mult = 1; @@ -374,9 +356,6 @@ static ssize_t store_up_threshold(struct dbs_data *dbs_data, const char *buf, input < MIN_FREQUENCY_UP_THRESHOLD) { return -EINVAL; } - /* Calculate the new adj_up_threshold */ - od_tuners->adj_up_threshold += input; - od_tuners->adj_up_threshold -= od_tuners->up_threshold; od_tuners->up_threshold = input; return count; @@ -525,8 +504,6 @@ static int od_init(struct dbs_data *dbs_data) if (idle_time != -1ULL) { /* Idle micro accounting is supported. Use finer thresholds */ tuners->up_threshold = MICRO_FREQUENCY_UP_THRESHOLD; - tuners->adj_up_threshold = MICRO_FREQUENCY_UP_THRESHOLD - - MICRO_FREQUENCY_DOWN_DIFFERENTIAL; /* * In nohz/micro accounting case we set the minimum frequency * not depending on HZ, but fixed (very low). The deferred @@ -535,8 +512,6 @@ static int od_init(struct dbs_data *dbs_data) dbs_data->min_sampling_rate = MICRO_FREQUENCY_MIN_SAMPLE_RATE; } else { tuners->up_threshold = DEF_FREQUENCY_UP_THRESHOLD; - tuners->adj_up_threshold = DEF_FREQUENCY_UP_THRESHOLD - - DEF_FREQUENCY_DOWN_DIFFERENTIAL; /* For correct statistics, we need 10 ticks for each measure */ dbs_data->min_sampling_rate = MIN_SAMPLING_RATE_RATIO * From 61c63e5ed3b9c472899d7152e961f2ffaafcf5a0 Mon Sep 17 00:00:00 2001 From: Stratos Karafotis Date: Wed, 5 Jun 2013 19:01:42 +0300 Subject: [PATCH 2/3] cpufreq: Remove unused APERF/MPERF support The target frequency calculation method in the ondemand governor has changed and it is now independent of the measured average frequency. Consequently, the APERF/MPERF support in cpufreq is not used any more, so drop it. [rjw: Changelog] Signed-off-by: Stratos Karafotis Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- arch/x86/include/asm/processor.h | 29 ------------------ drivers/cpufreq/Makefile | 2 +- drivers/cpufreq/acpi-cpufreq.c | 5 ---- drivers/cpufreq/mperf.c | 51 -------------------------------- drivers/cpufreq/mperf.h | 9 ------ 5 files changed, 1 insertion(+), 95 deletions(-) delete mode 100644 drivers/cpufreq/mperf.c delete mode 100644 drivers/cpufreq/mperf.h diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 24cf5aefb704..4f4a3d98c170 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -942,35 +942,6 @@ extern int set_tsc_mode(unsigned int val); extern u16 amd_get_nb_id(int cpu); -struct aperfmperf { - u64 aperf, mperf; -}; - -static inline void get_aperfmperf(struct aperfmperf *am) -{ - WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_APERFMPERF)); - - rdmsrl(MSR_IA32_APERF, am->aperf); - rdmsrl(MSR_IA32_MPERF, am->mperf); -} - -#define APERFMPERF_SHIFT 10 - -static inline -unsigned long calc_aperfmperf_ratio(struct aperfmperf *old, - struct aperfmperf *new) -{ - u64 aperf = new->aperf - old->aperf; - u64 mperf = new->mperf - old->mperf; - unsigned long ratio = aperf; - - mperf >>= APERFMPERF_SHIFT; - if (mperf) - ratio = div64_u64(aperf, mperf); - - return ratio; -} - extern unsigned long arch_align_stack(unsigned long sp); extern void free_init_pages(char *what, unsigned long begin, unsigned long end); diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile index d345b5a7aa71..ad5866c2ada0 100644 --- a/drivers/cpufreq/Makefile +++ b/drivers/cpufreq/Makefile @@ -23,7 +23,7 @@ obj-$(CONFIG_GENERIC_CPUFREQ_CPU0) += cpufreq-cpu0.o # powernow-k8 can load then. ACPI is preferred to all other hardware-specific drivers. # speedstep-* is preferred over p4-clockmod. -obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o mperf.o +obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index 39264020b88a..e673670d2321 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -45,7 +45,6 @@ #include #include #include -#include "mperf.h" MODULE_AUTHOR("Paul Diefenbaugh, Dominik Brodowski"); MODULE_DESCRIPTION("ACPI Processor P-States Driver"); @@ -861,10 +860,6 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) /* notify BIOS that we exist */ acpi_processor_notify_smm(THIS_MODULE); - /* Check for APERF/MPERF support in hardware */ - if (boot_cpu_has(X86_FEATURE_APERFMPERF)) - acpi_cpufreq_driver.getavg = cpufreq_get_measured_perf; - pr_debug("CPU%u - ACPI performance management activated.\n", cpu); for (i = 0; i < perf->state_count; i++) pr_debug(" %cP%d: %d MHz, %d mW, %d uS\n", diff --git a/drivers/cpufreq/mperf.c b/drivers/cpufreq/mperf.c deleted file mode 100644 index 911e193018ae..000000000000 --- a/drivers/cpufreq/mperf.c +++ /dev/null @@ -1,51 +0,0 @@ -#include -#include -#include -#include -#include -#include - -#include "mperf.h" - -static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf); - -/* Called via smp_call_function_single(), on the target CPU */ -static void read_measured_perf_ctrs(void *_cur) -{ - struct aperfmperf *am = _cur; - - get_aperfmperf(am); -} - -/* - * Return the measured active (C0) frequency on this CPU since last call - * to this function. - * Input: cpu number - * Return: Average CPU frequency in terms of max frequency (zero on error) - * - * We use IA32_MPERF and IA32_APERF MSRs to get the measured performance - * over a period of time, while CPU is in C0 state. - * IA32_MPERF counts at the rate of max advertised frequency - * IA32_APERF counts at the rate of actual CPU frequency - * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and - * no meaning should be associated with absolute values of these MSRs. - */ -unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, - unsigned int cpu) -{ - struct aperfmperf perf; - unsigned long ratio; - unsigned int retval; - - if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) - return 0; - - ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf); - per_cpu(acfreq_old_perf, cpu) = perf; - - retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; - - return retval; -} -EXPORT_SYMBOL_GPL(cpufreq_get_measured_perf); -MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/mperf.h b/drivers/cpufreq/mperf.h deleted file mode 100644 index 5dbf2950dc22..000000000000 --- a/drivers/cpufreq/mperf.h +++ /dev/null @@ -1,9 +0,0 @@ -/* - * (c) 2010 Advanced Micro Devices, Inc. - * Your use of this code is subject to the terms and conditions of the - * GNU general public license version 2. See "COPYING" or - * http://www.gnu.org/licenses/gpl.html - */ - -unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, - unsigned int cpu); From cffe4e0e7413eb29fb8bd035c8b12b33a4b8522a Mon Sep 17 00:00:00 2001 From: Stratos Karafotis Date: Wed, 5 Jun 2013 19:01:50 +0300 Subject: [PATCH 3/3] cpufreq: Remove unused function __cpufreq_driver_getavg() The target frequency calculation method in the ondemand governor has changed and it is now independent of the measured average frequency. Consequently, the __cpufreq_driver_getavg() function and getavg member of struct cpufreq_driver are not used any more, so drop them. [rjw: Changelog] Signed-off-by: Stratos Karafotis Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/cpufreq.c | 12 ------------ include/linux/cpufreq.h | 6 ------ 2 files changed, 18 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index a4ad7339588d..9a9d8ee9faec 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1593,18 +1593,6 @@ fail: } EXPORT_SYMBOL_GPL(cpufreq_driver_target); -int __cpufreq_driver_getavg(struct cpufreq_policy *policy, unsigned int cpu) -{ - if (cpufreq_disabled()) - return 0; - - if (!cpufreq_driver->getavg) - return 0; - - return cpufreq_driver->getavg(policy, cpu); -} -EXPORT_SYMBOL_GPL(__cpufreq_driver_getavg); - /* * when "event" is CPUFREQ_GOV_LIMITS */ diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 90d5a15120d5..e1fd215e16c9 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -216,10 +216,6 @@ extern int cpufreq_driver_target(struct cpufreq_policy *policy, extern int __cpufreq_driver_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation); - -extern int __cpufreq_driver_getavg(struct cpufreq_policy *policy, - unsigned int cpu); - int cpufreq_register_governor(struct cpufreq_governor *governor); void cpufreq_unregister_governor(struct cpufreq_governor *governor); @@ -258,8 +254,6 @@ struct cpufreq_driver { unsigned int (*get) (unsigned int cpu); /* optional */ - unsigned int (*getavg) (struct cpufreq_policy *policy, - unsigned int cpu); int (*bios_limit) (int cpu, unsigned int *limit); int (*exit) (struct cpufreq_policy *policy);