| /kernel/linux/linux-6.6/include/uapi/linux/sched/ |
| D | types.h | 74 * Task Utilization Attributes
77 * A subset of sched_attr attributes allows to specify the utilization
79 * the utilization boundaries within which it should schedule the task. These
83 * @sched_util_min represents the minimum utilization
84 * @sched_util_max represents the maximum utilization
86 * Utilization is a value in the range [0..SCHED_CAPACITY_SCALE]. It
89 * 20% utilization task is a task running for 2ms every 10ms at maximum
92 * A task with a min utilization value bigger than 0 is more likely scheduled
94 * A task with a max utilization value smaller than 1024 is more likely
97 * A task utilization boundary can be reset by setting the attribute to -1.
[all …]
|
| /kernel/linux/linux-6.6/Documentation/scheduler/ |
| D | sched-capacity.rst | 126 2. Task utilization
134 while task utilization is specific to CFS, it is convenient to describe it here
137 Task utilization is a percentage meant to represent the throughput requirements
142 On an SMP system with fixed frequencies, 100% utilization suggests the task is a
143 busy loop. Conversely, 10% utilization hints it is a small periodic task that
172 The task utilization signal can be made frequency invariant using the following
178 task utilization of 25%.
183 CPU capacity has a similar effect on task utilization in that running an
210 The task utilization signal can be made CPU invariant using the following
217 invariant task utilization of 25%.
[all …]
|
| D | sched-energy.rst | 75 normalized in a 1024 range, and are comparable with the utilization signals of
77 to capacity and utilization values, EAS is able to estimate how big/busy a
135 for the CPU with the highest spare capacity (CPU capacity - CPU utilization) in
143 looks at the current utilization landscape of the CPUs and adjusts it to
146 the given utilization landscape.
158 The current utilization landscape of the CPUs is depicted on the graph
188 compared to leaving P on CPU0. EAS assumes that OPPs follow utilization
253 bigs, for example. So, if the little CPUs happen to have enough utilization at
274 impact on throughput for high-utilization scenarios, EAS also implements another
275 mechanism called 'over-utilization'.
[all …]
|
| D | sched-deadline.rst | 183 the task's utilization must be removed from the previous runqueue's active
184 utilization and must be added to the new runqueue's active utilization.
192 its utilization is removed from the runqueue's active utilization.
195 its utilization is added to the active utilization of the runqueue where
219 - Umax is the maximum reclaimable utilization (subjected to RT throttling
221 - Uinact is the (per runqueue) inactive utilization, computed as
223 - Uextra is the (per runqueue) extra reclaimable utilization
343 The utilization of a real-time task is defined as the ratio between its
347 If the total utilization U=sum(WCET_i/P_i) is larger than M (with M equal
350 Note that total utilization is defined as the sum of the utilizations
[all …]
|
| D | sched-nice-design.rst | 46 a CPU utilization, but because it causes too frequent (once per
52 right minimal granularity - and this translates to 5% CPU utilization.
55 terms of CPU utilization, we only got complaints about it (still) being
99 the new scheduler makes nice(1) have the same CPU utilization effect on
102 utilization "split" between them as running a nice -5 and a nice -4
|
| D | sched-util-clamp.rst | 4 Utilization Clamping
10 Utilization clamping, also known as util clamp or uclamp, is a scheduler
22 point; hence the name. That is, by clamping utilization we are making the
39 the uclamp values as performance points rather than utilization is a better
83 how scheduler utilization signal is calculated**.
122 its utilization signal; acting as a bias mechanism that influences certain
125 The actual utilization signal of a task is never clamped in reality. If you
133 which have implications on the utilization value at CPU runqueue (rq for short)
136 When a task wakes up on an rq, the utilization signal of the rq will be
148 The way this is handled is by dividing the utilization range into buckets
[all …]
|
| D | schedutil.rst | 90 - Documentation/scheduler/sched-capacity.rst:"1. CPU Capacity + 2. Task utilization"
97 though when running their expected utilization will be the same, they suffer a
129 the frequency invariant utilization estimate of the CPU. From this we compute
163 will closely reflect utilization.
|
| /kernel/linux/linux-5.10/Documentation/scheduler/ |
| D | sched-capacity.rst | 126 2. Task utilization
134 while task utilization is specific to CFS, it is convenient to describe it here
137 Task utilization is a percentage meant to represent the throughput requirements
142 On an SMP system with fixed frequencies, 100% utilization suggests the task is a
143 busy loop. Conversely, 10% utilization hints it is a small periodic task that
172 The task utilization signal can be made frequency invariant using the following
178 task utilization of 25%.
183 CPU capacity has a similar effect on task utilization in that running an
210 The task utilization signal can be made CPU invariant using the following
217 invariant task utilization of 25%.
[all …]
|
| D | sched-energy.rst | 75 normalized in a 1024 range, and are comparable with the utilization signals of
77 to capacity and utilization values, EAS is able to estimate how big/busy a
135 for the CPU with the highest spare capacity (CPU capacity - CPU utilization) in
143 looks at the current utilization landscape of the CPUs and adjusts it to
146 the given utilization landscape.
158 The current utilization landscape of the CPUs is depicted on the graph
188 compared to leaving P on CPU0. EAS assumes that OPPs follow utilization
253 bigs, for example. So, if the little CPUs happen to have enough utilization at
274 impact on throughput for high-utilization scenarios, EAS also implements another
275 mechanism called 'over-utilization'.
[all …]
|
| D | sched-deadline.rst | 183 the task's utilization must be removed from the previous runqueue's active
184 utilization and must be added to the new runqueue's active utilization.
192 its utilization is removed from the runqueue's active utilization.
195 its utilization is added to the active utilization of the runqueue where
216 - Umax is the maximum reclaimable utilization (subjected to RT throttling
218 - Uinact is the (per runqueue) inactive utilization, computed as
220 - Uextra is the (per runqueue) extra reclaimable utilization
340 The utilization of a real-time task is defined as the ratio between its
344 If the total utilization U=sum(WCET_i/P_i) is larger than M (with M equal
347 Note that total utilization is defined as the sum of the utilizations
[all …]
|
| D | sched-nice-design.rst | 46 a CPU utilization, but because it causes too frequent (once per
52 right minimal granularity - and this translates to 5% CPU utilization.
55 terms of CPU utilization, we only got complaints about it (still) being
99 the new scheduler makes nice(1) have the same CPU utilization effect on
102 utilization "split" between them as running a nice -5 and a nice -4
|
| /kernel/linux/linux-5.10/include/uapi/linux/sched/ |
| D | types.h | 78 * Task Utilization Attributes
81 * A subset of sched_attr attributes allows to specify the utilization
83 * the utilization boundaries within which it should schedule the task. These
87 * @sched_util_min represents the minimum utilization
88 * @sched_util_max represents the maximum utilization
90 * Utilization is a value in the range [0..SCHED_CAPACITY_SCALE]. It
93 * 20% utilization task is a task running for 2ms every 10ms at maximum
96 * A task with a min utilization value bigger than 0 is more likely scheduled
98 * A task with a max utilization value smaller than 1024 is more likely
134 /* Utilization hints */
|
| /kernel/linux/linux-5.10/kernel/sched/ |
| D | cpufreq_schedutil.c | 3 * CPUFreq governor based on scheduler-provided CPU utilization data.
145 * @util: Current CPU utilization.
148 * If the utilization is frequency-invariant, choose the new frequency to be
153 * Otherwise, approximate the would-be frequency-invariant utilization by
181 * This function computes an effective utilization for the given CPU, to be
192 * The cfs,rt,dl utilization are the running times measured with rq->clock_task
194 * in the irq utilization.
197 * based on the task model parameters and gives the minimal utilization
224 * utilization (PELT windows are synchronized) we can directly add them in schedutil_cpu_util()
225 * to obtain the CPU's actual utilization. in schedutil_cpu_util()
[all …]
|
| /kernel/linux/linux-6.6/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/ |
| D | gk20a.c | 125 u32 utilization = 0; in gk20a_pmu_dvfs_work() local
138 utilization = div_u64((u64)status.busy * 100, status.total); in gk20a_pmu_dvfs_work()
140 data->avg_load = (data->p_smooth * data->avg_load) + utilization; in gk20a_pmu_dvfs_work()
142 nvkm_trace(subdev, "utilization = %d %%, avg_load = %d %%\n", in gk20a_pmu_dvfs_work()
143 utilization, data->avg_load); in gk20a_pmu_dvfs_work()
|
| /kernel/linux/linux-5.10/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/ |
| D | gk20a.c | 125 u32 utilization = 0; in gk20a_pmu_dvfs_work() local
138 utilization = div_u64((u64)status.busy * 100, status.total); in gk20a_pmu_dvfs_work()
140 data->avg_load = (data->p_smooth * data->avg_load) + utilization; in gk20a_pmu_dvfs_work()
142 nvkm_trace(subdev, "utilization = %d %%, avg_load = %d %%\n", in gk20a_pmu_dvfs_work()
143 utilization, data->avg_load); in gk20a_pmu_dvfs_work()
|
| /kernel/linux/linux-6.6/drivers/cpufreq/ |
| D | Kconfig | 151 changes frequency based on the CPU utilization.
195 This governor makes decisions based on the utilization data provided
197 the utilization/capacity ratio coming from the scheduler. If the
198 utilization is frequency-invariant, the new frequency is also
201 frequency tipping point is at utilization/capacity equal to 80% in
|
| /kernel/linux/linux-6.6/kernel/sched/ |
| D | cpufreq_schedutil.c | 3 * CPUFreq governor based on scheduler-provided CPU utilization data.
138 * @util: Current CPU utilization.
141 * If the utilization is frequency-invariant, choose the new frequency to be
146 * Otherwise, approximate the would-be frequency-invariant utilization by
220 * Each time a task wakes up after an IO operation, the CPU utilization can be
221 * boosted to a certain utilization which doubles at each "frequent and
222 * successive" wakeup from IO, ranging from IOWAIT_BOOST_MIN to the utilization
226 * otherwise we restart from the utilization of the minimum OPP.
265 * utilization boosted to speed up the completion of those IO operations.
271 * its IO boost utilization reset.
[all …]
|
| /kernel/linux/linux-5.10/drivers/cpufreq/ |
| D | Kconfig | 153 changes frequency based on the CPU utilization.
195 This governor makes decisions based on the utilization data provided
197 the utilization/capacity ratio coming from the scheduler. If the
198 utilization is frequency-invariant, the new frequency is also
201 frequency tipping point is at utilization/capacity equal to 80% in
|
| /kernel/linux/linux-6.6/drivers/gpu/drm/amd/include/ |
| D | kgd_pp_interface.h | 450 /* Utilization */
499 /* Utilization */
558 /* Utilization */
620 /* Utilization */
697 /* Utilization */
744 /* Utilization */
794 /* Utilization */
847 /* Utilization */
906 /* Utilization (unit: centi) */
|
| /kernel/linux/linux-5.10/drivers/devfreq/event/ |
| D | Kconfig | 12 (e.g., raw data, utilization, latency, bandwidth). The events
33 utilization of each module.
|
| /kernel/linux/linux-6.6/drivers/devfreq/event/ |
| D | Kconfig | 12 (e.g., raw data, utilization, latency, bandwidth). The events
33 utilization of each module.
|
| /kernel/linux/linux-6.6/tools/perf/pmu-events/arch/x86/elkhartlake/ |
| D | ehl-metrics.json | 39 "BriefDescription": "Average CPU Utilization",
49 "BriefDescription": "Average Frequency Utilization relative nominal frequency",
|
| /kernel/linux/linux-6.6/include/linux/ |
| D | energy_model.h | 211 * @max_util : highest utilization among CPUs of the domain
212 * @sum_util : sum of the utilization of all CPUs in the domain
221 * a capacity state satisfying the max utilization of the domain.
235 * In order to predict the performance state, map the utilization of in em_cpu_energy()
239 * max utilization to the allowed CPU capacity before calculating in em_cpu_energy()
|
| /kernel/linux/linux-5.10/include/linux/ |
| D | energy_model.h | 105 * @max_util : highest utilization among CPUs of the domain
106 * @sum_util : sum of the utilization of all CPUs in the domain
113 * a capacity state satisfying the max utilization of the domain.
123 * In order to predict the performance state, map the utilization of in em_cpu_energy()
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/broadwellde/ |
| D | bdwde-metrics.json | 87 …"BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by ins…
93 "BriefDescription": "Average CPU Utilization",
105 "BriefDescription": "Average Frequency Utilization relative nominal frequency",
|