Home
last modified time | relevance | path

Searched full:cpus (Results 1 – 25 of 2511) sorted by relevance

12345678910>>...101

/Linux-v5.10/tools/lib/perf/
Dcpumap.c15 struct perf_cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int)); in perf_cpu_map__dummy_new() local
17 if (cpus != NULL) { in perf_cpu_map__dummy_new()
18 cpus->nr = 1; in perf_cpu_map__dummy_new()
19 cpus->map[0] = -1; in perf_cpu_map__dummy_new()
20 refcount_set(&cpus->refcnt, 1); in perf_cpu_map__dummy_new()
23 return cpus; in perf_cpu_map__dummy_new()
50 struct perf_cpu_map *cpus; in cpu_map__default_new() local
57 cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int)); in cpu_map__default_new()
58 if (cpus != NULL) { in cpu_map__default_new()
62 cpus->map[i] = i; in cpu_map__default_new()
[all …]
Devlist.c42 * We already have cpus for evsel (via PMU sysfs) so in __perf_evlist__propagate_maps()
46 perf_cpu_map__put(evsel->cpus); in __perf_evlist__propagate_maps()
47 evsel->cpus = perf_cpu_map__get(evlist->cpus); in __perf_evlist__propagate_maps()
48 } else if (!evsel->system_wide && perf_cpu_map__empty(evlist->cpus)) { in __perf_evlist__propagate_maps()
49 perf_cpu_map__put(evsel->cpus); in __perf_evlist__propagate_maps()
50 evsel->cpus = perf_cpu_map__get(evlist->cpus); in __perf_evlist__propagate_maps()
51 } else if (evsel->cpus != evsel->own_cpus) { in __perf_evlist__propagate_maps()
52 perf_cpu_map__put(evsel->cpus); in __perf_evlist__propagate_maps()
53 evsel->cpus = perf_cpu_map__get(evsel->own_cpus); in __perf_evlist__propagate_maps()
58 evlist->all_cpus = perf_cpu_map__merge(evlist->all_cpus, evsel->cpus); in __perf_evlist__propagate_maps()
[all …]
/Linux-v5.10/drivers/cpuidle/
Dcoupled.c3 * coupled.c - helper functions to enter the same idle state on multiple cpus
24 * cpus cannot be independently powered down, either due to
31 * shared between the cpus (L2 cache, interrupt controller, and
33 * be tightly controlled on both cpus.
36 * WFI state until all cpus are ready to enter a coupled state, at
38 * cpus at approximately the same time.
40 * Once all cpus are ready to enter idle, they are woken by an smp
42 * cpus will find work to do, and choose not to enter idle. A
43 * final pass is needed to guarantee that all cpus will call the
46 * ready counter matches the number of online coupled cpus. If any
[all …]
/Linux-v5.10/Documentation/admin-guide/cgroup-v1/
Dcpusets.rst31 2.2 Adding/removing cpus
43 Cpusets provide a mechanism for assigning a set of CPUs and Memory
57 include CPUs in its CPU affinity mask, and using the mbind(2) and
60 CPUs or Memory Nodes not in that cpuset. The scheduler will not
67 cpusets and which CPUs and Memory Nodes are assigned to each cpuset,
75 The management of large computer systems, with many processors (CPUs),
113 Cpusets provide a Linux kernel mechanism to constrain which CPUs and
117 CPUs a task may be scheduled (sched_setaffinity) and on which Memory
122 - Cpusets are sets of allowed CPUs and Memory Nodes, known to the
126 - Calls to sched_setaffinity are filtered to just those CPUs
[all …]
/Linux-v5.10/Documentation/timers/
Dno_hz.rst19 2. Omit scheduling-clock ticks on idle CPUs (CONFIG_NO_HZ_IDLE=y or
23 3. Omit scheduling-clock ticks on CPUs that are either idle or that
65 Omit Scheduling-Clock Ticks For Idle CPUs
74 scheduling-clock interrupts to idle CPUs, which is critically important
82 idle CPUs. That said, dyntick-idle mode is not free:
104 Omit Scheduling-Clock Ticks For CPUs With Only One Runnable Task
109 Note that omitting scheduling-clock ticks for CPUs with only one runnable
110 task implies also omitting them for idle CPUs.
113 sending scheduling-clock interrupts to CPUs with a single runnable task,
114 and such CPUs are said to be "adaptive-ticks CPUs". This is important
[all …]
/Linux-v5.10/tools/perf/tests/
Dopenat-syscall-all-cpus.c26 struct perf_cpu_map *cpus; in test__openat_syscall_event_on_all_cpus() local
39 cpus = perf_cpu_map__new(NULL); in test__openat_syscall_event_on_all_cpus()
40 if (cpus == NULL) { in test__openat_syscall_event_on_all_cpus()
54 if (evsel__open(evsel, cpus, threads) < 0) { in test__openat_syscall_event_on_all_cpus()
61 for (cpu = 0; cpu < cpus->nr; ++cpu) { in test__openat_syscall_event_on_all_cpus()
66 * without CPU_ALLOC. 1024 cpus in 2010 still seems in test__openat_syscall_event_on_all_cpus()
69 if (cpus->map[cpu] >= CPU_SETSIZE) { in test__openat_syscall_event_on_all_cpus()
70 pr_debug("Ignoring CPU %d\n", cpus->map[cpu]); in test__openat_syscall_event_on_all_cpus()
74 CPU_SET(cpus->map[cpu], &cpu_set); in test__openat_syscall_event_on_all_cpus()
77 cpus->map[cpu], in test__openat_syscall_event_on_all_cpus()
[all …]
Dcpumap.c54 struct cpu_map_entries *cpus; in process_event_cpus() local
62 cpus = (struct cpu_map_entries *)data->data; in process_event_cpus()
64 TEST_ASSERT_VAL("wrong nr", cpus->nr == 2); in process_event_cpus()
65 TEST_ASSERT_VAL("wrong cpu", cpus->cpu[0] == 1); in process_event_cpus()
66 TEST_ASSERT_VAL("wrong cpu", cpus->cpu[1] == 256); in process_event_cpus()
80 struct perf_cpu_map *cpus; in test__cpu_map_synthesize() local
83 cpus = perf_cpu_map__new("0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19"); in test__cpu_map_synthesize()
86 !perf_event__synthesize_cpu_map(NULL, cpus, process_event_mask, NULL)); in test__cpu_map_synthesize()
88 perf_cpu_map__put(cpus); in test__cpu_map_synthesize()
91 cpus = perf_cpu_map__new("1,256"); in test__cpu_map_synthesize()
[all …]
/Linux-v5.10/include/linux/
Dstop_machine.h13 * function to be executed on a single or multiple cpus preempting all
14 * other processes and monopolizing those cpus until it finishes.
18 * cpus are online.
94 * stop_machine: freeze the machine on all CPUs and run this function
97 * @cpus: the cpus to run the @fn() on (NULL = any online cpu)
109 int stop_machine(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus);
112 * stop_machine_cpuslocked: freeze the machine on all CPUs and run this function
115 * @cpus: the cpus to run the @fn() on (NULL = any online cpu)
120 int stop_machine_cpuslocked(cpu_stop_fn_t fn, void *data, const struct cpumask *cpus);
123 const struct cpumask *cpus);
[all …]
Denergy_model.h32 * @cpus: Cpumask covering the CPUs of the domain. It's here
37 * In case of CPU device, a "performance domain" represents a group of CPUs
38 * whose performance is scaled together. All CPUs of a performance domain
40 * a 1-to-1 mapping with CPUFreq policies. In case of other devices the @cpus
46 unsigned long cpus[]; member
49 #define em_span_cpus(em) (to_cpumask((em)->cpus))
68 * In case of CPUs, the power is the one of a single CPU in the domain,
86 * em_cpu_energy() - Estimates the energy consumed by the CPUs of a
89 * @max_util : highest utilization among CPUs of the domain
90 * @sum_util : sum of the utilization of all CPUs in the domain
[all …]
/Linux-v5.10/tools/perf/arch/arm64/util/
Dheader.c17 static int _get_cpuid(char *buf, size_t sz, struct perf_cpu_map *cpus) in _get_cpuid() argument
26 cpus = perf_cpu_map__get(cpus); in _get_cpuid()
28 for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) { in _get_cpuid()
33 sysfs, cpus->map[cpu]); in _get_cpuid()
57 perf_cpu_map__put(cpus); in _get_cpuid()
67 struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); in get_cpuid() local
70 if (!cpus) in get_cpuid()
73 ret = _get_cpuid(buf, sz, cpus); in get_cpuid()
75 perf_cpu_map__put(cpus); in get_cpuid()
85 if (!pmu || !pmu->cpus) in get_cpuid_str()
[all …]
/Linux-v5.10/drivers/clk/sunxi/
Dclk-sun9i-cpus.c7 * Allwinner A80 CPUS clock driver
22 * sun9i_a80_cpus_clk_setup() - Setup function for a80 cpus composite clk
55 struct sun9i_a80_cpus_clk *cpus = to_sun9i_a80_cpus_clk(hw); in sun9i_a80_cpus_clk_recalc_rate() local
60 reg = readl(cpus->reg); in sun9i_a80_cpus_clk_recalc_rate()
155 struct sun9i_a80_cpus_clk *cpus = to_sun9i_a80_cpus_clk(hw); in sun9i_a80_cpus_clk_set_rate() local
162 reg = readl(cpus->reg); in sun9i_a80_cpus_clk_set_rate()
170 writel(reg, cpus->reg); in sun9i_a80_cpus_clk_set_rate()
188 struct sun9i_a80_cpus_clk *cpus; in sun9i_a80_cpus_setup() local
193 cpus = kzalloc(sizeof(*cpus), GFP_KERNEL); in sun9i_a80_cpus_setup()
194 if (!cpus) in sun9i_a80_cpus_setup()
[all …]
/Linux-v5.10/tools/lib/perf/tests/
Dtest-evlist.c30 struct perf_cpu_map *cpus; in test_stat_cpu() local
43 cpus = perf_cpu_map__new(NULL); in test_stat_cpu()
44 __T("failed to create cpus", cpus); in test_stat_cpu()
59 perf_evlist__set_maps(evlist, cpus, NULL); in test_stat_cpu()
65 cpus = perf_evsel__cpus(evsel); in test_stat_cpu()
67 perf_cpu_map__for_each_cpu(cpu, tmp, cpus) { in test_stat_cpu()
78 perf_cpu_map__put(cpus); in test_stat_cpu()
200 struct perf_cpu_map *cpus; in test_mmap_thread() local
245 cpus = perf_cpu_map__dummy_new(); in test_mmap_thread()
246 __T("failed to create cpus", cpus); in test_mmap_thread()
[all …]
Dtest-cpumap.c15 struct perf_cpu_map *cpus; in main() local
21 cpus = perf_cpu_map__dummy_new(); in main()
22 if (!cpus) in main()
25 perf_cpu_map__get(cpus); in main()
26 perf_cpu_map__put(cpus); in main()
27 perf_cpu_map__put(cpus); in main()
/Linux-v5.10/Documentation/admin-guide/
Dcputopology.rst41 internal kernel map of CPUs within the same core.
46 human-readable list of CPUs within the same core.
51 internal kernel map of the CPUs sharing the same physical_package_id.
56 human-readable list of CPUs sharing the same physical_package_id.
61 internal kernel map of CPUs within the same die.
65 human-readable list of CPUs within the same die.
137 offline: CPUs that are not online because they have been
139 of CPUs allowed by the kernel configuration (kernel_max
140 above). [~cpu_online_mask + cpus >= NR_CPUS]
142 online: CPUs that are online and being scheduled [cpu_online_mask]
[all …]
Dkernel-per-CPU-kthreads.rst13 - Documentation/core-api/irq/irq-affinity.rst: Binding interrupts to sets of CPUs.
15 - Documentation/admin-guide/cgroup-v1: Using cgroups to bind tasks to sets of CPUs.
18 of CPUs.
21 call to bind tasks to sets of CPUs.
50 2. Do all eHCA-Infiniband-related work on other CPUs, including
53 provisioned only on selected CPUs.
101 with multiple CPUs, force them all offline before bringing the
102 first one back online. Once you have onlined the CPUs in question,
103 do not offline any other CPUs, because doing so could force the
104 timer back onto one of the CPUs in question.
[all …]
/Linux-v5.10/Documentation/scheduler/
Dsched-energy.rst9 the impact of its decisions on the energy consumed by CPUs. EAS relies on an
10 Energy Model (EM) of the CPUs to select an energy efficient CPU for each task,
59 In short, EAS changes the way CFS tasks are assigned to CPUs. When it is time
64 knowledge about the platform's topology, which include the 'capacity' of CPUs,
72 differentiate CPUs with different computing throughput. The 'capacity' of a CPU
76 tasks and CPUs computed by the Per-Entity Load Tracking (PELT) mechanism. Thanks
79 energy trade-offs. The capacity of CPUs is provided via arch-specific code
99 Let us consider a platform with 12 CPUs, split in 3 performance domains
102 CPUs: 0 1 2 3 4 5 6 7 8 9 10 11
108 containing 6 CPUs. The two root domains are denoted rd1 and rd2 in the
[all …]
Dsched-domains.rst10 Each scheduling domain spans a number of CPUs (stored in the ->span field).
13 i. The top domain for each CPU will generally span all CPUs in the system
15 CPUs will never be given tasks to run unless the CPUs allowed mask is
17 CPUs".
23 to which the domain belongs. Groups may be shared among CPUs as they contain
27 shared between CPUs.
31 load of each of its member CPUs, and only when the load of a group becomes
49 If it succeeds, it looks for the busiest runqueue of all the CPUs' runqueues in
62 In SMP, the parent of the base domain will span all physical CPUs in the
80 CPUs using cpu_attach_domain.
/Linux-v5.10/drivers/cpufreq/
Dcpufreq-dt.c30 cpumask_var_t cpus; member
50 if (cpumask_test_cpu(cpu, priv->cpus)) in cpufreq_dt_find_data()
119 cpumask_copy(policy->cpus, priv->cpus); in cpufreq_init()
129 * Initialize OPP tables for all policy->cpus. They will be shared by in cpufreq_init()
130 * all CPUs which have marked their CPUs shared with OPP bindings. in cpufreq_init()
133 * before updating policy->cpus. Otherwise, we will end up creating in cpufreq_init()
134 * duplicate OPPs for policy->cpus. in cpufreq_init()
138 if (!dev_pm_opp_of_cpumask_add_table(policy->cpus)) in cpufreq_init()
180 dev_pm_opp_of_register_em(cpu_dev, policy->cpus); in cpufreq_init()
188 dev_pm_opp_of_cpumask_remove_table(policy->cpus); in cpufreq_init()
[all …]
/Linux-v5.10/Documentation/devicetree/bindings/csky/
Dcpus.txt5 The device tree allows to describe the layout of CPUs in a system through
6 the "cpus" node, which in turn contains a number of subnodes (ie "cpu")
9 Only SMP system need to care about the cpus node and single processor
10 needn't define cpus node at all.
13 cpus and cpu node bindings definition
16 - cpus node
20 The node name must be "cpus".
22 A cpus node must define the following properties:
59 cpus {
/Linux-v5.10/Documentation/power/
Dsuspend-and-cpuhotplug.rst27 |tasks | | cpus | | | | cpus | |tasks|
59 online CPUs
75 Note down these cpus in | P
100 | Call _cpu_up() [for all those cpus in the frozen_cpus mask, in a loop]
158 the non-boot CPUs are offlined or onlined, the _cpu_*() functions are called
177 update on the CPUs, as discussed below:
184 a. When all the CPUs are identical:
187 to apply the same microcode revision to each of the CPUs.
192 all CPUs, in order to handle case 'b' described below.
195 b. When some of the CPUs are different than the rest:
[all …]
/Linux-v5.10/tools/perf/util/
Dcpumap.c21 static struct perf_cpu_map *cpu_map__from_entries(struct cpu_map_entries *cpus) in cpu_map__from_entries() argument
25 map = perf_cpu_map__empty_new(cpus->nr); in cpu_map__from_entries()
29 for (i = 0; i < cpus->nr; i++) { in cpu_map__from_entries()
35 if (cpus->cpu[i] == (u16) -1) in cpu_map__from_entries()
38 map->map[i] = (int) cpus->cpu[i]; in cpu_map__from_entries()
83 struct perf_cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int) * nr); in perf_cpu_map__empty_new() local
85 if (cpus != NULL) { in perf_cpu_map__empty_new()
88 cpus->nr = nr; in perf_cpu_map__empty_new()
90 cpus->map[i] = -1; in perf_cpu_map__empty_new()
92 refcount_set(&cpus->refcnt, 1); in perf_cpu_map__empty_new()
[all …]
Dperf_api_probe.c63 struct perf_cpu_map *cpus; in perf_probe_api() local
66 cpus = perf_cpu_map__new(NULL); in perf_probe_api()
67 if (!cpus) in perf_probe_api()
69 cpu = cpus->map[0]; in perf_probe_api()
70 perf_cpu_map__put(cpus); in perf_probe_api()
128 struct perf_cpu_map *cpus; in perf_can_record_cpu_wide() local
131 cpus = perf_cpu_map__new(NULL); in perf_can_record_cpu_wide()
132 if (!cpus) in perf_can_record_cpu_wide()
134 cpu = cpus->map[0]; in perf_can_record_cpu_wide()
135 perf_cpu_map__put(cpus); in perf_can_record_cpu_wide()
/Linux-v5.10/tools/testing/selftests/cpu-hotplug/
Dcpu-on-off-test.sh44 echo -e "\t Cpus in online state: $online_cpus"
52 echo -e "\t Cpus in offline state: $offline_cpus"
56 # list all hot-pluggable CPUs
172 echo -e "\t run with -a option to offline all cpus"
206 echo "Full scope test: all hotplug cpus"
207 echo -e "\t online all offline cpus"
208 echo -e "\t offline all online cpus"
209 echo -e "\t online all offline cpus"
213 # Online all hot-pluggable CPUs
220 # Offline all hot-pluggable CPUs
[all …]
/Linux-v5.10/tools/testing/selftests/rcutorture/bin/
Djitter.sh4 # Alternate sleeping and spinning on randomly selected CPUs. The purpose
56 if cpus=`grep 1 /sys/devices/system/cpu/*/online 2>&1 |
61 cpus=
63 # Do not leave out non-hot-pluggable CPUs
64 cpus="$cpus $nohotplugcpus"
66 cpumask=`awk -v cpus="$cpus" -v me=$me -v n=$n 'BEGIN {
68 ncpus = split(cpus, ca);
/Linux-v5.10/Documentation/arm/
Dcluster-pm-race-avoidance.rst18 In a system containing multiple CPUs, it is desirable to have the
19 ability to turn off individual CPUs when the system is idle, reducing
22 In a system containing multiple clusters of CPUs, it is also desirable
27 of independently running CPUs, while the OS continues to run. This
92 CPUs in the cluster simultaneously modifying the state. The cluster-
104 referred to as a "CPU". CPUs are assumed to be single-threaded:
107 This means that CPUs fit the basic model closely.
216 A cluster is a group of connected CPUs with some common resources.
217 Because a cluster contains multiple CPUs, it can be doing multiple
272 which exact CPUs within the cluster play these roles. This must
[all …]

12345678910>>...101