summaryrefslogtreecommitdiff
path: root/tools/tracing/rtla/src
diff options
context:
space:
mode:
Diffstat (limited to 'tools/tracing/rtla/src')
-rw-r--r--tools/tracing/rtla/src/osnoise_top.c2
-rw-r--r--tools/tracing/rtla/src/timerlat_hist.c64
-rw-r--r--tools/tracing/rtla/src/timerlat_top.c54
-rw-r--r--tools/tracing/rtla/src/utils.c186
-rw-r--r--tools/tracing/rtla/src/utils.h15
5 files changed, 281 insertions, 40 deletions
diff --git a/tools/tracing/rtla/src/osnoise_top.c b/tools/tracing/rtla/src/osnoise_top.c
index 30e3853076a0..45647495ce3b 100644
--- a/tools/tracing/rtla/src/osnoise_top.c
+++ b/tools/tracing/rtla/src/osnoise_top.c
@@ -627,7 +627,7 @@ osnoise_top_apply_config(struct osnoise_tool *tool, struct osnoise_top_params *p
auto_house_keeping(&params->monitored_cpus);
}
- if (isatty(1) && !params->quiet)
+ if (isatty(STDOUT_FILENO) && !params->quiet)
params->pretty_output = 1;
return 0;
diff --git a/tools/tracing/rtla/src/timerlat_hist.c b/tools/tracing/rtla/src/timerlat_hist.c
index a3907c390d67..8b66387e5f35 100644
--- a/tools/tracing/rtla/src/timerlat_hist.c
+++ b/tools/tracing/rtla/src/timerlat_hist.c
@@ -55,6 +55,7 @@ struct timerlat_hist_params {
int entries;
int warmup;
int buffer_size;
+ int deepest_idle_state;
};
struct timerlat_hist_cpu {
@@ -62,9 +63,9 @@ struct timerlat_hist_cpu {
int *thread;
int *user;
- int irq_count;
- int thread_count;
- int user_count;
+ unsigned long long irq_count;
+ unsigned long long thread_count;
+ unsigned long long user_count;
unsigned long long min_irq;
unsigned long long sum_irq;
@@ -304,15 +305,15 @@ timerlat_print_summary(struct timerlat_hist_params *params,
continue;
if (!params->no_irq)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
data->hist[cpu].irq_count);
if (!params->no_thread)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
data->hist[cpu].thread_count);
if (params->user_hist)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
data->hist[cpu].user_count);
}
trace_seq_printf(trace->seq, "\n");
@@ -488,15 +489,15 @@ timerlat_print_stats_all(struct timerlat_hist_params *params,
trace_seq_printf(trace->seq, "count:");
if (!params->no_irq)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
sum.irq_count);
if (!params->no_thread)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
sum.thread_count);
if (params->user_hist)
- trace_seq_printf(trace->seq, "%9d ",
+ trace_seq_printf(trace->seq, "%9llu ",
sum.user_count);
trace_seq_printf(trace->seq, "\n");
@@ -655,7 +656,7 @@ static void timerlat_hist_usage(char *usage)
" [-t[file]] [-e sys[:event]] [--filter <filter>] [--trigger <trigger>] [-c cpu-list] [-H cpu-list]\\",
" [-P priority] [-E N] [-b N] [--no-irq] [--no-thread] [--no-header] [--no-summary] \\",
" [--no-index] [--with-zeros] [--dma-latency us] [-C[=cgroup_name]] [--no-aa] [--dump-task] [-u|-k]",
- " [--warm-up s]",
+ " [--warm-up s] [--deepest-idle-state n]",
"",
" -h/--help: print this menu",
" -a/--auto: set automatic trace mode, stopping the session if argument in us latency is hit",
@@ -695,6 +696,7 @@ static void timerlat_hist_usage(char *usage)
" -U/--user-load: enable timerlat for user-defined user-space workload",
" --warm-up s: let the workload run for s seconds before collecting data",
" --trace-buffer-size kB: set the per-cpu trace buffer size in kB",
+ " --deepest-idle-state n: only go down to idle state n on cpus used by timerlat to reduce exit from idle latency",
NULL,
};
@@ -732,6 +734,9 @@ static struct timerlat_hist_params
/* disabled by default */
params->dma_latency = -1;
+ /* disabled by default */
+ params->deepest_idle_state = -2;
+
/* display data in microseconds */
params->output_divisor = 1000;
params->bucket_size = 1;
@@ -772,13 +777,14 @@ static struct timerlat_hist_params
{"dump-task", no_argument, 0, '\1'},
{"warm-up", required_argument, 0, '\2'},
{"trace-buffer-size", required_argument, 0, '\3'},
+ {"deepest-idle-state", required_argument, 0, '\4'},
{0, 0, 0, 0}
};
/* getopt_long stores the option index here. */
int option_index = 0;
- c = getopt_long(argc, argv, "a:c:C::b:d:e:E:DhH:i:knp:P:s:t::T:uU0123456:7:8:9\1\2:\3",
+ c = getopt_long(argc, argv, "a:c:C::b:d:e:E:DhH:i:knp:P:s:t::T:uU0123456:7:8:9\1\2:\3:",
long_options, &option_index);
/* detect the end of the options. */
@@ -960,6 +966,9 @@ static struct timerlat_hist_params
case '\3':
params->buffer_size = get_llong_from_str(optarg);
break;
+ case '\4':
+ params->deepest_idle_state = get_llong_from_str(optarg);
+ break;
default:
timerlat_hist_usage("Invalid option");
}
@@ -1064,7 +1073,7 @@ timerlat_hist_apply_config(struct osnoise_tool *tool, struct timerlat_hist_param
* If the user did not specify a type of thread, try user-threads first.
* Fall back to kernel threads otherwise.
*/
- if (!params->kernel_workload && !params->user_workload) {
+ if (!params->kernel_workload && !params->user_hist) {
retval = tracefs_file_exists(NULL, "osnoise/per_cpu/cpu0/timerlat_fd");
if (retval) {
debug_msg("User-space interface detected, setting user-threads\n");
@@ -1152,6 +1161,7 @@ int timerlat_hist_main(int argc, char *argv[])
int return_value = 1;
pthread_t timerlat_u;
int retval;
+ int nr_cpus, i;
params = timerlat_hist_parse_args(argc, argv);
if (!params)
@@ -1201,6 +1211,28 @@ int timerlat_hist_main(int argc, char *argv[])
}
}
+ if (params->deepest_idle_state >= -1) {
+ if (!have_libcpupower_support()) {
+ err_msg("rtla built without libcpupower, --deepest-idle-state is not supported\n");
+ goto out_free;
+ }
+
+ nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+
+ for (i = 0; i < nr_cpus; i++) {
+ if (params->cpus && !CPU_ISSET(i, &params->monitored_cpus))
+ continue;
+ if (save_cpu_idle_disable_state(i) < 0) {
+ err_msg("Could not save cpu idle state.\n");
+ goto out_free;
+ }
+ if (set_deepest_cpu_idle_state(i, params->deepest_idle_state) < 0) {
+ err_msg("Could not set deepest cpu idle state.\n");
+ goto out_free;
+ }
+ }
+ }
+
if (params->trace_output) {
record = osnoise_init_trace_tool("timerlat");
if (!record) {
@@ -1332,6 +1364,13 @@ out_hist:
timerlat_aa_destroy();
if (dma_latency_fd >= 0)
close(dma_latency_fd);
+ if (params->deepest_idle_state >= -1) {
+ for (i = 0; i < nr_cpus; i++) {
+ if (params->cpus && !CPU_ISSET(i, &params->monitored_cpus))
+ continue;
+ restore_cpu_idle_disable_state(i);
+ }
+ }
trace_events_destroy(&record->trace, params->events);
params->events = NULL;
out_free:
@@ -1340,6 +1379,7 @@ out_free:
osnoise_destroy_tool(record);
osnoise_destroy_tool(tool);
free(params);
+ free_cpu_idle_disable_states();
out_exit:
exit(return_value);
}
diff --git a/tools/tracing/rtla/src/timerlat_top.c b/tools/tracing/rtla/src/timerlat_top.c
index 210b0f533534..059b468981e4 100644
--- a/tools/tracing/rtla/src/timerlat_top.c
+++ b/tools/tracing/rtla/src/timerlat_top.c
@@ -48,15 +48,16 @@ struct timerlat_top_params {
int pretty_output;
int warmup;
int buffer_size;
+ int deepest_idle_state;
cpu_set_t hk_cpu_set;
struct sched_attr sched_param;
struct trace_events *events;
};
struct timerlat_top_cpu {
- int irq_count;
- int thread_count;
- int user_count;
+ unsigned long long irq_count;
+ unsigned long long thread_count;
+ unsigned long long user_count;
unsigned long long cur_irq;
unsigned long long min_irq;
@@ -280,7 +281,7 @@ static void timerlat_top_print(struct osnoise_tool *top, int cpu)
/*
* Unless trace is being lost, IRQ counter is always the max.
*/
- trace_seq_printf(s, "%3d #%-9d |", cpu, cpu_data->irq_count);
+ trace_seq_printf(s, "%3d #%-9llu |", cpu, cpu_data->irq_count);
if (!cpu_data->irq_count) {
trace_seq_printf(s, "%s %s %s %s |", no_value, no_value, no_value, no_value);
@@ -447,7 +448,7 @@ static void timerlat_top_usage(char *usage)
"",
" usage: rtla timerlat [top] [-h] [-q] [-a us] [-d s] [-D] [-n] [-p us] [-i us] [-T us] [-s us] \\",
" [[-t[file]] [-e sys[:event]] [--filter <filter>] [--trigger <trigger>] [-c cpu-list] [-H cpu-list]\\",
- " [-P priority] [--dma-latency us] [--aa-only us] [-C[=cgroup_name]] [-u|-k] [--warm-up s]",
+ " [-P priority] [--dma-latency us] [--aa-only us] [-C[=cgroup_name]] [-u|-k] [--warm-up s] [--deepest-idle-state n]",
"",
" -h/--help: print this menu",
" -a/--auto: set automatic trace mode, stopping the session if argument in us latency is hit",
@@ -481,6 +482,7 @@ static void timerlat_top_usage(char *usage)
" -U/--user-load: enable timerlat for user-defined user-space workload",
" --warm-up s: let the workload run for s seconds before collecting data",
" --trace-buffer-size kB: set the per-cpu trace buffer size in kB",
+ " --deepest-idle-state n: only go down to idle state n on cpus used by timerlat to reduce exit from idle latency",
NULL,
};
@@ -518,6 +520,9 @@ static struct timerlat_top_params
/* disabled by default */
params->dma_latency = -1;
+ /* disabled by default */
+ params->deepest_idle_state = -2;
+
/* display data in microseconds */
params->output_divisor = 1000;
@@ -550,6 +555,7 @@ static struct timerlat_top_params
{"aa-only", required_argument, 0, '5'},
{"warm-up", required_argument, 0, '6'},
{"trace-buffer-size", required_argument, 0, '7'},
+ {"deepest-idle-state", required_argument, 0, '8'},
{0, 0, 0, 0}
};
@@ -726,6 +732,9 @@ static struct timerlat_top_params
case '7':
params->buffer_size = get_llong_from_str(optarg);
break;
+ case '8':
+ params->deepest_idle_state = get_llong_from_str(optarg);
+ break;
default:
timerlat_top_usage("Invalid option");
}
@@ -830,7 +839,7 @@ timerlat_top_apply_config(struct osnoise_tool *top, struct timerlat_top_params *
* If the user did not specify a type of thread, try user-threads first.
* Fall back to kernel threads otherwise.
*/
- if (!params->kernel_workload && !params->user_workload) {
+ if (!params->kernel_workload && !params->user_top) {
retval = tracefs_file_exists(NULL, "osnoise/per_cpu/cpu0/timerlat_fd");
if (retval) {
debug_msg("User-space interface detected, setting user-threads\n");
@@ -850,7 +859,7 @@ timerlat_top_apply_config(struct osnoise_tool *top, struct timerlat_top_params *
}
}
- if (isatty(1) && !params->quiet)
+ if (isatty(STDOUT_FILENO) && !params->quiet)
params->pretty_output = 1;
return 0;
@@ -922,6 +931,7 @@ int timerlat_top_main(int argc, char *argv[])
int return_value = 1;
char *max_lat;
int retval;
+ int nr_cpus, i;
params = timerlat_top_parse_args(argc, argv);
if (!params)
@@ -971,6 +981,28 @@ int timerlat_top_main(int argc, char *argv[])
}
}
+ if (params->deepest_idle_state >= -1) {
+ if (!have_libcpupower_support()) {
+ err_msg("rtla built without libcpupower, --deepest-idle-state is not supported\n");
+ goto out_free;
+ }
+
+ nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+
+ for (i = 0; i < nr_cpus; i++) {
+ if (params->cpus && !CPU_ISSET(i, &params->monitored_cpus))
+ continue;
+ if (save_cpu_idle_disable_state(i) < 0) {
+ err_msg("Could not save cpu idle state.\n");
+ goto out_free;
+ }
+ if (set_deepest_cpu_idle_state(i, params->deepest_idle_state) < 0) {
+ err_msg("Could not set deepest cpu idle state.\n");
+ goto out_free;
+ }
+ }
+ }
+
if (params->trace_output) {
record = osnoise_init_trace_tool("timerlat");
if (!record) {
@@ -1125,6 +1157,13 @@ out_top:
timerlat_aa_destroy();
if (dma_latency_fd >= 0)
close(dma_latency_fd);
+ if (params->deepest_idle_state >= -1) {
+ for (i = 0; i < nr_cpus; i++) {
+ if (params->cpus && !CPU_ISSET(i, &params->monitored_cpus))
+ continue;
+ restore_cpu_idle_disable_state(i);
+ }
+ }
trace_events_destroy(&record->trace, params->events);
params->events = NULL;
out_free:
@@ -1134,6 +1173,7 @@ out_free:
osnoise_destroy_tool(record);
osnoise_destroy_tool(top);
free(params);
+ free_cpu_idle_disable_states();
out_exit:
exit(return_value);
}
diff --git a/tools/tracing/rtla/src/utils.c b/tools/tracing/rtla/src/utils.c
index 9ac71a66840c..4995d35cf3ec 100644
--- a/tools/tracing/rtla/src/utils.c
+++ b/tools/tracing/rtla/src/utils.c
@@ -4,6 +4,9 @@
*/
#define _GNU_SOURCE
+#ifdef HAVE_LIBCPUPOWER_SUPPORT
+#include <cpuidle.h>
+#endif /* HAVE_LIBCPUPOWER_SUPPORT */
#include <dirent.h>
#include <stdarg.h>
#include <stdlib.h>
@@ -211,29 +214,25 @@ long parse_ns_duration(char *val)
/*
* This is a set of helper functions to use SCHED_DEADLINE.
*/
-#ifdef __x86_64__
-# define __NR_sched_setattr 314
-# define __NR_sched_getattr 315
-#elif __i386__
-# define __NR_sched_setattr 351
-# define __NR_sched_getattr 352
-#elif __arm__
-# define __NR_sched_setattr 380
-# define __NR_sched_getattr 381
-#elif __aarch64__ || __riscv
-# define __NR_sched_setattr 274
-# define __NR_sched_getattr 275
-#elif __powerpc__
-# define __NR_sched_setattr 355
-# define __NR_sched_getattr 356
-#elif __s390x__
-# define __NR_sched_setattr 345
-# define __NR_sched_getattr 346
+#ifndef __NR_sched_setattr
+# ifdef __x86_64__
+# define __NR_sched_setattr 314
+# elif __i386__
+# define __NR_sched_setattr 351
+# elif __arm__
+# define __NR_sched_setattr 380
+# elif __aarch64__ || __riscv
+# define __NR_sched_setattr 274
+# elif __powerpc__
+# define __NR_sched_setattr 355
+# elif __s390x__
+# define __NR_sched_setattr 345
+# endif
#endif
#define SCHED_DEADLINE 6
-static inline int sched_setattr(pid_t pid, const struct sched_attr *attr,
+static inline int syscall_sched_setattr(pid_t pid, const struct sched_attr *attr,
unsigned int flags) {
return syscall(__NR_sched_setattr, pid, attr, flags);
}
@@ -243,7 +242,7 @@ int __set_sched_attr(int pid, struct sched_attr *attr)
int flags = 0;
int retval;
- retval = sched_setattr(pid, attr, flags);
+ retval = syscall_sched_setattr(pid, attr, flags);
if (retval < 0) {
err_msg("Failed to set sched attributes to the pid %d: %s\n",
pid, strerror(errno));
@@ -519,6 +518,153 @@ int set_cpu_dma_latency(int32_t latency)
return fd;
}
+#ifdef HAVE_LIBCPUPOWER_SUPPORT
+static unsigned int **saved_cpu_idle_disable_state;
+static size_t saved_cpu_idle_disable_state_alloc_ctr;
+
+/*
+ * save_cpu_idle_state_disable - save disable for all idle states of a cpu
+ *
+ * Saves the current disable of all idle states of a cpu, to be subsequently
+ * restored via restore_cpu_idle_disable_state.
+ *
+ * Return: idle state count on success, negative on error
+ */
+int save_cpu_idle_disable_state(unsigned int cpu)
+{
+ unsigned int nr_states;
+ unsigned int state;
+ int disabled;
+ int nr_cpus;
+
+ nr_states = cpuidle_state_count(cpu);
+
+ if (nr_states == 0)
+ return 0;
+
+ if (saved_cpu_idle_disable_state == NULL) {
+ nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+ saved_cpu_idle_disable_state = calloc(nr_cpus, sizeof(unsigned int *));
+ if (!saved_cpu_idle_disable_state)
+ return -1;
+ }
+
+ saved_cpu_idle_disable_state[cpu] = calloc(nr_states, sizeof(unsigned int));
+ if (!saved_cpu_idle_disable_state[cpu])
+ return -1;
+ saved_cpu_idle_disable_state_alloc_ctr++;
+
+ for (state = 0; state < nr_states; state++) {
+ disabled = cpuidle_is_state_disabled(cpu, state);
+ if (disabled < 0)
+ return disabled;
+ saved_cpu_idle_disable_state[cpu][state] = disabled;
+ }
+
+ return nr_states;
+}
+
+/*
+ * restore_cpu_idle_disable_state - restore disable for all idle states of a cpu
+ *
+ * Restores the current disable state of all idle states of a cpu that was
+ * previously saved by save_cpu_idle_disable_state.
+ *
+ * Return: idle state count on success, negative on error
+ */
+int restore_cpu_idle_disable_state(unsigned int cpu)
+{
+ unsigned int nr_states;
+ unsigned int state;
+ int disabled;
+ int result;
+
+ nr_states = cpuidle_state_count(cpu);
+
+ if (nr_states == 0)
+ return 0;
+
+ if (!saved_cpu_idle_disable_state)
+ return -1;
+
+ for (state = 0; state < nr_states; state++) {
+ if (!saved_cpu_idle_disable_state[cpu])
+ return -1;
+ disabled = saved_cpu_idle_disable_state[cpu][state];
+ result = cpuidle_state_disable(cpu, state, disabled);
+ if (result < 0)
+ return result;
+ }
+
+ free(saved_cpu_idle_disable_state[cpu]);
+ saved_cpu_idle_disable_state[cpu] = NULL;
+ saved_cpu_idle_disable_state_alloc_ctr--;
+ if (saved_cpu_idle_disable_state_alloc_ctr == 0) {
+ free(saved_cpu_idle_disable_state);
+ saved_cpu_idle_disable_state = NULL;
+ }
+
+ return nr_states;
+}
+
+/*
+ * free_cpu_idle_disable_states - free saved idle state disable for all cpus
+ *
+ * Frees the memory used for storing cpu idle state disable for all cpus
+ * and states.
+ *
+ * Normally, the memory is freed automatically in
+ * restore_cpu_idle_disable_state; this is mostly for cleaning up after an
+ * error.
+ */
+void free_cpu_idle_disable_states(void)
+{
+ int cpu;
+ int nr_cpus;
+
+ if (!saved_cpu_idle_disable_state)
+ return;
+
+ nr_cpus = sysconf(_SC_NPROCESSORS_CONF);
+
+ for (cpu = 0; cpu < nr_cpus; cpu++) {
+ free(saved_cpu_idle_disable_state[cpu]);
+ saved_cpu_idle_disable_state[cpu] = NULL;
+ }
+
+ free(saved_cpu_idle_disable_state);
+ saved_cpu_idle_disable_state = NULL;
+}
+
+/*
+ * set_deepest_cpu_idle_state - limit idle state of cpu
+ *
+ * Disables all idle states deeper than the one given in
+ * deepest_state (assuming states with higher number are deeper).
+ *
+ * This is used to reduce the exit from idle latency. Unlike
+ * set_cpu_dma_latency, it can disable idle states per cpu.
+ *
+ * Return: idle state count on success, negative on error
+ */
+int set_deepest_cpu_idle_state(unsigned int cpu, unsigned int deepest_state)
+{
+ unsigned int nr_states;
+ unsigned int state;
+ int result;
+
+ nr_states = cpuidle_state_count(cpu);
+
+ for (state = deepest_state + 1; state < nr_states; state++) {
+ result = cpuidle_state_disable(cpu, state, 1);
+ if (result < 0)
+ return result;
+ }
+
+ return nr_states;
+}
+#endif /* HAVE_LIBCPUPOWER_SUPPORT */
+
#define _STR(x) #x
#define STR(x) _STR(x)
diff --git a/tools/tracing/rtla/src/utils.h b/tools/tracing/rtla/src/utils.h
index d44513e6c66a..101d4799a009 100644
--- a/tools/tracing/rtla/src/utils.h
+++ b/tools/tracing/rtla/src/utils.h
@@ -46,6 +46,7 @@ update_sum(unsigned long long *a, unsigned long long *b)
*a += *b;
}
+#ifndef SCHED_ATTR_SIZE_VER0
struct sched_attr {
uint32_t size;
uint32_t sched_policy;
@@ -56,6 +57,7 @@ struct sched_attr {
uint64_t sched_deadline;
uint64_t sched_period;
};
+#endif /* SCHED_ATTR_SIZE_VER0 */
int parse_prio(char *arg, struct sched_attr *sched_param);
int parse_cpu_set(char *cpu_list, cpu_set_t *set);
@@ -64,6 +66,19 @@ int set_comm_sched_attr(const char *comm_prefix, struct sched_attr *attr);
int set_comm_cgroup(const char *comm_prefix, const char *cgroup);
int set_pid_cgroup(pid_t pid, const char *cgroup);
int set_cpu_dma_latency(int32_t latency);
+#ifdef HAVE_LIBCPUPOWER_SUPPORT
+int save_cpu_idle_disable_state(unsigned int cpu);
+int restore_cpu_idle_disable_state(unsigned int cpu);
+void free_cpu_idle_disable_states(void);
+int set_deepest_cpu_idle_state(unsigned int cpu, unsigned int state);
+static inline int have_libcpupower_support(void) { return 1; }
+#else
+static inline int save_cpu_idle_disable_state(unsigned int cpu) { return -1; }
+static inline int restore_cpu_idle_disable_state(unsigned int cpu) { return -1; }
+static inline void free_cpu_idle_disable_states(void) { }
+static inline int set_deepest_cpu_idle_state(unsigned int cpu, unsigned int state) { return -1; }
+static inline int have_libcpupower_support(void) { return 0; }
+#endif /* HAVE_LIBCPUPOWER_SUPPORT */
int auto_house_keeping(cpu_set_t *monitored_cpus);
#define ns_to_usf(x) (((double)x/1000))
generated by cgit (git 2.34.1) at 2025-05-14 21:40:17 +0000