summaryrefslogtreecommitdiffstats
path: root/tools/perf/bench/futex-lock-pi.c
diff options
context:
space:
mode:
authorDavidlohr Bueso <dave@stgolabs.net>2015-07-07 01:55:53 -0700
committerArnaldo Carvalho de Melo <acme@redhat.com>2015-07-20 17:49:51 -0300
commitd2f3f5d2e9cae6e73f9642a5ddc8c8a07c35e79b (patch)
tree76994d160a27adaab0a7263c2697afdcae4f5df5 /tools/perf/bench/futex-lock-pi.c
parent52c0a18b9010fb19d10889e8a00aa784197d357c (diff)
downloadop-kernel-dev-d2f3f5d2e9cae6e73f9642a5ddc8c8a07c35e79b.zip
op-kernel-dev-d2f3f5d2e9cae6e73f9642a5ddc8c8a07c35e79b.tar.gz
perf bench futex: Add lock_pi stresser
Allows a way of measuring low level kernel implementation of FUTEX_LOCK_PI and FUTEX_UNLOCK_PI. The program comes in two flavors: (i) single futex (default), all threads contend on the same uaddr. For the sake of the benchmark, we call into kernel space even when the lock is uncontended. The kernel will set it to TID, any waters that come in and contend for the pi futex will be handled respectively by the kernel. (ii) -M option for multiple futexes, each thread deals with its own futex. This is a trivial scenario and only measures kernel handling of 0->TID transition. Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Cc: Mel Gorman <mgorman@suse.de> Link: http://lkml.kernel.org/r/1436259353.12255.78.camel@stgolabs.net Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to 'tools/perf/bench/futex-lock-pi.c')
-rw-r--r--tools/perf/bench/futex-lock-pi.c219
1 files changed, 219 insertions, 0 deletions
diff --git a/tools/perf/bench/futex-lock-pi.c b/tools/perf/bench/futex-lock-pi.c
new file mode 100644
index 0000000..bc6a16a
--- /dev/null
+++ b/tools/perf/bench/futex-lock-pi.c
@@ -0,0 +1,219 @@
+/*
+ * Copyright (C) 2015 Davidlohr Bueso.
+ */
+
+#include "../perf.h"
+#include "../util/util.h"
+#include "../util/stat.h"
+#include "../util/parse-options.h"
+#include "../util/header.h"
+#include "bench.h"
+#include "futex.h"
+
+#include <err.h>
+#include <stdlib.h>
+#include <sys/time.h>
+#include <pthread.h>
+
+struct worker {
+ int tid;
+ u_int32_t *futex;
+ pthread_t thread;
+ unsigned long ops;
+};
+
+static u_int32_t global_futex = 0;
+static struct worker *worker;
+static unsigned int nsecs = 10;
+static bool silent = false, multi = false;
+static bool done = false, fshared = false;
+static unsigned int ncpus, nthreads = 0;
+static int futex_flag = 0;
+struct timeval start, end, runtime;
+static pthread_mutex_t thread_lock;
+static unsigned int threads_starting;
+static struct stats throughput_stats;
+static pthread_cond_t thread_parent, thread_worker;
+
+static const struct option options[] = {
+ OPT_UINTEGER('t', "threads", &nthreads, "Specify amount of threads"),
+ OPT_UINTEGER('r', "runtime", &nsecs, "Specify runtime (in seconds)"),
+ OPT_BOOLEAN( 'M', "multi", &multi, "Use multiple futexes"),
+ OPT_BOOLEAN( 's', "silent", &silent, "Silent mode: do not display data/details"),
+ OPT_BOOLEAN( 'S', "shared", &fshared, "Use shared futexes instead of private ones"),
+ OPT_END()
+};
+
+static const char * const bench_futex_lock_pi_usage[] = {
+ "perf bench futex requeue <options>",
+ NULL
+};
+
+static void print_summary(void)
+{
+ unsigned long avg = avg_stats(&throughput_stats);
+ double stddev = stddev_stats(&throughput_stats);
+
+ printf("%sAveraged %ld operations/sec (+- %.2f%%), total secs = %d\n",
+ !silent ? "\n" : "", avg, rel_stddev_stats(stddev, avg),
+ (int) runtime.tv_sec);
+}
+
+static void toggle_done(int sig __maybe_unused,
+ siginfo_t *info __maybe_unused,
+ void *uc __maybe_unused)
+{
+ /* inform all threads that we're done for the day */
+ done = true;
+ gettimeofday(&end, NULL);
+ timersub(&end, &start, &runtime);
+}
+
+static void *workerfn(void *arg)
+{
+ struct worker *w = (struct worker *) arg;
+
+ pthread_mutex_lock(&thread_lock);
+ threads_starting--;
+ if (!threads_starting)
+ pthread_cond_signal(&thread_parent);
+ pthread_cond_wait(&thread_worker, &thread_lock);
+ pthread_mutex_unlock(&thread_lock);
+
+ do {
+ int ret;
+ again:
+ ret = futex_lock_pi(w->futex, NULL, 0, futex_flag);
+
+ if (ret) { /* handle lock acquisition */
+ if (!silent)
+ warn("thread %d: Could not lock pi-lock for %p (%d)",
+ w->tid, w->futex, ret);
+ if (done)
+ break;
+
+ goto again;
+ }
+
+ usleep(1);
+ ret = futex_unlock_pi(w->futex, futex_flag);
+ if (ret && !silent)
+ warn("thread %d: Could not unlock pi-lock for %p (%d)",
+ w->tid, w->futex, ret);
+ w->ops++; /* account for thread's share of work */
+ } while (!done);
+
+ return NULL;
+}
+
+static void create_threads(struct worker *w, pthread_attr_t thread_attr)
+{
+ cpu_set_t cpu;
+ unsigned int i;
+
+ threads_starting = nthreads;
+
+ for (i = 0; i < nthreads; i++) {
+ worker[i].tid = i;
+
+ if (multi) {
+ worker[i].futex = calloc(1, sizeof(u_int32_t));
+ if (!worker[i].futex)
+ err(EXIT_FAILURE, "calloc");
+ } else
+ worker[i].futex = &global_futex;
+
+ CPU_ZERO(&cpu);
+ CPU_SET(i % ncpus, &cpu);
+
+ if (pthread_attr_setaffinity_np(&thread_attr, sizeof(cpu_set_t), &cpu))
+ err(EXIT_FAILURE, "pthread_attr_setaffinity_np");
+
+ if (pthread_create(&w[i].thread, &thread_attr, workerfn, &worker[i]))
+ err(EXIT_FAILURE, "pthread_create");
+ }
+}
+
+int bench_futex_lock_pi(int argc, const char **argv,
+ const char *prefix __maybe_unused)
+{
+ int ret = 0;
+ unsigned int i;
+ struct sigaction act;
+ pthread_attr_t thread_attr;
+
+ argc = parse_options(argc, argv, options, bench_futex_lock_pi_usage, 0);
+ if (argc)
+ goto err;
+
+ ncpus = sysconf(_SC_NPROCESSORS_ONLN);
+
+ sigfillset(&act.sa_mask);
+ act.sa_sigaction = toggle_done;
+ sigaction(SIGINT, &act, NULL);
+
+ if (!nthreads)
+ nthreads = ncpus;
+
+ worker = calloc(nthreads, sizeof(*worker));
+ if (!worker)
+ err(EXIT_FAILURE, "calloc");
+
+ if (!fshared)
+ futex_flag = FUTEX_PRIVATE_FLAG;
+
+ printf("Run summary [PID %d]: %d threads doing pi lock/unlock pairing for %d secs.\n\n",
+ getpid(), nthreads, nsecs);
+
+ init_stats(&throughput_stats);
+ pthread_mutex_init(&thread_lock, NULL);
+ pthread_cond_init(&thread_parent, NULL);
+ pthread_cond_init(&thread_worker, NULL);
+
+ threads_starting = nthreads;
+ pthread_attr_init(&thread_attr);
+ gettimeofday(&start, NULL);
+
+ create_threads(worker, thread_attr);
+ pthread_attr_destroy(&thread_attr);
+
+ pthread_mutex_lock(&thread_lock);
+ while (threads_starting)
+ pthread_cond_wait(&thread_parent, &thread_lock);
+ pthread_cond_broadcast(&thread_worker);
+ pthread_mutex_unlock(&thread_lock);
+
+ sleep(nsecs);
+ toggle_done(0, NULL, NULL);
+
+ for (i = 0; i < nthreads; i++) {
+ ret = pthread_join(worker[i].thread, NULL);
+ if (ret)
+ err(EXIT_FAILURE, "pthread_join");
+ }
+
+ /* cleanup & report results */
+ pthread_cond_destroy(&thread_parent);
+ pthread_cond_destroy(&thread_worker);
+ pthread_mutex_destroy(&thread_lock);
+
+ for (i = 0; i < nthreads; i++) {
+ unsigned long t = worker[i].ops/runtime.tv_sec;
+
+ update_stats(&throughput_stats, t);
+ if (!silent)
+ printf("[thread %3d] futex: %p [ %ld ops/sec ]\n",
+ worker[i].tid, worker[i].futex, t);
+
+ if (multi)
+ free(worker[i].futex);
+ }
+
+ print_summary();
+
+ free(worker);
+ return ret;
+err:
+ usage_with_options(bench_futex_lock_pi_usage, options);
+ exit(EXIT_FAILURE);
+}
OpenPOWER on IntegriCloud