Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1757485AbcJPTIb (ORCPT <rfc822;w@1wt.eu>);
        Sun, 16 Oct 2016 15:08:31 -0400
Received: from Galois.linutronix.de ([146.0.238.70]:35219 "EHLO
        Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1754199AbcJPTIW (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Sun, 16 Oct 2016 15:08:22 -0400
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>,
        linux-kernel@vger.kernel.org, Davidlohr Bueso <dbueso@suse.de>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: [PATCH 2/2] perf bench futex: add NUMA support
Date: Sun, 16 Oct 2016 21:08:03 +0200
Message-Id: <20161016190803.3392-2-bigeasy@linutronix.de>
X-Mailer: git-send-email 2.9.3
In-Reply-To: <20161016190803.3392-1-bigeasy@linutronix.de>
References: <20161016190803.3392-1-bigeasy@linutronix.de>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 5770
Lines: 194

By default the application uses malloc() and all available CPUs. This
patch introduces NUMA support which means:
- memory is allocated node local via numa_alloc_local()
- all CPUs of the specified NUMA node are used. This is also true if the
  number of threads set is greater than the number of CPUs available on
  this node.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 tools/perf/bench/Build        |  4 ++
 tools/perf/bench/futex-hash.c | 87 ++++++++++++++++++++++++++++++++++++++-=
----
 2 files changed, 81 insertions(+), 10 deletions(-)

diff --git a/tools/perf/bench/Build b/tools/perf/bench/Build
index 60bf11943047..9e6e518d7d62 100644
--- a/tools/perf/bench/Build
+++ b/tools/perf/bench/Build
@@ -1,3 +1,7 @@
+ifdef CONFIG_NUMA
+CFLAGS_futex-hash.o   +=3D -DCONFIG_NUMA=3D1
+endif
+
 perf-y +=3D sched-messaging.o
 perf-y +=3D sched-pipe.o
 perf-y +=3D mem-functions.o
diff --git a/tools/perf/bench/futex-hash.c b/tools/perf/bench/futex-hash.c
index d9e5e80bb4d0..8db4a5bd6a4e 100644
--- a/tools/perf/bench/futex-hash.c
+++ b/tools/perf/bench/futex-hash.c
@@ -25,6 +25,9 @@
=20
 #include <err.h>
 #include <sys/time.h>
+#ifdef CONFIG_NUMA
+#include <numa.h>
+#endif
=20
 static unsigned int nthreads =3D 0;
 static unsigned int nsecs    =3D 10;
@@ -32,6 +35,7 @@ static unsigned int nsecs    =3D 10;
 static unsigned int nfutexes =3D 1024;
 static bool fshared =3D false, done =3D false, silent =3D false;
 static int futex_flag =3D 0;
+static int numa_node =3D -1;
=20
 struct timeval start, end, runtime;
 static pthread_mutex_t thread_lock;
@@ -55,9 +59,28 @@ static const struct option options[] =3D {
 	OPT_UINTEGER('f', "futexes", &nfutexes, "Specify amount of futexes per th=
reads"),
 	OPT_BOOLEAN( 's', "silent",  &silent,   "Silent mode: do not display data=
/details"),
 	OPT_BOOLEAN( 'S', "shared",  &fshared,  "Use shared futexes instead of pr=
ivate ones"),
+#ifdef CONFIG_NUMA
+	OPT_INTEGER( 'n', "numa",   &numa_node,  "Specify the NUMA node"),
+#endif
 	OPT_END()
 };
=20
+#ifndef CONFIG_NUMA
+static int numa_run_on_node(int node __maybe_unused) { return 0; }
+static int numa_node_of_cpu(int node __maybe_unused) { return 0; }
+static void *numa_alloc_local(size_t size) { return malloc(size); }
+static void numa_free(void *p, size_t size __maybe_unused) { return free(p=
); }
+#endif
+
+static bool cpu_is_local(int cpu)
+{
+	if (numa_node < 0)
+		return true;
+	if (numa_node_of_cpu(cpu) =3D=3D numa_node)
+		return true;
+	return false;
+}
+
 static const char * const bench_futex_hash_usage[] =3D {
 	"perf bench futex hash <options>",
 	NULL
@@ -123,6 +146,8 @@ int bench_futex_hash(int argc, const char **argv,
 	unsigned int i, ncpus;
 	pthread_attr_t thread_attr;
 	struct worker *worker =3D NULL;
+	char *node_str =3D NULL;
+	unsigned int cpunum;
=20
 	argc =3D parse_options(argc, argv, options, bench_futex_hash_usage, 0);
 	if (argc) {
@@ -136,18 +161,50 @@ int bench_futex_hash(int argc, const char **argv,
 	act.sa_sigaction =3D toggle_done;
 	sigaction(SIGINT, &act, NULL);
=20
-	if (!nthreads) /* default to the number of CPUs */
-		nthreads =3D ncpus;
+	if (!nthreads) {
+		/* default to the number of CPUs per NUMA node */
+		if (numa_node < 0) {
+			nthreads =3D ncpus;
+		} else {
+			for (i =3D 0; i < ncpus; i++) {
+				if (cpu_is_local(i))
+					nthreads++;
+			}
+			if (!nthreads)
+				err(EXIT_FAILURE, "No online CPUs for this node");
+		}
+	} else {
+		int cpu_available =3D 0;
=20
-	worker =3D calloc(nthreads, sizeof(*worker));
+		for (i =3D 0; i < ncpus && !cpu_available; i++) {
+			if (cpu_is_local(i))
+				cpu_available =3D 1;
+		}
+		if (!cpu_available)
+			err(EXIT_FAILURE, "No online CPUs for this node");
+	}
+
+	if (numa_node >=3D 0) {
+		ret =3D numa_run_on_node(numa_node);
+		if (ret < 0)
+			err(EXIT_FAILURE, "numa_run_on_node");
+		ret =3D asprintf(&node_str, " on node %d", numa_node);
+		if (ret < 0)
+			err(EXIT_FAILURE, "numa_node, asprintf");
+	}
+
+	worker =3D numa_alloc_local(nthreads * sizeof(*worker));
 	if (!worker)
 		goto errmem;
=20
 	if (!fshared)
 		futex_flag =3D FUTEX_PRIVATE_FLAG;
=20
-	printf("Run summary [PID %d]: %d threads, each operating on %d [%s] futex=
es for %d secs.\n\n",
-	       getpid(), nthreads, nfutexes, fshared ? "shared":"private", nsecs);
+	printf("Run summary [PID %d]: %d threads%s, each operating on %d [%s] fut=
exes for %d secs.\n\n",
+	       getpid(), nthreads,
+	       node_str ? : "",
+	       nfutexes, fshared ? "shared":"private",
+	       nsecs);
=20
 	init_stats(&throughput_stats);
 	pthread_mutex_init(&thread_lock, NULL);
@@ -157,14 +214,24 @@ int bench_futex_hash(int argc, const char **argv,
 	threads_starting =3D nthreads;
 	pthread_attr_init(&thread_attr);
 	gettimeofday(&start, NULL);
-	for (i =3D 0; i < nthreads; i++) {
+	for (cpunum =3D 0, i =3D 0; i < nthreads; i++, cpunum++) {
+
+		do {
+			if (cpu_is_local(cpunum))
+				break;
+			cpunum++;
+			if (cpunum > ncpus)
+				cpunum =3D 0;
+		} while (1);
+
 		worker[i].tid =3D i;
-		worker[i].futex =3D calloc(nfutexes, sizeof(*worker[i].futex));
+		worker[i].futex =3D numa_alloc_local(nfutexes *
+						   sizeof(*worker[i].futex));
 		if (!worker[i].futex)
 			goto errmem;
=20
 		CPU_ZERO(&cpu);
-		CPU_SET(i % ncpus, &cpu);
+		CPU_SET(cpunum % ncpus, &cpu);
=20
 		ret =3D pthread_attr_setaffinity_np(&thread_attr, sizeof(cpu_set_t), &cp=
u);
 		if (ret)
@@ -211,12 +278,12 @@ int bench_futex_hash(int argc, const char **argv,
 				       &worker[i].futex[nfutexes-1], t);
 		}
=20
-		free(worker[i].futex);
+		numa_free(worker[i].futex, nfutexes * sizeof(*worker[i].futex));
 	}
=20
 	print_summary();
=20
-	free(worker);
+	numa_free(worker, nthreads * sizeof(*worker));
 	return ret;
 errmem:
 	err(EXIT_FAILURE, "calloc");
--=20
2.9.3