perf bench: Add missing pthread.h include for CPU_*() macros

Cc: David Ahern <dsahern@gmail.com>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-48qbfv7tqs8n8ey74lbyfjtq@git.kernel.orgSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/bench/futex-hash.c

@@ -8,8 +8,10 @@
  * many threads and futexes as possible.
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
-#include "../util/util.h"
 #include "../util/stat.h"
 #include <subcmd/parse-options.h>
 #include "../util/header.h"
@@ -17,9 +19,7 @@
 #include "futex.h"
 
 #include <err.h>
-#include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 static unsigned int nthreads = 0;
 static unsigned int nsecs    = 10;

tools/perf/bench/futex-lock-pi.c

@@ -2,6 +2,9 @@
  * Copyright (C) 2015 Davidlohr Bueso.
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../util/util.h"
 #include "../util/stat.h"
@@ -13,7 +16,6 @@
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 struct worker {
 	int tid;

tools/perf/bench/futex-requeue.c

@@ -8,6 +8,9 @@
  * requeues without waking up any tasks -- thus mimicking a regular futex_wait.
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../util/util.h"
 #include "../util/stat.h"
@@ -19,7 +22,6 @@
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 static u_int32_t futex1 = 0, futex2 = 0;
 

tools/perf/bench/futex-wake-parallel.c

@@ -7,6 +7,9 @@
  * it can be used to measure futex_wake() changes.
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../util/util.h"
 #include "../util/stat.h"
@@ -18,7 +21,6 @@
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 struct thread_data {
 	pthread_t worker;

tools/perf/bench/futex-wake.c

@@ -8,6 +8,9 @@
  * one or more tasks, and thus the waitqueue is never empty.
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../util/util.h"
 #include "../util/stat.h"
@@ -19,7 +22,6 @@
 #include <err.h>
 #include <stdlib.h>
 #include <sys/time.h>
-#include <pthread.h>
 
 /* all threads will block on the same futex */
 static u_int32_t futex1 = 0;

tools/perf/bench/numa.c

@@ -4,6 +4,9 @@
  * numa: Simulate NUMA-sensitive workload and measure their NUMA performance
  */
 
+/* For the CLR_() macros */
+#include <pthread.h>
+
 #include "../perf.h"
 #include "../builtin.h"
 #include "../util/util.h"
@@ -21,7 +24,6 @@
 #include <stdlib.h>
 #include <string.h>
 #include <unistd.h>
-#include <pthread.h>
 #include <sys/mman.h>
 #include <sys/time.h>
 #include <sys/resource.h>