perf/core improvements and fixes

index 9e104a2e973d..2006485a2859 100644
@@ -34,7 +34,7 @@ int test__openat_syscall_event_on_all_cpus(void)
 	evsel = perf_evsel__newtp("syscalls", "sys_enter_openat");
 	if (IS_ERR(evsel)) {
 		tracing_path__strerror_open_tp(errno, errbuf, sizeof(errbuf), "syscalls", "sys_enter_openat");
-		pr_err("%s\n", errbuf);
+		pr_debug("%s\n", errbuf);
 		goto out_thread_map_delete;
 	}
 

index 473d3869727e..5e811cd8f1c3 100644
@@ -89,7 +89,7 @@ int test__syscall_openat_tp_fields(void)
 
 				err = perf_evsel__parse_sample(evsel, event, &sample);
 				if (err) {
-					pr_err("Can't parse sample, err = %d\n", err);
+					pr_debug("Can't parse sample, err = %d\n", err);
 					goto out_delete_evlist;
 				}
 

index 7b1db8306098..033b54797b8a 100644
@@ -22,7 +22,7 @@ int test__openat_syscall_event(void)
 	evsel = perf_evsel__newtp("syscalls", "sys_enter_openat");
 	if (IS_ERR(evsel)) {
 		tracing_path__strerror_open_tp(errno, errbuf, sizeof(errbuf), "syscalls", "sys_enter_openat");
-		pr_err("%s\n", errbuf);
+		pr_debug("%s\n", errbuf);
 		goto out_thread_map_delete;
 	}
 

index 0648b84a9171..636d7b42d844 100644
@@ -5,6 +5,7 @@
 #include <api/fs/fs.h>
 #include "tests.h"
 #include "debug.h"
+#include "util.h"
 #include <linux/hw_breakpoint.h>
 #include <api/fs/fs.h>
 

@@ -1753,6 +1754,17 @@ static int test_pmu_events(void)
 	return ret;
 }
 
+static void debug_warn(const char *warn, va_list params)
+{
+	char msg[1024];
+
+	if (!verbose)
+		return;
+
+	vsnprintf(msg, sizeof(msg), warn, params);
+	fprintf(stderr, " Warning: %s\n", msg);
+}
+
 int test__parse_events(void)
 {
 	int ret1, ret2 = 0;

@@ -1764,6 +1776,8 @@ do {							\
 		ret2 = ret1;				\
 } while (0)
 
+	set_warning_routine(debug_warn);
+
 	TEST_EVENTS(test__events);
 
 	if (test_pmu())

index 4007aca8e0ca..6adfa18cdd4e 100644
@@ -50,6 +50,11 @@ void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN)
 	die_routine = routine;
 }
 
+void set_warning_routine(void (*routine)(const char *err, va_list params))
+{
+	warn_routine = routine;
+}
+
 void usage(const char *err)
 {
 	usage_routine(err);

index 3d5b01e8978f..4cfb913aa9e0 100644
@@ -145,6 +145,7 @@ extern void warning(const char *err, ...) __attribute__((format (printf, 1, 2)))
 
 
 extern void set_die_routine(void (*routine)(const char *err, va_list params) NORETURN);
+extern void set_warning_routine(void (*routine)(const char *err, va_list params));
 
 extern int prefixcmp(const char *str, const char *prefix);
 extern void set_buildid_dir(const char *dir);

index d3dfb7936dcd..27606ff5c4f9 100644
@@ -1,7 +1,7 @@
 /*
  * mem-memcpy.c
  *
- * memcpy: Simple memory copy in various ways
+ * Simple memcpy() and memset() benchmarks
  *
  * Written by Hitoshi Mitake <[email protected]>
  */

@@ -61,20 +61,17 @@ struct routine {
 };
 
 struct routine memcpy_routines[] = {
-	{ .name = "default",
-	  .desc = "Default memcpy() provided by glibc",
-	  .fn.memcpy = memcpy },
-#ifdef HAVE_ARCH_X86_64_SUPPORT
-
-#define MEMCPY_FN(_fn, _name, _desc) {.name = _name, .desc = _desc, .fn.memcpy = _fn},
-#include "mem-memcpy-x86-64-asm-def.h"
-#undef MEMCPY_FN
+	{ .name		= "default",
+	  .desc		= "Default memcpy() provided by glibc",
+	  .fn.memcpy	= memcpy },
 
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+# define MEMCPY_FN(_fn, _name, _desc) {.name = _name, .desc = _desc, .fn.memcpy = _fn},
+# include "mem-memcpy-x86-64-asm-def.h"
+# undef MEMCPY_FN
 #endif
 
-	{ NULL,
-	  NULL,
-	  {NULL}   }
+	{ NULL, }
 };
 
 static const char * const bench_mem_memcpy_usage[] = {

@@ -89,8 +86,7 @@ static struct perf_event_attr cycle_attr = {
 
 static void init_cycle(void)
 {
-	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1,
-				       perf_event_open_cloexec_flag());
+	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, perf_event_open_cloexec_flag());
 
 	if (cycle_fd < 0 && errno == ENOSYS)
 		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");

@@ -111,12 +107,9 @@ static u64 get_cycle(void)
 
 static double timeval2double(struct timeval *ts)
 {
-	return (double)ts->tv_sec +
-		(double)ts->tv_usec / (double)1000000;
+	return (double)ts->tv_sec + (double)ts->tv_usec / (double)1000000;
 }
 
-#define pf (no_prefault ? 0 : 1)
-
 #define print_bps(x) do {					\
 		if (x < K)					\
 			printf(" %14lf B/Sec", x);		\

@@ -140,6 +133,7 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 	const struct routine *r = &info->routines[r_idx];
 	double result_bps[2];
 	u64 result_cycle[2];
+	int prefault = no_prefault ? 0 : 1;
 
 	result_cycle[0] = result_cycle[1] = 0ULL;
 	result_bps[0] = result_bps[1] = 0.0;

@@ -149,8 +143,8 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 	if (bench_format == BENCH_FORMAT_DEFAULT)
 		printf("# Copying %s Bytes ...\n\n", length_str);
 
-	if (!only_prefault && !no_prefault) {
-		/* show both of results */
+	if (!only_prefault && prefault) {
+		/* Show both results: */
 		if (use_cycle) {
 			result_cycle[0] = info->do_cycle(r, len, false);
 			result_cycle[1] = info->do_cycle(r, len, true);

@@ -160,14 +154,14 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 		}
 	} else {
 		if (use_cycle)
-			result_cycle[pf] = info->do_cycle(r, len, only_prefault);
+			result_cycle[prefault] = info->do_cycle(r, len, only_prefault);
 		else
-			result_bps[pf] = info->do_gettimeofday(r, len, only_prefault);
+			result_bps[prefault] = info->do_gettimeofday(r, len, only_prefault);
 	}
 
 	switch (bench_format) {
 	case BENCH_FORMAT_DEFAULT:
-		if (!only_prefault && !no_prefault) {
+		if (!only_prefault && prefault) {
 			if (use_cycle) {
 				printf(" %14lf Cycle/Byte\n",
 					(double)result_cycle[0]

@@ -184,16 +178,16 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 		} else {
 			if (use_cycle) {
 				printf(" %14lf Cycle/Byte",
-					(double)result_cycle[pf]
+					(double)result_cycle[prefault]
 					/ totallen);
 			} else
-				print_bps(result_bps[pf]);
+				print_bps(result_bps[prefault]);
 
 			printf("%s\n", only_prefault ? " (with prefault)" : "");
 		}
 		break;
 	case BENCH_FORMAT_SIMPLE:
-		if (!only_prefault && !no_prefault) {
+		if (!only_prefault && prefault) {
 			if (use_cycle) {
 				printf("%lf %lf\n",
 					(double)result_cycle[0] / totallen,

@@ -204,14 +198,14 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 			}
 		} else {
 			if (use_cycle) {
-				printf("%lf\n", (double)result_cycle[pf]
+				printf("%lf\n", (double)result_cycle[prefault]
 					/ totallen);
 			} else
-				printf("%lf\n", result_bps[pf]);
+				printf("%lf\n", result_bps[prefault]);
 		}
 		break;
 	default:
-		/* reaching this means there's some disaster: */
+		/* Reaching this means there's some disaster: */
 		die("unknown format: %d\n", bench_format);
 		break;
 	}

@@ -225,8 +219,7 @@ static int bench_mem_common(int argc, const char **argv,
 	size_t len;
 	double totallen;
 
-	argc = parse_options(argc, argv, options,
-			     info->usage, 0);
+	argc = parse_options(argc, argv, options, info->usage, 0);
 
 	if (no_prefault && only_prefault) {
 		fprintf(stderr, "Invalid options: -o and -n are mutually exclusive\n");

@@ -244,7 +237,7 @@ static int bench_mem_common(int argc, const char **argv,
 		return 1;
 	}
 
-	/* same to without specifying either of prefault and no-prefault */
+	/* Same as without specifying either of prefault and no-prefault: */
 	if (only_prefault && no_prefault)
 		only_prefault = no_prefault = false;
 

@@ -282,7 +275,8 @@ static void memcpy_alloc_mem(void **dst, void **src, size_t length)
 	*src = zalloc(length);
 	if (!*src)
 		die("memory allocation failed - maybe length is too large?\n");
-	/* Make sure to always replace the zero pages even if MMAP_THRESH is crossed */
+
+	/* Make sure to always prefault zero pages even if MMAP_THRESH is crossed: */
 	memset(*src, 0, length);
 }
 

@@ -308,8 +302,7 @@ static u64 do_memcpy_cycle(const struct routine *r, size_t len, bool prefault)
 	return cycle_end - cycle_start;
 }
 
-static double do_memcpy_gettimeofday(const struct routine *r, size_t len,
-				     bool prefault)
+static double do_memcpy_gettimeofday(const struct routine *r, size_t len, bool prefault)
 {
 	struct timeval tv_start, tv_end, tv_diff;
 	memcpy_t fn = r->fn.memcpy;

@@ -337,10 +330,10 @@ int bench_mem_memcpy(int argc, const char **argv,
 		     const char *prefix __maybe_unused)
 {
 	struct bench_mem_info info = {
-		.routines = memcpy_routines,
-		.do_cycle = do_memcpy_cycle,
-		.do_gettimeofday = do_memcpy_gettimeofday,
-		.usage = bench_mem_memcpy_usage,
+		.routines		= memcpy_routines,
+		.do_cycle		= do_memcpy_cycle,
+		.do_gettimeofday	= do_memcpy_gettimeofday,
+		.usage			= bench_mem_memcpy_usage,
 	};
 
 	return bench_mem_common(argc, argv, prefix, &info);

@@ -404,30 +397,26 @@ static const char * const bench_mem_memset_usage[] = {
 };
 
 static const struct routine memset_routines[] = {
-	{ .name ="default",
-	  .desc = "Default memset() provided by glibc",
-	  .fn.memset = memset },
-#ifdef HAVE_ARCH_X86_64_SUPPORT
-
-#define MEMSET_FN(_fn, _name, _desc) { .name = _name, .desc = _desc, .fn.memset = _fn },
-#include "mem-memset-x86-64-asm-def.h"
-#undef MEMSET_FN
+	{ .name		= "default",
+	  .desc		= "Default memset() provided by glibc",
+	  .fn.memset	= memset },
 
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+# define MEMSET_FN(_fn, _name, _desc) { .name = _name, .desc = _desc, .fn.memset = _fn },
+# include "mem-memset-x86-64-asm-def.h"
+# undef MEMSET_FN
 #endif
 
-	{ .name = NULL,
-	  .desc = NULL,
-	  .fn.memset = NULL   }
+	{ NULL, }
 };
 
-int bench_mem_memset(int argc, const char **argv,
-		     const char *prefix __maybe_unused)
+int bench_mem_memset(int argc, const char **argv, const char *prefix __maybe_unused)
 {
 	struct bench_mem_info info = {
-		.routines = memset_routines,
-		.do_cycle = do_memset_cycle,
-		.do_gettimeofday = do_memset_gettimeofday,
-		.usage = bench_mem_memset_usage,
+		.routines		= memset_routines,
+		.do_cycle		= do_memset_cycle,
+		.do_gettimeofday	= do_memset_gettimeofday,
+		.usage			= bench_mem_memset_usage,
 	};
 
 	return bench_mem_common(argc, argv, prefix, &info);

index 27606ff5c4f9..263f84171ae5 100644
@@ -24,7 +24,7 @@
 #define K 1024
 
 static const char	*length_str	= "1MB";
-static const char	*routine	= "default";
+static const char	*routine	= "all";
 static int		iterations	= 1;
 static bool		use_cycle;
 static int		cycle_fd;

@@ -35,7 +35,7 @@ static const struct option options[] = {
 	OPT_STRING('l', "length", &length_str, "1MB",
 		    "Specify length of memory to copy. "
 		    "Available units: B, KB, MB, GB and TB (upper and lower)"),
-	OPT_STRING('r', "routine", &routine, "default",
+	OPT_STRING('r', "routine", &routine, "all",
 		    "Specify routine to copy, \"all\" runs all available routines"),
 	OPT_INTEGER('i', "iterations", &iterations,
 		    "repeat memcpy() invocation this number of times"),

index 263f84171ae5..7acb9b83382c 100644
@@ -211,9 +211,7 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t l
 	}
 }
 
-static int bench_mem_common(int argc, const char **argv,
-		     const char *prefix __maybe_unused,
-		     struct bench_mem_info *info)
+static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *info)
 {
 	int i;
 	size_t len;

@@ -326,8 +324,7 @@ static double do_memcpy_gettimeofday(const struct routine *r, size_t len, bool p
 	return (double)(((double)len * iterations) / timeval2double(&tv_diff));
 }
 
-int bench_mem_memcpy(int argc, const char **argv,
-		     const char *prefix __maybe_unused)
+int bench_mem_memcpy(int argc, const char **argv, const char *prefix __maybe_unused)
 {
 	struct bench_mem_info info = {
 		.routines		= memcpy_routines,

@@ -336,7 +333,7 @@ int bench_mem_memcpy(int argc, const char **argv,
 		.usage			= bench_mem_memcpy_usage,
 	};
 
-	return bench_mem_common(argc, argv, prefix, &info);
+	return bench_mem_common(argc, argv, &info);
 }
 
 static void memset_alloc_mem(void **dst, size_t length)

@@ -419,5 +416,5 @@ int bench_mem_memset(int argc, const char **argv, const char *prefix __maybe_unu
 		.usage			= bench_mem_memset_usage,
 	};
 
-	return bench_mem_common(argc, argv, prefix, &info);
+	return bench_mem_common(argc, argv, &info);
 }

similarity index 100%
rename from tools/perf/bench/mem-memcpy.c
rename to tools/perf/bench/mem-functions.c

259 lines not on this page · raw message

index f67934d46d40..1b585213ba5a 100644
@@ -110,7 +110,7 @@ int bench_format = BENCH_FORMAT_DEFAULT;
 unsigned int bench_repeat = 10; /* default number of times to repeat the run */
 
 static const struct option bench_options[] = {
-	OPT_STRING('f', "format", &bench_format_str, "default", "Specify format style"),
+	OPT_STRING('f', "format", &bench_format_str, "default|simple", "Specify the output formatting style"),
 	OPT_UINTEGER('r', "repeat",  &bench_repeat,   "Specify amount of times to repeat the run"),
 	OPT_END()
 };

160 lines not on this page · raw message

index 6fe8667fbdf2..a76e57f3ab66 100644
@@ -24,7 +24,7 @@
 #define K 1024
 
 static const char	*length_str	= "1MB";
-static const char	*routine	= "all";
+static const char	*routine_str	= "all";
 static int		iterations	= 1;
 static bool		use_cycles;
 static int		cycles_fd;

@@ -33,8 +33,8 @@ static const struct option options[] = {
 	OPT_STRING('l', "length", &length_str, "1MB",
 		    "Specify length of memory to copy. "
 		    "Available units: B, KB, MB, GB and TB (upper and lower)"),
-	OPT_STRING('r', "routine", &routine, "all",
-		    "Specify routine to copy, \"all\" runs all available routines"),
+	OPT_STRING('r', "routine", &routine_str, "all",
+		    "Specify the routine to run, \"all\" runs all available routines"),
 	OPT_INTEGER('i', "iterations", &iterations,
 		    "repeat memcpy() invocation this number of times"),
 	OPT_BOOLEAN('c', "cycles", &use_cycles,

@@ -181,18 +181,18 @@ static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *
 		return 1;
 	}
 
-	if (!strncmp(routine, "all", 3)) {
+	if (!strncmp(routine_str, "all", 3)) {
 		for (i = 0; info->routines[i].name; i++)
 			__bench_mem_routine(info, i, len, totallen);
 		return 0;
 	}
 
 	for (i = 0; info->routines[i].name; i++) {
-		if (!strcmp(info->routines[i].name, routine))
+		if (!strcmp(info->routines[i].name, routine_str))
 			break;
 	}
 	if (!info->routines[i].name) {
-		printf("Unknown routine:%s\n", routine);
+		printf("Unknown routine: %s\n", routine_str);
 		printf("Available routines...\n");
 		for (i = 0; info->routines[i].name; i++) {
 			printf("\t%s ... %s\n",

258 lines not on this page · raw message

index 1605249d2912..318da3421e69 100644
@@ -32,13 +32,17 @@ static int		cycles_fd;
 static const struct option options[] = {
 	OPT_STRING('l', "size", &size_str, "1MB",
 		    "Specify the size of the memory buffers. "
-		    "Available units: B, KB, MB, GB and TB (upper and lower)"),
+		    "Available units: B, KB, MB, GB and TB (case insensitive)"),
+
 	OPT_STRING('r', "routine", &routine_str, "all",
-		    "Specify the routine to run, \"all\" runs all available routines"),
+		    "Specify the routine to run, \"all\" runs all available routines, \"help\" lists them"),
+
 	OPT_INTEGER('i', "iterations", &iterations,
-		    "repeat memcpy() invocation this number of times"),
+		    "Repeat the function this number of times"),
+
 	OPT_BOOLEAN('c', "cycles", &use_cycles,
 		    "Use a cycles event instead of gettimeofday() to measure performance"),
+
 	OPT_END()
 };
 

@@ -106,13 +110,13 @@ static double timeval2double(struct timeval *ts)
 
 #define print_bps(x) do {						\
 		if (x < K)						\
-			printf(" %14lf B/Sec\n", x);			\
+			printf(" %14lf bytes/sec\n", x);		\
 		else if (x < K * K)					\
-			printf(" %14lfd KB/Sec\n", x / K);		\
+			printf(" %14lfd KB/sec\n", x / K);		\
 		else if (x < K * K * K)					\
-			printf(" %14lf MB/Sec\n", x / K / K);		\
+			printf(" %14lf MB/sec\n", x / K / K);		\
 		else							\
-			printf(" %14lf GB/Sec\n", x / K / K / K);	\
+			printf(" %14lf GB/sec\n", x / K / K / K);	\
 	} while (0)
 
 struct bench_mem_info {

@@ -128,10 +132,10 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t s
 	double result_bps = 0.0;
 	u64 result_cycles = 0;
 
-	printf("Routine %s (%s)\n", r->name, r->desc);
+	printf("# Routine '%s' (%s)\n", r->name, r->desc);
 
 	if (bench_format == BENCH_FORMAT_DEFAULT)
-		printf("# Copying %s Bytes ...\n\n", size_str);
+		printf("# Copying %s bytes ...\n\n", size_str);
 
 	if (use_cycles) {
 		result_cycles = info->do_cycles(r, size);

@@ -142,7 +146,7 @@ static void __bench_mem_routine(struct bench_mem_info *info, int r_idx, size_t s
 	switch (bench_format) {
 	case BENCH_FORMAT_DEFAULT:
 		if (use_cycles) {
-			printf(" %14lf cycles/Byte\n", (double)result_cycles/size_total);
+			printf(" %14lf cycles/byte\n", (double)result_cycles/size_total);
 		} else {
 			print_bps(result_bps);
 		}

@@ -192,8 +196,9 @@ static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *
 			break;
 	}
 	if (!info->routines[i].name) {
-		printf("Unknown routine: %s\n", routine_str);
-		printf("Available routines...\n");
+		if (strcmp(routine_str, "help") && strcmp(routine_str, "h"))
+			printf("Unknown routine: %s\n", routine_str);
+		printf("Available routines:\n");
 		for (i = 0; info->routines[i].name; i++) {
 			printf("\t%s ... %s\n",
 			       info->routines[i].name, info->routines[i].desc);

index 1b585213ba5a..a8fc948c8ace 100644
@@ -49,9 +49,9 @@ static struct bench sched_benchmarks[] = {
 };
 
 static struct bench mem_benchmarks[] = {
-	{ "memcpy",	"Benchmark for memcpy()",			bench_mem_memcpy	},
-	{ "memset",	"Benchmark for memset() tests",			bench_mem_memset	},
-	{ "all",	"Test all memory benchmarks",			NULL			},
+	{ "memcpy",	"Benchmark for memcpy() functions",		bench_mem_memcpy	},
+	{ "memset",	"Benchmark for memset() functions",		bench_mem_memset	},
+	{ "all",	"Test all memory access benchmarks",		NULL			},
 	{ NULL,		NULL,						NULL			}
 };
 

index 318da3421e69..8d980d4a64e5 100644
@@ -58,25 +58,6 @@ struct routine {
 	} fn;
 };
 
-struct routine memcpy_routines[] = {
-	{ .name		= "default",
-	  .desc		= "Default memcpy() provided by glibc",
-	  .fn.memcpy	= memcpy },
-
-#ifdef HAVE_ARCH_X86_64_SUPPORT
-# define MEMCPY_FN(_fn, _name, _desc) {.name = _name, .desc = _desc, .fn.memcpy = _fn},
-# include "mem-memcpy-x86-64-asm-def.h"
-# undef MEMCPY_FN
-#endif
-
-	{ NULL, }
-};
-
-static const char * const bench_mem_memcpy_usage[] = {
-	"perf bench mem memcpy <options>",
-	NULL
-};
-
 static struct perf_event_attr cycle_attr = {
 	.type		= PERF_TYPE_HARDWARE,
 	.config		= PERF_COUNT_HW_CPU_CYCLES

@@ -278,6 +259,25 @@ static double do_memcpy_gettimeofday(const struct routine *r, size_t size)
 	return (double)(((double)size * iterations) / timeval2double(&tv_diff));
 }
 
+struct routine memcpy_routines[] = {
+	{ .name		= "default",
+	  .desc		= "Default memcpy() provided by glibc",
+	  .fn.memcpy	= memcpy },
+
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+# define MEMCPY_FN(_fn, _name, _desc) {.name = _name, .desc = _desc, .fn.memcpy = _fn},
+# include "mem-memcpy-x86-64-asm-def.h"
+# undef MEMCPY_FN
+#endif
+
+	{ NULL, }
+};
+
+static const char * const bench_mem_memcpy_usage[] = {
+	"perf bench mem memcpy <options>",
+	NULL
+};
+
 int bench_mem_memcpy(int argc, const char **argv, const char *prefix __maybe_unused)
 {
 	struct bench_mem_info info = {

index 8d980d4a64e5..c5dfabdbd8d0 100644
@@ -25,20 +25,20 @@
 
 static const char	*size_str	= "1MB";
 static const char	*routine_str	= "all";
-static int		iterations	= 1;
+static int		nr_loops	= 1;
 static bool		use_cycles;
 static int		cycles_fd;
 
 static const struct option options[] = {
-	OPT_STRING('l', "size", &size_str, "1MB",
+	OPT_STRING('s', "size", &size_str, "1MB",
 		    "Specify the size of the memory buffers. "
 		    "Available units: B, KB, MB, GB and TB (case insensitive)"),
 
 	OPT_STRING('r', "routine", &routine_str, "all",
 		    "Specify the routine to run, \"all\" runs all available routines, \"help\" lists them"),
 
-	OPT_INTEGER('i', "iterations", &iterations,
-		    "Repeat the function this number of times"),
+	OPT_INTEGER('l', "nr_loops", &nr_loops,
+		    "Specify the number of loops to run. (default: 1)"),
 
 	OPT_BOOLEAN('c', "cycles", &use_cycles,
 		    "Use a cycles event instead of gettimeofday() to measure performance"),

@@ -159,7 +159,7 @@ static int bench_mem_common(int argc, const char **argv, struct bench_mem_info *
 		init_cycles();
 
 	size = (size_t)perf_atoll((char *)size_str);
-	size_total = (double)size * iterations;
+	size_total = (double)size * nr_loops;
 
 	if ((s64)size <= 0) {
 		fprintf(stderr, "Invalid size:%s\n", size_str);

@@ -222,7 +222,7 @@ static u64 do_memcpy_cycles(const struct routine *r, size_t size)
 	fn(dst, src, size);
 
 	cycle_start = get_cycles();
-	for (i = 0; i < iterations; ++i)
+	for (i = 0; i < nr_loops; ++i)
 		fn(dst, src, size);
 	cycle_end = get_cycles();
 

@@ -247,7 +247,7 @@ static double do_memcpy_gettimeofday(const struct routine *r, size_t size)
 	fn(dst, src, size);
 
 	BUG_ON(gettimeofday(&tv_start, NULL));
-	for (i = 0; i < iterations; ++i)
+	for (i = 0; i < nr_loops; ++i)
 		fn(dst, src, size);
 	BUG_ON(gettimeofday(&tv_end, NULL));
 

@@ -256,7 +256,7 @@ static double do_memcpy_gettimeofday(const struct routine *r, size_t size)
 	free(src);
 	free(dst);
 
-	return (double)(((double)size * iterations) / timeval2double(&tv_diff));
+	return (double)(((double)size * nr_loops) / timeval2double(&tv_diff));
 }
 
 struct routine memcpy_routines[] = {

@@ -313,7 +313,7 @@ static u64 do_memset_cycles(const struct routine *r, size_t size)
 	fn(dst, -1, size);
 
 	cycle_start = get_cycles();
-	for (i = 0; i < iterations; ++i)
+	for (i = 0; i < nr_loops; ++i)
 		fn(dst, i, size);
 	cycle_end = get_cycles();
 

@@ -337,14 +337,14 @@ static double do_memset_gettimeofday(const struct routine *r, size_t size)
 	fn(dst, -1, size);
 
 	BUG_ON(gettimeofday(&tv_start, NULL));
-	for (i = 0; i < iterations; ++i)
+	for (i = 0; i < nr_loops; ++i)
 		fn(dst, i, size);
 	BUG_ON(gettimeofday(&tv_end, NULL));
 
 	timersub(&tv_end, &tv_start, &tv_diff);
 
 	free(dst);
-	return (double)(((double)size * iterations) / timeval2double(&tv_diff));
+	return (double)(((double)size * nr_loops) / timeval2double(&tv_diff));
 }
 
 static const char * const bench_mem_memset_usage[] = {

index 870b7e665a20..492df2752a2d 100644
@@ -164,8 +164,8 @@ static const struct option options[] = {
 	OPT_STRING('L', "mb_proc_locked", &p0.mb_proc_locked_str,"MB", "process serialized/locked memory access (MBs), <= process_memory"),
 	OPT_STRING('T', "mb_thread"	, &p0.mb_thread_str,	"MB", "thread  memory (MBs)"),
 
-	OPT_UINTEGER('l', "nr_loops"	, &p0.nr_loops,		"max number of loops to run"),
-	OPT_UINTEGER('s', "nr_secs"	, &p0.nr_secs,		"max number of seconds to run"),
+	OPT_UINTEGER('l', "nr_loops"	, &p0.nr_loops,		"max number of loops to run (default: unlimited)"),
+	OPT_UINTEGER('s', "nr_secs"	, &p0.nr_secs,		"max number of seconds to run (default: 5 secs)"),
 	OPT_UINTEGER('u', "usleep"	, &p0.sleep_usecs,	"usecs to sleep per loop iteration"),
 
 	OPT_BOOLEAN('R', "data_reads"	, &p0.data_reads,	"access the data via writes (can be mixed with -W)"),

index d7f281c2828d..d4ff1b539cfd 100644
@@ -33,7 +33,7 @@
 #define DATASIZE 100
 
 static bool use_pipes = false;
-static unsigned int loops = 100;
+static unsigned int nr_loops = 100;
 static bool thread_mode = false;
 static unsigned int num_groups = 10;
 

@@ -79,7 +79,7 @@ static void ready(int ready_out, int wakefd)
 		err(EXIT_FAILURE, "poll");
 }
 
-/* Sender sprays loops messages down each file descriptor */
+/* Sender sprays nr_loops messages down each file descriptor */
 static void *sender(struct sender_context *ctx)
 {
 	char data[DATASIZE];

@@ -88,7 +88,7 @@ static void *sender(struct sender_context *ctx)
 	ready(ctx->ready_out, ctx->wakefd);
 
 	/* Now pump to every receiver. */
-	for (i = 0; i < loops; i++) {
+	for (i = 0; i < nr_loops; i++) {
 		for (j = 0; j < ctx->num_fds; j++) {
 			int ret, done = 0;
 

@@ -213,7 +213,7 @@ static unsigned int group(pthread_t *pth,
 		/* Create the pipe between client and server */
 		fdpair(fds);
 
-		ctx->num_packets = num_fds * loops;
+		ctx->num_packets = num_fds * nr_loops;
 		ctx->in_fds[0] = fds[0];
 		ctx->in_fds[1] = fds[1];
 		ctx->ready_out = ready_out;

@@ -250,7 +250,7 @@ static const struct option options[] = {
 	OPT_BOOLEAN('t', "thread", &thread_mode,
 		    "Be multi thread instead of multi process"),
 	OPT_UINTEGER('g', "group", &num_groups, "Specify number of groups"),
-	OPT_UINTEGER('l', "loop", &loops, "Specify number of loops"),
+	OPT_UINTEGER('l', "nr_loops", &nr_loops, "Specify the number of loops to run (default: 100)"),
 	OPT_END()
 };
 

155 lines not on this page · raw message

index a8fc948c8ace..b17aed36ca16 100644
@@ -36,7 +36,7 @@ struct bench {
 #ifdef HAVE_LIBNUMA_SUPPORT
 static struct bench numa_benchmarks[] = {
 	{ "mem",	"Benchmark for NUMA workloads",			bench_numa		},
-	{ "all",	"Test all NUMA benchmarks",			NULL			},
+	{ "all",	"Run all NUMA benchmarks",			NULL			},
 	{ NULL,		NULL,						NULL			}
 };
 #endif

@@ -44,14 +44,14 @@ static struct bench numa_benchmarks[] = {
 static struct bench sched_benchmarks[] = {
 	{ "messaging",	"Benchmark for scheduling and IPC",		bench_sched_messaging	},
 	{ "pipe",	"Benchmark for pipe() between two processes",	bench_sched_pipe	},
-	{ "all",	"Test all scheduler benchmarks",		NULL			},
+	{ "all",	"Run all scheduler benchmarks",		NULL			},
 	{ NULL,		NULL,						NULL			}
 };
 
 static struct bench mem_benchmarks[] = {
 	{ "memcpy",	"Benchmark for memcpy() functions",		bench_mem_memcpy	},
 	{ "memset",	"Benchmark for memset() functions",		bench_mem_memset	},
-	{ "all",	"Test all memory access benchmarks",		NULL			},
+	{ "all",	"Run all memory access benchmarks",		NULL			},
 	{ NULL,		NULL,						NULL			}
 };
 

@@ -62,7 +62,7 @@ static struct bench futex_benchmarks[] = {
 	{ "requeue",	"Benchmark for futex requeue calls",            bench_futex_requeue	},
 	/* pi-futexes */
 	{ "lock-pi",	"Benchmark for futex lock_pi calls",            bench_futex_lock_pi	},
-	{ "all",	"Test all futex benchmarks",			NULL			},
+	{ "all",	"Run all futex benchmarks",			NULL			},
 	{ NULL,		NULL,						NULL			}
 };
 

index 36486eade1ef..a7d588bf3cdd 100644
@@ -463,7 +463,7 @@ int cmd_help(int argc, const char **argv, const char *prefix __maybe_unused)
 			builtin_help_subcommands, builtin_help_usage, 0);
 
 	if (show_all) {
-		printf("\n usage: %s\n\n", perf_usage_string);
+		printf("\n Usage: %s\n\n", perf_usage_string);
 		list_commands("perf commands", &main_cmds, &other_cmds);
 		printf(" %s\n\n", perf_more_info_string);
 		return 0;

index 9a38b05f0273..8aa7922397a9 100644
@@ -648,7 +648,7 @@ int usage_with_options_internal(const char * const *usagestr,
 	if (!usagestr)
 		return PARSE_OPT_HELP;
 
-	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	fprintf(stderr, "\n Usage: %s\n", *usagestr++);
 	while (*usagestr && **usagestr)
 		fprintf(stderr, "    or: %s\n", *usagestr++);
 	while (*usagestr) {

@@ -684,7 +684,7 @@ int parse_options_usage(const char * const *usagestr,
 	if (!usagestr)
 		goto opt;
 
-	fprintf(stderr, "\n usage: %s\n", *usagestr++);
+	fprintf(stderr, "\n Usage: %s\n", *usagestr++);
 	while (*usagestr && **usagestr)
 		fprintf(stderr, "    or: %s\n", *usagestr++);
 	while (*usagestr) {

index 5ef88f760b12..184057295970 100644
@@ -434,7 +434,7 @@ static void print_noise_pct(double total, double avg)
 
 static void print_noise(struct perf_evsel *evsel, double avg)
 {
-	struct perf_stat *ps;
+	struct perf_stat_evsel *ps;
 
 	if (run_count == 1)
 		return;

@@ -671,7 +671,7 @@ static void print_aggr_thread(struct perf_evsel *counter, char *prefix)
 static void print_counter_aggr(struct perf_evsel *counter, char *prefix)
 {
 	FILE *output = stat_config.output;
-	struct perf_stat *ps = counter->priv;
+	struct perf_stat_evsel *ps = counter->priv;
 	double avg = avg_stats(&ps->res_stats[0]);
 	int scaled = counter->counts->scaled;
 	double uval;

index 2d065d065b67..93e6d697e574 100644
@@ -67,7 +67,7 @@ double rel_stddev_stats(double stddev, double avg)
 bool __perf_evsel_stat__is(struct perf_evsel *evsel,
 			   enum perf_stat_evsel_id id)
 {
-	struct perf_stat *ps = evsel->priv;
+	struct perf_stat_evsel *ps = evsel->priv;
 
 	return ps->id == id;
 }

@@ -84,7 +84,7 @@ static const char *id_str[PERF_STAT_EVSEL_ID__MAX] = {
 
 void perf_stat_evsel_id_init(struct perf_evsel *evsel)
 {
-	struct perf_stat *ps = evsel->priv;
+	struct perf_stat_evsel *ps = evsel->priv;
 	int i;
 
 	/* ps->id is 0 hence PERF_STAT_EVSEL_ID__NONE by default */

@@ -100,7 +100,7 @@ void perf_stat_evsel_id_init(struct perf_evsel *evsel)
 void perf_evsel__reset_stat_priv(struct perf_evsel *evsel)
 {
 	int i;
-	struct perf_stat *ps = evsel->priv;
+	struct perf_stat_evsel *ps = evsel->priv;
 
 	for (i = 0; i < 3; i++)
 		init_stats(&ps->res_stats[i]);

@@ -110,7 +110,7 @@ void perf_evsel__reset_stat_priv(struct perf_evsel *evsel)
 
 int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel)
 {
-	evsel->priv = zalloc(sizeof(struct perf_stat));
+	evsel->priv = zalloc(sizeof(struct perf_stat_evsel));
 	if (evsel->priv == NULL)
 		return -ENOMEM;
 	perf_evsel__reset_stat_priv(evsel);

@@ -304,7 +304,7 @@ int perf_stat_process_counter(struct perf_stat_config *config,
 			      struct perf_evsel *counter)
 {
 	struct perf_counts_values *aggr = &counter->counts->aggr;
-	struct perf_stat *ps = counter->priv;
+	struct perf_stat_evsel *ps = counter->priv;
 	u64 *count = counter->counts->aggr.values;
 	int i, ret;
 

index 62448c8175d3..f8d9d5ce24a9 100644
@@ -20,7 +20,7 @@ enum perf_stat_evsel_id {
 	PERF_STAT_EVSEL_ID__MAX,
 };
 
-struct perf_stat {
+struct perf_stat_evsel {
 	struct stats		res_stats[3];
 	enum perf_stat_evsel_id	id;
 };

index 184057295970..abeb15aebd12 100644
@@ -479,6 +479,7 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr)
 			csv_sep);
 		break;
 	case AGGR_GLOBAL:
+	case AGGR_UNSET:
 	default:
 		break;
 	}

@@ -799,6 +800,8 @@ static void print_interval(char *prefix, struct timespec *ts)
 		case AGGR_GLOBAL:
 		default:
 			fprintf(output, "#           time             counts %*s events\n", unit_width, "unit");
+		case AGGR_UNSET:
+			break;
 		}
 	}
 

@@ -880,6 +883,7 @@ static void print_counters(struct timespec *ts, int argc, const char **argv)
 		evlist__for_each(evsel_list, counter)
 			print_counter(counter, prefix);
 		break;
+	case AGGR_UNSET:
 	default:
 		break;
 	}

@@ -960,6 +964,7 @@ static int perf_stat_init_aggr_mode(void)
 	case AGGR_NONE:
 	case AGGR_GLOBAL:
 	case AGGR_THREAD:
+	case AGGR_UNSET:
 	default:
 		break;
 	}

index 93e6d697e574..837374181ec7 100644
@@ -272,6 +272,7 @@ process_counter_values(struct perf_stat_config *config, struct perf_evsel *evsel
 			aggr->ena += count->ena;
 			aggr->run += count->run;
 		}
+	case AGGR_UNSET:
 	default:
 		break;
 	}

index f8d9d5ce24a9..da1d11c4f8c1 100644
@@ -31,6 +31,7 @@ enum aggr_mode {
 	AGGR_SOCKET,
 	AGGR_CORE,
 	AGGR_THREAD,
+	AGGR_UNSET,
 };
 
 struct perf_stat_config {

index c51c29fd0732..70ec8d031f9d 100644
@@ -258,8 +258,8 @@ static int cmp_ids(const void *a, const void *b)
 	return *(int *)a - *(int *)b;
 }
 
-static int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
-			      int (*f)(struct cpu_map *map, int cpu))
+int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
+		       int (*f)(struct cpu_map *map, int cpu))
 {
 	struct cpu_map *c;
 	int nr = cpus->nr;

index 8982d538da83..6e36fc35eca6 100644
@@ -87,4 +87,6 @@ static inline int cpu__get_node(int cpu)
 	return cpunode_map[cpu];
 }
 
+int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
+		       int (*f)(struct cpu_map *map, int cpu));
 #endif /* __PERF_CPUMAP_H */

index abeb15aebd12..91e793a76929 100644
@@ -944,6 +944,16 @@ static int stat__set_big_num(const struct option *opt __maybe_unused,
 	return 0;
 }
 
+static int perf_stat__get_socket(struct cpu_map *map, int cpu)
+{
+	return cpu_map__get_socket(map, cpu, NULL);
+}
+
+static int perf_stat__get_core(struct cpu_map *map, int cpu)
+{
+	return cpu_map__get_core(map, cpu, NULL);
+}
+
 static int perf_stat_init_aggr_mode(void)
 {
 	switch (stat_config.aggr_mode) {

@@ -952,14 +962,14 @@ static int perf_stat_init_aggr_mode(void)
 			perror("cannot build socket map");
 			return -1;
 		}
-		aggr_get_id = cpu_map__get_socket;
+		aggr_get_id = perf_stat__get_socket;
 		break;
 	case AGGR_CORE:
 		if (cpu_map__build_core_map(evsel_list->cpus, &aggr_map)) {
 			perror("cannot build core map");
 			return -1;
 		}
-		aggr_get_id = cpu_map__get_core;
+		aggr_get_id = perf_stat__get_core;
 		break;
 	case AGGR_NONE:
 	case AGGR_GLOBAL:

index c3aff53a976a..f5bb096c3bd9 100644
@@ -73,10 +73,10 @@ static int check_cpu_topology(char *path, struct cpu_map *map)
 
 	for (i = 0; i < map->nr; i++) {
 		TEST_ASSERT_VAL("Core ID doesn't match",
-			(session->header.env.cpu[map->map[i]].core_id == (cpu_map__get_core(map, i) & 0xffff)));
+			(session->header.env.cpu[map->map[i]].core_id == (cpu_map__get_core(map, i, NULL) & 0xffff)));
 
 		TEST_ASSERT_VAL("Socket ID doesn't match",
-			(session->header.env.cpu[map->map[i]].socket_id == cpu_map__get_socket(map, i)));
+			(session->header.env.cpu[map->map[i]].socket_id == cpu_map__get_socket(map, i, NULL)));
 	}
 
 	perf_session__delete(session);

index 70ec8d031f9d..b36845347f01 100644
@@ -241,7 +241,7 @@ int cpu_map__get_socket_id(int cpu)
 	return ret ?: value;
 }
 
-int cpu_map__get_socket(struct cpu_map *map, int idx)
+int cpu_map__get_socket(struct cpu_map *map, int idx, void *data __maybe_unused)
 {
 	int cpu;
 

@@ -259,7 +259,8 @@ static int cmp_ids(const void *a, const void *b)
 }
 
 int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
-		       int (*f)(struct cpu_map *map, int cpu))
+		       int (*f)(struct cpu_map *map, int cpu, void *data),
+		       void *data)
 {
 	struct cpu_map *c;
 	int nr = cpus->nr;

@@ -271,7 +272,7 @@ int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
 		return -1;
 
 	for (cpu = 0; cpu < nr; cpu++) {
-		s1 = f(cpus, cpu);
+		s1 = f(cpus, cpu, data);
 		for (s2 = 0; s2 < c->nr; s2++) {
 			if (s1 == c->map[s2])
 				break;

@@ -295,7 +296,7 @@ int cpu_map__get_core_id(int cpu)
 	return ret ?: value;
 }
 
-int cpu_map__get_core(struct cpu_map *map, int idx)
+int cpu_map__get_core(struct cpu_map *map, int idx, void *data)
 {
 	int cpu, s;
 

@@ -306,7 +307,7 @@ int cpu_map__get_core(struct cpu_map *map, int idx)
 
 	cpu = cpu_map__get_core_id(cpu);
 
-	s = cpu_map__get_socket(map, idx);
+	s = cpu_map__get_socket(map, idx, data);
 	if (s == -1)
 		return -1;
 

@@ -321,12 +322,12 @@ int cpu_map__get_core(struct cpu_map *map, int idx)
 
 int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp)
 {
-	return cpu_map__build_map(cpus, sockp, cpu_map__get_socket);
+	return cpu_map__build_map(cpus, sockp, cpu_map__get_socket, NULL);
 }
 
 int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep)
 {
-	return cpu_map__build_map(cpus, corep, cpu_map__get_core);
+	return cpu_map__build_map(cpus, corep, cpu_map__get_core, NULL);
 }
 
 /* setup simple routines to easily access node numbers given a cpu number */

index 6e36fc35eca6..f1bcd2cfa164 100644
@@ -19,9 +19,9 @@ struct cpu_map *cpu_map__dummy_new(void);
 struct cpu_map *cpu_map__read(FILE *file);
 size_t cpu_map__fprintf(struct cpu_map *map, FILE *fp);
 int cpu_map__get_socket_id(int cpu);
-int cpu_map__get_socket(struct cpu_map *map, int idx);
+int cpu_map__get_socket(struct cpu_map *map, int idx, void *data);
 int cpu_map__get_core_id(int cpu);
-int cpu_map__get_core(struct cpu_map *map, int idx);
+int cpu_map__get_core(struct cpu_map *map, int idx, void *data);
 int cpu_map__build_socket_map(struct cpu_map *cpus, struct cpu_map **sockp);
 int cpu_map__build_core_map(struct cpu_map *cpus, struct cpu_map **corep);
 

@@ -88,5 +88,6 @@ static inline int cpu__get_node(int cpu)
 }
 
 int cpu_map__build_map(struct cpu_map *cpus, struct cpu_map **res,
-		       int (*f)(struct cpu_map *map, int cpu));
+		       int (*f)(struct cpu_map *map, int cpu, void *data),
+		       void *data);
 #endif /* __PERF_CPUMAP_H */

index 837374181ec7..2d9d8306dbd3 100644
@@ -230,7 +230,7 @@ static int check_per_pkg(struct perf_evsel *counter,
 	if (!(vals->run && vals->ena))
 		return 0;
 
-	s = cpu_map__get_socket(cpus, cpu);
+	s = cpu_map__get_socket(cpus, cpu, NULL);
 	if (s < 0)
 		return -1;
 

index 8ce1c6bbfa45..2653c0273b89 100644
@@ -686,7 +686,10 @@ static int process_attr(struct perf_tool *tool, union perf_event *event,
 
 	set_print_ip_opts(&evsel->attr);
 
-	return perf_evsel__check_attr(evsel, scr->session);
+	if (evsel->attr.sample_type)
+		err = perf_evsel__check_attr(evsel, scr->session);
+
+	return err;
 }
 
 static int process_comm_event(struct perf_tool *tool,

index d1de9c4a7ddf..9419b944220f 100644
@@ -270,7 +270,7 @@ struct function memcpy_functions[] = {
 # undef MEMCPY_FN
 #endif
 
-	{ NULL, }
+	{ .name = NULL, }
 };
 
 static const char * const bench_mem_memcpy_usage[] = {

@@ -363,7 +363,7 @@ static const struct function memset_functions[] = {
 # undef MEMSET_FN
 #endif
 
-	{ NULL, }
+	{ .name = NULL, }
 };
 
 int bench_mem_memset(int argc, const char **argv, const char *prefix __maybe_unused)