[PATCH] perf/bench: Fix perf bench syscall XXXX loop count

[PATCH] perf/bench: Fix perf bench syscall XXXX loop count

[Thomas Richter]

Command 'perf bench syscall fork -l 100000' offers option
-l to run for a specified number of iterations. However this
option is not always observed. The number is silently limited to
10000 iterations as can be seen:

Output before:
 # perf bench syscall fork -l 100000
 # Running 'syscall/fork' benchmark:
 # Executed 10,000 fork() calls
     Total time: 23.388 [sec]

    2338.809800 usecs/op
            427 ops/sec
 #

When explicitly specified with option -l or --loops, also observe
higher number of iterations:

Output after:
 # perf bench syscall fork -l 100000
 # Running 'syscall/fork' benchmark:
 # Executed 100,000 fork() calls
     Total time: 716.982 [sec]

    7169.829510 usecs/op
            139 ops/sec
 #

This patch fixes the issue for basic execve fork and getpgid.

Fixes: ece7f7c0507c ("perf bench syscall: Add fork syscall benchmark")
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Cc: Tiezhu Yang <yangtiezhu@loongson.cn>
---
 tools/perf/bench/syscall.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/tools/perf/bench/syscall.c b/tools/perf/bench/syscall.c
index ea4dfc07cbd6..e7dc216f717f 100644
--- a/tools/perf/bench/syscall.c
+++ b/tools/perf/bench/syscall.c
@@ -22,8 +22,7 @@
 #define __NR_fork -1
 #endif
 
-#define LOOPS_DEFAULT 10000000
-static	int loops = LOOPS_DEFAULT;
+static	int loops;
 
 static const struct option options[] = {
 	OPT_INTEGER('l', "loop",	&loops,		"Specify number of loops"),
@@ -80,6 +79,18 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
 	const char *name = NULL;
 	int i;
 
+	switch (syscall) {
+	case __NR_fork:
+	case __NR_execve:
+		/* Limit default loop to 10000 times to save time */
+		loops = 10000;
+		break;
+	default:
+		loops = 10000000;
+		break;
+	}
+
+	/* Options -l and --loops override default above */
 	argc = parse_options(argc, argv, options, bench_syscall_usage, 0);
 
 	gettimeofday(&start, NULL);
@@ -94,16 +105,9 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
 			break;
 		case __NR_fork:
 			test_fork();
-			/* Only loop 10000 times to save time */
-			if (i == 10000)
-				loops = 10000;
 			break;
 		case __NR_execve:
 			test_execve();
-			/* Only loop 10000 times to save time */
-			if (i == 10000)
-				loops = 10000;
-			break;
 		default:
 			break;
 		}
-- 
2.47.0

Re: [PATCH] perf/bench: Fix perf bench syscall XXXX loop count

[Athira Rajeev]

> On 4 Mar 2025, at 2:53 PM, Thomas Richter <tmricht@linux.ibm.com> wrote:
> 
> Command 'perf bench syscall fork -l 100000' offers option
> -l to run for a specified number of iterations. However this
> option is not always observed. The number is silently limited to
> 10000 iterations as can be seen:
> 
> Output before:
> # perf bench syscall fork -l 100000
> # Running 'syscall/fork' benchmark:
> # Executed 10,000 fork() calls
>     Total time: 23.388 [sec]
> 
>    2338.809800 usecs/op
>            427 ops/sec
> #
> 
> When explicitly specified with option -l or --loops, also observe
> higher number of iterations:
> 
> Output after:
> # perf bench syscall fork -l 100000
> # Running 'syscall/fork' benchmark:
> # Executed 100,000 fork() calls
>     Total time: 716.982 [sec]
> 
>    7169.829510 usecs/op
>            139 ops/sec
> #
> 

Tested-by: Athira Rajeev <atrajeev@linux.ibm.com>

Thanks
Athira
> This patch fixes the issue for basic execve fork and getpgid.
> 
> Fixes: ece7f7c0507c ("perf bench syscall: Add fork syscall benchmark")
> Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
> Acked-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
> Cc: Tiezhu Yang <yangtiezhu@loongson.cn>
> ---
> tools/perf/bench/syscall.c | 22 +++++++++++++---------
> 1 file changed, 13 insertions(+), 9 deletions(-)
> 
> diff --git a/tools/perf/bench/syscall.c b/tools/perf/bench/syscall.c
> index ea4dfc07cbd6..e7dc216f717f 100644
> --- a/tools/perf/bench/syscall.c
> +++ b/tools/perf/bench/syscall.c
> @@ -22,8 +22,7 @@
> #define __NR_fork -1
> #endif
> 
> -#define LOOPS_DEFAULT 10000000
> -static int loops = LOOPS_DEFAULT;
> +static int loops;
> 
> static const struct option options[] = {
> OPT_INTEGER('l', "loop", &loops, "Specify number of loops"),
> @@ -80,6 +79,18 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
> const char *name = NULL;
> int i;
> 
> + switch (syscall) {
> + case __NR_fork:
> + case __NR_execve:
> + /* Limit default loop to 10000 times to save time */
> + loops = 10000;
> + break;
> + default:
> + loops = 10000000;
> + break;
> + }
> +
> + /* Options -l and --loops override default above */
> argc = parse_options(argc, argv, options, bench_syscall_usage, 0);
> 
> gettimeofday(&start, NULL);
> @@ -94,16 +105,9 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
> break;
> case __NR_fork:
> test_fork();
> - /* Only loop 10000 times to save time */
> - if (i == 10000)
> - loops = 10000;
> break;
> case __NR_execve:
> test_execve();
> - /* Only loop 10000 times to save time */
> - if (i == 10000)
> - loops = 10000;
> - break;
> default:
> break;
> }
> -- 
> 2.47.0
> 
> 

Re: [PATCH] perf/bench: Fix perf bench syscall XXXX loop count

[Namhyung Kim]

On Tue, 04 Mar 2025 10:23:49 +0100, Thomas Richter wrote:
> Command 'perf bench syscall fork -l 100000' offers option
> -l to run for a specified number of iterations. However this
> option is not always observed. The number is silently limited to
> 10000 iterations as can be seen:
> 
> Output before:
>  # perf bench syscall fork -l 100000
>  # Running 'syscall/fork' benchmark:
>  # Executed 10,000 fork() calls
>      Total time: 23.388 [sec]
> 
> [...]
Applied to perf-tools-next, thanks!

Best regards,
Namhyung