diff mbox series

[bpf-next,v6,3/5] selftests/bpf: Ensure that next_cpu() returns a valid CPU number

Message ID 20230613080921.1623219-4-houtao@huaweicloud.com (mailing list archive)
State Superseded
Headers show
Series Add benchmark for bpf memory allocator | expand

Commit Message

Hou Tao June 13, 2023, 8:09 a.m. UTC
From: Hou Tao <houtao1@huawei.com>

When using option -a without --prod-affinity or --cons-affinity, if the
number of producers and consumers is greater than the number of online
CPUs, the benchmark will fail to run as shown below:

  $ getconf _NPROCESSORS_ONLN
  8
  $ ./bench bpf-loop -a -p9
  Setting up benchmark 'bpf-loop'...
  setting affinity to CPU #8 failed: -22

Fix it by returning the remainder of next_cpu divided by the number of
online CPUs in next_cpu().

Signed-off-by: Hou Tao <houtao1@huawei.com>
---
 tools/testing/selftests/bpf/bench.c | 3 ++-
 tools/testing/selftests/bpf/bench.h | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)
diff mbox series

Patch

diff --git a/tools/testing/selftests/bpf/bench.c b/tools/testing/selftests/bpf/bench.c
index 0b5d2b5303c9..56f1c166a57b 100644
--- a/tools/testing/selftests/bpf/bench.c
+++ b/tools/testing/selftests/bpf/bench.c
@@ -469,7 +469,7 @@  static int next_cpu(struct cpu_set *cpu_set)
 		exit(1);
 	}
 
-	return cpu_set->next_cpu++;
+	return cpu_set->next_cpu++ % env.nr_cpus;
 }
 
 static struct bench_state {
@@ -659,6 +659,7 @@  static void collect_measurements(long delta_ns) {
 
 int main(int argc, char **argv)
 {
+	env.nr_cpus = get_nprocs();
 	parse_cmdline_args_init(argc, argv);
 
 	if (env.list) {
diff --git a/tools/testing/selftests/bpf/bench.h b/tools/testing/selftests/bpf/bench.h
index 402729c6a3ac..7ff32be3d730 100644
--- a/tools/testing/selftests/bpf/bench.h
+++ b/tools/testing/selftests/bpf/bench.h
@@ -27,6 +27,7 @@  struct env {
 	bool quiet;
 	int consumer_cnt;
 	int producer_cnt;
+	int nr_cpus;
 	struct cpu_set prod_cpus;
 	struct cpu_set cons_cpus;
 };