The bperf BPF counter code doesn't handle "any"(-1) CPU events, always
wanting to aggregate a count against a CPU, which avoids the need for
atomics so let's not change that. Force evsels used for BPF counters
to require a CPU when not in system-wide mode so that the "any"(-1)
value isn't used during map propagation and evsel's CPU map matches
that of the PMU.

Fixes: b91917c0c6fa ("perf bpf_counter: Fix handling of cpumap fixing hybrid")
Signed-off-by: Ian Rogers <irogers@google.com>
---
 tools/perf/builtin-stat.c     | 13 +++++++++++++
 tools/perf/util/bpf_counter.c |  7 ++++++-
 2 files changed, 19 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 7006f848f87a..f1c9d6c94fc5 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -2540,6 +2540,7 @@ int cmd_stat(int argc, const char **argv)
 	unsigned int interval, timeout;
 	const char * const stat_subcommands[] = { "record", "report" };
 	char errbuf[BUFSIZ];
+	struct evsel *counter;
 
 	setlocale(LC_ALL, "");
 
@@ -2797,6 +2798,18 @@ int cmd_stat(int argc, const char **argv)
 
 	evlist__warn_user_requested_cpus(evsel_list, target.cpu_list);
 
+	evlist__for_each_entry(evsel_list, counter) {
+		/*
+		 * Setup BPF counters to require CPUs as any(-1) isn't
+		 * supported. evlist__create_maps below will propagate this
+		 * information to the evsels. Note, evsel__is_bperf isn't yet
+		 * set up, and this change must happen early, so directly use
+		 * the bpf_counter variable and target information.
+		 */
+		if ((counter->bpf_counter || target.use_bpf) && !target__has_cpu(&target))
+			counter->core.requires_cpu = true;
+	}
+
 	if (evlist__create_maps(evsel_list, &target) < 0) {
 		if (target__has_task(&target)) {
 			pr_err("Problems finding threads of monitor\n");
diff --git a/tools/perf/util/bpf_counter.c b/tools/perf/util/bpf_counter.c
index ca5d01b9017d..a5882b582205 100644
--- a/tools/perf/util/bpf_counter.c
+++ b/tools/perf/util/bpf_counter.c
@@ -460,6 +460,7 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,
 	struct bperf_leader_bpf *skel = bperf_leader_bpf__open();
 	int link_fd, diff_map_fd, err;
 	struct bpf_link *link = NULL;
+	struct perf_thread_map *threads;
 
 	if (!skel) {
 		pr_err("Failed to open leader skeleton\n");
@@ -495,7 +496,11 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,
 	 * following evsel__open_per_cpu call
 	 */
 	evsel->leader_skel = skel;
-	evsel__open(evsel, evsel->core.cpus, evsel->core.threads);
+	assert(!perf_cpu_map__has_any_cpu_or_is_empty(evsel->core.cpus));
+	/* Always open system wide. */
+	threads = thread_map__new_by_tid(-1);
+	evsel__open(evsel, evsel->core.cpus, threads);
+	perf_thread_map__put(threads);
 
 out:
 	bperf_leader_bpf__destroy(skel);
-- 
2.51.0.710.ga91ca5db03-goog

On Thu, Oct 09, 2025 at 06:29:11AM -0700, Ian Rogers wrote:
> The bperf BPF counter code doesn't handle "any"(-1) CPU events, always
> wanting to aggregate a count against a CPU, which avoids the need for
> atomics so let's not change that. Force evsels used for BPF counters
> to require a CPU when not in system-wide mode so that the "any"(-1)
> value isn't used during map propagation and evsel's CPU map matches
> that of the PMU.
> 
> Fixes: b91917c0c6fa ("perf bpf_counter: Fix handling of cpumap fixing hybrid")
> Signed-off-by: Ian Rogers <irogers@google.com>

Applied to perf-tools-next, thanks!

Best regards,
Namhyung

On 2025/10/9 21:29, Ian Rogers wrote:
> The bperf BPF counter code doesn't handle "any"(-1) CPU events, always
> wanting to aggregate a count against a CPU, which avoids the need for
> atomics so let's not change that. Force evsels used for BPF counters
> to require a CPU when not in system-wide mode so that the "any"(-1)
> value isn't used during map propagation and evsel's CPU map matches
> that of the PMU.
> 
> Fixes: b91917c0c6fa ("perf bpf_counter: Fix handling of cpumap fixing hybrid")
> Signed-off-by: Ian Rogers <irogers@google.com>
> ---
>  tools/perf/builtin-stat.c     | 13 +++++++++++++
>  tools/perf/util/bpf_counter.c |  7 ++++++-
>  2 files changed, 19 insertions(+), 1 deletion(-)
> 
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 7006f848f87a..f1c9d6c94fc5 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -2540,6 +2540,7 @@ int cmd_stat(int argc, const char **argv)
>  	unsigned int interval, timeout;
>  	const char * const stat_subcommands[] = { "record", "report" };
>  	char errbuf[BUFSIZ];
> +	struct evsel *counter;
>  
>  	setlocale(LC_ALL, "");
>  
> @@ -2797,6 +2798,18 @@ int cmd_stat(int argc, const char **argv)
>  
>  	evlist__warn_user_requested_cpus(evsel_list, target.cpu_list);
>  
> +	evlist__for_each_entry(evsel_list, counter) {
> +		/*
> +		 * Setup BPF counters to require CPUs as any(-1) isn't
> +		 * supported. evlist__create_maps below will propagate this
> +		 * information to the evsels. Note, evsel__is_bperf isn't yet
> +		 * set up, and this change must happen early, so directly use
> +		 * the bpf_counter variable and target information.
> +		 */
> +		if ((counter->bpf_counter || target.use_bpf) && !target__has_cpu(&target))
> +			counter->core.requires_cpu = true;
> +	}
> +
>  	if (evlist__create_maps(evsel_list, &target) < 0) {
>  		if (target__has_task(&target)) {
>  			pr_err("Problems finding threads of monitor\n");
> diff --git a/tools/perf/util/bpf_counter.c b/tools/perf/util/bpf_counter.c
> index ca5d01b9017d..a5882b582205 100644
> --- a/tools/perf/util/bpf_counter.c
> +++ b/tools/perf/util/bpf_counter.c
> @@ -460,6 +460,7 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,
>  	struct bperf_leader_bpf *skel = bperf_leader_bpf__open();
>  	int link_fd, diff_map_fd, err;
>  	struct bpf_link *link = NULL;
> +	struct perf_thread_map *threads;
>  
>  	if (!skel) {
>  		pr_err("Failed to open leader skeleton\n");
> @@ -495,7 +496,11 @@ static int bperf_reload_leader_program(struct evsel *evsel, int attr_map_fd,
>  	 * following evsel__open_per_cpu call
>  	 */
>  	evsel->leader_skel = skel;
> -	evsel__open(evsel, evsel->core.cpus, evsel->core.threads);
> +	assert(!perf_cpu_map__has_any_cpu_or_is_empty(evsel->core.cpus));
> +	/* Always open system wide. */
> +	threads = thread_map__new_by_tid(-1);
> +	evsel__open(evsel, evsel->core.cpus, threads);
> +	perf_thread_map__put(threads);
>  
>  out:
>  	bperf_leader_bpf__destroy(skel);

LGTM.

Tengda