Re: [PATCH v6 07/26] perf trace: Start BPF summary before starting workload

[Aaron Tomlin]

On Mon, Sep 28, 2026 at 11:25:46AM -0700, Ian Rogers wrote:
> trace_start_bpf_summary() is called after evlist__start_workload(), so a
> short lived workload can make its syscalls, or even exit, before the BPF
> summary is enabled. It also ignores --delay. Enable it along with the
> evlist, before the workload starts or after the initial delay.
> 
> Assisted-by: Antigravity:gemini-3.1-pro
> Signed-off-by: Ian Rogers <irogers@xxxxxxxxxx>
> ---
>  tools/perf/builtin-trace.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
> index f94745a60f4a..1ffe824ba670 100644
> --- a/tools/perf/builtin-trace.c
> +++ b/tools/perf/builtin-trace.c
> @@ -4855,17 +4855,19 @@ static int trace__run(struct trace *trace, int argc, const char **argv)
>  	if (!target__none(&trace->opts.target) && !trace->opts.target.initial_delay)
>  		evlist__enable(evlist);
>  
> +	if (trace->summary_bpf && !trace->opts.target.initial_delay)
> +		trace_start_bpf_summary();
> +
>  	if (forks)
>  		evlist__start_workload(evlist);
>  
>  	if (trace->opts.target.initial_delay) {
>  		usleep(trace->opts.target.initial_delay * 1000);
>  		evlist__enable(evlist);
> +		if (trace->summary_bpf)
> +			trace_start_bpf_summary();
>  	}
>  
> -	if (trace->summary_bpf)
> -		trace_start_bpf_summary();
> -
>  	trace->multiple_threads = perf_thread_map__pid(evlist__core(evlist)->threads, 0) == -1 ||
>  		perf_thread_map__nr(evlist__core(evlist)->threads) > 1 ||
>  		evlist__first(evlist)->core.attr.inherit;
> -- 
> 2.56.0.rc1.315.gc6ed9934b7-goog
> 

Reviewed-by: Aaron Tomlin <atomlin@xxxxxxxxxxx>

-- 
Aaron Tomlin