Re: [PATCH] perf record: add dummy event during system wide synthesis

[Jiri Olsa]

On Wed, Apr 15, 2020 at 05:47:13PM -0700, Ian Rogers wrote:
> During the processing of /proc during event synthesis new processes may
> start. Add a dummy event if /proc is to be processed, to capture mmaps
> for starting processes. This reuses the existing logic for
> initial-delay.
> 
> Suggested-by: Stephane Eranian <eranian@xxxxxxxxxx>
> Signed-off-by: Ian Rogers <irogers@xxxxxxxxxx>
> ---
>  tools/perf/builtin-record.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index 1ab349abe904..bab4fc8568d1 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -805,16 +805,18 @@ static int record__open(struct record *rec)
>  	int rc = 0;
>  
>  	/*
> -	 * For initial_delay we need to add a dummy event so that we can track
> -	 * PERF_RECORD_MMAP while we wait for the initial delay to enable the
> -	 * real events, the ones asked by the user.
> +	 * For initial_delay or system wide, we need to add a dummy event so
> +	 * that we can track PERF_RECORD_MMAP to cover the delay (of waiting or
> +	 * synthesis) prior to the real user events being enabled.
>  	 */
> -	if (opts->initial_delay) {
> +	if (opts->initial_delay || target__has_cpu(&opts->target)) {

hum, how this works for system wide? IIRC the delay works
for monitoring tasks by setting dummy with enable_on_exec,
and enabling the rest by ioctl after the delay.. so without
the monitored task there's no enable_on_exec effect

jirka

>  		if (perf_evlist__add_dummy(evlist))
>  			return -ENOMEM;
>  
> +		/* Disable tracking of mmaps on lead event. */
>  		pos = evlist__first(evlist);
>  		pos->tracking = 0;
> +		/* Activate dummy event immediately. */
>  		pos = evlist__last(evlist);
>  		pos->tracking = 1;
>  		pos->core.attr.enable_on_exec = 1;
> -- 
> 2.26.0.110.g2183baf09c-goog
>