On 22 November 2016 at 11:41, Juri Lelli juri.lelli@arm.com wrote:
On 22/11/16 10:21, Juri Lelli wrote:
On 22/11/16 10:41, Vincent Guittot wrote:
[...]
There also the c_run metrics in another patch, does it mean that you want only one run event per phase ?
Nope. That is broken in my current set. I'll fix that as you suggested (summing up the durations).
Which basically should result in the following right?
--->8--- src/rt-app.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/rt-app.c b/src/rt-app.c index 655c43e06a7d..bec511121d77 100644 --- a/src/rt-app.c +++ b/src/rt-app.c @@ -287,7 +287,7 @@ static int run_event(event_data_t *event, int dry_run, { struct timespec t_start, t_end; log_debug("run %d ", event->duration);
ldata->c_run = event->duration;
ldata->c_run += event->duration; clock_gettime(CLOCK_MONOTONIC, &t_start); *perf += loadwait(event->duration); clock_gettime(CLOCK_MONOTONIC, &t_end);--->8---
I was actually thinking that we should also sum up durations of all event that specify a duration (sleep, runtime, etc.) and maybe modify the log header as s/c_run/c_duration/ ?
IMO adding runtime makes perfectly sense as commented in the related patch But not sleep which is a idle state just like timer