[PATCH i-g-t] tests/intel/xe_pm_residency: Print execution and sleep times

Riana Tauro riana.tauro at intel.com
Thu Sep 19 07:07:21 UTC 2024


Hi Karthik

you had sent a patch with assert prior to this. Don't we need
that?

Thanks,
Riana

On 9/19/2024 12:03 PM, Karthik Poosa wrote:
> Print the workload execution and sleep times of the test to help diagnose
> timeouts.
> This will show how long the workload ran and how long the test slept.
> 
> Signed-off-by: Karthik Poosa <karthik.poosa at intel.com>
> ---
>   tests/intel/xe_pm_residency.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/intel/xe_pm_residency.c b/tests/intel/xe_pm_residency.c
> index 0e687558b..3c3b33ab1 100644
> --- a/tests/intel/xe_pm_residency.c
> +++ b/tests/intel/xe_pm_residency.c
> @@ -139,7 +139,7 @@ static void exec_load(int fd, struct drm_xe_engine_class_instance *hwe, unsigned
>   		elapsed = igt_nsec_elapsed(&tv);
>   		igt_assert_eq(data->data, done[1]);
>   
> -		igt_debug("Execution took %.3fms (submit %.1fus, wait %.1fus)\n",
> +		igt_info("Execution took %.3fms (submit %.1fus, wait %.1fus)\n",
>   			  1e-6 * elapsed,
>   			  1e-3 * submit,
>   			  1e-3 * (elapsed - submit));
> @@ -150,6 +150,8 @@ static void exec_load(int fd, struct drm_xe_engine_class_instance *hwe, unsigned
>   		 * Execute the above workload for ~1% of the elapsed time and sleep for
>   		 * the rest of the time (~99%)
>   		 */
> +		igt_info("Sleeping for %ld ms\n", ((elapsed / 10)/1000));
> +
>   		usleep(elapsed / 10);
>   	} while (!READ_ONCE(*done));
>   


More information about the igt-dev mailing list