Op 20-07-16 om 14:56 schreef Ander Conselvan De Oliveira:
> On Wed, 2016-07-06 at 11:55 +0200, Maarten Lankhorst wrote:
>> Use for_each_pipe_with_valid_output instead.
>>
>> Signed-off-by: Maarten Lankhorst <maarten.lankho...@linux.intel.com>
>> ---
>> tests/kms_crtc_background_color.c | 3 +--
>> tests/kms_flip_tiling.c | 50
>> +++++++++++++++++++++++---------------
>> -
>> tests/kms_panel_fitting.c | 5 ++--
>> tests/kms_plane_scaling.c | 5 ++--
>> 4 files changed, 34 insertions(+), 29 deletions(-)
>>
>> diff --git a/tests/kms_crtc_background_color.c
>> b/tests/kms_crtc_background_color.c
>> index b496625c1693..b97c1142df6e 100644
>> --- a/tests/kms_crtc_background_color.c
>> +++ b/tests/kms_crtc_background_color.c
>> @@ -133,10 +133,9 @@ static void test_crtc_background(data_t *data)
>>
>> igt_require(data->display.has_universal_planes);
>>
>> - for_each_connected_output(display, output) {
>> + for_each_pipe_with_valid_output(display, pipe, output) {
>> igt_plane_t *plane;
>>
>> - pipe = output->config.pipe;
>> igt_output_set_pipe(output, pipe);
> Won't this cause the test to be executed more times? I.e., every output paired
> with every supported pipe vs. every output only with the pipe it is currently
> configured to use.
>
> None of the changed tests are part of BAT, so I'm not sure we care about a
> potential increase in execution time, though. But maybe add a note in the
> commit
> message for future reference?
It will, but it will also increase test coverage by running the test on each
crtc. If we only cared for the
test to run with a single combination, we should have just added break at the
end.
Good idea to put it in the commit message though, would be useful.
~Maarten
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
