Re: [igt-dev] [PATCH i-g-t] tests/kms_display_modes: Skip the negative test on MST Hub

["Lisovskiy, Stanislav" <stanislav.lisovskiy@intel.com>]

On Wed, Aug 23, 2023 at 12:07:52PM +0530, Modem, Bhanuprakash wrote:
> 
> 
> On Fri-18-08-2023 02:14 pm, Lisovskiy, Stanislav wrote:
> > On Fri, Jul 07, 2023 at 09:07:20PM +0530, Thasleem, Mohammed wrote:
> > > Skip the test on which MST hub which support bandwidth for 2k-4k
> > > and 4k-4k multi displays.
> > > 
> > > Signed-off-by: Thasleem, Mohammed <mohammed.thasleem@intel.com>
> > 
> > Okay I missed the MST check in the begining of the main function, so
> > 
> > Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
> > 
> > > ---
> > >   tests/kms_display_modes.c | 1 +
> > >   1 file changed, 1 insertion(+)
> > > 
> > > diff --git a/tests/kms_display_modes.c b/tests/kms_display_modes.c
> > > index 4bebb25e..8e117695 100644
> > > --- a/tests/kms_display_modes.c
> > > +++ b/tests/kms_display_modes.c
> > > @@ -310,6 +310,7 @@ static void run_extendedmode_negative(data_t *data, int pipe1, int pipe2)
> > >   	igt_require(i915_pipe_output_combo_valid(display));
> > >   	ret = igt_display_try_commit2(display, COMMIT_ATOMIC);
> > > +	igt_skip_on_f(ret == 0, "MST Hub supports bandwidth on multi display\n");
> 
> NACK
> 
> Failure reason can be anything, you need to check for the precise error
> code.

From my point of view, nack is for another reason: we can't skip the negative test case,
if it just happens to potentially pass.
Scenarios where it might pass and we dont consider that as a problem should
be filtered out by using some other checks to distinguish those scenarios more
explicitly.
Otherwise here we will get the situation that this negative test case always succeeds,
because we skip it always whenever it doesn't.

Stan

> 
> - Bhanu
> 
> > >   	igt_assert(ret != 0 && errno == ENOSPC);
> > >   }
> > > -- 
> > > 2.34.1
> > >