[igt-dev] [PATCH i-g-t v1 2/3] tests/kms_universal_plane: Use igt_get_assigned_primary to get the primary plane

[Rob Clark]

On Mon, May 9, 2022 at 9:30 AM Jessica Zhang <quic_jesszhan at quicinc.com> wrote:
>
>
>
> On 5/9/2022 2:10 AM, Petri Latvala wrote:
> > On Fri, May 06, 2022 at 11:14:31AM -0700, Jessica Zhang wrote:
> >> Currently, pageflip_test_pipe will use igt_output_get_plane_type to get
> >> the primary plane of the pipe and assumes that the primary plane marked
> >> by the driver will also be used by drmModePageFlip to perform the page flip.
> >>
> >> This is not necessarily true in cases where there are multiple possible
> >> primary planes.
> >>
> >> To fix this, let's use igt_get_assigned_primary to get the primary plane
> >> instead.
> >>
> >> Signed-off-by: Jessica Zhang <quic_jesszhan at quicinc.com>
> >> ---
> >>   tests/kms_universal_plane.c | 2 +-
> >>   1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/tests/kms_universal_plane.c b/tests/kms_universal_plane.c
> >> index 3cb6d704a6ef..7bd2c667b552 100644
> >> --- a/tests/kms_universal_plane.c
> >> +++ b/tests/kms_universal_plane.c
> >> @@ -483,7 +483,7 @@ pageflip_test_pipe(data_t *data, enum pipe pipe, igt_output_t *output)
> >>
> >>      pageflip_test_init(&test, output, pipe);
> >>
> >> -    primary = igt_output_get_plane_type(output, DRM_PLANE_TYPE_PRIMARY);
> >> +    primary = igt_get_assigned_primary(output, &data->display.pipes[pipe]);
> >
> > https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7059/shard-glk6/igt@kms_rotation_crc@primary-rotation-270.html
> >
> > The setcrtc call is getting ENOSPC there. Doing that unconditionally
> > is a no-go, and needs to be handled somehow cleanly.
> >
> > Check in igt_get_assigned_primary that there's multiple primary
> > planes? Check that in init and store that info in igt_output_t or
> > igt_pipe_t? Check for msm device specifically (I don't like this, but
> > it's the option if all else is unclean)?
>
> I see... The best option might be to check during init then storing it
> in igt_pipe_t.

Perhaps check if the CRTC is already enabled before doing the SETCRTC?
 I guess the test that is hitting the failure already has the CRTC
enabled?

BR,
-R

> Thanks,
>
> Jessica Zhang
>
> >
> >
> > --
> > Petri Latvala