Re: [PATCH i-g-t] tests/intel/kms:cdclk: skip cdclk tests for bmg

[Matt Roper <matthew.d.roper@intel.com>]

On Fri, Aug 30, 2024 at 02:27:05PM +0530, Sharma, Swati2 wrote:
> Hi Vinod,
> 
> Please fix the subject.
> ++Matt for his feedback

Yeah, due to Wa_15015413771 there's only a single possible cdclk value
on Xe2_HPD, so we won't be able to trigger any cdclk changes in igt
tests.  It might be worth actually putting that workaround number in the
comment to make it clear why BMG is special, but:

Acked-by: Matt Roper <matthew.d.roper@intel.com>

Matt

> 
> On 13-Aug-24 4:37 PM, Vinod Govindapillai wrote:
> > As there is only one cdclk option in BMG, these cdclk tests
> > which look for change in cdclk frequency when changing
> > between high and low resolution mode will fail. So skip
> > these tests for bmg
> > 
> > Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
> > ---
> >   tests/intel/kms_cdclk.c | 3 +++
> >   1 file changed, 3 insertions(+)
> > 
> > diff --git a/tests/intel/kms_cdclk.c b/tests/intel/kms_cdclk.c
> > index 6fe6e6c3f..24a0c4547 100644
> > --- a/tests/intel/kms_cdclk.c
> > +++ b/tests/intel/kms_cdclk.c
> > @@ -402,6 +402,9 @@ igt_main
> >   			      "Hardware doesn't support crawling/squashing.\n");
> >   		igt_display_require(&data.display, data.drm_fd);
> >   		igt_display_require_output(&data.display);
> > +
> > +		/* Only one cdclk option in bmg. Skip all these tests for BMG? */
> > +		igt_require(!IS_BATTLEMAGE(data.devid));
> >   	}
> >   	igt_describe("Plane scaling test to validate cdclk frequency change.");

-- 
Matt Roper
Graphics Software Engineer
Linux GPU Platform Enablement
Intel Corporation