[igt-dev] [PATCH i-g-t 2/2] tests/kms_vrr: Extend to cover full VRR range

[Petri Latvala]

On Tue, Mar 02, 2021 at 07:06:15AM +0200, Modem, Bhanuprakash wrote:
> > -----Original Message-----
> > From: Navare, Manasi <manasi.d.navare at intel.com>
> > Sent: Tuesday, March 2, 2021 5:10 AM
> > To: Modem, Bhanuprakash <bhanuprakash.modem at intel.com>
> > Cc: igt-dev at lists.freedesktop.org; Ville Syrjälä
> > <ville.syrjala at linux.intel.com>; Nicholas Kazlauskas
> > <nicholas.kazlauskas at amd.com>; Latvala, Petri <petri.latvala at intel.com>
> > Subject: Re: [PATCH i-g-t 2/2] tests/kms_vrr: Extend to cover full VRR
> > range
> >
> > Looks good to me.
> > While at it, lets also remove the test for refresh rates less than vmin
> > since
> > in that case the frame repetition is expected and we wont get the expected
> > target rate
> To validate the flipline mode:
> * Generate the Push before the decision boundary (vmin by default), means we
>   need to request flips with the refresh rate < vmin. Expected refresh rate
>   would be vmin because vblank termination starts at decision boundary.
> * Generate the Push before the decision boundary, means we need to request
>   flips with any refresh rate b/w vmin & vmax. Expected refresh rate should
>   match with the requested rate because vblank termination starts immediately
> 
> If we remove the test for refresh rates less than vmin, subtest "flipline"
> is not required anymore. Subtest "flip-basic" is already covering the vrr
> range (running test with midpoint of vrr range)
> 
> Please suggest to proceed further.

For the record, this has been discussed offline and the verdict is
that the patch as-is is fine and future work can happen
separately. Thus, merged now, thanks.


-- 
Petri Latvala