[igt-dev] [v3] tests/kms_color: Fix CRC mismatch issues with ctm test

[Ville Syrjälä]

On Wed, Mar 27, 2019 at 10:52:48AM +0000, Shankar, Uma wrote:
> 
> 
> >On Wed, Mar 27, 2019 at 06:20:10AM +0000, Shankar, Uma wrote:
> >>
> >>
> >> >-----Original Message-----
> >> >From: Ville Syrjälä [mailto:ville.syrjala at linux.intel.com]
> >> >Sent: Tuesday, March 26, 2019 10:16 PM
> >> >To: Shankar, Uma <uma.shankar at intel.com>
> >> >Cc: igt-dev at lists.freedesktop.org; Syrjala, Ville
> >> ><ville.syrjala at intel.com>; Lankhorst, Maarten
> >> ><maarten.lankhorst at intel.com>
> >> >Subject: Re: [igt-dev] [v3] tests/kms_color: Fix CRC mismatch issues
> >> >with ctm test
> >> >
> >> >On Tue, Mar 26, 2019 at 10:31:01AM +0000, Shankar, Uma wrote:
> >> >>
> >> >>
> >> >> >-----Original Message-----
> >> >> >From: Shankar, Uma
> >> >> >Sent: Monday, March 25, 2019 2:11 PM
> >> >> >To: igt-dev at lists.freedesktop.org
> >> >> >Cc: Shankar, Uma <uma.shankar at intel.com>
> >> >> >Subject: [v3] tests/kms_color: Fix CRC mismatch issues with ctm
> >> >> >test
> >> >> >
> >> >> >Due to Gamma/Degamma limitation with precision (lack of exact 1.0
> >> >> >representation) due to ABI restriction, applying linear gamma
> >> >> >affects crc. This patch fixes the same by making ctm tests
> >> >> >independant of
> >> >gamma/degamma.
> >> >> >
> >> >> >v2: Disable degamma/gamma programming for ctm max test as it leads
> >> >> >to crc mimsmatch. Limiting it to this test case alone as other
> >> >> >tests need it to be enabled, hence not touching those scenarios.
> >> >> >
> >> >> >v3: Fixed a fumble with compilation.
> >> >>
> >> >> Hi Ville,
> >> >> Can I keep your RB on this version. We have a clean CI result available now.
> >> >
> >> >I'm confused. Why didn't disabling degamma+gamma work?
> >>
> >> We need to disable degamma/gamma and also make sure we don't set them
> >> again with linear values, which was being done currently. Hence,
> >> disabled it and later made sure it's kept at that.
> >
> >Who is turning it back on and why?
> 
> When we call disable from IGT, it passes a NULL LUT. In this case, default linear LUT is
> getting applied in kernel. I am not sure if we really disable the gamma/degamma fully in
> driver. It just works with linear values and remains enabled.

We would disable after https://patchwork.freedesktop.org/series/58137/

Hmm. Looks like one patch missing r-b. I should probably go respin that
with Matt's comment added.

-- 
Ville Syrjälä
Intel