APPLIED[G/H]/cmnt: [SRU] [G/H/Unstable/OEM-5.10] [PATCH 0/2] Fix screen flickering on some panels

[Kleber Souza]

On 06.05.21 17:44, Kai-Heng Feng wrote:
> BugLink: https://bugs.launchpad.net/bugs/1925685
> 
> [Impact]
> We are having screen flickering and glitching issues with some of our
> hardware upon wake from suspend. We have worked with our ODM to figure
> out a fix for this. They had reported there were no issues when running
> Windows on the same hardware. As such, we tested, and found that this
> issue first began as a regression in kernel 5.0 (did not exist in kernel
> 4.18 per our tests).
> 
> We identified a patch via
> https://gitlab.freedesktop.org/drm/intel/-/issues/272 in which this has
> been solved. Per our testing by recompiling the kernel, this completely
> fixes our hardware issue. As of 4-21, this fix has been merged into the
> Linux kernel per Timo Aaltonen, and will be included in the kernel in
> the 5.13 release.
> 
> Our hardware ships with Kubuntu 20.04 LTS, which means the 5.13 kernel
> will not see these machines, at least in the forseeable future. As such,
> we request that this patch be added to the kernel. Per our testing, this
> patch is straightforward to add and works flawlessly.
> 
> The patch that works best was created in collaboration with Kai-Heng
> Feng, is attached to this bug report, and can be found here:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1925685/+attachment/5491421/+files/drm-intel-both.patch
> 
> Until this is fixed, we have a whole batch of hardware we cannot sell,
> so, as an OEM, this is a high priority.
> 
> [Test Plan]
> * Install Kubuntu on an affected Kubuntu Focus M2 computer (not in the
>    hands of customers at this time)
> 
> * Notice screen glitching/flickering after wake-from-suspend.
> 
> * Install kernel compiled with the attached patch. Boot from patched
>    kernel.
> 
> * Suspend laptop, resume. No glitches.
> 
> [Where problems could occur]
> Although this solves the problem for us and has been confirmed to
> work with the Intel developers that developed this patch, it could
> cause another display type to have problems. This is unlikely as it
> has been vetted by Intel and by the mainline Linux kernel
> developers.
> 
> Kai-Heng Feng (1):
>    UBUNTU: SAUCE: drm/i915/dp: Use slow and wide link training for
>      everything
> 
> Ville Syrjälä (1):
>    drm/i915: Try to use fast+narrow link on eDP again and fall back to
>      the old max strategy on failure
> 
>   .../drm/i915/display/intel_display_types.h    |  1 +
>   drivers/gpu/drm/i915/display/intel_dp.c       | 21 +++++++++++++------
>   2 files changed, 16 insertions(+), 6 deletions(-)
> 


Applied to groovy:linux and hirsute:linux, with the changes to patch 2/2.

Thanks,
Kleber