ACK/Cmnt: [SRU] [F/G/Unstable/OEM-5.6] [PATCH 0/3] Enable brightness control on HP DreamColor panel

Kai-Heng Feng kai.heng.feng at canonical.com
Thu Oct 8 13:30:18 UTC 2020



> On Oct 8, 2020, at 17:27, Stefan Bader <stefan.bader at canonical.com> wrote:
> 
> On 08.10.20 11:15, Kai-Heng Feng wrote:
>> 
>> 
>>> On Oct 8, 2020, at 17:07, Stefan Bader <stefan.bader at canonical.com> wrote:
>>> 
>>> On 07.10.20 13:53, Kai-Heng Feng wrote:
>>>> BugLink: https://bugs.launchpad.net/bugs/1898865
>>>> 
>>>> [Impact]
>>>> Brightness on HP DreamColor panel, which can be found on new HP Zbook
>>>> Studio, cannot be changed.
>>>> 
>>>> [Fix]
>>>> Let DRM know the panel should use DPCD instead of PWM to control
>>>> backlight.
>>>> 
>>>> [Test]
>>>> With the patch applied, we can change the brightness on HP Zbook Studio.
>>>> 
>>>> [Regression Potential]
>>>> If there's any panel in the quirk list depends on checking brightness
>>>> control capability then this change will have an impact. However it's
>>>> rather unlikely because it defeats the purpose of the quirk list.
>>>> 
>>>> Kai-Heng Feng (2):
>>>> UBUNTU: SAUCE: drm/i915/dpcd_bl: Skip testing control capability with
>>>>   force DPCD quirk
>>>> UBUNTU: SAUCE: drm/dp: HP DreamColor panel brigntness fix
>>>> 
>>>> Lyude Paul (1):
>>>> drm/i915/dpcd_bl: Unbreak enable_dpcd_backlight modparam
>>>> 
>>>> drivers/gpu/drm/drm_dp_helper.c                       | 1 +
>>>> drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c | 9 ++++++---
>>>> 2 files changed, 7 insertions(+), 3 deletions(-)
>>>> 
>>> For Focal I am not really happy to ack because it seems to add a lot of new
>>> SAUCE which could mean it is not accepted upstream or is rather new that there
>>> was not much baking time. All things that do not really comply with the concept
>>> of a stable release.
>> 
>> I hope users who want to stick to GA kernel can get fixes too.
> 
> Yes, of course. This was not so much about the change itself but more about the
> timing. I understand this causes more work but I think a staged approach (first
> oem and devel and later stable) would be better. In this case it is probably too
> simple to be a risk but as you say it might become a different approach upstream
> and so we should avoid back and forth.

Ok, will see if I can manage to do that next time...

Kai-Heng

> 
> -Stefan
> 
>> 
>>> 
>>> Though the first patch now is upstream:
>>> 
>>> commit d082119f4277ff4a63e44d293864aa9f2112b217
>>> Author: Lyude Paul <lyude at redhat.com>
>>> Date:   Mon Apr 13 17:44:06 2020 -0400
>>>   drm/i915/dpcd_bl: Unbreak enable_dpcd_backlight modparam
>>> 
>>> So the following should be put before your sign-off when things get committed:
>>> 
>>> (cherry picked from commit d7fb38ae36a2dc97924b075ad1d1a88792777ea9)
>>> Signed-off-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
>>> 
>>> (cherry picked from commit d082119f4277ff4a63e44d293864aa9f2112b217)
>>> Signed-off-by: ...
>>> 
>>> The other two patches, at least they appear harmless enough, are you planning on
>>> upstreaming those?
>> 
>> Yes, and it's under review. Will send another SRU if upstream version is different to this one.
>> 
>> Kai-Heng
>> 
>>> 
>>> For now, as this looks isolated enough...
>>> 
>>> Acked-by: Stefan Bader <stefan.bader at canonical.com>




More information about the kernel-team mailing list