[PATCH 1/2] Revert "drm/i915/dp: Reject HBR3 when sink doesn't support TPS4"

[Jani Nikula]

On Tue, 24 Jun 2025, Ville Syrjälä <ville.syrjala at linux.intel.com> wrote:
> On Tue, Jun 24, 2025 at 10:10:53AM +0530, Nautiyal, Ankit K wrote:
>> 
>> On 6/23/2025 8:12 PM, Ville Syrjälä wrote:
>> > On Fri, Jun 20, 2025 at 06:14:16PM +0530, Ankit Nautiyal wrote:
>> >> This reverts commit 584cf613c24a4250d9be4819efc841aa2624d5b6.
>> >> Commit 584cf613c24a ("drm/i915/dp: Reject HBR3 when sink doesn't support
>> >> TPS4") introduced a blanket rejection of HBR3 link rate when the sink does
>> >> not support TPS4. While this was intended to address instability observed
>> >> on certain eDP panels [1], the TPS4 requirement is only mandated for DPRX
>> >> and not for eDPRX.
>> > I see no exception given for eDP regarding this rule. The only exception
>> > allowed is that eDP can say DPCD_REV=1.4 + TPS4_SUPPORTED=0. So I still
>> > claim that these eDP sinks are violating the spec.
>> 
>> Hmm.. Yes the spec allows eDP sinks to report DPCD_REV=1.4 and 
>> TPS4_SUPPORTED, so perhaps eDPs claiming HBR3 with DPCD rev other than 
>> rev 1.4 and not supporting TPS4 are indeed violating the spec.
>> 
>> Would it make sense to add a condition that checks for DPCD_REV=1.4.
>> 
>> Specifically:
>> 
>> if DPCD_REV=1.4 and TPS4_SUPPORTED = 0, then do not prune the HBR3 rate?
>> 
>> Or otherway if DPCD_REV!=1.4 and TPS4_SUPPORTER = 0, prune the HBR3 rate
>> 
>> This way the patch need not be reverted, but modified to address 
>> instability issues for eDP panels that are not aligned with the spec.
>> 
>> That said, the gitlab issue#5969 [1] will still need another solution 
>> since it seems to have DPCD rev 14 as per logs:
>> 
>> DPCD: 14 1e 44 41 00 00 01 80 02 00 02 00 00 0b 80
>
> I think we do need the quirk. But the commit message should IMO be
> adjusted so that it doesn't claim that these panels are 100% legal.

Right. But are you okay with reverting 584cf613c24a ("drm/i915/dp:
Reject HBR3 when sink doesn't support TPS4") i.e. allowing HBR3
*without* TPS4 by default, and quirking the panel that can't handle
HBR3?

No matter what the spec says, this seems to be common. More regressing
systems are cropping up, e.g. [1].


BR,
Jani.


[1] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14517




-- 
Jani Nikula, Intel