[Bug 101144] [BAT][SKL] *ERROR* failed to enable link training

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=101144

--- Comment #26 from Manasi <manasi.d.navare at intel.com> ---
Hi Marta,

After looking through the dmesg logs, looks like in the modeset in
intel_ddi_pre_enable_dp(), after the edp panel is powered on, it gets two
/three HPD pulses after which it shows AUX ch timeouts. That could be because
the panel is not powered on or eDP connector is disconnected. 

The attached patch adds debug print for checking the connector status here and
also powers up panel again just before link training.

Could you please test it out and see what happens and attach the debug log?

Manasi



(In reply to Marta Löfstedt from comment #23)
> (In reply to Manasi from comment #22)
> > (In reply to Jani Nikula from comment #17)
> > > (In reply to Manasi from comment #14)
> > > > I am trying to find out at what point can we first check for dp aux
> > > > transactions and wait longer to try again?
> > > 
> > > Can you explain how we'd end up in link training if native aux was failing
> > > altogether? It must have worked for some time before link training, right?
> > > 
> > > Ville's explanation would cover this as well.
> > 
> > Actually it fails inside the function
> > intel_dp_link_training_clock_reciovery() when it tries to write the TP1 and
> > errors out saying "Failed to enable link training".
> > Yes but it worked fine for DPCD reads over AUX, so that's why I am confused
> > as to where we need to add the delay for the AUX reads?
> > 
> > Also, shouldn't the spec for that specific panel or the eDP spec talk about
> > how much delay should be given for the AUX read/writes?
> > Any thoughts on double checking this in the spec?
> > 
> > Manasi
> 
> Manasi, I believe that one theory is that device have actually become
> broken/unstable and that we should try to prove this somehow.
> 
> If you or Ville has some ideas on how to prove/disprove this, could you try
> this on the try-bot, or provide a patch that I can run on the machine early
> in the morning when the CI is usually idle.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20170614/7de92527/attachment.html>