[Bug 109522] drm:intel_dp_start_link_train [i915] *ERROR* failed to enable link training

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=109522

--- Comment #20 from howaboutsynergy at pm.me ---
Created attachment 144603
  --> https://bugs.freedesktop.org/attachment.cgi?id=144603&action=edit
dmesg with drm.debug=0x1e on both just console and within X

only while inside X (ie. startx) `drm.debug=0x1e` constantly generates dmesg
spam

I'm attaching full dmesg log (ring buffer didn't wrap around and thus no
overwrites) with drm.debug=0x1e

First I tried in console only (tty1) and the issue doesn't happen there, so
this is good:
[    6.032810] [drm:intel_dp_set_link_train [i915]] *ERROR* ret=5 len=5
however on console/tty1 , it requires running `monon` in order to unblank!
Pressing enter didn't do anything, so between these two messages I couldn't see
what I was typing:
[  222.564063] snd_hda_intel 0000:00:1f.3: power state changed by ACPI to D3hot
[  229.562650] user[1383]: '/home/user/bin/monon'(on /dev/tty1) is executing:
'sleep 1'

I guess at
[  395.795237] [drm:drm_atomic_state_init [drm]] Allocated atomic state
0000000023bfb3ad
I ran `startx` and that's when the spam started.

At about
[  415.457985] [drm:intel_dp_set_link_train [i915]] *ERROR* ret=5 len=5
I ran a normal monoff then turned on monitor test, so this wouldn't have caused
the -110 (link training issue) - I actually forgot (sort of) the steps I was
supposed to do here, so ignore this.

I did it right at:
[  443.388949] [drm:intel_dp_set_link_train [i915]] *ERROR* ret=-110 len=5

so between these two I waited: after pressing the key(433) to then turning on
monitor(441, 8 seconds later):
[  433.005944] [drm:__drm_atomic_state_free [drm]] Freeing atomic state
0000000076021904
[  441.745985] [drm:drm_atomic_state_init [drm]] Allocated atomic state
0000000076021904

and like 2 sec later is the -110 mentioned above.

Looks like it's those:
[  441.820989] [drm:intel_dp_aux_xfer [i915]] dp_aux_ch timeout status
0x7d4003ff
that Manasi mentions in https://bugs.freedesktop.org/show_bug.cgi?id=105998#c5
however since I've tried with `i915.disable_power_well=0` in my previous
comment(albeit without the `drm.debug=0x1e` - does it matter though?!) and it
had no effect, I'm out of ideas..

-------
monon is essentially this:
#!/bin/bash

roon() {
  local ec
  logger "'$0'(on $(tty)) is executing: '$*'"
  "$@" ; ec="$?"
  if test "$ec" != "0"; then
    logger "'$0' failed(ec='$ec') executing: '$*'"
  fi
  return "$ec"
}


roon sleep 1
roon xset dpms force on
roon setterm --blank poke

#to also get back from blanking
roon xset s reset

-------
monoff is this:
#!/bin/bash

roon() {                                                                        
  local ec
  logger "'$0'(on $(tty)) is executing: '$*'"
  "$@" ; ec="$?"
  if test "$ec" != "0"; then
    logger "'$0' failed(ec='$ec') executing: '$*'"
  fi
  return "$ec"
}

roon sleep 1
roon xset dpms force suspend
roon sleep 1
roon xset dpms force off
roon sleep 1
roon xset dpms force standby
roon setterm --blank force

-------

-- 
You are receiving this mail because:
You are the assignee for the bug.
You are on the CC list for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190621/9f4d7754/attachment.html>