Let's try this again.  I see this:

1) xrandr thinks the monitor is attached and enabled.
2) /sys/class/drm/card0/card0-VGA-1 says the monitor is connected, but NOT 
enabled.
3) intel_reg_dumper shows that pipeB is disabled. (LVDS is on pipeA and is 
enabled)

My kernel debug code shows that when the system is resuming, the first
couple of times it checks to see if the monitor is connected in
output_poll_execute(), intel_crt_detect() says it isn't.  The first time
it checks, it changes the status from 1 (connected) to 2 (not
connected).  This ends up causing drm_helper_disable_unused_functions()
to  remove the encoder from the connector, the crtc from the encoder,
and it disables the CRTC.  The driver does generate udev hotplug events.
This does not seem to cause the X server to sync with the driver and get
the new state of the output though.  Turning the monitor off and on
again via xrandr brings the monitor back, as does going down into a
different VT and back.

I'm going to try to build the X server and intel driver to see if I
can't debug it further.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/911045

Title:
  [GM45] xrandr does not enable external display after suspend/resume
  cycle

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/911045/+subscriptions

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to