Created attachment 112481 [details] full dmesg during boot Hi, On a HP EliteBook 840 with a Intel HD 4400, the VGA output only works once. Steps to reproduce: 1. boot 2. plug a project to the VGA output => works fine 3. unplug it 4. plug it again => doesn't work (shown as disconnected in xrandr) Software versions & hardware: Linux grr 3.18.0-trunk-amd64 #1 SMP Debian 3.18-1~exp1 (2014-12-11) x86_64 GNU/Linux $ pkg-config --modversion libdrm ==> 2.4.58 (II) Module intel: vendor="X.Org Foundation" compiled for 1.16.2.901, module version = 2.99.917 (--) intel(0): Integrated Graphics Chipset: Intel(R) HD Graphics 4400 All attachments with drm.debug=0x06 Thanks!
Created attachment 112482 [details] dmesg output when connecting projector to the VGA output
Created attachment 112483 [details] xrandr output with the VGA (DP2) connected
Created attachment 112484 [details] dmesg when disconnecting the VGA output
Created attachment 112485 [details] dmesg when reconnecting VGA. DP2 stays "disconnected" in xrandr.
Note that it could be the same as bug 87736, but there wasn't enough information in that bug to ensure that it's really the same one.
Yes, it is the same. It is a issue with this dock firmware. *** This bug has been marked as a duplicate of bug 87736 ***
oh, sorry, I misread bug 87736. In my case, the problem doesn't involve docking at all. (I'm booting out of the docking station, and using the laptop's VGA port)
This is interesting, and the first I've seen. HP has picked up HSW ULT which does *not* have VGA, and has (apparently, per reference to "laptop's VGA port" in comment #7) embedded a DP-to-VGA converter in the box. Unfortunately, the converters, or dongles, are still quite unreliable and frankly I'm not quite sure why. Please try v3.19-rc1 or later, or try cherry-picking commit 19a93f042fc241ecdf98543cedfe7c171f8cdf53 Author: Dave Airlie <airlied@redhat.com> Date: Wed Nov 26 13:13:09 2014 +1000 drm/dp: retry AUX transactions 32 times (v1.1) to v3.18. The above commit is marked cc: stable, so it should eventually get backported to stable kernels. It should give the DP downstream a little more slack in responding; the legacy analog things may need that.
(In reply to Jani Nikula from comment #8) > Please try v3.19-rc1 or later, or try cherry-picking > > commit 19a93f042fc241ecdf98543cedfe7c171f8cdf53 > Author: Dave Airlie <airlied@redhat.com> > Date: Wed Nov 26 13:13:09 2014 +1000 > > drm/dp: retry AUX transactions 32 times (v1.1) > > to v3.18. The above commit is marked cc: stable, so it should eventually get > backported to stable kernels. It should give the DP downstream a little more > slack in responding; the legacy analog things may need that. Hi, I tried with 3.19.0-rc5, and the problem still occurs. I'm adding the dmesg output with drm.debug=0x06 Some more information: There are two Ubuntu bug reports where users complain about the same problem (or at least some of the posters do -- others are commenting about seemingly unrelated issues) https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1300914 https://bugs.launchpad.net/ubuntu/+source/libdrm/+bug/1411552 Comment 13 describes exactly the behaviour I'm seeing: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1300914/comments/13 Changing window manager (from GNOME to OpenBox) changes the behaviour: - with OpenBox, switching to a TTY and then switching back to X "fixes" the problem. - with GNOME, it does not. However, I confirm that suspending/resuming "fixes" the problem (even with GNOME).
Created attachment 112663 [details] dmesg with 3.19.0-rc5
Please give this patch a try http://mid.gmane.org/1422373429-10137-1-git-send-email-simon.farnsworth@onelan.co.uk
(In reply to Jani Nikula from comment #11) > Please give this patch a try > > http://mid.gmane.org/1422373429-10137-1-git-send-email-simon. > farnsworth@onelan.co.uk It does not change anything for me.
Note that if you are running into this, an easy way out is to just buy an external DisplayPort to VGA adapter for 10-15 EUR.
We've since fixed plenty of DP aux/i2c-over-aux issues, so presumed fixed upstream. Please reopen if the problem persists with latest kernels.
> On a HP EliteBook 840 with a Intel HD 4400, the VGA output only works once. > > Steps to reproduce: > 1. boot > 2. plug a project to the VGA output => works fine > 3. unplug it > 4. plug it again => doesn't work (shown as disconnected in xrandr) > > Software versions & hardware: > Linux grr 3.18.0-trunk-amd64 #1 SMP Debian 3.18-1~exp1 (2014-12-11) x86_64 > GNU/Linux > > $ pkg-config --modversion libdrm ==> 2.4.58 > > (II) Module intel: vendor="X.Org Foundation" > compiled for 1.16.2.901, module version = 2.99.917 > (--) intel(0): Integrated Graphics Chipset: Intel(R) HD Graphics 4400 Hi, I confirm that the above still happens with: Linux grr 4.2.0-1-amd64 #1 SMP Debian 4.2.3-2 (2015-10-14) x86_64 GNU/Linux $ pkg-config --modversion libdrm 2.4.64 [ 12.284] (II) Module intel: vendor="X.Org Foundation" [ 12.284] compiled for 1.17.2, module version = 2.99.917 [ 12.284] Module class: X.Org Video Driver [ 12.284] ABI class: X.Org Video Driver, version 19.0 So, reopening as requested
Please try v4.3-rc7.
(In reply to Jani Nikula from comment #16) > Please try v4.3-rc7. same with 4.3-rc7
Seems we have neglected updating this bug more than a little; apologies. We have fixed a number of related issues, however. Is this still a problem with current kernels?
(In reply to Jani Nikula from comment #18) > Seems we have neglected updating this bug more than a little; apologies. We > have fixed a number of related issues, however. Is this still a problem with > current kernels? Yes (with 4.7 at least)
How about current as in v4.9-rc3 or so? ;)
actually, it seems that this was fixed in 4.7.0. I did not notice because I'm so used to working-around this bug ;) Closing.
(In reply to Lucas Nussbaum from comment #21) > actually, it seems that this was fixed in 4.7.0. I did not notice because > I'm so used to working-around this bug ;) Closing. Closing as fixed then
(In reply to Lucas Nussbaum from comment #21) > actually, it seems that this was fixed in 4.7.0. I did not notice because > I'm so used to working-around this bug ;) Closing. Awesome, thanks for the follow-up! \o/
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.