Bug 19370 - Blank laptop screen if DVI display plugged in on Xorg startup after "Fix connector table mapping for lvtma/kaleidoscope/uniphy" commit (d744dc01...)
Summary: Blank laptop screen if DVI display plugged in on Xorg startup after "Fix conn...
Status: RESOLVED INVALID
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/Radeon (show other bugs)
Version: 7.4 (2008.09)
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: xf86-video-ati maintainers
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-01-02 08:41 UTC by Tyson Whitehead
Modified: 2009-01-03 14:20 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

Attachments
xorg startup with previous commit driver (last working one) and DVI display attached (53.39 KB, text/x-log)
2009-01-02 08:43 UTC, Tyson Whitehead 		no flags Details
xorg startup with commit that broke it driver and DVI display attached (53.62 KB, text/x-log)
2009-01-02 08:45 UTC, Tyson Whitehead 		no flags Details
xorg startup with previous commit driver (last working one) (36.76 KB, text/x-log)
2009-01-02 08:46 UTC, Tyson Whitehead 		no flags Details
xorg startup with commit that broke it driver (37.45 KB, text/x-log)
2009-01-02 08:47 UTC, Tyson Whitehead 		no flags Details
xorg.conf for completeness (only specifies the radeon driver) (876 bytes, text/plain)
2009-01-02 09:06 UTC, Tyson Whitehead 		no flags Details
View All

Description Tyson Whitehead 2009-01-02 08:41:32 UTC 
I noticed my laptop's panel goes black/blank on Xorg startup if an external DVI display is plugged into the HDMI connect on the side of the laptop (via a HDMI to DVI cable) during Xorg startup.  Plugging it in after startup does not cause this problem.

Using git bisect shows the first commit this started happening at was "Fix connector table mapping for lvtma/kaleidoscope/uniphy" (d744dc013cf5ca3141c0fdf55e36d713fe9f53ba).  I'll attach a log file from Xorg starting up with the with the the external DVI display plugged in for this commit and the previous working one.  The difference between them is

tyson@whitehead2:~$ diff -u gooddvi.log baddvi.log 
--- gooddvi.log 2009-01-02 10:56:35.000000000 -0500
+++ baddvi.log  2009-01-02 10:53:02.000000000 -0500
@@ -11,7 +11,7 @@                                  
 Markers: (--) probed, (**) from config file, (==) default setting,
        (++) from command line, (!!) notice, (II) informational,   
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
-(++) Log file: "gooddvi.log", Time: Fri Jan  2 10:56:17 2009
+(++) Log file: "baddvi.log", Time: Fri Jan  2 10:52:44 2009
 (==) Using config file: "/etc/X11/xorg.conf"
 (==) ServerLayout "Default Layout"
 (**) |-->Screen "Default Screen" (0)
@@ -424,7 +424,7 @@
  Monitor   -- AUTO
  Connector -- LVDS
  DAC Type  -- None
- TMDS Type -- LVTMA
+ TMDS Type -- None
  DDC Type  -- 0xac0
 (II) RADEON(0): Port2:
  Monitor   -- AUTO
@@ -799,6 +799,9 @@
 Output 68 disable success
 (II) RADEON(0): ParseTable said: CD_SUCCESS
 (II) RADEON(0): Call to AtomBIOS Exec succeeded
+Output 23 disable success
+(II) RADEON(0): ParseTable said: CD_SUCCESS
+(II) RADEON(0): Call to AtomBIOS Exec succeeded
 Output 66 disable success
 (II) RADEON(0): ParseTable said: CD_SUCCESS
 (II) RADEON(0): Call to AtomBIOS Exec succeeded
@@ -943,6 +946,9 @@
 (II) RADEON(0): ParseTable said: CD_SUCCESS
 (II) RADEON(0): Call to AtomBIOS Exec succeeded
 Output 68 disable success
+(II) RADEON(0): ParseTable said: CD_SUCCESS
+(II) RADEON(0): Call to AtomBIOS Exec succeeded
+Output 23 disable success
 (II) RADEON(0): RandR 1.2 enabled, ignore the following RandR disabled message.
 (II) RADEON(0): ParseTable said: CD_SUCCESS
 (II) RADEON(0): Call to AtomBIOS Exec succeeded

I'll also attach logs for both driver versions starting up without the DVI display plugged in.
Comment 1 Tyson Whitehead 2009-01-02 08:43:18 UTC 
Created attachment 21631 [details]
xorg startup with previous commit driver (last working one) and DVI display attached
Comment 2 Tyson Whitehead 2009-01-02 08:45:31 UTC 
Created attachment 21632 [details]
xorg startup with commit that broke it driver and DVI display attached
Comment 3 Tyson Whitehead 2009-01-02 08:46:17 UTC 
Created attachment 21633 [details]
xorg startup with previous commit driver (last working one)
Comment 4 Tyson Whitehead 2009-01-02 08:47:09 UTC 
Created attachment 21634 [details]
xorg startup with commit that broke it driver
Comment 5 Tyson Whitehead 2009-01-02 09:06:50 UTC 
Created attachment 21635 [details]
xorg.conf for completeness (only specifies the radeon driver)
Comment 6 Tyson Whitehead 2009-01-03 14:20:19 UTC 
I've come to conclude that this is not a problem.  It seems the radeon driver disables the LVDS display (the local laptop screen) when it detects the DVI display plugged in.  What was tripping me up was that I didn't realize this as the DVI display does not work due to

http://bugs.freedesktop.org/show_bug.cgi?id=18564

If I go ahead and log in blind, start up a terminal, and do a "xrandr --output LVDS --auto" the local laptop display comes up fine (and, from the xrandr report, I also see that the radeon driver believes it has turned the DVI display on).

I'll close this bug report.

Cheers!  -Tyson

PS:  Actually, if I use the version prior to commit d744dc01... (i.e., the one I thought was working when I submitted this report) and check xrandr after logging in, I see that driver actually thinks that it has disabled the LVDS display despite the fact that it hasn't.

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.