Bug 15432

Summary:

X fails to start when connected to TV via DVI->HDMI when TV on alternate input.

Product:

xorg

Reporter:

Billy Macdonald <billymacdonald>

Component:

Driver/intel

Assignee:

Hong Liu <hong.liu>

Status:

RESOLVED DUPLICATE

QA Contact:

Xorg Project Team <xorg-team>

Severity:

normal

Priority:

medium

Keywords:

NEEDINFO

Version:

git

Hardware:

x86 (IA32)

OS:

Linux (All)

Whiteboard:

i915 platform:

i915 features:

Attachments:

Description	Flags
Xorg with untouched code where X fails	none
X log when X is able to start with code change. Shows that the driver does see the DVI connection and gets EDID info.	none
Xorg conf file	none
Xorg.0.log	none
force enable output -2nd	none

Description Billy Macdonald 2008-04-09 16:59:34 UTC

The situation I have run into is that when X restarts when the TV is switched to a different input the DVI connection is not connected and X fails to start.  I have not tested with the TV off yet.

However, a minor modification to the driver (which is not a fix, just a test from other tickets) allows X to start.  When X does start it is able to get EDID information from the TV.  The retrieval of EDID info tells me that part of the connection is seen by the driver.

I am currently using xf86-video-intel-2.2.99.902

Two Xorg logs are attached.  Xorg.0.log (unmodified code) and Xorg.0.log.Unknown (modified code to allow X to start, showing that it is able to get EDID info)

This may duplicate https://bugs.freedesktop.org/show_bug.cgi?id=14611 but in my case the display is connected just not switched to the proper input.

Thanks.

PS to get X to start I did the following since I know that the DVI is connected.
Changed XF86OutputStatusDisconnected to XF86OutputStatusUnknown on line 1098 of src/i830_sdvo.c

Comment 1 Billy Macdonald 2008-04-09 17:00:36 UTC

Created attachment 15790 [details]
Xorg with untouched code where X fails

Comment 2 Billy Macdonald 2008-04-09 17:01:31 UTC

Created attachment 15791 [details]
X log when X is able to start with code change.  Shows that the driver does see the DVI connection and gets EDID info.

Comment 3 Billy Macdonald 2008-04-09 17:02:35 UTC

Created attachment 15792 [details]
Xorg conf file

Comment 4 Hong Liu 2008-04-13 19:39:59 UTC

Would you please attach the xorg log with option modedebug turned on?
It should be the same problem as bug 14611.

When you switch the TV to other input, our driver can't detect any active output attached to SDVO tmds-1 output (using the hotplug detection mechanism but it seems that we can still get EDID from the DDC channel to your TV).

Thanks,
Hong

Comment 5 Hong Liu 2008-04-16 21:51:30 UTC

Please try the patch in bug 14611.

Comment 6 Billy Macdonald 2008-05-02 18:34:57 UTC

Created attachment 16317 [details]
Xorg.0.log

Logfile with ModeDebug set to true.

Comment 7 Billy Macdonald 2008-05-02 18:51:03 UTC

I tried the referenced patch and setting the enable option did not help.

Comment 8 Hong Liu 2008-05-03 18:52:45 UTC

Created attachment 16339 [details] [review]
force enable output -2nd

Would you please make sure that you are using the new X server (with patch applied)? I've tried you xorg.conf file with my patch, it works.

Attached patch is the same with the 1st, except I've added some debug messages.
This will make sure whether you are using the newly patched X server. Would you please try this patch and repost the xorg log?

Thanks,
Hong

Comment 9 Gordon Jin 2008-06-03 23:56:30 UTC

Billy, any feedback for comment#8? Thanks.

Comment 10 Michael Fu 2008-07-10 00:10:19 UTC


*** This bug has been marked as a duplicate of bug 14611 ***

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.