Bug 114144 - Screen becomes blank when switching between X11 and console
Summary: Screen becomes blank when switching between X11 and console
Status: RESOLVED FIXED
Alias: None
Product: SUSE LINUX 10.0
Classification: openSUSE
Component: X.Org (show other bugs)
Version: Beta 3
Hardware: Other All
: P2 - High : Major
Target Milestone: ---
Assignee: Stefan Dirsch
QA Contact: Stefan Dirsch
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-08-30 15:16 UTC by Gernot Payer
Modified: 2019-07-23 10:00 UTC (History)
1 user (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
hwinfo --gfxcard (1019 bytes, text/plain)
2005-08-30 15:16 UTC, Gernot Payer
Details
Xorg.0.log (33.32 KB, text/plain)
2005-09-01 10:55 UTC, Gernot Payer
Details
Xorg.0.log.old (33.90 KB, text/plain)
2005-09-01 10:55 UTC, Gernot Payer
Details
The requested patch (3.24 KB, patch)
2005-09-06 10:58 UTC, Egbert Eich
Details | Diff
Patch on top of the last one (484 bytes, patch)
2005-09-08 10:12 UTC, Egbert Eich
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Gernot Payer 2005-08-30 15:16:15 UTC
When switching to the console (Ctrl+Alt+F1) and switching back to X11 again, the screen gets blank 
and whatever is done afterwards, stays blank.
This used to work perfectly woth earlier SUSE versions.
Comment 1 Gernot Payer 2005-08-30 15:16:51 UTC
Created attachment 48190 [details]
hwinfo --gfxcard
Comment 2 Stefan Dirsch 2005-08-30 15:18:54 UTC
Please try again with the new driver on Beta4.
Comment 3 Gernot Payer 2005-08-31 14:22:06 UTC
Same behaviour with beta4.

Suspicious messages from /var/log/messages:

Aug 31 16:06:42 vogon kdm[21128]: X server for display :0 terminated unexpectedly
Aug 31 16:06:42 vogon kernel: mtrr: 0x88000000,0x8000000 overlaps existing 0x88000000,0x80000
Aug 31 16:06:49 vogon kdm[21128]: X server for display :0 terminated unexpectedly
Aug 31 16:06:49 vogon kernel: mtrr: 0x88000000,0x8000000 overlaps existing 0x88000000,0x80000
(and so on)
Comment 4 Stefan Dirsch 2005-08-31 14:40:23 UTC
Egbert, any ideas?
Comment 5 Egbert Eich 2005-08-31 17:47:26 UTC
Yeah, the Xserver crashed. 
Works for me on an ICH7 (i945). 
I don't have an i845.
Comment 6 Stefan Dirsch 2005-08-31 18:50:26 UTC
At least we need the logfile /var/log/Xorg.0.log* 
Comment 7 Egbert Eich 2005-09-01 06:53:50 UTC
> At least we need the logfile /var/log/Xorg.0.log* 
I assume that I also need access to the machine. I know we had discussions about
this with QA in the past. But without access to the machine I cannot get a clue
where to look for a segfault. The log file doesn't contain this information.
Comment 8 Gernot Payer 2005-09-01 10:55:15 UTC
Created attachment 48429 [details]
Xorg.0.log
Comment 9 Gernot Payer 2005-09-01 10:55:32 UTC
Created attachment 48430 [details]
Xorg.0.log.old
Comment 10 Gernot Payer 2005-09-01 10:57:44 UTC
You can have access to the machine, it's vogon. But as I also use this machine as workstation, please 
tell me (by phone or email or whatever) when you intend to test on this machine, so I can logout 
beforehand.
Comment 11 Egbert Eich 2005-09-01 13:34:53 UTC
I will call you within the next few minutes. The Xserver appearantly didn't crash.
Comment 12 Stefan Dirsch 2005-09-03 16:02:19 UTC
Meanwhile a fixed driver is available, a patch is *not*. :-( I think this is 
sth. for RC2. 
Comment 13 Egbert Eich 2005-09-05 08:17:43 UTC
What do you mean by 'fixed driver'? This has been available as a test version
for quite a while. I don't see a patch in CVS yet. I notified Alan about the
success and will ping him again later today if nothing has happened. 
Comment 14 Stefan Dirsch 2005-09-05 08:26:56 UTC
Ah. You've meant the binary-only test version by Alan. I thought you did build
the driver.
Comment 15 Egbert Eich 2005-09-06 10:58:02 UTC
Created attachment 48902 [details]
The requested patch

Attached is Alan's patch which fixes the problem.
This contains some fixes for gamma correction also. Really only the second last
hunk is related to the reported problem. Stefan, it's up to you what you do.
Comment 16 Stefan Dirsch 2005-09-06 15:34:53 UTC
package with (complete) patch submitted now.  
Comment 17 Stefan Dirsch 2005-09-07 02:41:20 UTC
fixed for RC2. 
Comment 18 Egbert Eich 2005-09-08 10:10:28 UTC
We need to add one more little thing.
Comment 19 Egbert Eich 2005-09-08 10:12:50 UTC
Created attachment 49172 [details]
Patch on top of the last one

This patch should prevent undesireable side effects when accel is disabled.
Comment 20 Stefan Dirsch 2005-09-08 10:20:53 UTC
I've seen this, but I wonder about the other location in 
I830BIOSSwitchMode(). Isn't this the same problem?

[...]
   /* Sync the engine before mode switch */
   if (!pI830->noAccel)
      (*pI830->AccelInfoRec->Sync)(pScrn);


Comment 21 Egbert Eich 2005-09-08 10:42:59 UTC
Good catch! Thanks!
Comment 22 Stefan Dirsch 2005-09-08 12:59:31 UTC
fixed for RC2.
Comment 23 Swamp Workflow Management 2019-07-16 12:10:07 UTC
This is an autogenerated message for OBS integration:
This bug (114144) was mentioned in
https://build.opensuse.org/request/show/715662 Factory / clementine
Comment 24 Swamp Workflow Management 2019-07-23 10:00:15 UTC
This is an autogenerated message for OBS integration:
This bug (114144) was mentioned in
https://build.opensuse.org/request/show/717812 Factory / clementine