Bug 12297 - [855GM] hang during VT switch
Summary: [855GM] hang during VT switch
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: git
Hardware: Other Linux (All)
: medium critical
Assignee: Jesse Barnes
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 13027
  Show dependency treegraph
 
Reported: 2007-09-05 21:39 UTC by CIJOML CIJOMLovic CIJOMLov
Modified: 2007-11-08 10:04 UTC (History)
4 users (show)

See Also:
i915 platform:
i915 features:


Attachments
Fix / workaround which helped many Ubuntu users seeing this bug (1.37 KB, patch)
2007-11-01 17:02 UTC, Peter Clifton
no flags Details | Splinter Review

Description CIJOML CIJOMLovic CIJOMLov 2007-09-05 21:39:58 UTC
Hi,

I use git 20070901 version and when CTRL+ALT+F1 switching to console or ALT+F& back to xorg, usually with this version I experience computer freeze. This can be tested very easy.

00:02.0 VGA compatible controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)
00:02.1 Display controller: Intel Corporation 82852/855GM Integrated Graphics Device (rev 02)

kernel vanilla 2.6.22.5, distro Debian testing

I have never seen this error with old version (i810)
Comment 1 CIJOML CIJOMLovic CIJOMLov 2007-09-17 15:47:10 UTC
with git 20070917 it works better, but still experiencing this crash in second or third switch X,console,X, console and so on...
Comment 2 CIJOML CIJOMLovic CIJOMLov 2007-10-23 07:56:28 UTC
Hi Eric,

any news about this problem?? It is really annoying! :(

Thanks a lot for reply
Comment 3 Eric Anholt 2007-10-23 10:12:10 UTC
If I thought your bug was fixed, I would have closed it.  If you wonder if it's been fixed without someone noticing it to close it, keep testing master.
Comment 4 Jesse Barnes 2007-10-31 12:20:49 UTC
There were some fixes for VT switch crashes for 8xx chips that went into master not long ago.  Can you test again and see if your problem still exists?
Comment 5 CIJOML CIJOMLovic CIJOMLov 2007-10-31 12:52:16 UTC
Hi, 

My version is pretty new:

drwxr-sr-x  6 root    src           4096 2007-10-26 18:45 xf86-video-intel
Comment 6 CIJOML CIJOMLovic CIJOMLov 2007-10-31 12:54:18 UTC
Hi,

my version is pretty new:

cijoml@notas:~/work/lingea$ ls -la /usr/src| grep xf86-video-intel
drwxr-sr-x  6 root    src           4096 2007-10-26 18:45 xf86-video-intel
Comment 7 Peter Clifton 2007-11-01 17:02:51 UTC
Created attachment 12305 [details] [review]
Fix / workaround which helped many Ubuntu users seeing this bug

This might not be the "right" thing to do, but it seems that people hitting the bug on VT-switching were seeing the crash when programming the PIPE[AB]CONF registers. It looked like the registers might be keyed to some hardware state-machine dependant on the DPLL clock to sync, and locks the chip up if this isn't running. More specifically, it seems to lock it up if you try to enable the PIPE whilst the corresponding DPLL is in VGA mode.

Perhaps the tests might be special cased to only stop this reg programming on the 855 chips?
Comment 8 Jesse Barnes 2007-11-08 10:04:08 UTC
I pushed Peter's fix upstream.  Please test and reopen if it doesn't work.

Thanks,
Jesse


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.