Bug 10037

Summary: Screen corruption for "SuperSavage/IXC 64" when "exa" is enabled
Product: xorg Reporter: Joachim Frieben <jfrieben>
Component: Driver/savageAssignee: Xorg Project Team <xorg-team>
Status: RESOLVED FIXED QA Contact: Xorg Project Team <xorg-team>
Severity: normal    
Priority: medium CC: alexdeucher
Version: 7.2 (2007.02)   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
"Xorg 7.2" configuration file for "SuperSavage/IXC 64"
none
"Xorg 7.2" log file for "SuperSavage/IXC 64"
none
"Xorg 7.2" screenshot for "SuperSavage/IXC 64" showing screen corruption none

Description Joachim Frieben 2007-02-20 00:40:22 UTC
After enabling "exa" acceleration on my "IBM ThinkPad T23", screen corruption occurs [see attached screenshot]. The system is running "Xorg 7.2" as part of the current "Fedora Core" development tree.
After commenting out the "exa" entry, the screen is rendered correctly again.
Comment 1 Joachim Frieben 2007-02-20 00:41:51 UTC
Created attachment 8782 [details]
"Xorg 7.2" configuration file for "SuperSavage/IXC 64"
Comment 2 Joachim Frieben 2007-02-20 00:42:29 UTC
Created attachment 8783 [details]
"Xorg 7.2" log file for "SuperSavage/IXC 64"
Comment 3 Joachim Frieben 2007-02-20 00:43:23 UTC
Created attachment 8784 [details]
"Xorg 7.2" screenshot for "SuperSavage/IXC 64" showing screen corruption
Comment 4 Alex Deucher 2007-02-21 06:34:14 UTC
Does the corruption go away if you return false from the copy() or UTS() functions?
Comment 5 Alex Deucher 2007-02-21 06:38:27 UTC
(In reply to comment #4)
> Does the corruption go away if you return false from the copy() or UTS()
> functions?
> 

Argh... not copy, solid.  doe sit still give you problem if you return false is in  SavagePrepareSolid() or SavageUploadToScreen().  you might as well check copy too...
Comment 6 Joachim Frieben 2007-02-21 10:19:40 UTC
(In reply to comment #4)
> Does the corruption go away if you return false from the copy() or UTS()

Huh? You mean in the driver source code?
Comment 7 Alex Deucher 2007-02-21 10:43:14 UTC
(In reply to comment #6)
> (In reply to comment #4)
> > Does the corruption go away if you return false from the copy() or UTS()
> 
> Huh? You mean in the driver source code?
> 

yes in the savage source.  see my post above.
Comment 8 Daniel Stone 2007-02-27 01:36:32 UTC
Sorry about the phenomenal bug spam, guys.  Adding xorg-team@ to the QA contact so bugs don't get lost in future.
Comment 9 Joachim Frieben 2007-03-08 11:10:15 UTC
(In reply to comment #5)
> Argh... not copy, solid. Does it still give you problem if you return false
> in SavagePrepareSolid() or SavageUploadToScreen()? You might as well check
> copy too...

No luck here. I've tried various combinations in "savage_exa.c" but the only result is a badly unresponsive system.
Comment 10 Alex Deucher 2007-04-02 14:34:31 UTC
This should be fixed in savage git head.
Comment 11 Matt Turner 2010-12-02 19:13:44 UTC
Closing as per Alex's comments. Reopen if not fixed.

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.