Summary: | driver compatibilty broken again ... | ||
---|---|---|---|
Product: | xorg | Reporter: | Stefan Dirsch <sndirsch> |
Component: | Server/General | Assignee: | Xorg Project Team <xorg-team> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | major | ||
Priority: | high | CC: | eich, eta, keithp, kem |
Version: | git | ||
Hardware: | All | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Bug Depends on: | |||
Bug Blocks: | 351 |
Description
Stefan Dirsch
2004-08-03 13:07:58 UTC
*** Bug 977 has been marked as a duplicate of this bug. *** I'm assuming Composite caused this to happen, by the driver stomping on some memory somewhere due to structure changes. Could you check to be sure that if you add #define BuildComposite NO to your host.def, it goes away? Thanks for the hint. I'll give it a try and let you know. Yes, disabling build of Composite fixes the problem. nvidia driver now works again. Hmm ... Someone needs to track this one down as maintaining ABI compatibility is one of the goals for this release. My only thought currently is the two extra fields in the VisualRec structure for alphaMask and offsetAlpha might be causing the problem, but I haven't looked very far into this yet. Adding Keith Packard and Eric Anholt on the CC list for their comments. I removed the new fields in the Visual structure related to making core color allocation automatically set alpha bits in returned pixel values. Aside from the new screen_x/screen_y fields in pixmaps (which we hope aren't an ABI issue), there shouldn't be any other ABI changes controlled by building with COMPOSITE defined. Thanks, Keith! I'll give it a try ASAP (this weekend for sure) and let you know about the results. Just verified. The nvidia driver works fine again. :-) |
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.