We are currently migrating Bugzilla to GitHub issues.
Any changes made to the bug tracker now will be lost, so please do not post new bugs or make changes to them.
When we're done, all bug URLs will redirect to their equivalent location on the new bug tracker.

Bug 204

Summary: YUV overlays fails with a BadMatch
Product: SDL Reporter: Sjoerd Simons <sjoerd>
Component: videoAssignee: Ryan C. Gordon <icculus>
Status: RESOLVED WORKSFORME QA Contact: Sam Lantinga <slouken>
Severity: normal    
Priority: P2    
Version: 1.2.10   
Hardware: PowerPC   
OS: Linux   

Description Sjoerd Simons 2006-04-24 08:43:20 UTC
Programs using the SDL YUV overlay crash with a BadMatch error.
The testoverlay programs gives the following output:
$  ./testoverlay2 
Created 64x88x1 hardware YUY2 overlay
  plane 0: pitch=128
X Error of failed request:  BadMatch (invalid parameter attributes)
  Major opcode of failed request:  141 (XVideo)
  Minor opcode of failed request:  19 ()
  Serial number of failed request:  24
  Current serial number in output stream:  25

This is on a Apple PowerBook5,2 (Ati radeon 9600 videocard, Xorg 7.0, Debian unstable). 
  
Sjoerd
Comment 1 Ryan C. Gordon 2006-06-21 03:11:38 UTC
This is working on amd64...powerpc or driver-specific issue?

--ryan.

Comment 2 Sjoerd Simons 2006-06-21 05:18:09 UTC
(In reply to comment #1)
> This is working on amd64...powerpc or driver-specific issue?

Yeah works for me on amd64 with an ati card too but not on my powerpc.. Things like the gstreamer's xvimagesink and videolan's own xv output work fine though, only SDL fails. So the driver is probably ok.

If i might do a wild guess, maybe an endianness error ?
Comment 3 Ryan C. Gordon 2006-06-21 05:27:59 UTC
I'll try this on the PowerPC Linux box in my office when I get some time. Ping me again if you haven't heard anything in a week.

--ryan.

Comment 4 Ryan C. Gordon 2007-02-14 07:52:55 UTC
Just tried this on an iBook G4 (Ubuntu 6.10, ATI Radeon 9600) with the latest revision of SDL...works fine.

Either x.org or the ATI driver was hosed, or we fixed this (I'm guessing the former here).

Flagging bug as WORKSFORME.

--ryan.