Bug 130

Summary: Automatic ATI WORKAROUND fails on Radeon at 2nd GLCanvas Context creation
Product: [JogAmp] Jogl Reporter: Sven Gothel <sgothel>
Component: coreAssignee: Sven Gothel <sgothel>
Status: VERIFIED DUPLICATE    
Severity: normal    
Priority: P3    
Version: 1   
Hardware: All   
OS: windows   
Type: DEFECT SCM Refs:
Workaround: ---
Attachments: Test case from submitter

Description Sven Gothel 2010-03-24 07:47:17 CET


---- Reported by fmpc77 2005-01-13 12:18:47 ----

I report an error that seems to be automatic ATI Card detection at Second 
GLCanvas GLContext creation (old issue dealing with dummyGL for correct pixel 
format selection). This automatic detection works fine at first GLCanvas 
creation, but creating an other Canvas in an other window fails with null 
Exception (of dummyGL if I remember well ;), atioglxx.dll ACCESS_VIOLATION JVM 
Crash, or as I saw some reports : simply hanging. 
I was well surprised that last release hided the dirty need to set 
ATI_WORKAROUND with a simple Application using a lonely GLCanvas. But it 
crashes my JVM if I don't set ATI_WORKAROUND System variable to true with a 2 
GLCanvas Apllication. I've seen many related issues I guess (109,118,128, a 
maybe others), and I think it needs little effort to fix that bug (I've read 
the code, it's pretty simple lol ;) and many users will love you after that as 
jogl is reaching good stability :)
For note I use Radeon 9700 (latest drivers), WinXP, Java 1.4.2 & 1.5, last Jogl 
release

PS : Thank you very much for FSAA Multisampling fix :) I'm really happy to see 
Jogl improving as it does, and so to have choosen it,... at your service



---- Additional Comments From kbr 2005-01-13 13:55:05 ----

Do you have a test case which illustrates this issue? If so, could you please
add it as an attachment?





---- Additional Comments From kbr 2005-01-17 16:29:20 ----

Created an attachment
Test case from submitter




---- Additional Comments From kbr 2005-01-28 19:10:53 ----

Is it possible for you to build JOGL from the CVS repository and try this test
again? I can't reproduce exactly the problem you mention; I can occasionally see
very slow performance while running your test, indicating something is going
wrong, but both windows come up on my Radeon 9800 Pro under both Windows and Linux.

Could you also try the test case from Issue 118 on your machine?




---- Additional Comments From kbr 2005-01-30 17:13:44 ----

The problems seen in this bug should be fixed with the workaround just checked
in for Issue 118. Please open a new bug or reopen this one if similar problems
persist. The fix is in the JOGL CVS repository and will be present in 1.1 b08.


*** This issue has been marked as a duplicate of 118 ***



--- Bug imported by sgothel@jausoft.com 2010-03-24 07:47 EDT  ---

This bug was previously known as _bug_ 130 at https://jogl.dev.java.net/bugs/show_bug.cgi?id=130
Imported an attachment (id=36)

The original submitter of attachment 36 [details] is unknown.
   Reassigning to the person who moved it here: sgothel@jausoft.com.