Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Felix Kühling
Hi, Forget the other patch, this one actually works. I tested it with a modified radeon driver. They way I tried to access per-screen glx extension data before didn't work. It used some X extension list to find the screen data structure for a given display and screen number. But while the DRI scre

Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Ian Romanick
Felix Kühling wrote: Forget the other patch, this one actually works. I tested it with a modified radeon driver. They way I tried to access per-screen glx extension data before didn't work. It used some X extension list to find the screen data structure for a given display and screen number. But w

Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Felix Kühling
On Mon, 11 Aug 2003 12:14:31 -0700 Ian Romanick <[EMAIL PROTECTED]> wrote: > Felix Kühling wrote: > > > So I moved the glx extension data to the end of __DRIscreenRec and added > > a pointer to __DRIscreenPrivateRec that points back to the > > __DRIscreenRec. This pointer is then passed to > > __

Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Ian Romanick
Felix Kühling wrote: On Mon, 11 Aug 2003 12:14:31 -0700 Ian Romanick <[EMAIL PROTECTED]> wrote: Felix Kühling wrote: So I moved the glx extension data to the end of __DRIscreenRec and added a pointer to __DRIscreenPrivateRec that points back to the __DRIscreenRec. This pointer is then passed to __

[Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Felix Kühling
Hello, this is a follow-up of the thread "per-screen client-side glx extensions". In addition to those changes the attached patch also removes __glXRegisterGLXFunction and __glXAddExtension. Therefore __driRegisterExtensions is no longer necessary and corresponding infrastructure was removed from

Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Ian Romanick
Felix Kühling wrote: So I moved the glx extension data to the end of __DRIscreenRec and added a pointer to __DRIscreenPrivateRec that points back to the __DRIscreenRec. This pointer is then passed to __glXScrEnable/DisableExtension by the driver's createScreen function. There may be binary compati

Re: [Dri-devel] [patch] dynamic glx extensions cleanup (the 2nd)

2003-08-14 Thread Felix Kühling
On Sun, 10 Aug 2003 13:04:49 -0700 Ian Romanick <[EMAIL PROTECTED]> wrote: > Felix Kühling wrote: > > > Forget the other patch, this one actually works. I tested it with a > > modified radeon driver. They way I tried to access per-screen glx > > extension data before didn't work. It used some X e