On Mon, 2002-05-27 at 00:02, Michael wrote: > On Sun, May 26, 2002 at 08:17:18PM +0200, Michel D�nzer wrote: > > I've updated the patch, unless someone comes up with an argument against it > > I'll commit it soon. > > I don't have anything to test, so if it doesn't break i386 it works for me :o)
Hehe, that would be good to know although I can hardly imagine how it could. > > > I'm going to try the TCL branch to see if those are still there. > > Assuming they are I'll have a look if no one beats me to it after I've > done this tuxkart hang. Great, thanks. > Some of them might appear fixed because the new optimisations in the TCL > driver may not use the error path. > > I typically ran tests (with things like mesa/demos mesa/samples/ > glut/progs/demos etc) > then with RADEON_NO_CODEGEN=1 > then with RADEON_NO_VTXFMT=1 > then with RADEON_NO_TCL=1 > > That makes it a bit easier to know where to start looking sometimes too. Well, the first and the last are implicit with an M6 on PPC, and the middle one didn't help with the glxgears hang. > Looking at it I'd inially suspect the celestia code (pull the window > around and the text displays ok. But then I think it always displayed fine with the XFree86 4.2 code. The evas_test problem was already there though. > I'd done a few 1 liner fixes for primitive rendering probs, if the > morph3d fails in TCL perhaps the best way is to a) see which primitive > it is b) look in the mesa/src/tnl_dd/t_dd_dmatmp2.h (or whichever is > relevent) for a winding error or similar for that primitive type. > > gdb can help avoid confusion about which path is actually being used > (avoiding the TAG() stuff) Thanks for the hints. -- Earthling Michel D�nzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast _______________________________________________________________ Don't miss the 2002 Sprint PCS Application Developer's Conference August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
