On 2002.02.11 01:01 Leif Delgass wrote: > The attached patch implements mach64_draw_point() and mach64_draw_line() > in the same vein as tris and quads (actually most of the code is just > copied from those functions), i.e. insecure client-side PIO/MMIO > register programming. This diff is against > xc/lib/GL/mesa/src/drv/mach64/mach64_tris.h. Could someone with cvs > write > access apply this please? From what I can tell, software fallback is > used > for GL_POINT_SMOOTH and GL_LINE_SMOOTH, since those seemed to be working > already without this patch. >
I've tested with the Nate Robins' OpenGL tutorials and it works great! If there is no objection I can submit it to CVS tonight. > Once we get the new branch with Frank's DMA code, these functions would > of > course need to be rewritten (along with tris and quads) to write to the > vertex buffer and use DMA. > > Here are some remaining Mesa driver bugs I've found: > > - fog not working (although setting fog color has an effect) > - multitexturing not working > - unimplemented texture environment modes: GL_DECAL, GL_BLEND, GL_REPLACE > > (GL_MODULATE seems to work) > - GL_CLAMP in GL_TEXTURE_WRAP_[S,T] not working (this seems to be a > problem in the r128 driver as well). > I think that we can start attacking these bugs (more like missing features ;-) while DMA work is in progress. > btw, I'm using the Mesa demos and Nate Robins' OpenGL tutorials for > testing. > I've check them and these tutorials are really great for interactively testing specific features. > Regards, > > Leif > Good work, Leif. Regards, José Fonseca _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
