On 2002.04.12 22:53 Brian Paul wrote: > ... > > As long as the code passes the glean and conform blend tests and the > 255*255=255 identity is satisfied then I'm happy with it. I'll apply > your new MMX patch. >
Good. It's still not complete. It misses updating the runin and runout duplicates. > BTW, are you interested in writing MMX functions for the blend_add > and blend_modulate functions? Those would be useful, and would > probably be trivial for you after doing blend_transparency. Sure! I had plan to this all along. Just a matter of letting the MMX code stabilizes, optimize it a little further, and get some free time while working on the mach64. The same goes for the C blending code. > > -Brian > José Fonseca _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
