| Re: [AD] Allegro x86 clear and blit optimizations - update |
[ Thread Index |
Date Index
| More lists.liballeg.org/allegro-developers Archives
]
> The main platform the clears and blitters were optimized for was
> PPlain/PMMX.
Optimizing for such half-superscalar, in-order execution CPUs is the very
essence of asm optimization :-)
> For PPro+, there's not a lot that can be done for the masked blits
> (expensive branch mispredictions).
Yes, all these hyper-pipelined things must go directly to the garbage
collector ;-)
> I considered working on the 16-bit code, but was waiting for a reply on
> the alignment issue in the clears.
Where was (is ?) exactly the problem ? MMX or non-MMX clear code ?
--
Eric Botcazou
ebotcazou@xxxxxxxxxx