Re: [AD] Feature request: Add missing GFX_VTABLE entries

[ Thread Index | Date Index | More lists.liballeg.org/allegro-developers Archives ]


On Sunday 26 September 2004 21:01, Robert Ohannessian wrote:
> I agree that the do_* and calc* functions don't belong there.

Of course... it doesn't make sense to try to accelerate these anyway.

> Stretch_sprite and masked_stretch_blit don't really differ, do they? Is
> one (both?) implemented by pivot_scaled_sprite_flip? If so, they don't
> need to be there.

stretch_blit, masked_stretch_blit and stretch_sprite all use the same 
do_stretch_blit function as a workhorse (src/i386/istretch.c), or 
_al_stretch_blit from src/c/cstretch.c. It's probably good enough to make 
a vtable entry for this function.

> For the set_*_blender functions, maybe we can just have one driver entry
> for all of them that takes an enum (or define) to select the blender?

Makes sense.
I'll see if I can make those changes later tonight.

Evert





Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/