RE: [AD] Use MMX to get fast

["Robert Ohannessian" <ROhannessian@xxxxxxxxxx>]

That's not the problem.

The problem is that Allegro doesn't special-case plain mem->mem copies.
Instead, it has all those code to deal with subbitmaps, video bitmaps,
etc, which add overhead.



> -----Original Message-----
> From: alleg-developers-admin@xxxxxxxxxx [mailto:alleg-
> developers-admin@xxxxxxxxxx] On Behalf Of Elias Pschernig
> Sent: Tuesday, September 28, 2004 3:27 PM
> To: alleg-developers
> Subject: Re: [AD] Use MMX to get fast
> 
> On Tue, 2004-09-28 at 18:31 -0300, RogerioUP wrote:
> >
> > I think it's very simple for you (allegro developers) to implement
> > this routine in Allegro (blitmmx for example) just to test, its
> > implementation is very similar to memcpy, but I use MM0 to move
> > quadwords (8 bytes).
> >
> 
> Hm, speeding up memory blits isn't on the todo, but sounds like a good
> idea. Did you look at allegro's code? It already uses MMX/SSE/3dnow
and
> so on in some places. src/i386/iblit*.s does the blitting it seems -
> maybe you can try adding the MMX code there at the relevant places?
> 
> --
> Elias Pschernig
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on
ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give
us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out
> more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> --
> https://lists.sourceforge.net/lists/listinfo/alleg-developers