[Mesa-dev] Re: [utah-glx-dev] 3Dnow asm broken on the K6-III?

[Victor Vedovato]

I can't really tell from the screen shot, but there seems to be portions
that render correctly. The obvious bug would be improper placement
of femms instructions resulting in a corrupted FP stack.

Another problem with 3dnow is that the math is not IEEE 754 compliant 
so there are problems with mixing FP and 3dnow results in the 
same data set. 

Also, PFRCP with 1.0 doesn't give 1.0 as a result. This showed
up on our windows driver with glOrtho ops. (I think Q2 status
items were done this way).

- Victor Vedovato

Ralph Giles wrote:
> 
> Hello all,
> 
> We've received several reports lately on the Utah-GLX list about the 3Dnow
> asm code not working for people with K6-IIIs. It shows up under software
> fallback, so our guess is that it's a problem in the Mesa code. (that, and
> I don't think we have any 3Dnow in our drivers--just mmx for texture
> conversion)
> 
> In particular see:
> 
> http://lists.sourceforge.net/pipermail/utah-glx-dev/2000-March/001914.html
> 
> And the associated screen shot for the fun effects on the projection
> matrix. :)
> 
> I didn't see a report of this in the list archives, so I wanted to bring
> it to your attention. I don't know if this is a bug in the new processors,
> or if they've changed the instruction set somewhat. Can any experts
> comment?
> 
>  -ralph
> 
> --
> [EMAIL PROTECTED]
> utah-glx.sourceforge.net
> 
> _______________________________________________
> utah-glx-dev mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/mailman/listinfo/utah-glx-dev


_______________________________________________
Mesa-dev maillist  -  [EMAIL PROTECTED]
http://lists.mesa3d.org/mailman/listinfo/mesa-dev