[XviD-devel] overflow being reported incorrectly

[suxen_drol]

On Sat, 4 Jan 2003 12:08:39 -0000 "Jim Hauxwell" <james@dattrax.co.uk> wrote:

> Hi,
> 
> I came across the femms problem this morning also.  seems like intel and
> microsoft compilers cant build for it.  Even the intrinsics version
> _m_femms() dosn't work so I've just commented it out.

bleh. my msvc6 + sp5 + pp setup compiles it, as does gcc.
( i guess i should've test icc also...)
> 
> As a matter of interest, why has someone added this?  According to AMD doc I
> read this morning, the instructions femms and emms have zero latency and
> dont do anything.
> 
> http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/2262
> 1.pdf
> 
> Maybe someone can shed some light on it.

on the k6-2(3) series, femms is faster. the speed increase is not
noticible, but ive included it for "completeness".

the emms_mmx and emms_3dn functions could be moved to nasm to avoid
compiler problems... and avoid having to defined EMMS() macros in
portab.h. 

or, would y'all rather switch intrinsics?

-- pete