Re: Re : [Bug-gnubg] USE_SSE2

[Jonathan Kinsey]

Michael Petch wrote:
>
>
> On 18/08/09 1:28 AM, "Massimiliano Maini" wrote:
>
>> Just to be sure I get this right: your recent code checks for sse2 (if
>> USE_SSE2)
>> and does or doesn't falback gracefully if check fails ?
>>
>>
>
> If you build with USE_SSE2 (and -msse2) it will look for SSE2 instruction
> set. If it doesn't find SSE2 capability it falls back to NO SSE.
>
> If you build the SSE version (and -msse) it will check for SSE instructions
> and fall back to none if the capability is not found.

Are you sure? The only link I could find about this says different:

http://www.network-theory.co.uk/docs/gccintro/gccintro_64.html

I suppose the only way to find out for sure is to test it, if someone can find a
machine that doesn't support sse...

It does seem unlikely that two lots of code would get generated for every
optimised part of code, but maybe there is some magic going on...

Jon

---

Internet Explorer 8 - accelerate your Hotmail. Download Internet Explorer 8