Subject: Re: accuracy of "long double"
To: None <M.Drochner@fz-juelich.de>
From: Vincent <10.50@free.fr>
List: tech-toolchain
Date: 09/03/2007 22:45:38
Matthias,

> It is initialized in machdep.c:setregs().

Ok, viel Dank.

> That's just an initial check to tell whether there is an FPU at all.
> (could probably be skipped if it is reported by cpuid)

Yup, I saw that, since the state is not altered.

> It is complex matter, and there are some tradeoffs. Imho, one is better
> off using just "double" -- the code is portable, and even on i386 it
> can benefit from SSE2 and successors.

I agree (after some hard head ache). Specifically, I have added a little 
extra asm to allow for full resolution just in the critical section of 
the code, and #if'ed with GNUC and i386. So, if we have extra precision, 
good, if not, doesn't matter much.

Now, is this thread safe?

Viel Dank nocheinmal,
Vincent