Re: gcc/libm floating-point bug?

From: David Schultz <das_at_FreeBSD.ORG>
Date: Wed, 21 May 2003 00:49:31 -0700
On Tue, May 20, 2003, Morten Rodal wrote:
> On Tue, May 20, 2003 at 11:03:36AM -0700, David Schultz wrote:
> > That's for an Alpha-specific problem.  The i386-specific problem
> > is due to a bug gcc has that involves Pentium 4 code generation
> > and aliasing.  The latter can be fixed by removing
> > -march=pentium4, disabling optimization, or upgrading to gcc 3.3.
> > The former is intentional and not expected to change; gcc
> > generates non-standard FP math code on Alpha because it saves a
> > few instructions to deal with NaNs, etc.  We work around it by
> > setting -mieee in bsd.cpu.mk for Alpha.
> >
> 
> The upgrade from gcc 3.2 to 3.3 isn't scheduled before after 5.1 is
> released? (I think I saw some post about that someplace)

Correct.  I've been poking various people to try to get this
nit documented at the very least.
Received on Tue May 20 2003 - 22:49:34 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:08 UTC