trouble using gfortran

[Toon Moene]

Steve Kargl wrote:

> On Wed, Apr 19, 2006 at 09:40:53PM +0200, Toon Moene wrote:

>>We might want to discern between "initialize with zero" (that g77 
>>already supports) and "initialize with bogus values - i.e. SNAN" (that 
>>g77 doesn't support, but is really, really useful).

> SNAN doesn't help with uninitialized integers.

No, but it is a nice "won't happen integer" due to its bit pattern.  Of 
course, I'd rather initialize integers with something like 5a5a5a5a hex :-)

> Additionally, gcc is very broken in its backend when inf and nan occur in
> complex types.  See PR 24581.

My reading of Joseph Myers' comment is that the C Standard hadn't made 
up its mind at that time ...

Da funny thing is that I tried to teach the RTL complex expanders to use 
R*C in the past ( ~ 2001) under the -funsafe-math-optimizations regime, 
and now it might turn out to be standard conforming C ? HAHAHAHAHA

Cheers,

-- 
Toon Moene - e-mail: toon@moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
A maintainer of GNU Fortran 95: http://gcc.gnu.org/fortran/
My next laptop will be pedal-powered