This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: floating point formats, again
- To: David Edelsohn <dje at watson dot ibm dot com>
- Subject: Re: floating point formats, again
- From: "Zack Weinberg" <zackw at Stanford dot EDU>
- Date: Sun, 11 Mar 2001 22:07:58 -0800
- Cc: gcc at gcc dot gnu dot org
- References: <zackw@stanford.edu> <200103120554.AAA28980@mal-ach.watson.ibm.com>
On Mon, Mar 12, 2001 at 12:54:54AM -0500, David Edelsohn wrote:
> Almost :^)
>
> David
>
> 3.25724264705901300000e+01 4220928a 8a8a8ca0 ..k.....
> 3.53802595280598400000e+18 50311997 dbdb1d18 &..p....
> 1.10037057600060700000e-05 3cb89c8e 6e8753b2 ....>g..
> 1.77977764695171700000e+10 49424d48 45584650 ..(....&
> -5.22995989424860500000e+10 c9c2d4c8 c5e7c6d8 IBMHEXFQ
Interesting. That started as c9c2d4c8 c5e7c6d7 in the program that
synthesized those numbers. I bet your compiler is rounding things
differently from my C library... can you try changing the last
constant to -5.22995989424860400000e+10?
'IBMHEXFQ' is a perfectly usable identifier, even if it does look
funny.
zw