[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Chicken-users] Bug in the numbers egg?
From: |
John.Cowan |
Subject: |
Re: [Chicken-users] Bug in the numbers egg? |
Date: |
Wed, 10 Aug 2005 18:20:29 -0400 |
User-agent: |
Mutt/1.4.2.1i |
Thomas Chust scripsit:
> This result is apparently wrong. Nevertheless it has the correct
> magnitude. I wonder whether this is some kind of amplified rounding
> error or a more serious problem.
On my system (Cygwin, gcc 3.4.4) the result is even worse:
179769313486231590772930519078902473361797697894230657273430081157732675805500963132708477322407536021120113879871393357658789768814416622492847430639474124377767893424865485276302219601246094119453082952085005768838150682342462881473913110540827237163350510684586298239947245938479716304835356329624224137216
which is again the right magnitude but has only 1 significant digit correct!
> Any idea or work around is appreciated.
According to the GMP home page at http://swox.com/gmp/ , many versions of gcc
miscompile GMP due to too-aggressive optimizations. We may need another
library (or, Ghu forbid, a different C compiler).
Try reinstalling libgmp from source.
--
John Cowan <address@hidden>
http://www.ccil.org/~cowan http://www.reutershealth.com
Charles li reis, nostre emperesdre magnes,
Set anz totz pleinz ad ested in Espagnes.