This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: new release candidate (gsl-1.5.90)
- From: Brian Gough <bjg at network-theory dot co dot uk>
- To: Giulio Bottazzi <giulio dot bottazzi at libero dot it>
- Cc: gsl-discuss at sources dot redhat dot com
- Date: Wed, 5 Jan 2005 15:46:12 +0000
- Subject: Re: new release candidate (gsl-1.5.90)
- References: <E1Cji3x-0008FV-00@localhost><Pine.LNX.4.50.0412291331150.29540-100000@grus.astro.umd.edu><16852.16896.863758.963158@network-theory.co.uk><20041231012413.4f080104.giulio.bottazzi@libero.it><16853.23210.503801.259972@network-theory.co.uk><20050101011706.241c5c61.giulio.bottazzi@libero.it>
- Reply-to: gsl-discuss at sources dot redhat dot com
Giulio Bottazzi writes:
> Is this a design choice? If it is so, probably it would be useful
> to add some warning about it in the documentation.
Thanks, I will add a note to the manual. The default in GSL is
"floating-point exceptions on" (for safety!). In particular,
underflow is worth flagging as it usually goes unnoticed.
Something to be aware of: some versions of the Linux kernel do not
(did not?) preserve the floating point mode correctly if signals
occur--it reverts to the default (not sure if this is fixed in recent
kernels). Interest in FP support generally seems to be lacking among
its developers.
--
Brian Gough
Network Theory Ltd,
Commercial support for GSL --- http://www.network-theory.co.uk/gsl/