This is the mail archive of the
cygwin-developers
mailing list for the Cygwin project.
RE: IEEE floating point routines
- From: "Tommaso Tagliapietra \(EXT VE SYS\)" <tommaso dot tagliapietra at enel dot it>
- To: <cygwin-developers at cygwin dot com>
- Date: Wed, 17 Oct 2007 17:42:34 +0200
- Subject: RE: IEEE floating point routines
- Reply-to: <tommaso dot tagliapietra at enel dot it>
Can I write an application with a BSD revised license that use internal
initialization functions and mantain my license (BSD)?
-----Original Message-----
From: cygwin-developers-owner@cygwin.com
[mailto:cygwin-developers-owner@cygwin.com] On Behalf Of Corinna
Vinschen
Sent: Wednesday, October 17, 2007 5:30 PM
To: cygwin-developers@cygwin.com
Subject: Re: IEEE floating point routines
On Oct 17 08:06, Burkhardt, Glenn wrote:
> I recommend that the standard Cygwin DLL include the standard routines
> that query/control the FPU, such as
>
> fpsetmask()
> fpsetround()
> fpsetsticky()
>
> These are available as a package at
> http://www.macaulay.ac.uk/fearlus/floating-point/download.html
These sources are GPLed. They can't be included into Cygwin for
licensing reasons. Additionally I don't think that Cygwin is the right
place to include math functions. This should be provided by the
standard c library, which is newlib in our case
(http://sourceware.org/newlib/). Please note that, even *if* these
functions are included in newlib, Cygwin would still not use them for
licensing reasons. If you need theses functions, nothing keeps you from
contributing your own code to newlib, provided you put the source into
the PD, or under a relaxed license like BSD.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat