[uClinux-dev] software floating point mcf5282

David McCullough davidm at snapgear.com
Thu Aug 25 08:18:04 EDT 2005


Jivin H?kan Kvist lays it down ...
> On Thu, Aug 25, 2005 at 08:19:17AM +1000, David McCullough wrote:
> > 
> > Jivin Michael Broughton lays it down ...
> > > Hi all,
> > > 
> > > I have an app that uses double to represent numbers and I'm having 
> > > trouble getting it to work properly. When I run the app it fails when it 
> > > tries to cast an int to double (pid: ## failed: 4). I haven't tried 
> > > doing any other fp operations yet, if that matters.
> > > 
> > > Is there something I need to do to configure uClinux, uClibc, gcc, or 
> > > the kernel (2.6 in this case) to get soft fp to work? My platform is a 
> > > Motorola mcf5282 eval board and I'm using gcc 2.95.
> > 
> > Be sure to include <math.h>, but I don't know why it has an affect.
> > Do not NOT use the -msoft-float option ever on coldfire :-)
>
> Does "Do not NOT use the -msoft-float option" apply to other platforms
> aswell, like armnommu?

On m68k the -msoft-float overrides any other -m command line options,
for example -m5307,  and so you stop generating code for a coldfire
board if you add -msoft-float.

I suspect that arm is ok with this,  but I don't know for sure.

Cheers,
Davidm

-- 
David McCullough, davidm at cyberguard.com.au, Custom Embedded Solutions + Security
Ph:+61 734352815 Fx:+61 738913630 http://www.uCdot.org http://www.cyberguard.com



More information about the uClinux-dev mailing list