More ARM binutils fuckage

[Nicolas Pitre]

On Thu, 7 Dec 2006, Michael K. Edwards wrote:

> On 12/7/06, Nicolas Pitre <nico@cam.org> wrote:
> > Probably.  But floating point in the kernel is forbiden anyway and the
> > -msoft-float will ensure that no FP instructions are issued at all (and
> > give you a link error) ifyou slip some floats in the code by mistake.
> 
> Right.  So it's only an issue when compiling a kernel to support OABI
> hard-float binaries?  Or is the floating point emulation code in the
> kernel insensitive to the -mfpu and -mfpe settings of the compiler?

The floating point emulation code is of course free of any floating 
point code itself, otherwise you'd end up in an infinite recursive 
state.  So even for the FP emulator this is not an issue.


Nicolas

--
For unsubscribe information see http://sourceware.org/lists.html#faq