A long double patch for ia64.

[H . J . Lu]

On Sat, Sep 16, 2000 at 11:20:22PM -0700, Ulrich Drepper wrote:
> "H . J . Lu" <hjl@valinux.com> writes:
> 
> > 	* sysdeps/ia64/fpu/s_isinfl.c: New. Copied from
> > 	sysdeps/i386/fpu.
> > 
> > 	* sysdeps/ia64/fpu/s_isnanl.c: New. Copied from
> > 	sysdeps/i386/fpu.
> > 
> > 	* sysdeps/ia64/fpu/s_nextafterl.c: New. Copied from
> > 	sysdeps/i386/fpu.

I don't know for sure if they should be the same. Someone has to
look up ia32 and ia64 floating point manuals. From my commone
senses, 80bit long double on ia64 should be identical with ia32.

> > 
> > 	* sysdeps/ia64/fpu/bits/mathdef.h: New. Copied from
> > 	sysdeps/i386/fpu/bits/mathdef.h.
> 
> If these files are identical they should be simple #includes and not
> be copied.
> 

I don't know if I can use #include in sysdeps/ia64/fpu/bits/mathdef.h
since it is an installed header files.

Please feel free to fix it right.

Thanks.

H.J.