[PATCH] math: fix isinf/isnan on SPU after move to math.h

[Ulrich Weigand]

> On Apr  6 20:35, Yaakov Selkowitz wrote:
> > Signed-off-by: Yaakov Selkowitz <yselkowi@redhat.com>
> > ---
> >  newlib/libm/common/s_isinf.c | 2 ++
> >  newlib/libm/common/s_isnan.c | 2 ++
> >  2 files changed, 4 insertions(+)
> >
> > diff --git a/newlib/libm/common/s_isinf.c b/newlib/libm/common/s_isinf.c
> > index f22acb7..fe18e2a 100644
> > --- a/newlib/libm/common/s_isinf.c
> > +++ b/newlib/libm/common/s_isinf.c
> > @@ -13,6 +13,8 @@
> >
> >  #ifndef _DOUBLE_IS_32BITS
> >
> > +#undef isinf
> > +
> >  int
> >  _DEFUN (isinf, (x),
> >  	double x)
> > diff --git a/newlib/libm/common/s_isnan.c b/newlib/libm/common/s_isnan.c
> > index 64deacd..bc0a916 100644
> > --- a/newlib/libm/common/s_isnan.c
> > +++ b/newlib/libm/common/s_isnan.c
> > @@ -190,6 +190,8 @@ QUICKREF
> >
> >  #ifndef _DOUBLE_IS_32BITS
> >
> > +#undef isnan
> > +
> >  #ifdef __STDC__
> >  	int isnan(double x)
> >  #else

This does indeed fix the build problem for me on SPU.
Are you planning to check the patch in?

Thanks,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU/Linux compilers and toolchain
  Ulrich.Weigand@de.ibm.com