patch for PR/23522

[Alexey Starovoytov]

On Wed, 28 Sep 2005, Ian Lance Taylor wrote:

> > > >   if (TREE_CODE (arg1_unw) != INTEGER_CST
> > > >       || TREE_CODE (shorter_type) != INTEGER_TYPE
> > > >       || !int_fits_type_p (arg1_unw, shorter_type))
> > > >     return NULL_TREE;

> Oh, sorry, you're right.  != INTEGER_TYPE is correct.  Forget what I
> said.  Please test your patch above:

c/c++ testsuites reported no regressions on sparc-solaris 9 and 10.
I don't have other systems to test on.

Alex.