Re: [PATCH] ARM: Don't define _SYS_AUXV_H in sysdep.h

["Joseph S. Myers" <joseph at codesourcery dot com>]

On Tue, 16 Sep 2014, Will Newton wrote:

> On 12 September 2014 10:04, Joseph S. Myers <joseph@codesourcery.com> wrote:
> > On Fri, 12 Sep 2014, Will Newton wrote:
> >
> >> sysdep.h was defining _SYS_AUXV_H in order to avoid an include guard check
> >> in hwcap.h. Unfortunately it didn't undefine it so it could leak out into
> >> code and caused a build failure with -Wimplicit-function-declaration
> >> building tst-auxv on ARM.
> >>
> >> ChangeLog:
> >>
> >> 2014-09-11  Will Newton  <will.newton@linaro.org>
> >>
> >>       * sysdeps/unix/sysv/linux/arm/bits/hwcap.h: Check for
> >>       _LINUX_ARM_SYSDEP_H include guard too.
> >
> > Rather than referring to a guard for an internal header in an installed
> > header, I think it would be better to test for _LIBC there.
> 
> This approach (testing for the sysdep.h guard) is used in sparc and
> powerpc at present. Testing _LIBC would work but

OK for the original patch then, but please file a bug in Bugzilla to 
eliminate such ad hoc conditionals from installed headers.  In general 
installed headers should have as little as possible that's not actually 
relevant for use of the installed library, and the particular set of 
conditionals used in installed headers in such cases should be minimal 
(_LIBC and _LIBC_TEST?).

> stdio-common/scanf15.c and stdio-common/scanf17.c both #undef _LIBC
> and break things.

Why does sysdep.h get included from these testcases anyway?  As well as 
moving away from defining _LIBC (and other internal defines) for tests 
unless needed for a particular test, it would also be good to move away 
from tests using the internal includes (if such an include is responsible 
here), instead making them use the versions of headers that actually get 
installed before looking for any other headers in the source tree or 
sysdeps directories.

-- 
Joseph S. Myers
joseph@codesourcery.com