This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [libstdc++] Build system cleanup 8/n
- From: Benjamin Kosnik <bkoz at redhat dot com>
- To: Kelley Cook <kcook at gcc dot gnu dot org>
- Cc: libstdc++ at gcc dot gnu dot org, GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Thu, 7 Apr 2005 10:58:33 -0500
- Subject: Re: [libstdc++] Build system cleanup 8/n
- Organization: Red Hat / Chicago
- References: <4255404B.9000709@gcc.gnu.org>
> But it is actually checked for in both various crossconfig and native
> (line 121 configure.ac). Itâlike most other headersâis not checked in
> newlib crossbuilds, which seems wrong. Regardless this patch seems safe.
Ok. I didn't catch the configure.ac line when I did the review of your
other patch, sorry.
> I actually don't see the reason for why any AC_CHECK_HEADERS would be
> split into host or target specific checks as checking if a header file
> exists is quick and wouldn't hurt if it doesn't exist, but I'll queue
> that cleanup to the next round after acconfig.h makes its long awaited
> departure.
Agreed about checking headers. In general, I think the goal should be to
try and move as much as possible into one place only, non-link checks
that happen for all builds, regardless of native/cross.
However, this can wait for a bit.
-benjamin