This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Don't build abi_check for crosses


> Date: Mon, 8 Sep 2003 16:18:02 -0400
> From: Daniel Jacobowitz <drow@mvista.com>

> On Mon, Sep 08, 2003 at 01:09:21PM -0700, Geoff Keating wrote:
> > Daniel Jacobowitz <drow@mvista.com> writes:
> > 
> > > This is the 3.3 version of the patch.  3.4 the test is negated and against
> > > GLIBCXX_IS_NATIVE:
> > > if test x"$GLIBCPP_IS_NATIVE" = xfalse || test x$enable_symvers = xno; then
> > > 
> > > This is necessary because automake can't build abi_check for the host. 
> > > Right now it's listed as a target program, and will be linked to the newly
> > > built libstdc++; if $build != $host or --with-cross-host, we won't be able
> > > to run it even if we were able to build it.  Ben, CC'd you because you
> > > removed the comment saying to build it native only; was there a reason?
> > > 
> > > OK?
> > 
> > Just because it can't be run is no reason not to double-check that it
> > can link...
> 
> But it can't link, necessarily.  We go through some perversions to make
> sure that CXX is a C compiler; then testsuite/Makefile.am defines
> CXXLINK with a cute little sed expression that might or might not find
> you the right C++ compiler.  It does for native bootstraps, but I don't
> know if it will for non-bootstraps; it definitely doesn't for
> host-x-host builds where $CXX at this point is "mipsel-linux-gcc".

Isn't that a problem anyway?  I mean, the C++ compiler is used for
other things building libstdc++, not just for abi-check.cc (like the
entire rest of the library).  Likewise, the testsuite seems to know
how to link C++ programs.

-- 
- Geoffrey Keating <geoffk@geoffk.org>


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]