This is the mail archive of the java-patches@gcc.gnu.org mailing list for the Java project.


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

Re: autoconf 2.50 updates - libjava


On Sat, May 26, 2001 at 04:28:55PM -0600, Tom Tromey wrote:
> >>>>> "Zack" == Zack Weinberg <zackw@stanford.edu> writes:
> 
> Zack> Continuing to break things up by directory.  libjava isn't set
> Zack> up for cross compilation, which makes life easier (no
> Zack> LIB_AC_PROG_CC to deal with).  These are pretty straightforward.
> 
> Actually, libjava can be cross-compiled.  I don't do it frequently,
> but it has been done in the past, and I know that Anthony has done
> some work on that recently.
> 
> We don't use LIB_AC_PROG_CC.  Instead we do this:
> 
>     dnl This shouldn't be needed, as long as top-level dependencies are
>     dnl defined correctly and shared-library paths are set up so that
>     dnl execution tests succeed.  FIXME.
>     define([AC_PROG_CC_WORKS],[])
>     define([AC_PROG_CXX_WORKS],[])
> 
> You will also get warnings from configure about not using certain
> macros when to cross-compiling.  Those macros are actually only
> invoked in the native case.
> 
> Does this approach continue to work with 2.50?

No it doesn't.  The _WORKS macros are now named something else.  I had
missed this hack or I wouldn't have submitted the patch - it has
exactly the same problems that LIB_AC_PROG_CC does.

There's a new macro called AC_NO_EXECUTABLES that does the equivalent
of that, but it also causes any use of AC_TRY_LINK or AC_TRY_RUN to be
rejected when autoconf runs.  All the affected configure scripts use
one or both of these at least once.  They are technically incorrect in
this context, but may have worked in the past (particularly if we do
link tests with the C compiler and it's the Java or C++ compiler that
won't link stuff).

-- 
zw      ...It's not easy for cognitive scientists to get grants if they
        are working on questions of any theoretical interest.  (To ensure
        this is a main function of the institution of peer review.)
        	-- Jerry Fodor, _The Mind Doesn't Work That Way_


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