This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: egcs-2.93.10 19990228: cc1 built with --enable-c-cpplib doesn't handle -MD
- To: egcs-bugs at egcs dot cygnus dot com
- Subject: Re: egcs-2.93.10 19990228: cc1 built with --enable-c-cpplib doesn't handle -MD
- From: Raja R Harinath <harinath at cs dot umn dot edu>
- Date: 01 Mar 1999 16:59:48 -0600
- References: <d9pv6sga8u.fsf@peca.cs.umn.edu> <36DB19E4.753953AD@cygnus.com>
Dave Brolley <brolley@cygnus.com> writes:
> Raja R Harinath wrote:
> > Hi,
> >
> > A `cc1' binary built with --enable-c-cpplib doesn't handle the `-MD file'
> > option. This feature is used by `automake', and breaks virtually
> > every package using automake.
> >
> > >From what I could figure out, the -M options are missing from
> > documented_lang_options in toplev.c. A simple fix appears to be just
> > to add
>
> I tried it with the latest source and found that it works fine for
> me. I did some digging and found that it's because the -M* options
> are added via java/lang-options.h.
Ah. That makes sense. I use
--enable-languages="c,c++,f77,objc"
So, my "options.h" is
#include "cp/lang-options.h"
#include "f/lang-options.h"
Notably missing is java/lang-options.h.
> I'm wondering if they should be moved into toplev.c?
They probably should. However, if the java compiler has intrinsic
support for -M* rather than through the pre-processor, I guess that
fact will have to be documented somewhere. (I think it currently is
documented by being in java/lang-options.h.)
- Hari
--
Raja R Harinath ------------------------------ harinath@cs.umn.edu
"When all else fails, read the instructions." -- Cahn's Axiom
"Our policy is, when in doubt, do the right thing." -- Roy L Ash