This is the mail archive of the
java-discuss@sourceware.cygnus.com
mailing list for the Java project.
Re: GCJ Compiler error building current CVS trees
- To: Tom Tromey <tromey at cygnus dot com>
- Subject: Re: GCJ Compiler error building current CVS trees
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Thu, 23 Sep 1999 13:44:36 -0600
- cc: Glenn Chambers <GChambers at provsol dot com>, "'java-discuss at sourceware dot cygnus dot com'" <java-discuss at sourceware dot cygnus dot com>, "'gcc-bugs at gcc dot gnu dot org'" <gcc-bugs at gcc dot gnu dot org>
- Reply-To: law at cygnus dot com
In message <199909211955.MAA29869@ferrule.cygnus.com.>you write:
> Jeff> Well, it looks like the C front-end is being called with
> Jeff> -fuse-divide-subroutine which isn't any option I'm aware of.
> Jeff> I'd put my money that it's a bug on the java side.
>
> -fuse-divide-subroutine is a Java option.
> This bug occurs because we have %{f*} in the "jvgenmain" spec (see
> gcc/java/jvspec.c). (gcj can generate a C program and then compile
> it; this is how we generate a "main" when given the --main option.)
>
> I'm not entirely sure how to fix this. Presumably some of the -f
> options given to gcj should be passed down to cc. I don't see how to
> filter them except to list all of the C front end's -f options
> explicitly. (Which, of course, I don't want to do.)
>
> Any advice?
Not sure. There may already be some support in the specs for passing around
language specific options to the right place.
Some folks have suggested that unknown -f/-W options should be a warning;
that wouldn't really solve your problem, but it would make it less pressing.
jeff