This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: CVS MinGW (Crossed-)Native Compiler is Toast
- From: Mohan Embar <gnustuff at thisiscool dot com>
- To: Andrew Haley <aph at redhat dot com>
- Cc: GCJ Java <java at gcc dot gnu dot org>
- Date: Sun, 21 Sep 2003 17:46:45 -0500
- Subject: Re: CVS MinGW (Crossed-)Native Compiler is Toast
- Reply-to: gnustuff at thisiscool dot com
Hi Andrew,
> > >---------------------------------8<---------------------------------------------
> > >D:\Program Files\thisiscool-gcc\examples\HelloWorld>gcj --main=HelloWorld -o HelloWorld.exe HelloWorld.java
> > >HelloWorld.java: In class `HelloWorld':
> > >HelloWorld.java: In method `HelloWorld.main(java.lang.String[])':
> > >HelloWorld.java:5: internal compiler error: tree check: expected class 'e', have 't' (complex_type) in issue_warning_error_from_context, at java/parse.y:3051
> > >Please submit a full bug report,
> > >with preprocessed source if appropriate.
> > >See <URL:http://gcc.gnu.org/bugs.html> for instructions.
> > >---------------------------------8<---------------------------------------------
> > >
> > >This doesn't happen with the cross compiler
> > >((i686-pc-linux-gnu,i686-pc-linux-gnu,i686-pc-mingw32)).
> >
> > Does anyone have any suggestions as to how I should investigate this?
>
>Open it in gdb and use debug_tree() on the expr that causes the error.
I usually have terrible luck with gdb on Windows.
I modified tree_class_check_failed in gcc/tree.c and added the
line debug_tree(node); before the call to internal_error.
I get this:
<complex_type 41f180
...and then a segfault.
Something seems horribly wrong. I'm not sure whether a complex_type node
is really being built or whether something is just plain hosed. I'll keep digging
but would appreciate any tips or insight. I can't understand why the cross compiler
doesn't behave this way.
-- Mohan
http://www.thisiscool.com/
http://www.animalsong.org/