>
> On Wed, Jun 09, 1999 at 12:40:46AM +0100, Brian Somers wrote:
> > >
> > > Can someone comment please? Is this a bug in the way the gcc2.8 is
> > > installed, or is it a bug in my understanding? (probably the latter).
> >
> > Perhaps you need a gcc-compiled version of libstdc++. It's just a
> > guess, but when we shifted to egcs, there were all sorts of problems
> > linking against the gcc-compiled version.
> >
>
> Ok. I've compiled up a 4.0-CURRENT box, with EGCS native, and recompiled the
> program. It still crashes, this time with:
>
> Core was generated by `search'.
> Program terminated with signal 11, Segmentation fault.
> Reading symbols from /usr/lib/libstdc++.so.3...done.
> Reading symbols from /usr/lib/libm.so.2...done.
> Reading symbols from /usr/lib/libc.so.3...done.
> Reading symbols from /usr/libexec/ld-elf.so.1...done.
> #0 0x8053169 in __get_eh_info () at /usr/include/ctype.h:149
> 149 }
> (gdb) bt
> #0 0x8053169 in __get_eh_info () at /usr/include/ctype.h:149
> #1 0x8053156 in __get_eh_info () at /usr/include/ctype.h:149
> #2 0x8053132 in __get_eh_context () at /usr/include/ctype.h:149
> #3 0x8059fda in my_set<char const *>::my_set (this=0x805ff9c) at search.c:64
> #4 0x804d3b9 in global constructors keyed to files () at search.c:64
> #5 0x804a5d8 in _start ()
> #6 0x804a25d in _init ()
> (gdb)
>
> >From search.c:
> 60 char const* me; // executable name
> 61 file_index files;
> 62 word_index words, stop_words, meta_names;
> 63 bool stem_words;
> 64 string_set stop_words_found;
> 65
> 66 void dump_single_word( char const *word );
> 67 void dump_word_window( char const *word, int window_size, int match
> );
>
> I'm very confused... the programmer is convinced that it works under
> other platforms, but I'm not getting any joy out of it :(
>
> Joe
Sorry I can't be of more help...
I'd have to add that I have a suspicion that something is still not
right "library-wise"... that is, the G++ library isn't built right,
or the program isn't linking right.
For what it's worth - the inline __XXX functions in ctype.h do work
correctly... So, I don't believe the problem isn't in the "source" per
se...
I'd suggest building the library with debugging enabled, linking
with that and determining what is wrong.
- Dave Rivers -
To Unsubscribe: send mail to majord...@freebsd.org
with "unsubscribe freebsd-hackers" in the body of the message
