Re: [Gcl-devel] Problems building gcl-2.6.1-18 on Solaris 8 (SPARC)

[Camm Maguire]

Greetings!  Have to say I've never seen this before, and that it is
likely independent of GCL.  I would suspect your libc, not gcc, no?
In any case, you can narrow this down a bit further by executing the
(bye) at the initial lisp prompt before anything else, and then
(seoncdly) looking in main.c where fLbye is defined, and putting those
commands in a dummy C file, compile with -g and run.  I'm guessing
that a simple int main() {exit(0);} might even show the same error.

Take care,

John Tang Boyland <address@hidden> writes:

> It looks like a fault deep in gcc-lib.
> So probably not GCL's fault?  (Unless some very serious
> pointer problem happened).  Maybe I just need a new gcc.
> (argh... it takes so long to install gcc.)
> 
> ....
> ;; Note: Tail-recursive call of SEQUENCE-CURSOR was replaced by iteration.
> End of Pass 2.  
> OPTIMIZE levels: Safety=2, Space=3, Speed=3
> Finished compiling gcl_arraylib.lsp.
> #p"gcl_arraylib.o"
> 
> >(bye)
> 
> Program received signal SIGSEGV, Segmentation fault.
> 0x28fe4 in __do_global_dtors_aux ()
> (gdb) where
> #0  0x28fe4 in __do_global_dtors_aux ()
> #1  0x1755c8 in _fini ()
> #2  0xff19bc94 in _exithandle ()
> #3  0xff21f8d4 in exit ()
> ...
> 
> The full output is in
>       http://www.cs.uwm.edu/~boyland/temp/gcl-2.6.1-18-debug-build.txt
> (This time, I shut it off the build before it got any further.)
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah