On Wed, May 22, 2002 at 10:43:23PM +0200, Antonio Martínez wrote:
> William Harold Newman <william.newman@...> writes:
> > As part of an unhealthy Lisp obsession, I'm trying to build SBCL 0.7.3
> > with CMU CL 2.4.19 under Debian 2.2 (on an AMD K7) and I'd appreciate
> > it a lot if anyone has time to send along any help with the following
> > problem.
> [ Good debugging advice snipped. ]
> CMUCL's COMPILE-FILE seems to return a failure code as the fourth
> value on compiling a file which has unused variables. This sets
> FAILURE-P, as you suggested, which breaks the proposed build.
> I don't know how to change this behaviour of CMUCL, so my simplest
> workaround was to just push ahead with the compilation, using an
> interactive lisp (i.e. using 'lisp -noinit' rather than 'lisp -batch
> -noinit' as the compiler) and choose the CONTINUE restart when the
> debugger was entered.
Another, more persistent way to work around bogus warnings is to set
the :IGNORE-FAILURE-P flag in build-order.lisp-expr.
> - added appropriate IGNORE declaration to src/code/late-type.lisp.
That is of course an even better way to avoid warnings. (CMU CL is
still wrong to treat unused variables as full warnings, since ANSI
specifically forbids that behavior; but that doesn't mean it's not
good for us to declare our unused variables.)
> Here are the diffs, which I may have buggered up, since I don't do
> diffs very often. If so, just flame me, and I'll see if I can send
> better diffs.
The patch looks reasonable to me. I'll test it and presumably merge it
sometime after I release 0.7.4 (which will probably be on Friday).
> Thanks for the help and for SBCL!
You're welcome, and thank you for the bug report and the patch.
William Harold Newman <william.newman@...>
"When I am picking problems to work on, ones that stumped John von
Neumann go at the bottom of the stack."
-- <http://www.daviddfriedman.com/laws_order/index.shtml>, ch. 8
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C B9 25 FB EE E0 C3 E5 7C