Re: Yow: Maciej.

[Jim Blandy <jimb@red-bean.com>]

> This is a minor correction, but there are also additional warnings if
> you configure --with-threads. There are a bunch of missing prototypes,
> but some of the ones the compiler complains about seem to exist. I
> could not untangle the mess in my limited time, because I have much
> more broken builds to fix at work. :-)

Right --- Greg Harvey has sent in patches for some of those, which
I'll look at tonight.

> Someone should probably try building with egcs or gcc 2.8 as well, one
> of these might well become the standard compiler for building things
> like Red Hat packages in the foreseeable future.

I have something similar to EGCS (the Cygnus 97r1 release) on my
work machine, so I can hack on that.  I've amended the HACKING
guidelines to something that acknowledges this, and tries to set a
policy that's easy to follow:


    - The Guile tree should compile without warnings under the following
    GCC switches, which are the default in the current configure script:
	-Wall -Wpointer-arith -Wmissing-prototypes
    The only exceptions are the warnings about variables being clobbered
    by longjmp/vfork in eval.c.  (Tho' if you can figure out how to get
    rid of those, too, I'd be happy.)

    Note that the warnings generated vary from one version of GCC to the
    next, and from one architecture to the next (apparently).  To provide
    a concrete common standard, Guile should compile without warnings from
    GCC 2.7.2.3 in a Red Hat 5.0 i386 Linux machine.  Furthermore, each
    developer should pursue any additional warnings noted by on their
    compiler.  This means that people using more stringent compilers will
    have more work to do, and assures that everyone won't switch to the
    most lenient compiler they can find.  :)