Re: GNUMach 1.x fixed (autoconf+kb)

[Marco Gerards]

Jeff Bailey <jbailey@nisa.net> writes:

> Marco, thanks for posting the patch on Savannah.  I did this work before
> as part of the automake'ifying, so I'll compare it against what I have
> and also test a bootstrap with it.

Thanks a lot!  It is weird that there was no report send to bug-hurd
(yet) about the patch I've uploaded.  Anyway, the patch that I've
uploaded is without the generated files.

Jeff, what is the status of your automake'ifying work for the Hurd?

> It'll take me a couple of days.
> 
> An initial comment is that:
> 
>  AC_ARG_ENABLE(kmsg,
>  [  --enable-kmsg            enable use of kmsg device [default=yes]],
> 
> Should be updated to look like
> 
>  AC_ARG_ENABLE([kmsg],
>       AC_HELP_STRING([--enable-kmsg],
>               [enable use of kmsg device [default=yes]], ...

Ok, I will fix that.

> Also, I think the _LOCAL functions may have to be preserved to permit
> bootstrapping.  GCC faced the same problem, so they may have gotten the
> fixes into upstream autoconf.

Ok, I must admit that I did not have a close look at these _LOCAL
functions.  They were removed in oskit-mach.  I assume Roland would
not remove it if it was not ok to do so.

Thanks,
Marco