a few short observations on latest BB

[Rob Landley]

On Wednesday 01 March 2006 5:19 am, Robert P. J. Day wrote:
>   first (and most pedantically), when you originally install the
> source and do your first "make *config", every single option is
> labelled as "(NEW)", simply because there is no previous record of it.
>
> from scripts.config.conf.c:
>
> =================================
> static void conf_askvalue(struct symbol *sym, const char *def)
> {
>         enum symbol_type type = sym_get_type(sym);
>         tristate val;
>
>         if (!sym_has_value(sym))
>                 printf("(NEW) ");
> ...
> =================================
>
>   but labelling everything as NEW is not all that informative that
> first time, and doesn't follow the tradition of the linux kernel in
> *not* doing that.  would it be a big deal to add an extra conditional
> to not label everything as NEW if there's no existing .config file?

We inherited that from the linux kernel.  I'll yank it when I resync with 
2.6.16 to get something I can apply the miniconfig patch against.

>   next, under "Installation Options," the selection of "Don't use
> /usr" might benefit from a slightly more expressive help message to
> explain what that option really represents, for the sake of newbies
> who don't know what it means.

I have no idea what it's for either.  (I recently made it so defconfig doesn't 
select that one, by the way, since it says not to unless you know you need 
it...)

Rob
-- 
Never bet against the cheap plastic solution.