[BusyBox] Re: newlib's libbb/printf.c [PATCH]

[Rob Landley]

On Wednesday 27 July 2005 08:55, Mike Frysinger wrote:
> On Wednesday 27 July 2005 03:07 am, Rob Landley wrote:
> > On Tuesday 26 July 2005 18:05, Mike Frysinger wrote:
> > > On Monday 25 July 2005 02:33 pm, Shaun Jackman wrote:
> > > > > 2005-04-05  Shaun Jackman  <sjackman at gmail.com>
> > > > >
> > > > >         * libbb/printf.c: Check for __NEWLIB_H__ before __GLIBC__.
> > >
> > > added, thanks
> > > -mike
> >
> > A) I don't suppose there's an easy way to centralize all the libc
> > dependencies into a compiler.h header file, somewhere?
>
> maybe ... that would make life easier

Not sure it's doable, but it would be nice.

Mostly it looks like we need a bb_fork() (since half the tests are to see if 
we should fork() or vfork(),) and SET_FERROR_UNLOCKED().

The loop.c mess I'm taking yet another stab at untangling now.  (It's an 
annual event, it seems.)

libbb/module_syscalls.c is...  um, icky.  (Which "very old versions" of glibc 
is it talking about?  Do we support these very old versions?  I know we 
_don't_ support libc5...  It seems that test _should_ be for the old versions 
of glibc known not to include this...)

> > B) If newlib is lying and claiming to be glibc, yet doesn't work like
> > glibc for something that we care about, why isn't it newlib that's
> > broken?
>
> you could say the same exact thing for uclibc

I agree.  But uclibc gets fixed when people point out cases where there are 
things that don't compile against it, unless those things are doing something 
actively stupid...

However, in this case all we're defining is SET_FERROR_UNLOCKED(), which is a 
lot more palatable than the four or five things I thought we were defining 
(which I hadn't noticed were commented out...)

> > C) Half this patch is whitespace noise adding spaces between the # and
> > the preprocessor directive (which I personally consider a regression, and
> > would have applied a patch to _undo_.  But that's just me...)
>
> yes, i added some whitespace changes that Shaun did not include ... they
> brought the file in sync with the rest of the whitespacing usage
>
> clearly you and i disagree on this point ... i find indented preprocessor
> macros to greatly improve readability ... i do agree it should have been a
> sep commit though

I'm being a bit grumpy lately.  I shouldn't be...

> -mike

Rob