cvs commit: src/usr.bin/make make.1

[Ceri Davies]

On Wed, Oct 12, 2005 at 10:43:00AM -0600, Warner Losh wrote:
> From: Ruslan Ermilov <ru at freebsd.org>
> Subject: Re: cvs commit: src/usr.bin/make make.1
> Date: Wed, 12 Oct 2005 18:27:10 +0300
> 
> > On Wed, Oct 12, 2005 at 09:13:30AM -0600, M. Warner Losh wrote:
> > > In message: <200510121009.j9CA9aE3026075 at repoman.freebsd.org>
> > >             Yar Tikhiy <yar at FreeBSD.org> writes:
> > > : yar         2005-10-12 10:09:36 UTC
> > > : 
> > > :   FreeBSD src repository
> > > : 
> > > :   Modified files:
> > > :     usr.bin/make         make.1 
> > > :   Log:
> > > :   __MAKE_CONF doesn't really belong here because it is
> > > :   a FreeBSD extension of sys.mk.  A xref to make.conf(5)
> > > :   will be enough here.
> > > :   
> > > :   Requested by:   ru
> > > 
> > > I disagree.  It is already hard enough to find info about __MAKE_CONF,
> > > and since it is part of the base system, this seems like an artificial
> > > distinction.
> > > 
> > __MAKE_CONF doesn't fall under "make sets or knows about the following
> > internal variables or environment variables".  Rather, it's a FreeBSD
> > specific feature, it doesn't have any direct connection to the make
> > utility (as well as CPUTYPE, CFLAGS, etc.).  As such, it shouldn't
> > be documented in the make(1) manpage.  OTOH, build(7) could benefit
> > from talking more about make.conf(5), while having __MAKE_CONF only
> > documented in make.conf(5) is fine.  We really don't need any more
> > duplication.
> 
> I disagree. It is directly connected to the make(1) utility, just like
> all the FreeBSD extentions we've added over the years for things like
> 'expand this variable, and make it upper case'.  Every single
> invocation of make(1) will cause __MAKE_CONF to be evalutated, and
> /etc/make.conf included if __MAKE_CONF isn't defined.  It is very much
> unlike CPUTYPE, CFLAGS, etc because of this.  As such, it should be in
> the make(1) man page.  Every user of make potentially has to know
> about it.  It is a variable that we've made fundamental to our make
> system, so it should be at least mentioned in make(1).  There's no
> harm in putting it make(1), and people absolutely will look there
> first for this information.  A simple xref to make.conf isn't
> sufficient.

It's more important than CPUTYPE, etc. for the reasons outlined above.

Ceri
-- 
Only two things are infinite, the universe and human stupidity, and I'm
not sure about the former.			  -- Einstein (attrib.)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/cvs-src/attachments/20051012/f90f0368/attachment.bin