Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 25 Sep 1995 13:58:01 -0600
From:      Nate Williams <nate@rocky.sri.MT.net>
To:        Terry Lambert <terry@lambert.org>
Cc:        wollman@lcs.mit.edu (Garrett A. Wollman), current@freebsd.org
Subject:   Re: kernel versions and config's rm -rf
Message-ID:  <199509251958.NAA10311@rocky.sri.MT.net>
In-Reply-To: <199509251859.LAA05603@phaeton.artisoft.com>
References:  <9509251502.AA12642@halloran-eldar.lcs.mit.edu> <199509251859.LAA05603@phaeton.artisoft.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Terry Lambert writes:
> > > The reason for this behaviour in the first place was an issue of
> > > dependencies not being calculated correctly
> > 
> > There is nothing wrong with the way the dependencies are calculated,
....

> But make no mistake: this is not a "fix", it is a "workaround".
> 
> *ONLY* the affected modules should have been rebuilt.

The major problem faced is that is near impossible to determine
dependencies on #ifdef'd code.  When an options value changes, or is
added/removed there is no way given the current scheme to know which
files should be changed.  This problem is faced by *ANY* C code
developed, and it's only solution is to make everything dependent on the
Makefile, which in term make re-compiling everything necessary for every
new config, which is a non-solution.

The issue of PTY's and others is a nice discussion point, but the you
are still only solving one problem with many more exist which can't be
solved as easily.



Nate



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199509251958.NAA10311>