-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tuesday June 17 2003 12:02, Allen Briggs wrote:
> On Tue, Jun 17, 2003 at 12:02:59PM -0500, Nate Hill wrote:
> > I'm a recent -current user (running first build now) and I'm wondering
> > about the frequency of toolchain updating versus distribution building.
> > Due to some ambiguities in documentation, I'm wondering if the toolchain
> > I built the first time would survive another source update/build.  In
> > general, I'm wondering how often I should update the toolchain if I plan
> > on updating the source tree and distribution approximately every two
> > weeks.
>
> If you're looking for a time-based method, "every time".
>
> A simple optimization to that is, "when any of the tools change."
>
> The most accurate answer is, "when any of the tools change in such a way
> that it affects the build."
> Obviously, it's not always obvious to identify the last--especially if
> you're not reading everything on source-changes.  Major changes will be
> noted and probably discussed here, though.
>
> To ramble a bit more: If you have a reasonably fast box, and you're
> updating every two weeks, just go ahead and build the toolchain every
> time.  On x86, the kernel complies take longer than anything else.
>
> -allen

Well, that's the answer I already knew.  I guess I'm wondering if there is an
easy way to identify the last answer.  I've always found that kernel compiles
are relatively fast on my machine and that the toolchain was an enormously
large build.  I guess a second question at this point is wether make is smart
enough to only update parts of the toolchain that change - anyone have an
answer to that?

- - --
Nate Hill <vugdeox@freeshell.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (NetBSD)

iD8DBQE+71Ox69yeq3Xq/RARAmgkAJwKVyWbHsR3cqKL4DTt9NW+0/U8wgCghfmw
Ac5IXxRFrTZEaAyyhPgT60U=
=5tWM
-----END PGP SIGNATURE-----