Date: Thu, 29 Jan 2004 16:28:54 +0200 From: Ruslan Ermilov <ru@FreeBSD.org> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/vm vm_glue.c Message-ID: <20040129142854.GG19899@FreeBSD.org.ua> In-Reply-To: <6364.1075380217@critter.freebsd.dk> References: <200401291235.i0TCZBDG036160@repoman.freebsd.org> <6364.1075380217@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
--VbfcI4OLZ4XW0yH2 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Jan 29, 2004 at 01:43:37PM +0100, Poul-Henning Kamp wrote: > In message <200401291235.i0TCZBDG036160@repoman.freebsd.org>, Bruce Evans= write > s: >=20 > > A more complete fix would remove the SYSINIT obfuscation. >=20 > I don't think SYSINIT is very easy to get a grip on and that it > could be improved, but removing it would not improve the situation > IMO. >=20 > I agree that the FIRST/SECOND/ANY order thing is obscure, and that > the general ordering thing is badly thoughtout or at least horribly > implemented. >=20 > The correct solution IMO is probably to record the dependencies > explicitly, but that would require a nomenclature where we can refer > to other sysinits in a sane manner. >=20 > One short-range improvement for the contiguous kernel would be to > abandon linker-sets, and instead generate a C-function at compile > time which people could inspect to see what's going on. >=20 Time for rcorder(9)? ;) Cheers, --=20 Ruslan Ermilov FreeBSD committer ru@FreeBSD.org --VbfcI4OLZ4XW0yH2 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFAGRimUkv4P6juNwoRAg7HAJ0RkinzoIl9bLIZ6lMlSQGA+l6sUwCfV3aR k8ZY5Qzim1EavS2PPsrAuTM= =w18X -----END PGP SIGNATURE----- --VbfcI4OLZ4XW0yH2--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040129142854.GG19899>