From owner-freebsd-ports@FreeBSD.ORG Sat Apr 14 00:00:33 2007 Return-Path: X-Original-To: ports@freebsd.org Delivered-To: freebsd-ports@FreeBSD.ORG Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 77E5A16A410; Sat, 14 Apr 2007 00:00:33 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id 568D213C46E; Sat, 14 Apr 2007 00:00:33 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id 0FCAD1A4D93; Fri, 13 Apr 2007 17:00:43 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 0F50E51565; Fri, 13 Apr 2007 20:00:32 -0400 (EDT) Date: Fri, 13 Apr 2007 20:00:31 -0400 From: Kris Kennaway To: "[LoN]Kamikaze" Message-ID: <20070414000031.GA90090@xor.obsecurity.org> References: <200704100452.40574.mail@maxlor.com> <1176391950.1820.3.camel@rnoland-ibm.acs.internap.com> <461E5F39.1030603@gmx.de> <200704122056.51123.mail@maxlor.com> <461E8919.1080702@gmx.de> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="MGYHOYXEY6WxJCY8" Content-Disposition: inline In-Reply-To: <461E8919.1080702@gmx.de> User-Agent: Mutt/1.4.2.2i Cc: Benjamin Lutz , ports@freebsd.org, Robert Noland , pav@freebsd.org Subject: Re: parallel builds revisited X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 14 Apr 2007 00:00:33 -0000 --MGYHOYXEY6WxJCY8 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Apr 12, 2007 at 09:31:37PM +0200, [LoN]Kamikaze wrote: > > Btw, do you think it's possible that a port can only be built with, n= =20 > > parallel make jobs, but will fail with n+1? >=20 > No. I do not think this can be the case. It certainly is the case. If a makefile has incorrectly specified dependencies then it may build by accident for certain values of n. There are also lots of ports with race conditions which means they may sometimes build but not always. For both of these reasons, once the infrastructure support is in place, I don't want to see an indiscriminate rush to mark ports as "make -j safe", because it will cause more problems than it solves. It should only be done after careful testing by a port's maintainer who promises to investigate and fix the race conditions that may be exposed. Kris --MGYHOYXEY6WxJCY8 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFGIBmfWry0BWjoQKURAoF2AKChz3FzH4qqfK1wobdiq9s4Zw1N1ACfUijg dvf8/hlneDvlM6kwZydgGRs= =wUXL -----END PGP SIGNATURE----- --MGYHOYXEY6WxJCY8--