Date: Fri, 13 Apr 2007 20:03:47 -0400 From: Kris Kennaway <kris@obsecurity.org> To: youshi10@u.washington.edu Cc: freebsd-ports@freebsd.org Subject: Re: parallel builds revisited Message-ID: <20070414000347.GB90090@xor.obsecurity.org> In-Reply-To: <Pine.LNX.4.43.0704131651330.27171@hymn09.u.washington.edu> References: <20070413154354.GP27736@potato.chello.upc.cz> <Pine.LNX.4.43.0704131651330.27171@hymn09.u.washington.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
--hHWLQfXTYDoKhP50 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Apr 13, 2007 at 04:51:33PM -0700, youshi10@u.washington.edu wrote: > My opinion is that there should be a threshold value empirically derived = by=20 > the developer / retrieved by bug reports, as well as a knob, to specify t= he=20 > maximum number of parallel jobs to be used for a particular port, that wa= y=20 > you don't get people accidentally specifying, say 10 jobs when it can onl= y=20 > handle 2-3. That's a false dichotomy; either a port correctly specifies its dependency ordering in which case it will build under all circumstances, or it does not, in which case it may randomly fail due to factors other than -j value, including system load, CPU vs I/O speed, etc. Kris --hHWLQfXTYDoKhP50 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD4DBQFGIBpjWry0BWjoQKURAlQDAJQKPATRODtEft2P5Mdx8SCOg+olAJ4qltrC 1W0lf+67IMkuJBmSBwTTIw== =BLcC -----END PGP SIGNATURE----- --hHWLQfXTYDoKhP50--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070414000347.GB90090>