Date: 17 Jul 2000 02:50:08 +0200 From: Cyrille Lefevre <clefevre%no-spam@citeweb.net> To: Kris Kennaway <kris@FreeBSD.org> Cc: Joachim =?iso-8859-1?q?Str=F6mbergson?= <watchman@ludd.luth.se>, clefevre@citeweb.net, stable@FreeBSD.org Subject: Re: Parallel kernel make Was Re: kernel.debug (was Re: HEADS UP! Always use the 'make buildkernel' target to make your kernels) Message-ID: <d7kdzikf.fsf@pc166.gits.fr> In-Reply-To: Kris Kennaway's message of "Sun, 16 Jul 2000 15:27:04 -0700 (PDT)" References: <Pine.BSF.4.21.0007161525200.89152-100000@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Kris Kennaway <kris@FreeBSD.org> writes: > On Sun, 16 Jul 2000, Joachim [iso-8859-1] Str=F6mbergson wrote: >=20 > > Just for kicks I tried doing a 'make -j4 buildkernel' with KERNEL=3DSNP= 40 > > in /etc/make.conf and /boot/loader.conf. It works, that is the make > > finished without crashing. I dunno if the binary works, nor if the > > compile was substantially faster. So, superficially, parallel make of a > > kernel does work. >=20 > Yes, there is no reason why this should not work. >=20 > > It sure consumed a lot of resources for a few minutes though. My SETI > > run ran sloowly. >=20 > Well, wouldn't you kind of expect this given what you asked the system to > do (namely run 4 tasks at once)? However, -j4 is often slightly faster > even on uniprocessor systems than -j1 (or "no -j") because it doesn't have > to wait on Disk I/O. YMMV, of course. I do make -j4 buildworld and kernel for month w/o any problems and is really faster than -j1. even on a P166 uniprocessor. Cyrille. --=20 home:mailto:clefevre%no-spam@citeweb.net Supprimer "%no-spam" pour me repon= dre. work:mailto:Cyrille.Lefevre%no-spam@edf.fr Remove "%no-spam" to answer me b= ack. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d7kdzikf.fsf>