Date: Thu, 13 Mar 2008 13:30:05 +0100 From: "Luca Presotto" <Luca.Presotto@cern.ch> To: "Julius Huang" <juliushuang@gmail.com> Cc: marcin.koziuk@planet.nl, freebsd-questions@freebsd.org Subject: RE: Tuning make.conf Message-ID: <9A6A62B6B84859469F3EBB5F09D818CA219E9B@cernxchg50.cern.ch> References: <Pine.LNX.4.58.0803130937530.5944@lxplus098.cern.ch> <47D9004D.5070407@planet.nl> <9A6A62B6B84859469F3EBB5F09D818CA219E98@cernxchg50.cern.ch> <87F851FC-750A-478F-BD01-1B7FED69BF7A@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>Isn't -j depend on how many cpu/core and a faster harddisk / raid? Yes, it is. But with j1 you have only one job running at a time. On a = dual core you can really easily running at least two jobs at the same = time. Then I've read a number of ideas about which is the relation between the = number of cores and the optimal number of jobs. I have been suggested something between n+1 and 2n+1 Of course the optimal number of jobs depends on the disk speed and = similar. But switching between 1 and 3 gives something like halving the time = needed to compile everything. Maybe it's possible that when building the kernel it gives some problem, = but I'm thinking about compiling ports. Or does portupgrade automatically chooses which -j to use?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9A6A62B6B84859469F3EBB5F09D818CA219E9B>