Date: Tue, 20 Mar 2001 20:05:26 -0800 From: Kent Stewart <kstewart@urx.com> To: nomad@netrail.net Cc: freebsd-questions@freebsd.org Subject: Re: Making & forking processes to accomplish faster compiles Message-ID: <3AB82886.27BE3136@urx.com> References: <MPEGJCJPPBKNCNBGOHGDIEECCNAA.cschreiber@netrail.net>
next in thread | previous in thread | raw e-mail | index | archive | help
"Christian S." wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Greets, > I was told by a friend of mine that he had a way of "make"ing and > having some sort of option to fork the resultant processes so that > instead of 1 make process running, one could split it into 4 separate > processes that would accomplish the "make" faster. I read the man > page for make, but did not find anything really related to it, > although it is entirely possible that I could have missed it. If > someone could kindly point me to a URL or let me know if it was even > possible, I would appreciate it. I would ask the friend, but I have > since lost touch with him. See if the "-j4" parameter fits the description of what they did. BTW, that only works when the HD I/O is faster than the cpu. When the cpu is starved for I/O and you use a -j4, the build takes much longer. To test what it does do your buildworld using "time make -j4 buildworld". You have to time the build with out the -j4 and with the -j4. When I timed both, the results I saw were as follows: buildworld with files on 3 controllers and -j4 with softupdates 1539.114u 521.486s 45:54.82 74.7% 1209+1431k 48857+129907io 1858pf+0w buildworld with files on 3 controllers with softupdates and no -j4 1524.149u 431.967s 38:41.07 84.2% 1245+1453k 49804+11665io 1915pf+0w It is fairly repeatable. The build without -j4 was slightly faster for both u&s times, and it ran 17% faster by the wall clock. The system is an AMD 900 Thunderbird with 256MB of PC-133 memory and 3x30GB ATA-100 drives on separate controllers. I have a 2nd system with dual 866 coppermines and they will do a buildworld with -j4 in 34 minutes. Anything different compiles with a longer time. Kent > > Thanks! > Christian > > "...we have only twice as many genes as a fruit fly, or roughly the > same number as an ear of corn, about 30,000." > Ergo, we are all corn. > > -----BEGIN PGP SIGNATURE----- > Version: PGPfreeware 6.5.8 for non-commercial use <http://www.pgp.com>; > > iQA/AwUBOrgiTikK9qTvGvteEQIO1gCeLV9Fts96IMfhLUNheOp6jwow+vIAoLVJ > 4OKxfZC3XsJTnaVsNfAWWmc4 > =jzRH > -----END PGP SIGNATURE----- > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message -- Kent Stewart Richland, WA mailto:kbstew99@hotmail.com http://kstewart.urx.com/kstewart/index.html FreeBSD News http://daily.daemonnews.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3AB82886.27BE3136>