From owner-freebsd-stable Thu Sep 20 9:31:29 2001 Delivered-To: freebsd-stable@freebsd.org Received: from clyde.goodleaf.net (piscator.seanet.com [199.181.165.218]) by hub.freebsd.org (Postfix) with ESMTP id 9535437B414 for ; Thu, 20 Sep 2001 09:31:20 -0700 (PDT) Received: by clyde.goodleaf.net (Postfix, from userid 1001) id 8064D5C74; Thu, 20 Sep 2001 09:32:47 -0700 (PDT) References: <200109170009.f8H09Y921584@gits.dyndns.org> <20010920153546.8D69937B411@hub.freebsd.org> In-Reply-To: <20010920153546.8D69937B411@hub.freebsd.org> From: "J. Goodleaf" To: stable@freebsd.org Subject: Re: make -j4 vs -j8... 4 works, but 8 does not Date: Thu, 20 Sep 2001 16:32:47 GMT Mime-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <20010920163247.8064D5C74@clyde.goodleaf.net> Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I'm sorry to interject into this cool benchmark discussion. I have a non-critical issue relating to this. I have a dual PII 350 machine with 256 RAM (once badass, now...) Interestingly, make -j4, 6, 8 all return errors while building the perl libraries. make -j10, however, works flawlessly. I'm not really alarmed by this, but I am curious. Any ideas? (Sadly, I did not keep the error log from my last attempt, or I would have attached it.) -John Gerald A.Speak writes: > On Sunday 16 September 2001 08:09 pm, Cyrille Lefevre wrote: >> Erik Trulsson wrote: >> > On Sat, Sep 15, 2001 at 06:06:10PM -0400, Garance A Drosihn wrote: >> > > [aside on those multiple builds that I did: It was interesting that >> > > even though it was on a dual-processor system, there was not much of >> > > a speed improvement (on 4.3-stable) when going from -j4 to -j10. Big >> > > improvement going from -j1 to -j4, but after that it didn't help much] >> > >> > That sounds exactly like what I would expect. More than 2 (or maybe 3) >> > jobs per CPU is normally not going to make things go faster but might >> > actually slow it down. >> >> I'm not right w/ you. let's try : >> >> # setup a reusable environment >> cd /sys/compile/MYKERNEL >> make depend >> make >> # run the jobs >> for i in 1 2 4 8; do >> (date;touch opt_global.h;time make -j$i all;date) >| .j$i 2>&1 >> done >> >> results are (on my old P200 processor) : >> >> j1 12:36 >> j2 12:46 >> j4 11:36 >> j8 11:21 >> >> so, for a 12 min job, you could win 1 min using -j4, which may gives >> you 1 hour for a 12 hours job (such as buildworld :). ok, it's not so >> many, but that may help a lot. for instace, I currently use -j4 since >> -j8 is too CPU and memory intensive (the machine slow down significally). >> >> PS : I'm using SCSI disks. using IDE disks probably gives you different >> results. I/O are very differents between the two. >> >> Cyrille. > > > I had some time to spend and came up with the following results for doing > make build world. Before each build I made sure that /usr/obj/ was empty. > > time make buildworld... > 1800.325u 818.652s 44:35.38 97.8% 1322+1456k 9130+5803io 555pf+0w > > time make -j 2 buildworld ... > 1852.919u 957.381s 44:09.71 106.0% 1305+1471k 9231+5677io 555pf+0w > > time make -j 4 buildworld ... > 1881.497u 965.330s 36:17.07 130.7% 1314+1472k 9307+6329io 626pf+0w > > time make -j 6 buildworld ... > 1884.092u 973.077s 35:25.23 134.4% 1314+1473k 9246+6096io 555pf+0w > > time make -j 8 buildworld ... > 1879.997u 977.772x 35:27.64 134.3% 1310+1469k 9240+6310io 555pf+0w > > As you can see there was only a matter of a couple minutes improvement when > using -j 8 instead of -j 4. > > The build machine is a Dual PIII 1000, with 1GB RAM and SCSI Ultra160 Drive > (too bad FreeBSD only runs the Adaptec 29160N at 80MHz. > dmesg below for those interested. > > Copyright (c) 1992-2001 The FreeBSD Project. > Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 > The Regents of the University of California. All rights reserved. > FreeBSD 4.4-STABLE #0: Thu Sep 20 10:17:57 EDT 2001 > gaspeak@sampson.gaspeak-p-va.prestige.net:/usr/obj/usr/src/sys/SAMPSON > Timecounter "i8254" frequency 1193182 Hz > CPU: Pentium III/Pentium III Xeon/Celeron (1004.50-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x68a Stepping = 10 > > Features=0x383fbff > real memory = 1073725440 (1048560K bytes) > config> q > avail memory = 1041256448 (1016852K bytes) > Programming 24 pins in IOAPIC #0 > IOAPIC #0 intpin 2 -> irq 0 > FreeBSD/SMP: Multiprocessor motherboard > cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 > cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 > io0 (APIC): apic id: 2, version: 0x00178011, at 0xfec00000 > Preloaded elf kernel "kernel" at 0xc0464000. > Preloaded userconfig_script "/boot/kernel.conf" at 0xc046409c. > Preloaded elf module "splash_bmp.ko" at 0xc04640ec. > Preloaded elf module "vesa.ko" at 0xc0464190. > Preloaded splash_image_data "/boot/splash.bmp" at 0xc046422c. > Preloaded elf module "vn.ko" at 0xc046427c. > Preloaded elf module "snd_es137x.ko" at 0xc0464318. > Preloaded elf module "snd_pcm.ko" at 0xc04643bc. > Preloaded elf module "usb.ko" at 0xc046445c. > Preloaded elf module "ugen.ko" at 0xc04644f8. > Preloaded elf module "uhid.ko" at 0xc0464594. > Preloaded elf module "ulpt.ko" at 0xc0464630. > Preloaded elf module "ums.ko" at 0xc04646cc. > Preloaded elf module "umass.ko" at 0xc0464768. > Preloaded elf module "uscanner.ko" at 0xc0464808. > Preloaded elf module "agp.ko" at 0xc04648a8. > Preloaded elf module "vpo.ko" at 0xc0464944. > VESA: v3.0, 16384k memory, flags:0x1, mode table:0xc0338437 (1000117) > VESA: 3dfx Interactive, Inc. > Pentium Pro MTRR support enabled > md0: Malloc disk > Using $PIR table, 7 entries at 0xc00f12d0 > npx0: on motherboard > npx0: INT 16 interface > pcib0: on motherboard > IOAPIC #0 intpin 18 -> irq 2 > IOAPIC #0 intpin 17 -> irq 5 > IOAPIC #0 intpin 16 -> irq 9 > IOAPIC #0 intpin 19 -> irq 10 > pci0: on pcib0 > agp0: mem 0xf8000000-0xfbffffff > at device 0.0 on pci0 > pcib2: at device 1.0 on > pci0 > pci1: on pcib2 > pci1: <3Dfx Voodoo 3 graphics accelerator> at 0.0 irq 9 > isab0: at device 4.0 on pci0 > isa0: on isab0 > atapci0: port 0xb800-0xb80f at device 4.1 on > pci0 > ata0: at 0x1f0 irq 14 on atapci0 > ata1: at 0x170 irq 15 on atapci0 > uhci0: port 0xb400-0xb41f irq 2 at device 4.2 on > pci0 > usb0: on uhci0 > usb0: USB revision 1.0 > uhub0: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 > uhub0: 2 ports with 2 removable, self powered > uhub1: Microsoft product 0x001c, class 9/0, rev 1.10/5.00, addr 2 > uhub1: 3 ports with 2 removable, bus powered > uhci1: port 0xb000-0xb01f irq 2 at device 4.3 on > pci0 > usb1: on uhci1 > usb1: USB revision 1.0 > uhub2: VIA UHCI root hub, class 9/0, rev 1.00/1.00, addr 1 > uhub2: 2 ports with 2 removable, self powered > pci0: (vendor=0x11d1, dev=0x01f7) at 10.0 irq 2 > pcm0: port 0xa800-0xa83f irq 5 at device 11.0 on pci0 > de0: port 0xa400-0xa47f mem > 0xf1000000-0xf100007f irq 9 at device 12.0 on pci0 > de0: 21140A [10-100Mb/s] pass 2.2 > de0: address 00:40:05:a2:d8:ea > ahc0: port 0xa000-0xa0ff mem > 0xf0800000-0xf0800fff irq 10 at device 13.0 on pci0 > aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/255 SCBs > pcib1: on motherboard > pci2: on pcib1 > orm0: