From owner-freebsd-hackers Tue Nov 4 23:28:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id XAA19738 for hackers-outgoing; Tue, 4 Nov 1997 23:28:46 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from vinyl.quickweb.com (vinyl.quickweb.com [209.112.4.14]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id XAA19713; Tue, 4 Nov 1997 23:28:33 -0800 (PST) (envelope-from mark@quickweb.com) Received: (from mark@localhost) by vinyl.quickweb.com (8.8.7/8.6.12) id CAA15682; Wed, 5 Nov 1997 02:29:26 -0500 (EST) Message-ID: <19971105022925.47573@vmunix.com> Date: Wed, 5 Nov 1997 02:29:25 -0500 From: Mark Mayo To: Julian Elischer Cc: dyson@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: Re: FreeBSD shines..[Fwd: Re: semaphore speed] References: <199711040736.CAA00454@dyson.iquest.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.85e In-Reply-To: ; from Julian Elischer on Tue, Nov 04, 1997 at 12:03:09AM -0800 X-Operating-System: FreeBSD 2.2.5-STABLE i386 Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, Nov 04, 1997 at 12:03:09AM -0800, Julian Elischer wrote: > > On Tue, 4 Nov 1997, John S. Dyson wrote: > > > Julian Elischer said: > > > oops I sent this to the wrong address before.. > > > > > > participants: > > > jallison: Samba guru: > > > Andrew Tridgell: wrote Sambe.. Linux hacker > > > Julian: yours truely: > > > > > FreeBSD 2.2.2 on a 200MHz pentium > > > fcntl: 3.16579 secs > > > ipc: 2.63504 secs > > > > > > Linux 2.0.30 on a PPro200 > > > fcntl: 12.2177 secs > > > ipc: 50.4559 secs > > > > > My results with FBSD-current on PPro200 > > > > UP kernel: > > fcntl: 10.735 secs > > ipc: 9.17823 secs > > > > SMP kernel: > > fcntl: 17.8355 secs > > ipc: 9.10841 secs > > > > It appears that the 200MHz Pentium results are for a really fast > > machine :-). Is there a chance that there was an error in the > > Pentium test? Or are the results correct? (Sometimes Pentiums > > are paradoxically faster than PPro's.) I just want to make sure > > that people aren't misinformed that FreeBSD is *that* much faster > > than Linux... > > I just tried 2.2.0 (approx) on a pentium 90 > I didn't have IPC compiled in, so I just did fcntl. > yielding 7.22 secs. > looks like we slowed it down more than a bit.. Hmm. To verify, I just ran it on a PPro 150, 2.2-STABLE circa Oct 1st: fcntl: 13.9273 secs ipc: 11.8795 secs On a PPro 200, 2.2.5-STABLE from today: fcntl: 10.9368 secs ipc: 9.40272 secs Now for the interesting part, a Pentium 166MMX: fcntl: 9.64589 ipc: 7.66043 Could it be a weird PPro thing? The only other difference in the 166 kernel is that the FASTER_5X86_FPU (something like that) is turned on. Oh yeah, the kernel is 2.2.5-STABLE from today as well. I'll see if I have an older machine/kernel around and try out some more tests.. -Mark > > > > > > John > > -- ------------------------------------------------------------------------ Mark Mayo mark@vmunix.com RingZero Comp. http://www.vmunix.com/mark finger mark@vmunix.com for my PGP key and GCS code ------------------------------------------------------------------------ Win95/NT - 32 bit extensions and a graphical shell for a 16 bit patch to an an 8 bit operating system originally coded for a 4 bit microprocessor, written by a 2 bit company that can't stand 1 bit of competition. -UGU