From owner-freebsd-hackers Sun Sep 23 12:19:46 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from critter.freebsd.dk (critter.freebsd.dk [212.242.86.163]) by hub.freebsd.org (Postfix) with ESMTP id 25C3737B421 for ; Sun, 23 Sep 2001 12:19:43 -0700 (PDT) Received: from critter (localhost [127.0.0.1]) by critter.freebsd.dk (8.11.4/8.11.4) with ESMTP id f8NJJUv01622; Sun, 23 Sep 2001 21:19:30 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Matt Dillon Cc: David Greenman , Seigo Tanimura , bright@wintelcom.net, hackers@FreeBSD.ORG Subject: Re: Conclusions on... was Re: More on the cache_purgeleafdirs() routine In-Reply-To: Your message of "Sat, 22 Sep 2001 23:42:25 PDT." <200109230642.f8N6gPj84955@earth.backplane.com> Date: Sun, 23 Sep 2001 21:19:30 +0200 Message-ID: <1620.1001272770@critter> From: Poul-Henning Kamp Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG I ran one of my trivial benchmarks here: "make -j 12 buildworld" on a dual 866MHz P3 with 640M RAM. Most of the stuff reported by "/usr/bin/time -l" here is useless: almost all the numbers are all inside the standard deviation I have recorded for them on this box. Block input operations is the one notable exception and it tells a very interesting story: Matts patch results in a 4% increase, but combined with vmdirioenable it results in a 21.5% decrease. That's pretty darn significant: one out of every five I/O have been saved. The reason it has not manifested itself in the "real" number is probably the high degree of parallelism in the task which practically ensures that the CPU will not go idle. I suggest we let Matt's patch depend on the vmiodirenable sysctl and change the default for that. If there are no bad side effects found in the next couple of months, then kill the sysctl and lets be done with it. Poul-Henning -current | Matts patch | Matts patch | | vmiodirenable=1 -------------+-------------+------------------------------------------- 2206.27| 2277.97| 2190.35|real 2118.62| 2119.07| 2128.68|user 828.77| 825.03| 835.24|system 15220 | 15220 | 15216 |maximum resident set size 1006 | 1006 | 1005 |average shared memory size 1036 | 1035 | 1034 |average unshared data size 128 | 128 | 128 |average unshared stack size 10800112 | 10812895 | 10799144 |page reclaims 2530 | 3002 | 2686 |page faults 0 | 0 | 0 |swaps 36971 | 38462 | 29030 |block input operations 6694 | 6637 | 6610 |block output operations 0 | 0 | 0 |messages sent 0 | 0 | 0 |messages received 6 | 6 | 6 |signals received 17857329 | 17161716 | 17914470 |voluntary context switches 843181 | 898530 | 778408 |involuntary context switches -------------+-------------+------------------------------------------- -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message