From owner-freebsd-hackers Thu May 24 23:20:12 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from dt051n37.san.rr.com (dt051n37.san.rr.com [204.210.32.55]) by hub.freebsd.org (Postfix) with ESMTP id 325F437B423 for ; Thu, 24 May 2001 23:20:09 -0700 (PDT) (envelope-from DougB@DougBarton.net) Received: from DougBarton.net (master [10.0.0.2]) by dt051n37.san.rr.com (8.9.3/8.9.3) with ESMTP id XAA21046; Thu, 24 May 2001 23:19:44 -0700 (PDT) (envelope-from DougB@DougBarton.net) Message-ID: <3B0DF980.EDA844F7@DougBarton.net> Date: Thu, 24 May 2001 23:19:44 -0700 From: Doug Barton Organization: Triborough Bridge & Tunnel Authority X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.12 i386) X-Accept-Language: en MIME-Version: 1.0 To: Ed Hudson Cc: hackers@FreeBSD.ORG Subject: Re: general speed differences between 4.1.1-RELEASE and 4.3-RELEASE References: <200105250025.f4P0Pu905553@m44.spnet.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit 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 Ed Hudson wrote: > > fyi, here's another hw.ata.wc=1 vs hw.ata.wc=0 comparison: > > 4.3-RELEASE install, ASUS A7V, 800mhz, hw.ata.wc=0, express install +all, > 60gig wd-600b udma100 drive, partitioned as: > > / 8192m > swap 1024m > /xtra 48023m > > installed from 52x cdrom. > install time: 27minutes, 38 seconds (1658 seconds). > > a modified 4.3-RELEASE install iso (modified by replacing /floppies/boot.flp > loader.rc/support.4th and adding a loader.conf, setting hw.ata.wc=1) file, > written to cd. > > same partition, same install (express + all) > > install time: 8minutes, 59 seconds ( 539 seconds) > > 4.3-with-wc=1 / 4.3-with-wc=0 : 3.07x > > in both installation cases both filesystems were newfs'd, but the > speed difference is evident everywhere during the install. > > i would urge the FreeBSD lords to consider mr. silbersack's proposal > of either remaking hw.ata.wc=1 the default, The current mood (which I agree with) is to make softupdates the default after installation. The problem with the combo of write caching and softupdates is that if the power actually goes off the meta-data writes that softupdates postpones and are further postponed by the write cache will never happen, therefore leaving the file system in a potentially unrecoverable state. > or at least, consider making this the default for the base install. This would work for situations where it's being installed to a clean disk, but might be trouble when installing over an existing installation. The problem (as I understand it) is that the write cache option can only be enabled at boot time. If I'm wrong about this, it'd be a good thing to enable if we're sure we're newfs'ing the partitions we're installing to. Doug -- I need someone really bad. Are you really bad? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message