From owner-freebsd-current Thu May 14 01:26:24 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA06170 for freebsd-current-outgoing; Thu, 14 May 1998 01:26:24 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA06164 for ; Thu, 14 May 1998 01:26:21 -0700 (PDT) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id IAA02122; Thu, 14 May 1998 08:25:16 GMT Date: Thu, 14 May 1998 17:25:15 +0900 (JST) From: Michael Hancock To: Mikhail Teterin cc: current@FreeBSD.ORG Subject: Re: soft updates and async In-Reply-To: <199805140430.AAA05562@rtfm.ziplink.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, 14 May 1998, Mikhail Teterin wrote: > Ok, it turns out you need the filesystem mount sync to use softupdates. > > My understanding was: softupdates are faster then sync and safer > then sync. Is not async still faster, even at the expense of safety? Softupdates are sometimes faster and sometimes slower. It's faster when the combination of operations allows you to skip writing to disk at all. It's slower when you have to undo-then-write and redo-then-write. But I don't think it is really ever significantly slower than async writes. Softupdates does use more memory so this would probably be the only reason to use async. > Will I be able to combine async and softupdates in the future and why > would I want to? No, they are mutually exclusive. If you do a write it is either sync (default), async, or delayed (softupdates). You'd have no reason to combine them unless you want to get undefined behavior. Regards, Mike Hancock To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message