From owner-freebsd-stable Tue Jul 11 19:13:58 2000 Delivered-To: freebsd-stable@freebsd.org Received: from mail2.uniserve.com (mail2.uniserve.com [204.244.156.10]) by hub.freebsd.org (Postfix) with ESMTP id 69E0E37B673 for ; Tue, 11 Jul 2000 19:13:55 -0700 (PDT) (envelope-from tom@uniserve.com) Received: from shell.uniserve.ca ([204.244.186.218]) by mail2.uniserve.com with esmtp (Exim 3.13 #1) id 13CC22-0009C3-00; Tue, 11 Jul 2000 19:13:51 -0700 Date: Tue, 11 Jul 2000 19:13:45 -0700 (PDT) From: Tom X-Sender: tom@shell.uniserve.ca To: Paul Coyne Cc: freebsd-stable@freebsd.org Subject: Re: Cached versus non cached disk I/O In-Reply-To: <396B8DC4.28CD9727@geodesic.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, 11 Jul 2000, Paul Coyne wrote: > Perhaps not really a STABLE question, sorry - but the expertise level is > good here. > > Can anyone point me to where I might find information if and when disk > I/O can/will be cached by FreeBSD, e.g.: > > Is a mkdir cached? mkdir() updates metadata. All metadata updates are written to disk before returning, whether on a local or NFS filesystem. NFS can be put into async mode, so the server will ack all requests (metadata updates, as well as file writes) as soon as they are received (sysctl -w vfs.nfs.async=1). The filesystem can be put into async (or softupdates) mode so that local metadata update functions return before the metadata is actualy written to disk. Tom Uniserve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message