From owner-freebsd-hackers Tue Dec 15 07:30:35 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA09227 for freebsd-hackers-outgoing; Tue, 15 Dec 1998 07:30:35 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from terra.Sarnoff.COM (terra.sarnoff.com [130.33.11.203]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id HAA09222 for ; Tue, 15 Dec 1998 07:30:33 -0800 (PST) (envelope-from rminnich@Sarnoff.COM) Received: (from rminnich@localhost) by terra.Sarnoff.COM (8.6.12/8.6.12) id KAA22633; Tue, 15 Dec 1998 10:29:37 -0500 Date: Tue, 15 Dec 1998 10:29:36 -0500 (EST) From: "Ron G. Minnich" X-Sender: rminnich@terra To: zhihuizhang cc: hackers Subject: Re: Page fault on a read-only apge In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, 14 Dec 1998, zhihuizhang wrote: > Hi, I read in the mailing list archive something like "Also the VM tends > to not consult the underlying object for a number of cases, the worst > one being that if a write fault occurs on a read-only page, the VM will > tend to add write access without asking the VFS if that is ok". That sounds like something I might have written. It's certainly a problem I've had. A quick scan of current sources for 3.0 indicates it is still broken. The problem: you read-fault on a page and it is brought in by the vfs. But, the vfs wishes to know if you ever write fault on a page so it can support cache coherency (a la MNFS). On SunOS and Solaris, the vfs is called when a write-fault happens on a read-only page. On FreeBSD and Linux and AIX and Irix and ... it doesn't happen correctly. I'm not quite sure about NetBSD. ron To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message