From owner-freebsd-hackers Mon Nov 2 11:37:47 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA22878 for freebsd-hackers-outgoing; Mon, 2 Nov 1998 11:37:47 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from smtp03.primenet.com (smtp03.primenet.com [206.165.6.133]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA22870 for ; Mon, 2 Nov 1998 11:37:45 -0800 (PST) (envelope-from tlambert@usr05.primenet.com) Received: (from daemon@localhost) by smtp03.primenet.com (8.8.8/8.8.8) id MAA29498; Mon, 2 Nov 1998 12:37:38 -0700 (MST) Received: from usr05.primenet.com(206.165.6.205) via SMTP by smtp03.primenet.com, id smtpd029398; Mon Nov 2 12:37:26 1998 Received: (from tlambert@localhost) by usr05.primenet.com (8.8.5/8.8.5) id MAA02997; Mon, 2 Nov 1998 12:37:25 -0700 (MST) From: Terry Lambert Message-Id: <199811021937.MAA02997@usr05.primenet.com> Subject: Re: sync writes with softupdates enabled To: peter.jeremy@auss2.alcatel.com.au (Peter Jeremy) Date: Mon, 2 Nov 1998 19:37:25 +0000 (GMT) Cc: hackers@FreeBSD.ORG In-Reply-To: <98Nov2.163001est.40354@border.alcanet.com.au> from "Peter Jeremy" at Nov 2, 98 04:30:29 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > >Now that this has been committed, can we have some feedback from people > >using mmap heavily on -current? > > There are no longer any sync writes as a result of using an MMAP'd > history file in inn. There is no obvious corruption of history file > or associated .pag file. > > Note that there is a similar bug in 2.2.x. The fix is similar, but > needs some thought since there appears to be a problem with the > parameters to vm_object_page_clean(). In particular, `syncio' is > defined as boolean_t, as passed as the `waitfor' argument to > VOP_FSYNC() - which is MNT_ASYNC or MNT_SYNC elsewhere. I haven't > had the time to work out the correct definitions. I have a problem with this. Has msync(2) been updated to "do the right thing" with regard to uncommitted data in an mmapped region? If so, how? I was pretty sure that it was impossible to identify the mapping object backing ranges -- or so the comments in the msync code imply. I don't think you can arbitrarily change these to async references; specifically, I think there is a problem with dirty page write through. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message