From owner-freebsd-stable  Sun Dec  7 19:54:39 1997
Return-Path: <owner-freebsd-stable>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id TAA24524
          for stable-outgoing; Sun, 7 Dec 1997 19:54:39 -0800 (PST)
          (envelope-from owner-freebsd-stable)
Received: from arg1.demon.co.uk (arg1.demon.co.uk [194.222.34.166])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA24514;
          Sun, 7 Dec 1997 19:54:28 -0800 (PST)
          (envelope-from arg@arg1.demon.co.uk)
Received: (from arg@localhost)
	by arg1.demon.co.uk (8.8.5/8.8.5) id DAA22961;
	Mon, 8 Dec 1997 03:54:02 GMT
Date: Mon, 8 Dec 1997 03:54:01 +0000 (GMT)
From: Andrew Gordon <arg@arg1.demon.co.uk>
X-Sender: arg@server.arg.sj.co.uk
To: Bruce Evans <bde@zeta.org.au>
cc: "John S. Dyson" <dyson@freebsd.org>, freebsd-stable@freebsd.org
Subject: Re: kern/2573 - NFS hangs with mmap()
In-Reply-To: <199712071716.EAA06587@godzilla.zeta.org.au>
Message-ID: <Pine.BSF.3.91.971208034817.22360B-100000@server.arg.sj.co.uk>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-stable@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Mon, 8 Dec 1997, Bruce Evans wrote:
> The PR is referenced properly at least in the main fix, so the
> following was easy to find (there was also a useful date in
> the closure message for the PR - everything was done right
> except for closing it while it was open in 2 branches).
> 
>   Revision  Changes    Path
>   1.116     +102 -50   src/sys/kern/vfs_bio.c
>   1.27      +2 -2      src/sys/nfs/nfs.h
>   1.38      +83 -2     src/sys/nfs/nfs_bio.c
>   1.50      +5 -4      src/sys/nfs/nfs_vnops.c
>   1.21      +2 -1      src/sys/nfs/nfsnode.h
>   1.69      +2 -2      src/sys/vm/vm_fault.c
>   1.71      +6 -1      src/sys/vm/vnode_pager.c
> Good luck merging the parts of this that apply to -stable (if any) :-).

Turns out that the original fix above was already in -stable.  Looking at 
the commit log for kern/vfs_bio.c, there were a couple more fixes a few 
days later that hadn't made it into -stable; stable appears to have fixes 
up to rev 1.117 (and later 1.120), but 1.118 and 1.119 were missing.

I added these (via cvs update -j1.117 -j1.119 and fixing up one conflict),
but sadly the problem still seems to be there.

I will set a make world going to bring my -current machine up to date, 
and test tomorrow whether my problem is actually fixed in -current after 
all.  I certainly have the same sort of symptoms as the original report.