From owner-freebsd-current Mon Apr 15 15:36:39 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id PAA13400 for current-outgoing; Mon, 15 Apr 1996 15:36:39 -0700 (PDT) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id PAA13394 for ; Mon, 15 Apr 1996 15:36:34 -0700 (PDT) Received: (from root@localhost) by dyson.iquest.net (8.7.5/8.6.9) id RAA09839; Mon, 15 Apr 1996 17:35:22 -0500 (EST) From: "John S. Dyson" Message-Id: <199604152235.RAA09839@dyson.iquest.net> Subject: Re: NFS timing (Was: Just how stable is current) To: joerg_wunsch@uriah.heep.sax.de Date: Mon, 15 Apr 1996 17:35:22 -0500 (EST) Cc: freebsd-current@FreeBSD.org In-Reply-To: <199604151859.UAA12685@uriah.heep.sax.de> from "J Wunsch" at Apr 15, 96 08:59:52 pm Reply-To: dyson@FreeBSD.org X-Mailer: ELM [version 2.4 PL24 ME8] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > For me, it required me to revert to 1 KB blocksize (even though the > NFS server is also only a lame 8-bit 3c503), and we've got at least > one installation problem report where the installation totally starved > at the first lengthy file, even _with_ the ``NFS slow'' option. > > I really wonder what might have caused this change. > Try adding an spl0(); at the beginning of vm_daemon in vm_pageout.c. This was causing some unpleasant interrupt latencies. BDE figured this one out. Might work, might not. John