From owner-freebsd-current@FreeBSD.ORG Tue Nov 22 00:41:40 2011 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 4CC57106566B for ; Tue, 22 Nov 2011 00:41:40 +0000 (UTC) (envelope-from rmacklem@uoguelph.ca) Received: from esa-jnhn.mail.uoguelph.ca (esa-jnhn.mail.uoguelph.ca [131.104.91.44]) by mx1.freebsd.org (Postfix) with ESMTP id 096718FC12 for ; Tue, 22 Nov 2011 00:41:39 +0000 (UTC) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApwEAALvyk6DaFvO/2dsb2JhbABDhQGmRYIcgQsCDRkCnVWOAJIFgTCFUoF/gRYEiBqMIYkCiSE X-IronPort-AV: E=Sophos;i="4.69,550,1315195200"; d="scan'208";a="146382956" Received: from erie.cs.uoguelph.ca (HELO zcs3.mail.uoguelph.ca) ([131.104.91.206]) by esa-jnhn-pri.mail.uoguelph.ca with ESMTP; 21 Nov 2011 19:41:38 -0500 Received: from zcs3.mail.uoguelph.ca (localhost.localdomain [127.0.0.1]) by zcs3.mail.uoguelph.ca (Postfix) with ESMTP id D4823B3F03 for ; Mon, 21 Nov 2011 19:41:38 -0500 (EST) Date: Mon, 21 Nov 2011 19:41:38 -0500 (EST) From: Rick Macklem To: freebsd-current@freebsd.org Message-ID: <1190103347.107832.1321922498854.JavaMail.root@erie.cs.uoguelph.ca> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [172.17.91.203] X-Mailer: Zimbra 6.0.10_GA_2692 (ZimbraWebClient - FF3.0 (Win)/6.0.10_GA_2692) Subject: Heads Up: NFS server will now use LK_SHARED vnode locks X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 22 Nov 2011 00:41:40 -0000 Hi, I have just committed r227809 to head/current which enables the new/default NFS server's use of shared vnode locks for Read, Readdir, Readlink, Getattr and Access. Although it is hoped that this will improve performance for these operations when multiple ones are performed concurrently on the same file/vnode, I thought I should give a heads up. Why? Well it is conceivable that this may have negative issues that I haven't seen in testing along the lines of overloading a server, due to the lack of serialization of the above RPCs for the same file/vnode. If anyone encounters problems with their NFS server after upgrading to post-r227809, please email and let me know. rick