From: "J. Bruce Fields" Subject: Re: [PATCH 0/7] Intro: convert lockd to kthread and fix use-after-free (try #3) Date: Sat, 5 Jan 2008 11:44:03 -0500 Message-ID: <20080105164403.GA12928@fieldses.org> References: <1198009182-27895-1-git-send-email-jlayton@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: Jeff Layton , linux-nfs@vger.kernel.org, nfsv4@linux-nfs.org, linux-kernel@vger.kernel.org, richterd@citi.umich.edu To: Tomasz =?utf-8?Q?K=C5=82oczko?= Return-path: Received: from mail.fieldses.org ([66.93.2.214]:36307 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755449AbYAEQoa (ORCPT ); Sat, 5 Jan 2008 11:44:30 -0500 In-Reply-To: Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, Dec 19, 2007 at 08:15:43PM +0100, Tomasz K=C5=82oczko wrote: > On Tue, 18 Dec 2007, Jeff Layton wrote: > >> This is the third patchset to fix the use-after-free problem in lock= d, >> and to convert lockd to use the kthread API instead of kernel_thread= =2E >> The main change from the last patchset is the elimination of the >> svc_create_kthread helper function, and having lockd_up call kthread= _run >> directly. I've also made some changes to the function prototype for >> the lockd function to eliminate the need for casting the function >> pointer. > > BTW kernel, NFS and lockd: SuSE kernel IIRC is distributed with patch= which=20 > adds kernel-based status monitor (statd). > > Q to NFS people: any plans about merge this in main kernel tree ? No. There was opposition to the idea. I can't remember the argument, though, and a quick google search isn't turning it up. > And another question related to NFS: any progress on add clear NFS=20 > statistics in kernel (nfsstat -z) ? No. We do have some patches that add roughly equivalent functionality in userspace. --b.