From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764273AbdEWTGl (ORCPT ); Tue, 23 May 2017 15:06:41 -0400 Received: from mail-yw0-f182.google.com ([209.85.161.182]:34196 "EHLO mail-yw0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751088AbdEWTGj (ORCPT ); Tue, 23 May 2017 15:06:39 -0400 Date: Tue, 23 May 2017 15:06:36 -0400 From: Tejun Heo To: Shaohua Li Cc: linux-kernel@vger.kernel.org, gregkh@linuxfoundation.org, viro@zeniv.linux.org.uk, Kernel-team@fb.com Subject: Re: [PATCH 0/5] kernfs: add exportfs operations Message-ID: <20170523190636.GD13222@htj.duckdns.org> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.8.2 (2017-04-18) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, On Mon, May 22, 2017 at 03:53:04PM -0700, Shaohua Li wrote: > The goal isn't to export kernfs to NFS. The intention is to make tracing cgroup > aware. To do this, tracing will record an id for cgroup and use the id to find > cgroup name later. The best id is the cgroup directory inode number. Further to > filter out stale cgroup directory, fhandle is the best to identify a cgroup. So > this is what this series try to do. Generally looks good to me. Had some review points in other replies. There is a related issue which can be a nice follow-up - replacing cgrp->id with kernfs's fid. cgrp->id is allocated on its own idr and its only use is to accelerate testing for ancestry and we can easily replace that with the unique file id. Thanks a lot! -- tejun