From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756574Ab2IUPam (ORCPT ); Fri, 21 Sep 2012 11:30:42 -0400 Received: from youngberry.canonical.com ([91.189.89.112]:54139 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753684Ab2IUPal (ORCPT ); Fri, 21 Sep 2012 11:30:41 -0400 Date: Fri, 21 Sep 2012 10:30:33 -0500 From: Serge Hallyn To: Feng Hong Cc: "Serge E. Hallyn" , "akpm@linux-foundation.org" , "gorcunov@openvz.org" , "keescook@chromium.org" , "ebiederm@xmission.com" , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH V2] poweroff: fix bug in orderly_poweroff Message-ID: <20120921153033.GB13764@sergelap> References: <1348036677-28626-1-git-send-email-hongfeng@marvell.com> <20120920170647.GA26278@mail.hallyn.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Quoting Feng Hong (hongfeng@marvell.com): > Hi, Serge, > > I am just a graduate and it's my first time to send a patch to opensource, so thank you very much for reminding me the "changelog affairs", it seems this patch has been added to -mm tree as attached mail, and I have no chance to change the comments, right ? Then I must remember this and be careful next time. Thanks again for reminding me ! Sorry, your description was fine, what i meant was something below your patch description that looks like Change since v1: [date] Per Eric's sugestion, switch from UMH_WAIT_PROC to UMH_WAIT_EXEC. > >Is this actually sufficient for you? The exec will have started, but may for whatever (very unlikely) reason fail. If you're happy with it, > I think UMH_WAIT_EXEC is sufficient for me, as in our system there is no "/sbin/poweroff" existed. On the other hand, UMH_WAIT_PROC is not suitable here as Eric analysis; if using UMH_WAIT_EXEC, and the user application fail, I'd prefer to complain bad application. So using UMH_WAIT_EXEC and UMH_WAIT_PROC has a tradeoff here, what do you think so ? Yup, that sounds fine to me, I just wanted to make sure you were ok with the fact that application failure (after successful exec) will be ignored. thanks, -serge