From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 20 May 2006 21:18:54 +0900 From: Yasunori Goto Subject: Re: [PATCH] Register sysfs file for hotpluged new node take 2. In-Reply-To: <1148058107.6623.160.camel@localhost.localdomain> References: <20060518143742.E2FB.Y-GOTO@jp.fujitsu.com> <1148058107.6623.160.camel@localhost.localdomain> Message-Id: <20060520104215.BBB9.Y-GOTO@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Dave Hansen Cc: Andrew Morton , Linux Kernel ML , linux-mm List-ID: > On Thu, 2006-05-18 at 14:50 +0900, Yasunori Goto wrote: > > + if (new_pgdat) { > > + ret = register_one_node(nid); > > + /* > > + * If sysfs file of new node can't create, cpu on the node > > + * can't be hot-added. There is no rollback way now. > > + * So, check by BUG_ON() to catch it reluctantly.. > > + */ > > + BUG_ON(ret); > > + } > > How about we register the node in sysfs _before_ it is > set_node_online()'d? Effectively an empty node with no memory and no > CPUs. It might be a wee bit confusing to any user tools watching the > NUMA sysfs stuff, but I think it beats a BUG(). Hmmm. I'm not sure what will happen when sysfs file is accessed by user at this time. I think this issue should be going to be solved when __remove_memory() and pgdat offline will be created. Thanks for your comment. -- Yasunori Goto -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org