Subject: Re: [Patch 4/4] cpusets top mask just online, not all possible
From: Dave Hansen <haveblue@us.ibm.com>
To: Paul Jackson <pj@sgi.com>
Cc: Anton Blanchard <anton@samba.org>, Andrew Morton <akpm@osdl.org>,
       Simon.Derr@bull.net,
       Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
       Andi Kleen <ak@suse.de>, IWAMOTO Toshihiro <iwamoto@valinux.co.jp>,
       "Martin J. Bligh" <mbligh@aracnet.com>
In-Reply-To: <20040911100731.2f400271.pj@sgi.com>
References: <20040911082810.10372.86008.84920@sam.engr.sgi.com>
	 <20040911082834.10372.51697.75658@sam.engr.sgi.com>
	 <20040911141001.GD32755@krispykreme>  <20040911100731.2f400271.pj@sgi.com>
Content-Type: text/plain
Message-Id: <1094924372.3997.16.camel@localhost>
Mime-Version: 1.0
Date: Sat, 11 Sep 2004 10:39:32 -0700
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 876
Lines: 20

Now that I've actuallly _looked_ at your code, I noticed that the
mems_allowed really is just a nodemask_t.  

So, you shouldn't have much to worry about from the current memory
hotplug stuff because we don't mess with hotplugging actual NUMA nodes
yet.  As I eluded to before, you might need notification of we ever get
down to a zero-sized node, but that should be id.  You'll certainly need
notification when entire nodes go offline, so I'm adding Martin Bligh to
the cc list (we just talked about NUMA hotplug yesterday).

Martin, you might want to look back in your LKML archives for where this
discussion came from.

-- Dave

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/