From mboxrd@z Thu Jan 1 00:00:00 1970 From: Simon Horman Date: Thu, 10 May 2007 12:49:58 +0000 Subject: Re: [request-for-test] [patch 0/3] [IA64] Allow DISCONTIGMEM without NUMA Message-Id: <20070510124956.GA3746@verge.net.au> List-Id: References: <20070510101945.495818588@tabatha.lab.ultramonkey.org> In-Reply-To: <20070510101945.495818588@tabatha.lab.ultramonkey.org> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org On Thu, May 10, 2007 at 12:45:48PM +0100, Christoph Hellwig wrote: > On Thu, May 10, 2007 at 07:19:45PM +0900, Simon Horman wrote: > > The following patches relate to allowing DISCONTIG memory to compile > > (and hopefully work) without NUMA. I've tested that they boot, but not much > > beyond that. > > Why? I though we wanted to use virtual mem_map and sparsemem for > holes inside a single node? There was a breif dicusion on linux-ia64 which I interpated as Tony saying it was a good idea. http://thread.gmane.org/gmane.linux.ports.ia64/16325/focus327 With regards to sparse, the current code needs most if not all these fixes in order to use arch/ia64/mm/discontig.c. Is the idea to rework that? -- Horms H: http://www.vergenet.net/~horms/ W: http://www.valinux.co.jp/en/