From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753228AbbCZSi3 (ORCPT ); Thu, 26 Mar 2015 14:38:29 -0400 Received: from verein.lst.de ([213.95.11.211]:53634 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751018AbbCZSiI (ORCPT ); Thu, 26 Mar 2015 14:38:08 -0400 Date: Thu, 26 Mar 2015 19:38:06 +0100 From: Christoph Hellwig To: Boaz Harrosh Cc: Christoph Hellwig , linux-nvdimm@ml01.01.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, x86@kernel.org, ross.zwisler@linux.intel.com, axboe@kernel.dk Subject: Re: another pmem variant V2 Message-ID: <20150326183806.GA27181@lst.de> References: <1427358764-6126-1-git-send-email-hch@lst.de> <55143A8B.2060304@plexistor.com> <20150326171858.GA25575@lst.de> <55144257.8050004@plexistor.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <55144257.8050004@plexistor.com> User-Agent: Mutt/1.5.17 (2007-11-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Mar 26, 2015 at 07:31:03PM +0200, Boaz Harrosh wrote: > But I hope you are not ignoring my real problem. any two memmap= ranges > will halt the boot. Specially if they are dis-contiguous. not the case here, verified it in various cofigurations. > Also I need the contiguous variant split into two devices because > they might belong to two NUMA nodes. It is very hard to manage > if a NUMA crossing is in a middle of a single pmemX device > The way we like to configure it is that each /dev/pmem belongs to > a single NUMA node. And in a multy device setup each CPU node > allocates from "his" pmem device If there is space. > (And it lets me set application affinity if need to) The hack below ensures two separate type 12 entries stay separate, but I'm not sure I really want this. so far it seems like very special hacks for your very specialized fake-pmem config. > BTW: Will device mapper let me call ->direct_access() Right now it doesn't, but it's not hard to add..