From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753820AbbC0IKQ (ORCPT ); Fri, 27 Mar 2015 04:10:16 -0400 Received: from verein.lst.de ([213.95.11.211]:55616 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932084AbbC0IKK (ORCPT ); Fri, 27 Mar 2015 04:10:10 -0400 Date: Fri, 27 Mar 2015 09:10:07 +0100 From: Christoph Hellwig To: Yinghai Lu Cc: Christoph Hellwig , Ingo Molnar , linux-nvdimm@ml01.01.org, linux-fsdevel@vger.kernel.org, Linux Kernel Mailing List , the arch/x86 maintainers , ross.zwisler@linux.intel.com, Jens Axboe , boaz@plexistor.com Subject: Re: [PATCH 2/3] x86: add a is_e820_ram() helper Message-ID: <20150327081007.GA12909@lst.de> References: <1427358764-6126-1-git-send-email-hch@lst.de> <1427358764-6126-3-git-send-email-hch@lst.de> <20150326090215.GA11520@gmail.com> <20150326093424.GA28217@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: 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 03:59:28PM -0700, Yinghai Lu wrote: > What do you want to get here? > > You did not modify memblock_x86_fill() to treat > E820_PRAM as E820_RAM, so memblock will not have any > entry for E820_PRAM, so you do not need to call memblock_reserve > there. > > And the same time, init_memory_mapping() will call > init_range_memory_mapping/for_each_mem_pfn_range() to > set kernel mapping for memory range in memblock only. > So here calling init_memory_mapping will not do anything. > then just drop calling to that init_memory_mapping. > --- so will not kernel mapping pmem, is that what you intended to have? I think the intent of the old Intel code was to indeed map the pmem into KVA space. That got broken when I forward ported it to use memblocks. However the current pmem infrastructure doesn't need the KVA mapping, so I can remove it for now. However we have heated discussions about how to do I/O to pmem, and KVA mapping is one of the options. If we got with that option I might bring this code back in a fixed up version.