From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753123AbcCGJxn (ORCPT ); Mon, 7 Mar 2016 04:53:43 -0500 Received: from mx2.suse.de ([195.135.220.15]:41272 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753090AbcCGJxh (ORCPT ); Mon, 7 Mar 2016 04:53:37 -0500 Subject: Re: [PATCH] mm: ZONE_DEVICE depends on SPARSEMEM_VMEMMAP To: Dan Williams , akpm@linux-foundation.org References: <20160305004214.12356.32017.stgit@dwillia2-desk3.jf.intel.com> Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org From: Vlastimil Babka Message-ID: <56DD4F9F.303@suse.cz> Date: Mon, 7 Mar 2016 10:53:35 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <20160305004214.12356.32017.stgit@dwillia2-desk3.jf.intel.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 03/05/2016 01:42 AM, Dan Williams wrote: > The primary use case for devm_memremap_pages() is to allocate an > memmap array from persistent memory. That capabilty requires > vmem_altmap which requires SPARSEMEM_VMEMMAP. Some if the ifdefs around to_vmem_altmap() could now be simplified, but that can also be done later... > Also, without SPARSEMEM_VMEMMAP the addition of ZONE_DEVICE expands > ZONES_WIDTH and triggers the: > > "Unfortunate NUMA and NUMA Balancing config, growing page-frame for > last_cpupid." > > ...warning in mm/memory.c. SPARSEMEM_VMEMMAP=n && ZONE_DEVICE=y is not > a configuration we should worry about supporting. > > Reported-by: Vlastimil Babka > Signed-off-by: Dan Williams Acked-by: Vlastimil Babka > --- > mm/Kconfig | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/mm/Kconfig b/mm/Kconfig > index 664fa2416909..b95322ba542b 100644 > --- a/mm/Kconfig > +++ b/mm/Kconfig > @@ -654,6 +654,7 @@ config ZONE_DEVICE > bool "Device memory (pmem, etc...) hotplug support" if EXPERT > depends on MEMORY_HOTPLUG > depends on MEMORY_HOTREMOVE > + depends on SPARSEMEM_VMEMMAP > depends on X86_64 #arch_add_memory() comprehends device memory > > help >