From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Hansen Subject: Re: [PATCHv3 1/2] proc: mm: export PTE sizes directly in smaps Date: Thu, 26 Oct 2017 07:25:37 -0700 Message-ID: <10bc2525-fccc-81e6-1ec0-27d63cff89b4@intel.com> References: <1508891255-15427-1-git-send-email-fan.du@intel.com> <20171025092838.4j4u4eeppzh6uto2@dhcp22.suse.cz> <5A90DA2E42F8AE43BC4A093BF06788481AA50489@SHSMSX104.ccr.corp.intel.com> <20171026141943.fsmd4goiec32qjkf@dhcp22.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20171026141943.fsmd4goiec32qjkf@dhcp22.suse.cz> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Michal Hocko , "Du, Fan" Cc: "akpm@linux-foundation.org" , "hch@lst.de" , "Williams, Dan J" , "linux-kernel@vger.kernel.org" , "linux-api@vger.kernel.org" List-Id: linux-api@vger.kernel.org On 10/26/2017 07:19 AM, Michal Hocko wrote: >> Current vm_normal_page implementation doesn't pick up page with DEVMAP pfn. >> The second patch fix this and export DAX mappings into counters introduced in the >> first patch. >> >> IMO, the user care more about how much persistent memory they used, how about >> a small tweak with smaps_account, and report the total mapping size into RSS/PSS, >> which user are usually more familiar with? > No! Rss and pss is already used by many tools to evaluate misbehaving > tasks. If you start accounting the memory which is not bound to the > process life time then you can break those usecases. This is the reason > why hugetlb is not accounted to rss as well. Yeah, folks (rightly) assume Rss is memory and eaten from MemTotal. DAX mappings do not use RAM.