From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Srivatsa S. Bhat" Subject: Re: [Results] [RFC PATCH v4 00/40] mm: Memory Power Management Date: Fri, 27 Sep 2013 00:26:00 +0530 Message-ID: <52448340.7060802@linux.vnet.ibm.com> References: <20130925231250.26184.31438.stgit@srivatsabhat.in.ibm.com> <52437128.7030402@linux.vnet.ibm.com> <20130925164057.6bbaf23bdc5057c42b2ab010@linux-foundation.org> <20130925234734.GK18242@two.firstfloor.org> <52438AA9.3020809@linux.intel.com> <20130925182129.a7db6a0fd2c7cc3b43fda92d@linux-foundation.org> <20130926015016.GM18242@two.firstfloor.org> <20130925195953.826a9f7d.akpm@linux-foundation.org> <524439D5.8020306@linux.vnet.ibm.com> <52445993.7050608@linux.intel.com> <52446841.2030301@linux.vnet.ibm.com> <524477AC.9090400@linux.intel.com> <52447DED.5080205@linux.vnet.ibm.com> <3908561D78D1C84285E8C5FCA982C28F31D1BA12@ORSMSX106.amr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <3908561D78D1C84285E8C5FCA982C28F31D1BA12@ORSMSX106.amr.corp.intel.com> Sender: owner-linux-mm@kvack.org To: "Luck, Tony" Cc: Arjan van de Ven , Andrew Morton , Andi Kleen , "mgorman@suse.de" , "dave@sr71.net" , "hannes@cmpxchg.org" , "matthew.garrett@nebula.com" , "riel@redhat.com" , "srinivas.pandruvada@linux.intel.com" , "willy@linux.intel.com" , "kamezawa.hiroyu@jp.fujitsu.com" , "lenb@kernel.org" , "rjw@sisk.pl" , "gargankita@gmail.com" , "paulmck@linux.vnet.ibm.com" , "svaidy@linux.vnet.ibm.com" List-Id: linux-pm@vger.kernel.org On 09/27/2013 12:20 AM, Luck, Tony wrote: >> And that's it! No other case for page movement. And with this conservative >> approach itself, I'm getting great consolidation ratios! >> I am also thinking of adding more smartness in the code to be very choosy in >> doing the movement, and do it only in cases where it is almost guaranteed to >> be beneficial. For example, I can make the kmempowerd kthread more "lazy" >> while moving/reclaiming stuff; I can bias the page movements such that "cold" >> pages are left around (since they are not expected to be referenced much >> anyway) and only the (few) hot pages are moved... etc. > > Can (or should) this migrator coordinate with khugepaged - I'd hate to see them > battling over where to move pages ... or undermining each other (your daemon > frees up a 512MB area ... and khugepaged immediately grabs a couple of 2MB > pages from it to upgrade some process with a scattershot of 4K pages). > That's a very good point! I need to look into it to see how such sub-optimal behavior can be avoided (perhaps by making khugepaged region-aware)... Hmmm.. Thanks for bringing it up! Regards, Srivatsa S. Bhat -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org