From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail202.messagelabs.com (mail202.messagelabs.com [216.82.254.227]) by kanga.kvack.org (Postfix) with SMTP id 051526B00B4 for ; Tue, 26 Jan 2010 14:50:26 -0500 (EST) Date: Tue, 26 Jan 2010 20:49:47 +0100 From: Andrea Arcangeli Subject: Re: [PATCH 21 of 31] split_huge_page_mm/vma Message-ID: <20100126194947.GV30452@random.random> References: <9cb2a8a61d32163dced8.1264513936@v2.random> <20100126173450.GE16468@csn.ul.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100126173450.GE16468@csn.ul.ie> Sender: owner-linux-mm@kvack.org To: Mel Gorman Cc: linux-mm@kvack.org, Marcelo Tosatti , Adam Litke , Avi Kivity , Izik Eidus , Hugh Dickins , Nick Piggin , Rik van Riel , Andi Kleen , Dave Hansen , Benjamin Herrenschmidt , Ingo Molnar , Mike Travis , KAMEZAWA Hiroyuki , Christoph Lameter , Chris Wright , Andrew Morton , bpicco@redhat.com, Christoph Hellwig , KOSAKI Motohiro List-ID: On Tue, Jan 26, 2010 at 05:34:51PM +0000, Mel Gorman wrote: > I guess this is the part that breaks huge pages when smaps is read. That > is a bit of a snag as a normal user could cause a lot of churn by > reading those files a lot. yep, this is the highest priority split_huge_page to remove, along with fixing lru stats. mprotect and mremap are much lower prio. I think I mentioned the removal of split_huge_page from smaps in earlier emails too. > In the event that gets fixed up, it's worth considering what KernelPageSize: > and MMUPageSize: should be printing in smaps for regions of memory backed > by a mix of base and huge pages. Suggestions and patches welcome. I just deferred it for later and no app on my system seems to cat those files but it's definitely something to fix or it's just spurious overhead given to khugepaged... -- 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