From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751330AbcB2VDo (ORCPT ); Mon, 29 Feb 2016 16:03:44 -0500 Received: from mail-wm0-f49.google.com ([74.125.82.49]:37400 "EHLO mail-wm0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750732AbcB2VDn (ORCPT ); Mon, 29 Feb 2016 16:03:43 -0500 Date: Tue, 1 Mar 2016 00:03:40 +0300 From: "Kirill A. Shutemov" To: Johannes Weiner Cc: Linus Torvalds , Andrew Morton , linux-mm , Linux Kernel Mailing List , kernel-team Subject: Re: [PATCH] mm: readahead: do not cap readahead() and MADV_WILLNEED Message-ID: <20160229210339.GA15095@node.shutemov.name> References: <1456277927-12044-1-git-send-email-hannes@cmpxchg.org> <20160229194159.GB29896@cmpxchg.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160229194159.GB29896@cmpxchg.org> User-Agent: Mutt/1.5.23.1 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Feb 29, 2016 at 02:41:59PM -0500, Johannes Weiner wrote: > That, or switch to read() from a separate thread for cache priming. mmap(MAP_POPULATE) would save you some copy_to_user() overhead, comparing to read(). -- Kirill A. Shutemov