From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755062Ab0CAFjF (ORCPT ); Mon, 1 Mar 2010 00:39:05 -0500 Received: from mga03.intel.com ([143.182.124.21]:19911 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750861Ab0CAFh5 (ORCPT ); Mon, 1 Mar 2010 00:37:57 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.49,557,1262592000"; d="scan'208";a="249208731" Message-Id: <20100301053620.966418452@intel.com> User-Agent: quilt/0.48-1 Date: Mon, 01 Mar 2010 13:26:56 +0800 From: Wu Fengguang To: Andrew Morton CC: Jens Axboe , Matt Mackall , Rik van Riel , Christian Ehrhardt , Wu Fengguang CC: Chris Mason CC: Peter Zijlstra CC: Clemens Ladisch CC: Olivier Galibert cc: Vivek Goyal CC: Nick Piggin cc: Linux Memory Management List CC: Cc: LKML Subject: [PATCH 05/16] readahead: limit read-ahead size for small memory systems References: <20100301052651.857984880@intel.com> Content-Disposition: inline; filename=readahead-small-memory-limit-readaround.patch Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When lifting the default readahead size from 128KB to 512KB, make sure it won't add memory pressure to small memory systems. For read-ahead, the memory pressure is mainly readahead buffers consumed by too many concurrent streams. The context readahead can adapt readahead size to thrashing threshold well. So in principle we don't need to adapt the default _max_ read-ahead size to memory pressure. For read-around, the memory pressure is mainly read-around misses on executables/libraries. Which could be reduced by scaling down read-around size on fast "reclaim passes". This patch presents a straightforward solution: to limit default read-ahead size proportional to available system memory, ie. 512MB mem => 512KB read-around size limit 128MB mem => 128KB read-around size limit 32MB mem => 32KB read-around size limit This will allow power users to adjust read-ahead/read-around size at once, while saving the low end from unnecessary memory pressure, under the assumption that low end users have no need to request a large read-around size. CC: Matt Mackall Acked-by: Rik van Riel Acked-by: Christian Ehrhardt Signed-off-by: Wu Fengguang --- mm/filemap.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) --- linux.orig/mm/filemap.c 2010-02-26 10:11:41.000000000 +0800 +++ linux/mm/filemap.c 2010-02-27 13:05:16.000000000 +0800 @@ -1431,7 +1431,9 @@ static void do_sync_mmap_readahead(struc /* * mmap read-around */ - ra_pages = max_sane_readahead(ra->ra_pages); + ra_pages = min_t(unsigned long, + ra->ra_pages, + roundup_pow_of_two(totalram_pages / 1024)); if (ra_pages) { ra->start = max_t(long, 0, offset - ra_pages/2); ra->size = ra_pages;