From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marco Stornelli <marco.stornelli@gmail.com>
Subject: Readahead pages
Date: Fri, 17 Sep 2010 09:11:34 +0200
Message-ID: <AANLkTi=zfLxYQK8C3GQSacVYNOAeX=zO=gFcADyNjMKc@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
To: linux-fsdevel@vger.kernel.org
Return-path: <linux-fsdevel-owner@vger.kernel.org>
Received: from mail-qy0-f174.google.com ([209.85.216.174]:46755 "EHLO
	mail-qy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751212Ab0IQHLf (ORCPT
	<rfc822;linux-fsdevel@vger.kernel.org>);
	Fri, 17 Sep 2010 03:11:35 -0400
Received: by qyk36 with SMTP id 36so510752qyk.19
        for <linux-fsdevel@vger.kernel.org>; Fri, 17 Sep 2010 00:11:34 -0700 (PDT)
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

Hi,

testing my fs (pramfs), I can see that VFS calls my read_page function
for a single read on a file. I'm using the do_sync_read path and
direct IO to read/write on a file. The read page is called from
readahead path (from page_cache_sync_readahead), but in the backing
dev info, the ra pages is zero, so how is it possible? Is the fs in
charge of doing some call to avoid readahead or is not possible avoid
it?

Thanks.

Marco