From owner-freebsd-geom@FreeBSD.ORG  Thu Apr 22 21:59:31 2010
Return-Path: <owner-freebsd-geom@FreeBSD.ORG>
Delivered-To: freebsd-geom@FreeBSD.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 673FD1065672
	for <freebsd-geom@FreeBSD.org>; Thu, 22 Apr 2010 21:59:31 +0000 (UTC)
	(envelope-from avg@icyb.net.ua)
Received: from citadel.icyb.net.ua (citadel.icyb.net.ua [212.40.38.140])
	by mx1.freebsd.org (Postfix) with ESMTP id AD78F8FC13
	for <freebsd-geom@FreeBSD.org>; Thu, 22 Apr 2010 21:59:30 +0000 (UTC)
Received: from porto.topspin.kiev.ua (porto-e.starpoint.kiev.ua
	[212.40.38.100])
	by citadel.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id AAA16055
	for <freebsd-geom@FreeBSD.org>; Fri, 23 Apr 2010 00:59:29 +0300 (EEST)
	(envelope-from avg@icyb.net.ua)
Received: from localhost.topspin.kiev.ua ([127.0.0.1])
	by porto.topspin.kiev.ua with esmtp (Exim 4.34 (FreeBSD))
	id 1O54Qi-0008bd-Sp
	for freebsd-geom@FreeBSD.org; Fri, 23 Apr 2010 00:59:28 +0300
Message-ID: <4BD0C6C0.30008@icyb.net.ua>
Date: Fri, 23 Apr 2010 00:59:28 +0300
From: Andriy Gapon <avg@icyb.net.ua>
User-Agent: Thunderbird 2.0.0.24 (X11/20100321)
MIME-Version: 1.0
To: freebsd-geom@FreeBSD.org
X-Enigmail-Version: 0.96.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Cc: 
Subject: d_maxsize .. si_iosize_max
X-BeenThere: freebsd-geom@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: GEOM-specific discussions and implementations
	<freebsd-geom.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-geom>,
	<mailto:freebsd-geom-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-geom>
List-Post: <mailto:freebsd-geom@freebsd.org>
List-Help: <mailto:freebsd-geom-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-geom>,
	<mailto:freebsd-geom-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Thu, 22 Apr 2010 21:59:31 -0000


I see that many disk drivers are careful to set an optimal value for d_maxsize.
On the other hand, si_iosize_max is hardcoded to MAXPHYS in geom_dev.c, which
leads to suboptimal I/O for devices with actual maximum I/O size less than MAXPHYS.
And right now d_maxsize seems to be used only for dumperinfo.maxiosize, which is
a waste.

So, perhaps, d_maxsize should be passed up through GEOM layer as a provider
property, e.g. pp->maxiosize.  So that si_iosize_max could be set correctly.
Or, more importantly, so that final consumers (e.g. filesystems) could now
correct optimal I/O size.

Perhaps, this should even be split into max_read_size and max_write_size for
geoms like gmirror where reads can be done in parallel.

-- 
Andriy Gapon