From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932357AbaHGPku (ORCPT ); Thu, 7 Aug 2014 11:40:50 -0400 Received: from mail-we0-f182.google.com ([74.125.82.182]:47603 "EHLO mail-we0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932086AbaHGPkt (ORCPT ); Thu, 7 Aug 2014 11:40:49 -0400 Message-ID: <53E39DFE.6050501@plexistor.com> Date: Thu, 07 Aug 2014 18:40:46 +0300 From: Boaz Harrosh User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0 MIME-Version: 1.0 To: Karel Zak CC: Boaz Harrosh , Ross Zwisler , Jens Axboe , Matthew Wilcox , linux-kernel , linux-fsdevel Subject: Re: [PATCH 4/4] brd: Request from fdisk 4k alignment References: <53E21125.9080308@plexistor.com> <53E21317.7080403@plexistor.com> <1407362634.2869.2.camel@rzwisler-mobl1.amr.corp.intel.com> <53E36E53.3090008@gmail.com> <20140807130042.GB9401@x2.net.home> <53E385BE.40100@plexistor.com> <20140807152112.GD9401@x2.net.home> In-Reply-To: <20140807152112.GD9401@x2.net.home> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/07/2014 06:21 PM, Karel Zak wrote: <> > > fdisk uses physical sector size or minimal I/O size (greater value wins) > OK >> I was trying to play with the heads, sectors, cylinders; values but none I tried > > don't play with CHS, that's waste of time and it's completely ignored > by fdisk by default > >> would cause an alignment of 4k, not even of the first partition start. >> >> Please advise what I can do? > > IMHO you're right with your patch (alignment offset is IMHO bad way). > It's all (brd) about pages, is there any reason to use something else > for I/O limits? > > It would be also nice to set minimal and optimal io size, zero values > in this case means (for userspace) that the device does not provide > any I/O information to system. It's normal for old hw disks and then > we use some built-in defaults, but I don't see a reason to do the > same for virtual devices. > Hi Ross I have by now read the all code, and Karel also confirms this from fdisk side. The best for us is the use of "physical sector size" but with our "minimal I/O size" set to 512. The later has actual bad effects in the Kernel code itself. But the "physical sector size" has no effect on Kernel code, and actually has a very good affect on fdisk which now works the way we would like it. Please send your review-by so Jens can pick these up for mainline > Karel > Thanks Boaz