From mboxrd@z Thu Jan  1 00:00:00 1970
From: Guoqing Jiang <gqjiang@suse.com>
Subject: Re: [RFC PATCH 1/2] RAID1: a new I/O barrier implementation to remove
 resync window
Date: Wed, 30 Nov 2016 14:37:06 +0800
Message-ID: <583E7392.9020308@suse.com>
References: <1479765241-15528-1-git-send-email-colyli@suse.de>
 <58369819.1070203@suse.com> <0bfdd1c7-5953-a282-dd37-3a8595c90844@suse.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Return-path: <linux-raid-owner@vger.kernel.org>
In-Reply-To: <0bfdd1c7-5953-a282-dd37-3a8595c90844@suse.de>
Sender: linux-raid-owner@vger.kernel.org
To: Coly Li <colyli@suse.de>, linux-raid@vger.kernel.org
Cc: Shaohua Li <shli@fb.com>, Neil Brown <neilb@suse.de>, Johannes Thumshirn <jthumshirn@suse.de>
List-Id: linux-raid.ids



On 11/28/2016 03:33 PM, Coly Li wrote:
> In raid1_sync_request(), I see,
> conf->cluster_sync_low = mddev->curr_resync_completed;
> conf->cluster_sync_high = conf->cluster_sync_low +
> CLUSTER_RESYNC_WINDOW_SECTORS;
>
> Is it possible that LBA range [conf->cluster_sync_low,
> conf->cluster_sync_high] goes across the border of a barrier unit size ?

Not pretty sure about it, but since cluster's resync window is 32M which is
less than BARRIER_UNIT_SECTOR_SIZE, I guess it would not cause trouble.

Thanks,
Guoqing