From mboxrd@z Thu Jan 1 00:00:00 1970 From: Guoqing Jiang Subject: Re: [PATCH 09/14] md-cluster: set MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD in metadata_update_start Date: Wed, 1 Mar 2017 10:58:37 +0800 Message-ID: <58B638DD.5070805@suse.com> References: <1487906124-20107-1-git-send-email-gqjiang@suse.com> <1487906124-20107-10-git-send-email-gqjiang@suse.com> <20170228182444.ggwl7dnmhw2zi3n4@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20170228182444.ggwl7dnmhw2zi3n4@kernel.org> Sender: linux-raid-owner@vger.kernel.org To: Shaohua Li Cc: linux-raid@vger.kernel.org, shli@fb.com, neilb@suse.de List-Id: linux-raid.ids On 03/01/2017 02:24 AM, Shaohua Li wrote: > On Fri, Feb 24, 2017 at 11:15:19AM +0800, Guoqing Jiang wrote: >> Since we have switched to sync way to handle METADATA_UPDATED >> msg, then process_metadata_update can only continue with either >> get the reconfig_mutex or MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD is >> set. > same here. Either the patch should be put into the 'METADATA_UPDATED' patch or > this patch should be before the 'METADATA_UPDATED' patch. > > We shouldn't introduce a broken patch then fix it in latter patch. Yes, for both 8 and 9, it is possible to merge with 6 (I resend it though it was not posted successfully for some reason). I seperated them since it would be more easy for review and don't want mess up header of the 6th patch. Thanks, Guoqing