From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from esa5.hgst.iphmx.com ([216.71.153.144]:39717 "EHLO esa5.hgst.iphmx.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751180AbdFEVdC (ORCPT ); Mon, 5 Jun 2017 17:33:02 -0400 From: Bart Van Assche To: "hch@infradead.org" , "linux-block@vger.kernel.org" , "axboe@fb.com" , "ming.lei@redhat.com" Subject: Re: [PATCH v4 07/11] blk-mq: use QUEUE_FLAG_QUIESCED to quiesce queue Date: Mon, 5 Jun 2017 21:32:59 +0000 Message-ID: <1496698378.2623.28.camel@sandisk.com> References: <20170605155925.7896-1-ming.lei@redhat.com> <20170605155925.7896-8-ming.lei@redhat.com> In-Reply-To: <20170605155925.7896-8-ming.lei@redhat.com> Content-Type: text/plain; charset="iso-8859-1" MIME-Version: 1.0 Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Mon, 2017-06-05 at 23:59 +0800, Ming Lei wrote: > It is required that no dispatch can happen any more once > blk_mq_quiesce_queue() returns, and we don't have such requirement > on APIs of stopping queue. >=20 > But blk_mq_quiesce_queue() still may not block/drain dispatch in the > the case of BLK_MQ_S_START_ON_RUN, so use the new introduced flag of > QUEUE_FLAG_QUIESCED and evaluate it inside RCU read-side critical > sections for fixing this issue. >=20 > Also blk_mq_quiesce_queue() is implemented via stopping queue, which > limits its uses, and easy to cause race, because any queue restart in > other paths may break blk_mq_quiesce_queue(). With the introduced > flag of QUEUE_FLAG_QUIESCED, we don't need to depend on stopping queue > for quiescing any more. Hello Ming, Since this patch depends on patch 1 I will wait with reviewing this patch until there is agreement about patch 1 in this series. Bart.=