From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Anish Jhaveri To: Guan Junxiong , Christoph Hellwig CC: Jens Axboe , "linux-block@vger.kernel.org" , "Shenhong (C)" , "Sagi Grimberg" , Mike Snitzer , "linux-nvme@lists.infradead.org" , "Keith Busch" , Hannes Reinecke , "Johannes Thumshirn" , niuhaoxin Subject: RE: [PATCH 06/17] block: introduce GENHD_FL_HIDDEN Date: Sun, 29 Oct 2017 08:00:00 +0000 Message-ID: References: <20171023145126.2471-1-hch@lst.de> <20171023145126.2471-7-hch@lst.de> <20171024213959.GB23307@redhat.com> <20171028063825.GA14261@lst.de> <20171028074245.GA15510@lst.de> <99f3c781-71ce-f915-d6f4-88d6dd115876@huawei.com> In-Reply-To: <99f3c781-71ce-f915-d6f4-88d6dd115876@huawei.com> Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 List-ID: On Saturday, October 28, 2017 3:10 AM, Guan Junxiong wrote: >Does it mean some extra works such as: >1) showing the path topology of nvme multipath device Isn't connection to the target suppose guide the host to the shortest and f= astest path available path or so called most optimized path. Sysfs entry ca= n easily store that as we store path related info over there.=20 >2) daemon to implement immediate and delayed failback This is also based on target, whenever target is ready for immediate or del= ayed failback it will let the connect from host succeed. Until then host is= in reconnecting state across all path until it finds an optimized or un-op= timized path. Why is this extra daemon needed ? =20 >4) grouping paths besides ANA Why we cannot use NS Group here. Would be good idea to move away from legacy way of doing things for NVME de= vices. NVME Multipath implementation by Christoph is very simple. How about= not making it super complicated. =20 Best regards, Anish From mboxrd@z Thu Jan 1 00:00:00 1970 From: anish.jhaveri@paviliondata.com (Anish Jhaveri) Date: Sun, 29 Oct 2017 08:00:00 +0000 Subject: [PATCH 06/17] block: introduce GENHD_FL_HIDDEN In-Reply-To: <99f3c781-71ce-f915-d6f4-88d6dd115876@huawei.com> References: <20171023145126.2471-1-hch@lst.de> <20171023145126.2471-7-hch@lst.de> <20171024213959.GB23307@redhat.com> <20171028063825.GA14261@lst.de> <20171028074245.GA15510@lst.de> <99f3c781-71ce-f915-d6f4-88d6dd115876@huawei.com> Message-ID: On Saturday, October 28, 2017 3:10 AM, Guan Junxiong wrote: >Does it mean some extra works such as: >1) showing the path topology of nvme multipath device Isn't connection to the target suppose guide the host to the shortest and fastest path available path or so called most optimized path. Sysfs entry can easily store that as we store path related info over there. >2) daemon to implement immediate and delayed failback This is also based on target, whenever target is ready for immediate or delayed failback it will let the connect from host succeed. Until then host is in reconnecting state across all path until it finds an optimized or un-optimized path. Why is this extra daemon needed ? >4) grouping paths besides ANA Why we cannot use NS Group here. Would be good idea to move away from legacy way of doing things for NVME devices. NVME Multipath implementation by Christoph is very simple. How about not making it super complicated. Best regards, Anish