From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Fri, 20 Jul 2018 10:23:21 -0600 From: Keith Busch To: Bart Van Assche Cc: "linux-scsi@vger.kernel.org" , "hch@lst.de" , "keith.busch@intel.com" , "linux-block@vger.kernel.org" , "linux-nvme@lists.infradead.org" , "axboe@kernel.dk" , "jianchao.w.wang@oracle.com" Subject: Re: [PATCH 2/2] scsi: set timed out out mq requests to complete Message-ID: <20180720162321.GG4093@localhost.localdomain> References: <20180719212618.2406-1-keith.busch@intel.com> <20180719212618.2406-2-keith.busch@intel.com> <1073d0d2902327970c4e28a4c7c97a21fd8885c8.camel@wdc.com> <20180720155646.GE4093@localhost.localdomain> <2b01f0a8d5864b2563573d46ce5b4ec5f593f538.camel@wdc.com> <20180720161240.GF4093@localhost.localdomain> <092dd4f2dae0cc533018c9fb829a50b4cbb6cb0b.camel@wdc.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <092dd4f2dae0cc533018c9fb829a50b4cbb6cb0b.camel@wdc.com> List-ID: On Fri, Jul 20, 2018 at 04:20:01PM +0000, Bart Van Assche wrote: > On Fri, 2018-07-20 at 10:12 -0600, Keith Busch wrote: > > On Fri, Jul 20, 2018 at 04:03:18PM +0000, Bart Van Assche wrote: > > > On Fri, 2018-07-20 at 09:56 -0600, Keith Busch wrote: > > > > SCSI is the only block driver that wants this behavior. Moving it back > > > > to generic where it used to be breaks other block drivers. > > > > > > That's new to me. What would break in the NVMe driver if the above change would be > > > present in the block layer? > > > > This is what causes the block layer to lose completions, and most drivers > > don't want the kernel to lose their completions. > > Hello Keith, > > Have you considered to introduce a fourth state for block layer requests to > avoid that completions that occur while a timeout handler is in progress get > lost? That would avoid that completions get lost not only for the NVMe driver > but also for SCSI drivers. See e.g. the MQ_RQ_TIMED_OUT state in > https://www.mail-archive.com/linux-block@vger.kernel.org/msg22196.html Yes, I've considered that, and I really want to use it, but scsi may still reference a freed request in scmd_eh_abort_handler that way.