On Wed, 2015-08-19 at 13:12 -0500, Mike Christie wrote: > On 08/19/2015 11:16 AM, Alex Gorbachev wrote: > > What is the difference, and is there willingness to allow LIO to be > > modified to work with this ESXi behavior? Or should we ask Vmware to > > do something for ESXi to play better with LIO? I cannot fix the code, > > but would be happy to be the voice of the issue via any available > > channels. > > I think we want to: > > 1. Allow lio to do more than wait for a command during aborts. For lio > we will want to add callouts similar to how we can override > discard/unmap behavior. > > 2. In the block layer add callouts/cmds so that we can abort > requests/bios at the LLD level. An API for explicit backend I/O cancellation might be useful. > > 3. For rbd, we will implement support for #2. In ceph then we would need > to add code to be able to track down commands and kill them if we can or > at least figure out what is going on and log a message so we do not have > these mysterious hung commands. > For the older make_request_fn() based raw block drivers using bio, the internal I/O timeout handling is completely implementation dependent. With modern blk-mq code, I/O timeout is driven by blk_mq_rq_timer() walking HW queues to complete request via blk_mq_ops->timeout() callback. AFAICT, blk-mq drivers are already expected to provide this callback to perform internal descriptor cleanup after request timeout. -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
