On Tue, 2018-08-14 at 11:42 +-0000, Stanislav Nijnikov wrote: +AD4- From: Bart Van Assche +AD4- +AD4- Are you perhaps referring to the transport+AF8-class+AF8-register() calls in SCSI +AD4- +AD4- transport drivers? From what I see in existing SCSI transport drivers the +AD4- +AD4- transport+AF8-class+AF8-register() function is used to register link, port, host, +AD4- +AD4- vport, rport and other objects. I don't think that a SCSI transport driver +AD4- +AD4- is required to register host and port objects. +AD4- +AD4- +AD4- +AD4- Maybe we should take a step back and discuss first why the new bsg queues +AD4- +AD4- are registered by a transport driver? Since in case of UFS as far as I can +AD4- +AD4- see there is no real need to introduce a transport driver other than for +AD4- +AD4- creating the bsg device nodes, have you considered to add the code for +AD4- +AD4- creating bsg device nodes to the UFS driver instead of in a UFS transport +AD4- +AD4- driver? I think transport drivers were introduced as a way to share code +AD4- +AD4- between multiple SCSI LLDs that use the same transport mechanism. In the +AD4- +AD4- case of UFS there is only one SCSI LLD. Hence I'm wondering whether we +AD4- +AD4- really need an UFS transport driver. +AD4- +AD4- At the moment, the SCSI transport related code could be found at driver/scsi/scsi+AF8-transport+AF8AKg- files. +AD4- What is a point of hiding the UFS transport code inside the UFS driver? Have you tried to implement the approach I proposed? If so, did you encounter any issues that made it impossible to implement that approach? If you have not yet tried to implement the above proposal, what are you waiting for? Bart.
