On Mon, Jul 21, 2014 at 01:51:26PM +0200, Hannes Reinecke wrote: > Yes, I have. How well does the SCSI layer recover when you write outside the write pointer? ZBC defeinds new sense codes that we don't handle, so I'd expect we'd get long loops of unrecovered errors. > I was actually considering implementing a module option for sd > (eg attach_zbc) to allow zac/zbc devices to be attached to the sd driver. > Which will be off per default, so your concern should be addressed. That sounds fine to. > However, zbc devices _are_ similar to normal 'sd' devices; it's only that > some commands might fail unexpectedly. > But all the usual commands from sbc are supported, so I found it a bit odd > having to implement my own device driver (which would be a clone of 'sd' > anyway). They are similar in some ways and very different in others. I defintively agree we should share as much code as possible, but I'm not sure exposing them as plain sd devices is a good idea. I'm defintively fine with an optional attachment to sd to start playing around with different ways to make use of them. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
