On Thu, 01 Oct 2015 07:42:13 +0200, Hannes Reinecke wrote: > > Would there be any benefit in making this a per-target, per-tpg, or > > per-lun attribute? (I can't think of any but wanted to raise the > > question.) > > > The vendor ID is describing the target implementation. > And it'll be rather hard to have different implementations for LUNs > being exported from the same host ... > > We need to be careful to not clutter the vendor ID namespace. > Most OS uses it to distinguish target implementation details > (think of multipathing) and it will become an absolute mess > if we have different vendor ID per LUN, all using the same > implementations. The inquiry PRODUCT IDENTIFICATION (model) and PRODUCT REVISION LEVEL fields can currently differ across LUs with differing backstores. To remain consistent with these fields, it may make sense to also allow VENDOR ID changes on a per LU basis. SPC-4 (r37) isn't very clear: 6.6.2 Standard INQUIRY data ... The T10 VENDOR IDENTIFICATION field contains eight bytes of left-aligned ASCII data (see 4.3.1) identifying the manufacturer of the logical unit. 7.8.6.4 T10 vendor ID based designator format ... The T10 VENDOR IDENTIFICATION field contains eight bytes of left-aligned ASCII data (see 4.3.1) identifying the manufacturer of the product. Cheers, David -- 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
