On Mon, 2011-01-17 at 18:18 +0100, Tejun Heo wrote:

Hello, James.

On Mon, Jan 17, 2011 at 10:18:51AM -0600, James Bottomley wrote:

quoted

quoted

The problem is that the PMP device itself is currently not allocated a
userland visible device, so it doesn't have any /dev/* node.  Hmmm...

So perhaps it should be.  If you look at the equivalent topology on SAS,
our expanders have a bsg device node precisely so that we can do this.

That said, SAS expanders have a defined protocol (SAS Management
Protocol) to talk to the outside world, so they are real visible objects
always in our topology ... I'm not sure PMP has this ... it seems that
all PMP visibility is an extension to the standard?

SATA PMP is mostly a dumb switch and there isn't much which can be
done by issuing custom commands (and IIRC we didn't have bsg back
then), so it was never made visible to userland, but yeah probably
exporting a bsg node is a good idea.  Inside libata, the PMP device
has its device representation and all so it shouldn't be too difficult
either.  Not really sure how the inquiry and stuff should be handled
tho.

Expanders aren't SCSI devices either ... that means they don't appear as
visible to standard SCSI mechanisms like INQUIRY (SMP isn't a SCSI
protocol it's a SAS extension).  They just appear as part of the
topology in the device tree.

James