[t10] 13 cases in bus annex why how
LAVARRE at iomega.com
Mon Mar 17 10:12:05 PST 2003
* From the T10 Reflector (t10 at t10.org), posted by:
* "Pat LaVarre" <LAVARRE at iomega.com>
Nuts I forgot on this reflector I have to add my own prefix.
From: Pat LaVarre
Sent: Mon 3/17/2003 11:11 AM
To: t10 at t10.org
Subject: 13 cases in bus annex why how
Can someone say more clearly why T10 wants to know more of the 13 cases and what is it that we want to know?
At length ...
I hear T10 is working Now to understand more of how generic USB Storage (i.e. bInterfaceProtocol 'P' = "bulk only transport") is a SCSI-over-whatever protocol.
People attending the 12/13 Mar Dallas TX USA meeting saw more or less of "USB_MMC4_Annex_Rev10.doc" which said:
"D.5.6 The 13 Cases"
"The host software that decides the CBWCB owns the job of making sure the device's interpretation of the CBWCB agrees with bmCBWFlags.Direction and dCBWDataTransferLength. Usb Mass, if implemented as specified, works even if the host software fails to deliver this guarantee, but then the details are complex. See section '6.7 The Thirteen Cases' in the full USB Mass Storage Class Specification."
In response paraphrased, people attending said:
"The 13 Cases" needs more explanation. It's not clear what the 13 Cases are.
What's not clear? The claim here is that we don't have to care what the 13 cases are any time we have a host that agrees out-of-band in advance with the device over how many bytes of data the device will ask to copy In or Out in response to the Cdb copied Out.
How can that be in dispute?
Why does T10 want to know more of the 13 cases and what is it that we want to know?
Cluelessly, curiously, thankfully yours, Pat LaVarre
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org
More information about the T10