Yoni.Shternhell at sandisk.com
Tue Feb 12 22:59:12 PST 2008
Formatted message: <A HREF="r0802130_f.htm">HTML-formatted message</A>
Upon first review of the UAS draft (T10/2095-D) one cannot escape several
1. Even though one of the triggers leading to this (UAS) project was the
introduction of USB 3.0 (see T10/07-400r1), the initial draft carefully
avoids any direct mention or reference to it. While this approach might be
considered as a valid workaround avoiding the need to execute the USB 3.0
Contributors Agreement required by the USB 3.0 Promoters Group in order to
gain access to the evolving USB 3.0 draft specifications, it does not make
any technical sense: one does not have to be a USB 3.0 Contributor to
understand that USB 3.0 will be significantly different from USB 2.0,
therefore, in order for UAS to be compatible or at least non-conflicting with
USB 3.0, the UAS specification must reference the USB 3.0 specification.
2. One of the key promises of UAS (according to the initial draft) is:
"does not interfere with the USB Mass Storage Class (MSC) bulk-only
transport"). This is a non-trivial promise, deserving significant attention;
therefore I suggest including in the UAS document a dedicated chapter about
the envisioned coexistence/non-interference/inter-operability between a
device and host, taking into considerations all the valid/envisioned
configuration combinations on both the host and device side (UAS enabled vs.
"classical" USB host protocol stack or both). This suggested new chapter in
the UAS spec, IMHO, must be discussed and agreed upon even before further
embarking into the bits and pieces of the UAS protocol details. Furthermore,
this chapter must be cooperatively coordinated with both USB IF and the USB
3.0 Promoter Group in order to ensure fulfillment of the above cited promise.
Yoni Shternhell | Standardization Engineer- CTO Office | SanDisk | o.
+972.(0)9.7632594 m. +972.(0)54.9222041| f. +972.(0)3.548-8666|
More information about the T10