> At least, according to the spec, a port reset is the "more drastic action"
> that I should try next.
I'm writing to confirm that people appreciate a port reset is a POWER CYCLE to a
compliant device that draws its power from the bus? With bInterfaceClass ...Protocol
x08:MassStorage x50:BBB, class-specific reset is an option, so the port reset is very
much the last resort.
Suspend is another form of power cycle for a compliant bus-powered device.
And I think Set Configuration too?
I think maybe those three are the list of ops whose meaning changes when you work with
compliant bus-powered mass storage rather than other forms of storage?
To work cleanly with bus-powered mass storage, the host has to try class-specific
reset before resorting to Port Reset, and arrange for a spin down before any Suspend
(e.g. via bInterfaceClass x06 Scsi bCWBCB[0] = x1B Stop Unit).
The nasty thing about bus-powered mass storage is that people are writing host drivers
that load for generic mass storage without bothering to check bmAttributes &
x40:Self-Power and MaxPower of the Device Descriptor. This means they are innocently
shipping incompatible drivers, whoooops.
x4402 Pat LaVarre of iOmega [EMAIL PROTECTED] http://members.aol.com/ppaatt/
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
