Matt D:
I don't understand.
You don't meant to say there exist devices which
react to a NAKed Out? Devices that spin up only
when the first CBW arrives, but then wait to
complete spin up before ACKing that CBW? Core
USB allows this possibility ... but this would
be Very Rude?
Perhaps you mean to say that there are devices
rude enough that they announce their presence
on the bus (e.g. via raising D+) but then
they NAK arriving CBWs until after the initial
spin up?
Or do you mean something else?
Help?!?!? Thanks again in advance.
x4402 Pat LaVarre of iOmega [EMAIL PROTECTED] http://members.aol.com/ppaatt/
>>> Matthew Dharm <[EMAIL PROTECTED]> 05/06/00 08:19PM >>>
If a device has to spin up, it can either NAK during the data stage while
the devices comes to speed, or it can NAK the CBW while it's spinning up.
The Bulk-only spec doesn't say when the delay should be done... and some
devices do it in the CBW and others during the data stage. Since timeouts
are done spearately for the CBW and for the data stage, it's hard to get
good values, since allowing enough time for spinup during each means a
possible 32 second delay -- which makes the SCSI midlayers a bit upset.
Matt Dharm
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
