[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[usb] RXValid signal
Hello everybody,
If anyone can help me, I have some questions about RXValid signal of USB
2.0 Transceiver.
1) In USB Trans spec (page 21) explains that "Each time 8 stuffed bits
are accumulated the state machine will enter the RX Data Wait state,
negating RXValid thus skipping a byte time". My questions:
- A bit stuff error supposes that after 6 ones in NRZ there is another
one (not a cero as supossed if well stuffed) ???
- "8 stuffed bits" means consecutive bits?
- This 8 stuffed bits would be a NRZ = 111111 0 111111 0 111111 0 111111
0 ... until 8 zeros ... ?
2) The assertion of RXValid differs in HS and FS modes. For HS is
explained in Figure 5 of referred specs (CLK could be 30 or 60 MHz).
- For FS would be like this?: CLK=48MHz --> RXValid is only asserted
during 1st cycle of the 32 for each byte, isn't it?
- So, for FS mode, RXValid must be negated just after cahnging to RX
Data state in the FSM of Figure 6 (Transceiver specs).
3) About Figure 6:
- It must work @ 48 MHz (FS) or 30-60 MHz (HS), just the CLOCKOUT, isn't
it?
- What does the "!Data" mean? Does it mean the hold register is not
full? There is a stuffed bit?
Well, I think that's enough for the first time I write to the list.
As you have seen, I'm designing a transceiver in FS mode to use it with
the USB 2.0 core of OPENCORES. This will help me (making some little
changes to do a HS mode model) to test this core (I wish). If anyone
needs help...
Thanks,
========================================================
Luis Jose Perez Lafuente [email protected]
Design Engineer
Digital Design Department
Design of Systems on Silicon http://www.ds2.es
Av. Charles Robert Darwin, 2 Phone. +34-96-136 60 04
Parc Tecnologic Ext. 152
46980 Paterna (VALENCIA) FAX +34-96-136 62 50
SPAIN
========================================================
--
To unsubscribe from usb mailing list please visit http://www.opencores.org/mailinglists.shtml