|
Message
From: S.Georges<info@s...>
Date: Fri Nov 26 14:21:51 CET 2004
Subject: [usb] Enumeration...
> > hello,
> > now i got the USB1.1 function core working with the SMSC GT3200
> PHY
> > chip. (little changings in usb1_utmi_if.v and usb_pa.v, because the
> > tx-
> > ready signals have different length)
> > but windows dont accepts the enumeration :-(
> > it asks for 64Byte device-descriptor, but since this is only 18Byte
> > long, i
> > am sending these 18 bytes (the original usb1_ctrl.v does so)
> > after PC had recieved these 18 bytes it sends again an IN-Token, to
> > wich the fpga replies with Zero-length-packet.
> > a lot of miliseconds later the PC sets an USB-RESET and asks again
> > for
> > 64 byte device-descriptor...
> > this repeats 3 times.. then it stops...
> > what do i have to do?
> > sending these 18 bytes only in packets a 8 byte?
> > sending 64 bytes? wich bytes?
> > or could it be, that my device-descriptor is wrong?
> I have got exactly the same problem.
> Do you send the 18 bytes device descriptor in chunks of 8 bytes
> packets?
> I realised that you also should transmit the zero len packet after each 8
> byte packet.
> How do you send the last 2 bytes?
> Do you receive the reset command from the host after sending the 1rst
> 8 bytes? Do you receive the SET_ADDRESS request?
> _______________________________________________
Hello,
now, i made a little progress..
i found, that i had some problems with the handshaking with the GT3200
because the gt3200 assumes, that the first 3 bytes come with each clock. (it
sets the first txReady for 3-4 clocks and not only one clock, like the PHY-Core
form opencores does.
after correcting these issues, the procedure was following:
PC asks for 64bytes of descr.
FPGA sends 18 byte in one packet.
PC sets USB-reset
PC sends SET-ADress-comand.
FPGA ack
PC asks for 18 bytes dev descriptor with new adress
fpga replys
pc asks for 9 byte of 2nd descriptor
fpga replys
pc asks for 255 bytes of 2nd descriptor
fpga replies..
an then there are 2 USB-resets.. and nothing else...
pc says device not working corrctly..
i musst say, that while i measured the procedure above, there were CRC-errors. i
fixed this today(usb1_idma.v and fifo2.v), but could not test it on a
logicanalyzer yet.. pc still doesn't accept it :-(
simulation is ok. :-/
best regards
suheel
|
 |
Browse
