On Fri, 8 Mar 2002 10:20, Leif Jensen wrote:
> On Wed, 6 Mar 2002, Brad Hards wrote:
> > On Wed, 6 Mar 2002 08:30, Leif Jensen wrote:
> > > T:  Bus=01 Lev=01 Prnt=01 Port=01 Cnt=02 Dev#=  4 Spd=12  MxCh= 0
> > > D:  Ver= 1.10 Cls=02(comm.) Sub=02 Prot=00 MxPS= 8 #Cfgs=  1
> >
> > This is an illegal configuration from the Communication Device Class
> > spec, Table 20. Sub has to be 00.
>
> I was going by a list at the linux-usb site, but it must have contained
> the table 15-17 info, which I now presume applies only to the I: line.
> Is the spec on line?
Yes. http://www.usb.org/developers/docs.html
http://www.usb.org/developers/data/usb_20.zip is the main USB spec,
and http://www.usb.org/developers/data/devclass/usbcdc11.pdf is the 
Communication Device Class spec.
The USB spec is big.

<snip>
> I knew this couldn't be a proper way to operate the modem.  It does at
> least allow AT commands and responses to be exchanged.  This may be the
> underlying reason I could not exchange data on incoming calls before,
> since now the acm driver can do that (yay!).  How can I find out what
> the E: lines mean?
They are endpoint information. Look at the specs, and in the Documentation/usb 
directory in your linux source directory for a file called proc_usb_info.txt

<snip>
> Since then I've been able to get a working incoming data connection
> up at 14.4Kbps.  The fact that this basically works but I have no luck
> making outgoing calls is very frustrating.  Out is the direction I want.
> The incoming connection cannot be escaped by +++ as usual.  Is the
> escape sequence not going over the right usb interface?
Should be transparent. 

<snip>
> The two possibilites I see are
>
> 1) Data dialing is a premium service I'm not subscribed to.
Seems unlikely, unless your phone is locked in some way.

> 2) Additional communication needed to enable dialing from data connection.
Probably. 

> On the other hand there are no doubt many undocumented AT or other commands
> and no way to discover them.
Maybe you could try the "tridge" approach. Set up a program to exhaustively 
seach the AT space, recording which ones errored. Build up a tree of options, 
then try to visually assess it.

> The other important issue is that the responses from the phone are often
> garbled or accompanied by hundreds to thousands of repeated characters,
> usually 'A' or newline or some charater from the command it received.
> This is the same whether operated by the acm driver or a forced serial
> driver.  Assuming there is no bug in the acm driver, this means the
> phone must deviate somehow from the spec and will need its own driver
> to communicate properly.
That could be bad. Have you tried contacting the manufacturer?

Brad

_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users

Reply via email to