Mailing List archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[linux-dvb] Re: Call for help on CI problems



Szymon Polom writes:
 > Hi,
 > 
 > Quoting Henning Holtschneider (hh@holtschneider.com):
 > 
 > > > I run my linux dvb box now for more then a year .. as the driver switches
 > > > to a new CAM-API the problems started .. the
 > > > well-known-encrypt-after-replay problem.
 > > >
 > > > When will you try to fix thise problem ??? It started 10-11 Months ago.
 > > 
 > > Don't cry, my AlphaCrypt module doesn't work at all ...
 > > 
 > > > P.S: Don't tell me it's open source and I could fix it by myself .. I
 > > > CAN'T
 > 
 > > I discussed this with Martin Springer from Convergence recently. He told me
 > > that CI support simply had (and still has) no priority for Convergence. The
 > > driver for the Siemens/Technotrend line of DVB cards is more or less a
 > > test-bed for the software they are developing for embedded systems and there
 > > simply has been no need for a Common Interface yet. If I understood Martin
 > > correctly (BTW, thank you for your extensive answers to my questions!), this
 > > situation will not change in the near future. Better CI support will
 > > probably come to you DVB card once Convergence has sold its software to a
 > > receiver manufacturer that has a need for a working Common Interface ;-)
 > 
 > This is partially true. We're working on a CA/CI framework in the area of
 > device drivers, middleware and MHP now and will implement it as soon as
 > possible.
 > 
 > AFAIR there were no discussions at convergence if and how to publish such a
 > development until now.
 > 
 > As always, stay tuned...


One problem I see is that the documents describing the CI standard are,
AFAIK, normally internal to DVB members. If you look really carefully
you can find them on the internet but I don't know what kind of legal 
status those documents have.
Since IANAL I also cannot even speculate about the legality of an
open source CI implementation by either DVB members or non-members.

Since the CI protocol itself does not contribute anything to the
actual security of the system this is purely about security through
obscurity (which IMHO is worthless) but, again, IANAL. The publication
of information and code related to such sensitive areas has to be 
considered carefully or it could hurt us very much.
Maybe you now understand why I personally don't want to go into many 
details in discussions about CI, Soft-CAMs, etc. 

Regarding an open source firmware you have to consider 
that TI owns the library (RTSL) which is needed to talk to the
hardware. You would need them to agree to use the include files 
and library binaries to write open source software. 
There were indications last year that they might do this but I 
don't know how this developed. You would still need a >$3000 
ARM compiler from ARM which is the only one supporting the 
binaries TI provides.
If you want to use binutils/gcc you either need the RTSL sources 
(very unlikely), a complete register descriptions (also very unlikely),
or you have to completely reverse engineer the fimware 
binary (have fun :-).
The Technotrend software sources (which build on the RTSL) would 
of course also be helpful but not absolutely necessary. 


Ralph


-- 
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe linux-dvb" as subject.


Home | Main Index | Thread Index