[linux-dvb] dib0700 problems (nova-T usb stick)

Soeren Moch Soeren.Moch at stud.uni-hannover.de
Mon Jun 11 14:02:57 CEST 2007


Patrick Boettcher wrote:
> Hi,
> 
> The biggest problem is, that we have no time to investigate the issue 
and 
> it is no problem on the Windows-platform (because of a different call 
> sequence).
> 
> It has nothing to with the I2C-gating used. It is a basic problem of 
doing 
> I2C while streaming is enabled. Not 100% what is exactly the problem.
> 
> I filed the problem to the correct guys at DiBcom - they will see 
(sooner 
> or later).

Hi Patrick,

maybe the DiBcom guys can also have a look on the disconnect problem,
which also seems to be related to the REQUEST_ENABLE_VIDEO firmware call.
The first urb received after switching video streaming on has in almost
all cases a error return state, sometimes a disconnect occurs (only
directly after a REQUEST_ENABLE_VIDEO call, Version 02-rc2 firmware used).
The firmware seems to deliver corrupt data after stream-on.

Since the i2c gating problem can be solved with the driver patch, the
disconnect problem is the last real problem with this stick.
As the 0700 bridge seems to be part of the shiny new 7770 device, fixing
this firmware shouldn't be wasted time for DiBcom.

Thank you again,
S:oren

> Sorry,
> Patrick.
> 
> On Thu, 7 Jun 2007, Soeren Moch wrote:
>> Hello Patrick!
>>
>> Patrick Boettcher wrote:
>>> It is _no_ hardware restriction.
>>>
>>> The symptoms described by Soeren is correct, but the conclusion is 
>> wrong. 
>>> In fact this behaviour seems to be specific to the open source driver, 
>> we 
>>> don't have this problem in our reference driver.
>> The "I2C gating feature" seems to be related to the 
REQUEST_ENABLE_VIDEO 
>> firmware call. If the same hardware with same firmware does not show 
this 
>> behavior with your reference drivers, are there other parameters in 
your 
>> reference REQUEST_ENABLE_VIDEO call?
>> Maybe this I2C gating is done externally via some gpio pins?
>> (Without schematic, datasheets and firmware code it is difficult to get
>> the big picture.)
>>
>>> If I only would have time, I would take a look. 
>> Sorry for asking again. Maybe you can solve tuning and frontend status
>> problems by "simply" changing the firmware call parameters. If I can
>> be helpful in any way, don't hesitate to ask.
>>
>>> Did you try the new firmware?
>>>
>>> Patrick.
>>>
>> Thank you,
>> S:oren
>>




More information about the linux-dvb mailing list