[cpia] still hoping

[Peter Pregler]

On 14-Mar-2000 Bastian wrote:
> At 08:52 14.3.2000 +0100, Peter Pregler wrote:
> 
>>On 13-Mar-2000 Bastian wrote:
>> > anybody made some new experiences concerning the "10-10-..." problem
>> > ?
>>
>>Not directly. I try to get some updated documentation for the 1.30
> 
> my camera seems to have firmware version 1.00

Actually it has version 1.02 (1.0) according to your postings.

>>firmware and some hints about the correct camera-initialisation from
>>VLSI.
>>In the meantime you can try out the following:
>>
>>- did you try to run the camera under windows and a soft-boot (i.e. no
>>power down) to linux?
> 
> under windows it works.

I guessed that. What I wanted to know is if windows does any
initialisation that our driver does not.

I checked out the mailing list archive and I cannot think of any good point
to start debugging on my side anymore. Can you try out the following:

- try out the driver version 0.4.4, that is the first one which
should work reliable with 1.02 firmware, so we can see if any bug was
introduced after that
(http://download.sourceforge.net/webcam/cpia-0.4.4.tgz), do not activate
dma please

- can you boot your box fresh in emergency mode, load the parport-modules,
v4l and the driver and _no_ other fency stuff, especially not the ppp/isdn
stuff you got in the ooops, use vidcat or the like to grab a single frame
to disk (or /dev/null ;), if it fails send me the syslog from boot to end,
and also the output of the relevant proc-file (interrupt, parport, cpia).

Sorry for not being of more help. But something in your setup is very
special and I have no idea what it is.

-Peter


-------------------------------
Email: Peter.Pregler@risc.uni-linz.ac.at
WWW:   http://www.risc.uni-linz.ac.at/people/ppregler