Driver V4.0 - My Findings

[Peter Pregler]

On 17-Nov-99 Chris Whiteford wrote:
> 
> [root@tennille /home/chris/cpia/client]# ./gqcam 
> /dev/video: No such device
> 
> When infact the /dev/video device does actually exist.  hmmmm

The message has nothing to do with the existence of the inode. It does
mean that no kernel-routine feels responsible to handle the open, i.e. if
the module seems load it means the kernel driver 'crashed'. As for the
errors I try to reproduce them. They seem to be just by some unclean
close/release of the module. I use some new allocation stuff that I will
put together for an update. That might solve your problems too.

Some more comments:

- I was able to try out isabel - a video conferencing tool - (Austria
  to Spain, TEN-155 as the basic network provider). It was quite nice. :)
- The kernel did not crash on my SMP-test machine during operation
  with the gqcam-application. Interrupts were taken even by both
  processors. Actually I think that that part of the driver _is_ SMP-safe.
  However, I will some more nasty tests to find possible race conditions
  etc. Has anyone a good (graphics) tool where one can see the cpu-usage of
  each of the processors?
- Regarding SMP mmap is still a killer. Do not try it out. I will fix this
  until next week.

Greetings, Peter
 

-------------------------------
Email: Peter.Pregler@risc.uni-linz.ac.at
WWW:   http://www.risc.uni-linz.ac.at/people/ppregler
-----------------------------------------------------------------------------
To unsubscribe from this mailinglist, send the line "unsubscribe vision-webcam" in the
body of a message to "majordomo@errors.no".