Hello David,

  There's a patch that has been discussed to do with 
resetting "hung" Logitec cameras;
 
https://lists.berlios.de/pipermail/linux-uvc-devel/2007-March/001476.html

  It does a check to see if the device is unhappy;

+       /* leave usb device in a clean state */
+       if (video->dev->state & UVC_DEV_IOERROR)
+               uvc_video_reinit(video);

.. and then re-initialises it on demand.  Depending on
what you know about the device when it enters and/or
leaves this calibration period, you could either put a 
timer in there, or leave it event driven based on the
device state (if you knew what that state was when the
calibration was complete -vs- when it was performing
calibration).

  It wouldn't fix all of your problems tho - that code will
re-initialise the device, cutting off your user-space stream
(I believe).


  Following up on your earlier thread;

http://www.mail-archive.com/[email protected]/msg00708.html

.. along with the info below.  What "hex file" did you
end up using?  I'm interested for two reasons;

  Initially because I'm about to buy either a heap of EZ USB
or a heap of CP2101 devices to use as USB-Serial/TTL 
bridges .. and I wasn't expecting to have to work with the
the local logic (from Linux) in order to use it as a pass-thru
device.

  Also, because I'm interested in doing thermal/IR imaging.  
I looked at the Thermoteknix gear, but couldn't see 
anything that was both small and would integrate into 
Linux -- if there's a camera that works, then I'd be keen to
try it out.


----- Original Message -----
>From: "David C Trotz Jr" <[EMAIL PROTECTED]>
>To: <[email protected]>
>Subject:  [Linux-uvc-devel] Timeout?
>Date: Mon, 26 Mar 2007 23:02:56 -0700
>
> Hi all,
> Its been a while since I last posted but here is a quick
update. I  
> last asked about why my UVC device is coming up as not
being a UVC  
> device and a couple of you correctly pointed out that my
cameras,  
> Thermoteknix Miricle IR cameras must use EZ USB. I got a
hold of the  
> hex file and I can now hot plug the cameras and udev does
the rest.  
> So this is good news, thanks for the advice.
> 
> Now I am having a problem with the hardware. It seems the
cameras  
> like to do a calibration sequence from time to time to
make sure that  
> the warmer areas are distinct from the cooler areas. But
when it does  
> this the video stream dies and I have to restart the
program in some  
> cases and in others completely reload the uvc kernel
driver. I am  
> thinking that the uvc driver times out while the cameras
do their  
> calibration because the driver thinks the camera is down
when in fact  
> it is on hold. Is there somewhere specifically in the code
where I  
> can look in the uvc driver for any timeouts for video
streaming, or  
> do any of you have any other ideas for what might cause
this behavior?
> Thanks!
> --
> David Trotz
> Associate Member of Technical Staff, Computer Vision Group
> Jet Propulsion Laboratory
> Mail Stop 198-235
> 4800 Oak Grove Drive
> Pasadena, CA 91109
> Office 818-354-1897, Fax 818-393-5007
> 
> 
> _______________________________________________
> Linux-uvc-devel mailing list
> [email protected]
> https://lists.berlios.de/mailman/listinfo/linux-uvc-devel
> 


--
Ian Latter
Late night coder ..
http://midnightcode.org/
_______________________________________________
Linux-uvc-devel mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/linux-uvc-devel

Reply via email to