Re: real-world experience with QuickCam in the kernel
>I found a couple of small bugs in my driver (no big deal), but I'm
>now convinced that without some serious head-scratching, doing this
>in kernel mode is a big lose. In fact, due to the extra copies and
>syscall overhead, I'm at least 5% slower than xfqcam.
This will be a problem (how big?) for motion video, but won't matter
at all for someone who just wants to make snapshots. I really want to
be able to do "ppmtogif < /dev/quickcam > snapshot.pgm".
Maybe the solution is to put in a device driver, but have one of the
options somehow be to hand over control of the camera to a user
process if they want to go to the trouble of hacking the camera
themselves. Something would have to be done to insure they don't
screw up the camera, though.
I looked at an SGI's camera yesterday, by the way, and got seriously
bummed. 30fps 640x480 full colour, no noticeable load on the machine.
I guess that's what you get when you have real IO. Of course, my Linux
box with QuickCam was 3x cheaper :-)