Jeroen Vreeken <[EMAIL PROTECTED]> writes:

> I have attached version 0.21 and 0.22 you can just drop in the two files
> and recompile the driver.

o I can insmod/rmmod 0.20, 0.21, and 0.22 at will, and they all work.

o Doing:

    # insmod /lib/modules/2.4.9-ac12/kernel/drivers/usb/se401.o
    # tail -f /var/log/messages
    se401.c: SE401 usb camera driver version 0.23 registering
    usb.c: registered new driver se401
    se401.c: SE401 camera found: Kensington VideoCAM 67016
    se401.c: firmware version: 46
    se401.c: ExtraFeatures: 0 Sizes: 160x120 200x152 176x144 320x240 352x288
400x300 640x480 704x576 800x600
    se401.c: registered new video device: video0
    usb-uhci.c: interrupt, status 3, frame# 636
    usb-uhci.c: interrupt, status 3, frame# 660
    usb-uhci.c: interrupt, status 3, frame# 684
    ...

o This goes on forever until killed with:

    # rmmod se401
    # tail -f /var/log/messages
    usb.c: deregistering driver se401
    se401.c: ohoh: device vapourished
    se401.c: Kensington VideoCAM 67016 disconnected
    se401.c: SE401 driver deregistered


o A couple of things ...

joey tsai <[EMAIL PROTECTED]> writes:

> I'm using the 0.22 driver you've attached.
> 
> However, when cam programs are running, they eventually die, and I get
> these messages:
> kernel: usb-uhci.c: interrupt, status 2, frame# 138
> kernel: usb-uhci.c: interrupt, status 2, frame# 1228
> kernel: usb-uhci.c: interrupt, status 2, frame# 304

o I also get this failure eventually, at least with 0.20 (haven't
  tested the others enough).  It doesn't seem to happen if I keep
  capturing images in a single run of my apps (which don't close the
  device between captures), but happens if I run the apps enough
  times (each run opens, captures, and closes the device).


o Also ... Doing `modprobe usb-uhci` always loads the default 0.23
  driver that was installed with `make modules_install`.  It does
  this (loads 0.23) even if I do:

    # cd /lib/modules/2.4.9-ac12/kernel/drivers/usb
    # mv se401.o se401-0.23.o
    # mv se401-0.20.o se401.o
    # depmod -ae

o Is this something sort of LILO-like, "the kernel loads directly
  from disk cylinder/head/sector instead of from filesystem
  semantics" behavior?  I always have to `rmmod se401` before
  insmod'ing se401-0.20.


--
MARK
[EMAIL PROTECTED]

_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users

Reply via email to