> On Sat, 14 Jul 2007, Haines Brown wrote: > > > My basic problem is that my kernel is unable to see my camera when it > > is connected, although it can see other USB devices. I assume there is > > no driver specific to a camera other than the lib provided by gphoto2, > > and so how can the modules work for all USB devices except one? > > "unable to see" isn't specific enough. There are lots of steps > involved, each with its own opportunity for something to go wrong.
By "unable to see", what I meant was that connecting and disonnecting my camera had no effect on kernlog or syslog (no messages there as a result of the actions), and the camera being on the bus is not reported by $ lsusb, nor do I see my camera in /proc/bus/usb/devices. The opposite is the case for all other USB devices. "Lots of steps" I suppose was the intent of my question. In searching for information, I've yet to see a description of the sequence of these steps. > For instance, with your old kernel did you use the gphoto2 library to > access the camera? It's not a question of an "old kernel", but of the kernel I've been using for months suddenly no longer seeing the camera being connected. Yes, I did use gphoto2 library to access the camera until the kernel no longer saw the camera, and then of course gphoto2 was no longer able to access it. > If you did, then the only module needed by the camera is usbcore. That's what I suspected, but then when one compiles a kernel, it seems you have an opportunity to specify what kind of camera chip is to be supported. Does this not refer to a module? > > CONFIG_USB_DEBUG=y > > > > Does this simply provide debug information in kernlog when the kernel > > fails to see my camera? > > It provides debug information in the kernel log whenever something > significant happens. It might reveal, for example, that the kernel > really _does_ see your camera but something goes wrong during an > initialization procedure. Thanks. > > If this reveals that the kernel USB support is buggy, what can I do > > about it? If a bug, there might be a patch. How would I know? If such > > a patch exists, I suppose I'd simply recompile my kernel with it. > > The best way to find the answers to these questions is to post the > debugging log. > > > Or I could upgrade the kernel. However, I gather doing so would give > > rise to dependency problems. Debian Etch uses kernel 2.6.18, and if I > > compile and install, say, 2.6.22 (current stable), won't I experience > > such problems? > > Probably not. But I am not at all familiar with Debian. The most > reliable way to find out is to try it and see what happens. On my wife's machine (running Debian Sarge), I find that usbdevfs is mounted, but on my current machine (runing Debian Etch), it is not mounted as best I can make out. From what I read, usbdevfs is necessary for the kernel to mount USB devices. Is that so, and how would I know I've got it mounted if it does not show up with the $ mount command? Haines Brown ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Linux-usb-users@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
