On Mon, May 13, 2002, Nemosoft Unv. <[EMAIL PROTECTED]> wrote: > I tried to jump to kernel 2.5.15 today, but with not much success... After > the usual boot I load the USB modules. The attached webcams are detected, but > when I try to get a picture, the program hangs. I can ^C it, but a second > attempt causes a 'D' process that's hanging in "down_i" (that's all 'ps' > gives me). Maybe a lock too much? I tried both uhci and usb-uhci. > > Also it doesn't matter which devices are attached, the CPiA cam or my Philips > cams. 'devices' does show all the info, so maybe its an ISOC issue...
That would be weird. Isoc doesn't have a different path, it's all the same code. I wonder if there's a v4l bug, that's the only thing I can think of that's common. > There have been other oddities. Once, /proc/bus/usb/devices showed only a few > webcams. When I unplugged a cam, for testing, my kernel log suddenly started > scrolling with detection routines of the rest of the devices. Weird. Sounds like khubd got stuck or something. > One other time I had loaded uhci, tried the cam, then unloaded it, with an > Oops as a result. I'm not sure if it's relevant. Probably not. This is the result of the reference counting changes in 2.5. I sent a patch out this morning which fixes that. > Since the last 2.5. kernel I got working was 2.5.6, maybe I'm missing > something? I'm not sure where to begin debugging this time... Any help, hints > and suggestions are appreciated. I haven't tried a camera with 2.5. I'll try to reproduce the problem. JE _______________________________________________________________ Have big pipes? SourceForge.net is looking for download mirrors. We supply the hardware. You get the recognition. Email Us: [EMAIL PROTECTED] _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
