On Thursday 04 July 2002 00:17, Nemosoft Unv. wrote: > On Wednesday 03 July 2002 16:01, Duncan Sands wrote: > > On Tuesday 02 Jul 2002 8:34 am, Greg KH wrote: > > > On Sun, Jun 30, 2002 at 07:17:19PM +0200, Duncan Sands wrote: > > > > PS: The taintedness and the "System.map says 2.5.24, pwcx-i386 says > > > > 2.4.18" is coming from the pwcx-i386 binary module that enables high > > > > resolution transfers from the camera. If anyone cares I will try > > > > reproducing the oops without it. > > > > > > Does the same oops happen in 2.5.24, without the pwcx-i386 module? > > > > I haven't been able to reproduce it without the pwcx-i386 module. > > So let's blame it on that binary module and forget it! > > I'm sorry, this is pissing me off. This is no way of handling a bug > report....
Hi Nemosoft, good to hear from you. First of all let me just thank you for writing the drivers for the Philips cameras. You've clearly put a lot of effort into them, and I see from the way you're telling me off that you want them to work perfectly. But don't you agree that this is a perfectly fine way of handling a bug report for an UNSTABLE kernel series: ignore it and hope it goes away? An oops seen once in a subsystem in flux is not something to spend a lot of time on. If I see it again in later unstable kernels, then I will certainly start chasing it seriously. > Did you try to reproduce it *exactly* like withthe pwcx module? Because you > wrote: > * Just got the following oops by plugging in my Philips Toucam Pro > * webcam, and unplugging it a few seconds later (before everything > * finished getting set up. > ...which leads me to believe there's a race condition somewhere. Yes, I tried to reproduce it exactly the same (of course). By the way, I haven't been able to get it again with or without the binary pwcx module. So it does indeed smell like a race condition in the USB subsystem. > If there really is a problem with the pwcx module, that's important because > a lot of people use that module. But I'm a bit skeptical since the pwcx > module is really simple, and does nothing with the kernel except for a > printk(). So it would be hard to trample over your system... Good to know it just does printk()s! By the way, I always wanted to ask you: does it use floating point for doing the decompression? I presume the answer is "no", but I just wanted to check. > Your Oops stacktrace points to somewhere in usbcore, and because you are > unplugging the camera, it may be a simple dangling pointer in there... Sure. All the best, Duncan. ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Caffeinated soap. No kidding. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
