On Sun, Jul 14, 2002 at 09:50:22PM -0700, David Brownell wrote: > I hit this oops recently on 2.5.25. Is this familiar? > Maybe fixed in that recent set of patches? I've seen > a few HID-related disconnect problems before, this is the > first one that dropped out as such a polite oops. > > What seems to be happening is that the hid-core code > is unlinking an urb that's already been unlinked; > that "5a" byte pattern says the memory has already > been poisoned. One fix might be to free urbs as soon > as they're unlinked, so such confusion can't arise.
I think I fixed this in my last patch to hid-core. If it happens with 2.5.26, bug me again, please. > > - Dave > > > ksymoops 2.4.4 on i586 2.5.25. Options used > -V (default) > -k /proc/ksyms (default) > -l /proc/modules (default) > -o /lib/modules/2.5.25/ (default) > -m /root/kernel/2.5.25/System.map (specified) > > Unable to handle kernel paging request at virtual address 5a5a5b26 > c882dd6f > *pde = 00000000 > Oops: 0000 > CPU: 0 > EIP: 0010:[<c882dd6f>] Not tainted > Using defaults from ksymoops -t elf32-i386 -a i386 > EFLAGS: 00010206 > eax: 5a5a5a5a ebx: c6e67f64 ecx: c6e67f64 edx: c65071fc > esi: 00000000 edi: c799b010 ebp: c6342000 esp: c685bea0 > ds: 0018 es: 0018 ss: 0018 > Stack: c01d0135 c65071fc c6e67f64 c884329b c6e67f64 00000000 c01d0ab2 c6e67f64 > c6e67f64 c799a000 c8854e40 c7f96dcc c88514f3 c799b010 c8854e58 c8829fa6 > c6342000 c799a000 00000000 00000000 c62f2400 c62f25d0 c7f96bbc c62f2400 > Call Trace: [<c01d0135>] [<c884329b>] [<c01d0ab2>] [<c8854e40>] [<c88514f3>] > [<c8854e58>] [<c8829fa6>] [<c882a01d>] [<c882a01d>] [<c882be04>] [<c882c1aa>] > [<c8835720>] [<c882c3ab>] [<c0111f80>] [<c0105596>] [<c882c380>] > Code: 8b 80 cc 00 00 00 85 c0 74 0d 8b 40 1c 85 c0 74 06 52 ff 50 > > >>EIP; c882dd6f <[usbcore]usb_unlink_urb+f/40> <===== > Trace; c01d0135 <input_close_device+15/20> > Trace; c884329b <[keybdev]keybdev_disconnect+b/20> > Trace; c01d0ab2 <input_unregister_device+52/d0> > Trace; c8854e40 <[hid]hid_driver+0/40> > Trace; c88514f3 <[hid]hid_disconnect+63/80> > Trace; c8854e58 <[hid]hid_driver+18/40> > Trace; c8829fa6 <[usbcore]usb_disconnect+86/160> > Trace; c882a01d <[usbcore]usb_disconnect+fd/160> > Trace; c882a01d <[usbcore]usb_disconnect+fd/160> > Trace; c882be04 <[usbcore]usb_hub_port_connect_change+94/310> > Trace; c882c1aa <[usbcore]usb_hub_events+12a/300> > Trace; c8835720 <[usbcore].rodata.end+1d15/7d55> > Trace; c882c3ab <[usbcore]usb_hub_thread+2b/f0> > Trace; c0111f80 <default_wake_function+0/40> > Trace; c0105596 <kernel_thread+26/30> > Trace; c882c380 <[usbcore]usb_hub_thread+0/f0> > Code; c882dd6f <[usbcore]usb_unlink_urb+f/40> > 00000000 <_EIP>: > Code; c882dd6f <[usbcore]usb_unlink_urb+f/40> <===== > 0: 8b 80 cc 00 00 00 mov 0xcc(%eax),%eax <===== > Code; c882dd75 <[usbcore]usb_unlink_urb+15/40> > 6: 85 c0 test %eax,%eax > Code; c882dd77 <[usbcore]usb_unlink_urb+17/40> > 8: 74 0d je 17 <_EIP+0x17> c882dd86 ><[usbcore]usb_unlink_urb+26/40> > Code; c882dd79 <[usbcore]usb_unlink_urb+19/40> > a: 8b 40 1c mov 0x1c(%eax),%eax > Code; c882dd7c <[usbcore]usb_unlink_urb+1c/40> > d: 85 c0 test %eax,%eax > Code; c882dd7e <[usbcore]usb_unlink_urb+1e/40> > f: 74 06 je 17 <_EIP+0x17> c882dd86 ><[usbcore]usb_unlink_urb+26/40> > Code; c882dd80 <[usbcore]usb_unlink_urb+20/40> > 11: 52 push %edx > Code; c882dd81 <[usbcore]usb_unlink_urb+21/40> > 12: ff 50 00 call *0x0(%eax) > > > -- Vojtech Pavlik SuSE Labs ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
