On Fri, 2004-05-14 at 04:05, Cedric Pradalier wrote: > Hi, > > Everything works fine on my ibook with 2.6.5. I just tested 2.6.6. A > small hic appears when awaking from sleep. I use a speedtouch modem, > with the modem_run tool but not the kernel module. > > With 2.6.5, when I want to put the machine to sleep, I stop pppd and > modem_run. Then I call snooze. When awaking, I just restart modem_run > and launch pppd.. > > With 2.6.6, I got a oops restarting modem_run. I tried also to remove > usb modules, after awaking. The oops occurs when removing ohci_hcd.
Can you report that to the linux-usb list ? > May 13 19:41:13 ceruse kernel: usbcore: deregistering driver hid > May 13 19:41:21 ceruse kernel: ohci_hcd 0001:01:18.0: remove, state 1 > May 13 19:41:21 ceruse kernel: usb usb1: USB disconnect, address 1 > May 13 19:41:21 ceruse kernel: usb 1-1: USB disconnect, address 2 > May 13 19:41:21 ceruse kernel: Oops: kernel access of bad area, sig: 11 [#1] > May 13 19:41:21 ceruse kernel: NIP: DA189A28 LR: DA180114 SP: D7553CC0 REGS: > d7553c10 TRAP: 0301 Not tainted > May 13 19:41:21 ceruse kernel: MSR: 00001032 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 11 > May 13 19:41:21 ceruse kernel: DAR: 00000000, DSISR: 42000000 > May 13 19:41:21 ceruse kernel: TASK = d689a4e0[11968] 'rmmod' THREAD: > d7552000Last syscall: 129 > May 13 19:41:21 ceruse kernel: GPR00: C020DA10 D7553CC0 D689A4E0 D68C2A78 > D68C2A98 00000000 00000000 D78F8EE4 > May 13 19:41:21 ceruse kernel: GPR08: 00009032 C020DA10 00000000 00000000 > FFFFFFFA > May 13 19:41:21 ceruse kernel: NIP [da189a28] destroy_async+0x3c/0xdc > [usbcore] > May 13 19:41:21 ceruse kernel: LR [da180114] usb_unbind_interface+0x88/0x8c > [usbcore] > May 13 19:41:21 ceruse kernel: Call trace: > May 13 19:41:21 ceruse kernel: [da180114] usb_unbind_interface+0x88/0x8c > [usbcore] > May 13 19:41:21 ceruse kernel: [c00f2868] device_release_driver+0x84/0x88 > May 13 19:41:21 ceruse kernel: [c00f2a08] bus_remove_device+0x74/0xd0 > May 13 19:41:21 ceruse kernel: [c00f13e4] device_del+0xa8/0x114 > May 13 19:41:21 ceruse kernel: [c00f1468] device_unregister+0x18/0x30 > May 13 19:41:21 ceruse kernel: [da186968] usb_disable_device+0x9c/0xd8 > [usbcore] > May 13 19:41:21 ceruse kernel: [da180e58] usb_disconnect+0x9c/0x134 [usbcore] > May 13 19:41:21 ceruse kernel: [da180ee0] usb_disconnect+0x124/0x134 > [usbcore] > May 13 19:41:21 ceruse kernel: [da1892b0] usb_hcd_pci_remove+0x7c/0x164 > [usbcore] > May 13 19:41:21 ceruse kernel: [c00d4758] pci_device_remove+0x60/0x64 > May 13 19:41:21 ceruse kernel: [c00f2868] device_release_driver+0x84/0x88 > May 13 19:41:21 ceruse kernel: [c00f2898] driver_detach+0x2c/0x50 > May 13 19:41:21 ceruse kernel: [c00f2b94] bus_remove_driver+0x50/0xa8 > May 13 19:41:21 ceruse kernel: [c00f2f7c] driver_unregister+0x18/0x78 > May 13 19:41:21 ceruse kernel: [c00d49a8] pci_unregister_driver+0x1c/0x34 > > I can make some other tests if needed. Any idea ? > > The strange thing is that the oops reset the backlight... Yes, this is a "feature" of the oops code ;) Ben.