As of 2.5.44 (and in 44-bk2), if I unplug a USB device, then plug it in
again, it registers itself twice. This does not happen in 2.5.43.
drivers/usb/core/hcd-pci.c: uhci-hcd @ 00:07.2, Intel Corp. 82371AB/EB/MB PIIX4 USB
drivers/usb/core/hcd-pci.c: irq 11, io base 0000cce0
drivers/usb/core/hcd.c: new USB bus registered, assigned bus number 1
drivers/usb/host/uhci-hcd.c: detected 2 ports
drivers/usb/core/hub.c: USB hub found at 0
drivers/usb/core/hub.c: 2 ports detected
drivers/usb/core/usb.c: registered new driver hid
drivers/usb/input/hid-core.c: v2.0:USB HID core driver
drivers/usb/input/hid-core.c: v2.0:USB HID core driver
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 2
Manufacturer: Logitech
input: USB HID v1.00 Mouse [Logitech] on usb-00:07.2-1
drivers/usb/core/usb.c: USB disconnect on device 2
<unplug mouse here, then plug it back in>
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 3
Manufacturer: Logitech
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 4
Manufacturer: Logitech
This causes the following oops on reboot:
Unable to handle kernel NULL pointer dereference at virtual address
00000000
printing eip:
c01b9765
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0060:[<c01b9765>] Not tainted
EFLAGS: 00010246
EIP is at device_shutdown+0x75/0x8c
eax: 00000001 ebx: 00000000 ecx: 00000000 edx: ffffffff
esi: c0346ad0 edi: fee1dead ebp: bffffd8c esp: c90c3ea0
ds: 0068 es: 0068 ss: 0068
Process reboot (pid: 302, threadinfo=c90c2000 task=c9c0c6c0)
Stack: c90c2000 01234567 c011b10d c03a12e8 00000001 00000000 c90c2000 08049960
c90a6400 400deb60 c0121a9d c90c2000 c90a6400 c9f40000 c13c9000 000013c9
c13c9740 00031768 00000005 c0128523 c119f930 c13c9000 00000246 c13c976c
Call Trace:
[<c011b10d>] sys_reboot+0xdd/0x2a0
[<c0121a9d>] do_no_page+0x219/0x288
[<c0128523>] kmem_cache_free+0x19b/0x244
[<c0128523>] kmem_cache_free+0x19b/0x244
[<c0147ec9>] dput+0x19/0x184
[<c0136a96>] __fput+0xba/0xdc
[<c01369db>] fput+0x13/0x14
[<c01351e5>] filp_close+0x99/0xa4
[<c0135248>] sys_close+0x58/0x80
[<c0106eb3>] syscall_call+0x7/0xb
Code: 8b 1b 81 fb c0 6a 34 c0 75 c1 89 f1 ff 05 d0 6a 34 c0 7e 35
Here is the verbose USB Debugging messages:
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 301, change 1, 1.5 Mb/s
drivers/usb/core/hub.c: hub 0 port 1 connection change
drivers/usb/core/hub.c: hub 0 port 1, portstatus 301, change 1, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 303, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 2
drivers/usb/core/usb.c: new device strings: Mfr=1, Product=0, SerialNumber=0
Manufacturer: Logitech
drivers/usb/core/usb.c: usb_new_device - registering 1-1:0
drivers/usb/core/usb.c: usb_device_probe
drivers/usb/core/usb.c: usb_device_probe - got id
input: USB HID v1.00 Mouse [Logitech] on usb-00:07.2-1
drivers/usb/core/hub.c: port 2, portstatus 100, change 0, 12 Mb/s
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 100, change 3, 12 Mb/s
drivers/usb/core/hub.c: hub 0 port 1 connection change
drivers/usb/core/hub.c: hub 0 port 1, portstatus 100, change 3, 12 Mb/s
drivers/usb/core/usb.c: USB disconnect on device 2
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/usb.c: unregistering interfaces on device 2
drivers/usb/core/hcd.c: (no bus?): hcd_unlink_urb fail -22
drivers/usb/core/hcd.c: (no bus?): hcd_unlink_urb fail -22
drivers/usb/core/usb.c: unregistering the device 2
drivers/usb/core/hub.c: port 2, portstatus 100, change 0, 12 Mb/s
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 100, change 2, 12 Mb/s
drivers/usb/core/hub.c: hub 0 port 1 enable change, status 100
drivers/usb/core/hub.c: port 2, portstatus 100, change 0, 12 Mb/s
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 301, change 1, 1.5 Mb/s
drivers/usb/core/hub.c: hub 0 port 1 connection change
drivers/usb/core/hub.c: hub 0 port 1, portstatus 301, change 1, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: port 1, portstatus 301, change 0, 1.5 Mb/s
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 303, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 3
drivers/usb/core/usb.c: new device strings: Mfr=1, Product=0, SerialNumber=0
Manufacturer: Logitech
drivers/usb/host/uhci-hcd.c: cce0: suspend_hc
drivers/usb/host/uhci-hcd.c: cce0: wakeup_hc
drivers/usb/core/hub.c: port 1, portstatus 303, change 0, 1.5 Mb/s
drivers/usb/core/hub.c: new USB device 00:07.2-1, assigned address 4
drivers/usb/core/usb.c: new device strings: Mfr=1, Product=0, SerialNumber=0
Manufacturer: Logitech
drivers/usb/core/hub.c: port 2, portstatus 100, change 0, 12 Mb/s
--
Burton Windle [email protected]
Linux: the "grim reaper of innocent orphaned children."
from /usr/src/linux-2.4.18/init/main.c:461
On Wed, Oct 30, 2002 at 10:41:43AM -0500, Burton Windle wrote:
> As of 2.5.44 (and in 44-bk2), if I unplug a USB device, then plug it in
> again, it registers itself twice. This does not happen in 2.5.43.
Thanks for testing again with bk2, but I'm waiting for the fixed
reference counting logic from Pat Mochel to go into the tree before I
dig into this one.
Now if we had that Bugzilla instance that the IBM people mentioned a
while ago working, this report could be entered into that...
thanks for your patience,
greg k-h