usb serial/visor oops in 2.4.22-pre2

usb serial/visor oops in 2.4.22-pre2

[Justin A]

basically, pilot-xfer and all don't work anymore, dmesg reports:

hub.c: new USB device 00:11.2-1, assigned address 3
usb.c: USB device 3 (vend/prod 0x82d/0x100) is not claimed by any active
driver.
usb.c: registered new driver serial
usbserial.c: USB Serial support registered for Generic
usbserial.c: USB Serial Driver core v1.4
usbserial.c: USB Serial support registered for Handspring Visor / Treo /
Palm 4.0 / Clié 4.x
usbserial.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x converter
detected
visor.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x: Number of ports:
2
visor.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x: port 1, is for
Generic use and is bound to ttyUSB0
visor.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x: port 2, is for
HotSync use and is bound to ttyUSB1
usbserial.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x converter now
attached to ttyUSB0 (or usb/tts/0 for devfs)
usbserial.c: Handspring Visor / Treo / Palm 4.0 / Clié 4.x converter now
attached to ttyUSB1 (or usb/tts/1 for devfs)
usbserial.c: USB Serial support registered for Sony Clié 3.5
visor.c: USB HandSpring Visor, Palm m50x, Treo, Sony Clié driver v1.7
usb-uhci.c: interrupt, status 3, frame# 1201
usb.c: USB disconnect on device 00:11.2-1 address 3
visor.c: Bytes In = 0  Bytes Out = 0
Unable to handle kernel NULL pointer dereference at virtual address
000009a4
 printing eip:
e2302d96
*pde = 00000000
Oops: 0002
CPU:    0
EIP:    0010:[<e2302d96>]    Tainted: P 
EFLAGS: 00013246
eax: 00000000   ebx: dd66d01c   ecx: ddf30000   edx: 00000001
esi: 00000001   edi: dd66d000   ebp: 00000000   esp: defdbf4c
ds: 0018   es: 0018   ss: 0018
Process khubd (pid: 61, stackpage=defdb000)
Stack: e2303fa0 df44bb40 e2303fc0 e097f09a de963200 dd66d000 00000100
00000001 
       df0b4600 00000000 00000000 de963200 e09814da df0b4710 00000000
00000001 
       df0b4600 de8e62c0 00000000 de8e62e8 0000000a e098177d de8e62c0
00000000 
Call Trace:    [<e2303fa0>] [<e2303fc0>] [<e097f09a>] [<e09814da>]
[<e098177d>]
  [<e0981945>] [<c01055b8>]

Code: c7 80 a4 09 00 00 00 00 00 00 8d 4b 5c ff 43 5c 0f 8e 06 04 

which is
>>EIP; e2302d96 <[usbserial]usb_serial_disconnect+66/1f0>   <=====

>>ebx; dd66d01c <_end+1d325db0/205f3df4>
>>ecx; ddf30000 <_end+1dbe8d94/205f3df4>
>>edi; dd66d000 <_end+1d325d94/205f3df4>
>>esp; defdbf4c <_end+1ec94ce0/205f3df4>

Trace; e2303fa0 <[usbserial]usb_serial_driver+0/3c>
Trace; e2303fc0 <[usbserial]usb_serial_driver+20/3c>
Trace; e097f09a <[usbcore]usb_disconnect+8a/130>
Trace; e09814da <[usbcore]usb_hub_port_connect_change+4a/210>
Trace; e098177d <[usbcore]usb_hub_events+dd/270>
Trace; e0981945 <[usbcore]usb_hub_thread+35/b0>
Trace; c01055b8 <arch_kernel_thread+28/40>

Code;  e2302d96 <[usbserial]usb_serial_disconnect+66/1f0>
00000000 <_EIP>:
Code;  e2302d96 <[usbserial]usb_serial_disconnect+66/1f0>   <=====
   0:   c7 80 a4 09 00 00 00      movl   $0x0,0x9a4(%eax)   <=====
Code;  e2302d9d <[usbserial]usb_serial_disconnect+6d/1f0>
   7:   00 00 00 
Code;  e2302da0 <[usbserial]usb_serial_disconnect+70/1f0>
   a:   8d 4b 5c                  lea    0x5c(%ebx),%ecx
Code;  e2302da3 <[usbserial]usb_serial_disconnect+73/1f0>
   d:   ff 43 5c                  incl   0x5c(%ebx)
Code;  e2302da6 <[usbserial]usb_serial_disconnect+76/1f0>
  10:   0f 8e 06 04 00 00         jle    41c <_EIP+0x41c>

This is very strange since the visor has always worked perfectly, but
recently stopped, I've tried messing with older versions of pilot-link,
switching usb slots...  I have a feeling it's probably heat related.

I don't know where the Oops is coming from, before it would just hang,
now it oopses.
CC: replies :)
-- 
Justin A <justin@bouncybouncy.net>

Re: usb serial/visor oops in 2.4.22-pre2

[Greg KH]

On Tue, Jul 01, 2003 at 08:33:40PM -0400, Justin A wrote:
> basically, pilot-xfer and all don't work anymore, dmesg reports:

But 2.4.21 works just fine for you?

What USB host driver are you using?

thanks,

greg k-h

Re: usb serial/visor oops in 2.4.22-pre2

[Justin A]

On Tue, 2003-07-01 at 23:32, Greg KH wrote:
> Any reason for not sending this to linux-kernel too?

hmmm i think you mean the linux-usb list too?

> 
> On Tue, Jul 01, 2003 at 09:03:10PM -0400, Justin A wrote:
> > On Tue, 2003-07-01 at 20:57, Greg KH wrote:
> > > On Tue, Jul 01, 2003 at 08:33:40PM -0400, Justin A wrote:
> > > > basically, pilot-xfer and all don't work anymore, dmesg reports:
> > > 
> > > But 2.4.21 works just fine for you?
> > dunno, I went from 2.4.21-rc6 to 2.2.22-pre2
> > it sorta worked once or twice in 2.2.21-rc6, but then stopped working.
> 
> With the same oops?
yeah, that or it would just timeout but no oops.
The thing about the oops is that it happens when it times out after not
establishing a connecting, it's not that I hit the sync button and it
immediately oopses.

> > I had been out of batteries for a while, so I can't remember the last
> > kernel it worked perfectly with, but I think it was 2.4.21-pre5-ac3
> > 
> > If you want I'll test a specific config.
> > > What USB host driver are you using?
> > uhci
> 
> Can you try usb-uhci and see if it works better?
I am using usb-uhci.. forgot there was a driver named just 'uhci' :)
> thanks,
> 
> greg k-h

I will reboot tomorrow when I get a chance, and verify that it still
works on my old kernels and if it does try with 2.4.21.

I wouldn't be surprised if the heat killed it or something.

-- 
Justin A <justin@bouncybouncy.net>

Re: usb serial/visor oops in 2.4.22-pre2

[Greg KH]

On Wed, Jul 02, 2003 at 12:17:30AM -0400, Justin A wrote:
> On Tue, 2003-07-01 at 23:32, Greg KH wrote:
> > Any reason for not sending this to linux-kernel too?
> 
> hmmm i think you mean the linux-usb list too?

No, you sent to me privately, and not CC lkml.

> > On Tue, Jul 01, 2003 at 09:03:10PM -0400, Justin A wrote:
> > > On Tue, 2003-07-01 at 20:57, Greg KH wrote:
> > > > On Tue, Jul 01, 2003 at 08:33:40PM -0400, Justin A wrote:
> > > > > basically, pilot-xfer and all don't work anymore, dmesg reports:
> > > > 
> > > > But 2.4.21 works just fine for you?
> > > dunno, I went from 2.4.21-rc6 to 2.2.22-pre2
> > > it sorta worked once or twice in 2.2.21-rc6, but then stopped working.
> > 
> > With the same oops?
> yeah, that or it would just timeout but no oops.
> The thing about the oops is that it happens when it times out after not
> establishing a connecting, it's not that I hit the sync button and it
> immediately oopses.

It oopses when the close() happens, right?  That's a common bug right
now in 2.4 :(

It's fixed in 2.5, if you want to try that.

> > > I had been out of batteries for a while, so I can't remember the last
> > > kernel it worked perfectly with, but I think it was 2.4.21-pre5-ac3
> > > 
> > > If you want I'll test a specific config.
> > > > What USB host driver are you using?
> > > uhci
> > 
> > Can you try usb-uhci and see if it works better?
> I am using usb-uhci.. forgot there was a driver named just 'uhci' :)

Then try uhci, people have found that this sometimes works.

thanks,

greg k-h