Re: [Bluez-users] krfcommd or kernel bug?

[Marcel Holtmann <marcel@holtmann.org>]

Hi Aquamala,

> > > $uname -a
> > > Linux localhost.localdomain 2.6.2 #16 Tue May 18 02:08:00 CEST 2004
> > > i586 i586 i386 GNU/Linux
> > 
> > and here is the problem. Your kernel is buggy and you should better use
> > the latest 2.6 kernel with our Bluetooth patch. For exmaple 2.6.6-mh1.
> 
> Ok, I've just upgreaded to 2.6.6-mh1.

and I found that the 2.6.6 has still a module reference count bug in the
RFCOMM layer. To prevent this bug you must start at least one service
that uses RFCOMM.

> Now, I can't even hciattach my device :/
> 
> #modprobe serport
> #modprobe hci_uart
> #hciattach ttyS0 ericsson 57600 flow
> 
> ...and then:
> 
> May 22 19:42:40 localhost kernel: Unable to handle kernel NULL
> pointer dereference at virtual address 00000014
> May 22 19:42:40 localhost kernel:  printing eip:
> May 22 19:42:40 localhost kernel: c6817289
> May 22 19:42:40 localhost kernel: *pde = 00000000
> May 22 19:42:40 localhost kernel: Oops: 0002 [#1]
> May 22 19:42:40 localhost kernel: PREEMPT
> May 22 19:42:40 localhost kernel: CPU:    0
> May 22 19:42:40 localhost kernel: EIP:    0060:[<c6817289>]    Not
> tainted
> May 22 19:42:40 localhost kernel: EFLAGS: 00010286   (2.6.6-mh1)
> May 22 19:42:40 localhost kernel: EIP is at hci_uart_close+0x9/0x40
> [hci_uart]
> May 22 19:42:40 localhost kernel: eax: 00000002   ebx: c1320d3c
> ecx: c0274b7c   edx: 00000000
> May 22 19:42:40 localhost kernel: esi: 00000000   edi: c5dbfa00
> ebp: 00000000   esp: c5315ebc
> May 22 19:42:40 localhost kernel: ds: 007b   es: 007b   ss: 0068
> May 22 19:42:40 localhost kernel: Process hciattach (pid: 1238,
> threadinfo=c5314000 task=c5f1a5f0)
> May 22 19:42:40 localhost kernel: Stack: c6817403 00000000 c52de000
> 00000000 c01c1a00 c52de000 00000000 00000000
> May 22 19:42:40 localhost kernel:        00000001 00000000 c02fd108
> 0000000a c02d6a00 c01151fc c02fd108 00000046
> May 22 19:42:40 localhost kernel:        c5314000 00000000 00000046
> c5314000 c0105a59 c0276424 c523d768 00000126
> May 22 19:42:40 localhost kernel: Call Trace:
> May 22 19:42:40 localhost kernel:  [<c6817403>]
> hci_uart_tty_close+0x23/0x60 [hci_uart]
> May 22 19:42:40 localhost kernel:  [<c01c1a00>]
> release_dev+0x4e0/0x5e0
> May 22 19:42:40 localhost kernel:  [<c01151fc>]
> __do_softirq+0x7c/0x80
> May 22 19:42:40 localhost kernel:  [<c0105a59>] do_IRQ+0xf9/0x140
> May 22 19:42:40 localhost kernel:  [<c01c1ea1>]
> tty_release+0x21/0x60
> May 22 19:42:41 localhost kernel:  [<c014313b>] __fput+0xfb/0x120
> May 22 19:42:41 localhost kernel:  [<c0141a43>] filp_close+0x43/0x80
> May 22 19:42:41 localhost kernel:  [<c0112cd5>]
> put_files_struct+0x55/0xc0
> May 22 19:42:41 localhost kernel:  [<c0113910>] do_exit+0x1b0/0x3e0
> May 22 19:42:41 localhost kernel:  [<c0113c30>]
> do_group_exit+0x70/0xc0
> May 22 19:42:41 localhost kernel:  [<c0103ed7>] syscall_call+0x7/0xb
> May 22 19:42:41 localhost kernel:
> May 22 19:42:41 localhost kernel: Code: 0f b3 42 14 19 c0 85 c0 75
> 0d b8 00 00 00 00 c3 8d b4 26 00
> 
> I have no idea what to do... 

I've never seen this before and actually I don't know why get a NULL
pointer dereference here.

Regards

Marcel




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
_______________________________________________
Bluez-users mailing list
Bluez-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-users