New: BUG at kernel/softirq.c:105

I've been trying to trace this callpath, and have hit a dead end in release_dev(). It seems to go weird after there, which makes it difficult to audit the code path.

Anyone got any ideas?

I'm not too bothered about this bug since I don't look after the ppp or tty (line discipline) code, so I'm also looking for someone else to assign this to.

----- Forwarded message from [email protected] -----

Date: Wed, 16 Apr 2003 03:20:37 -0700
From: [email protected]
To: [email protected]
Subject: [Bug 592] New: BUG at kernel/softirq.c:105

http://bugme.osdl.org/show_bug.cgi?id=592

Summary: BUG at kernel/softirq.c:105
Kernel Version: 2.5.67
Status: NEW
Severity: normal
Owner: [email protected]
Submitter: [email protected]

Problem Description:
At shutdown, when pppd gets killed this BUG shows up. Here is the syslog:

Apr 14 22:10:28 iris kernel: kernel BUG at kernel/softirq.c:105!
Apr 14 22:10:28 iris kernel: invalid operand: 0000 [#1]
Apr 14 22:10:28 iris kernel: CPU:    0
Apr 14 22:10:28 iris kernel: EIP:    0060:[<c011f8a3>]    Not tainted
Apr 14 22:10:28 iris kernel: EFLAGS: 00010002
Apr 14 22:10:28 iris kernel: EIP is at local_bh_enable+0x43/0x50
Apr 14 22:10:28 iris kernel: eax: 00000001   ebx: de840e00   ecx: 00000000  
edx: de840eb2
Apr 14 22:10:28 iris kernel: esi: 00000000   edi: 00000000   ebp: 00000000  
esp: de599e28
Apr 14 22:10:28 iris kernel: ds: 007b   es: 007b   ss: 0068
Apr 14 22:10:28 iris kernel: Process pptp (pid: 388, threadinfo=de598000
task=de90e740)
Apr 14 22:10:28 iris kernel: Stack: c0237910 dfddaf40 c01a95db dfddaf40 00000001
de8b8000 dede74c0 de8b8000 
Apr 14 22:10:28 iris kernel:        de840e00 de8b89e4 00000000 c0237231 de840e00
de8b8000 deca3000 c01fd15f 
Apr 14 22:10:28 iris kernel:        de8b8000 deca3000 de8b8000 c01f9db3 deca3000
c01f9e33 deca3000 deca3000 
Apr 14 22:10:28 iris kernel: Call Trace:
Apr 14 22:10:28 iris kernel:  [<c0237910>] ppp_async_push+0x90/0x170
Apr 14 22:10:28 iris kernel:  [<c01a95db>] _devfs_unregister+0x5b/0xa0
Apr 14 22:10:28 iris kernel:  [<c0237231>] ppp_asynctty_wakeup+0x31/0x70
Apr 14 22:10:28 iris kernel:  [<c01fd15f>] pty_unthrottle+0x5f/0x70
Apr 14 22:10:28 iris kernel:  [<c01f9db3>] check_unthrottle+0x33/0x40
Apr 14 22:10:28 iris kernel:  [<c01f9e33>] n_tty_flush_buffer+0x13/0x60
Apr 14 22:10:28 iris kernel:  [<c01fd53c>] pty_flush_buffer+0x6c/0x70
Apr 14 22:10:28 iris kernel:  [<c01f70fe>] do_tty_hangup+0x2fe/0x340
Apr 14 22:10:28 iris kernel:  [<c01f83b7>] release_dev+0x637/0x680
Apr 14 22:10:28 iris kernel:  [<c0138239>] slab_destroy+0xa9/0xd0
Apr 14 22:10:28 iris kernel:  [<c013cf63>] unmap_page_range+0x43/0x70
Apr 14 22:10:28 iris kernel:  [<c0138e73>] cache_flusharray+0xe3/0x100
Apr 14 22:10:28 iris kernel:  [<c01f87df>] tty_release+0xf/0x20
Apr 14 22:10:28 iris kernel:  [<c014bad1>] __fput+0xc1/0xd0
Apr 14 22:10:28 iris kernel:  [<c014a31d>] filp_close+0x4d/0x80
Apr 14 22:10:28 iris kernel:  [<c011d577>] put_files_struct+0x57/0xc0
Apr 14 22:10:28 iris kernel:  [<c011e012>] do_exit+0x102/0x300
Apr 14 22:10:28 iris kernel:  [<c014a31d>] filp_close+0x4d/0x80
Apr 14 22:10:28 iris kernel:  [<c011e2c2>] do_group_exit+0x52/0x80
Apr 14 22:10:28 iris kernel:  [<c010b14b>] syscall_call+0x7/0xb
Apr 14 22:10:28 iris kernel: 
Apr 14 22:10:28 iris kernel: Code: 0f 0b 69 00 e9 d7 32 c0 eb d3 8d 76 00 53 89

c1 9c 5b fa b8

Steps to reproduce:
Start pppd
Shutdown the system (I haven't tried to do a killall pppd)

------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.

----- End forwarded message -----

Russell King ([email protected])                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html

Andrew Morton wrote something about this BUG, his mail
is here: http://marc.theaimsgroup.com/?l=linux-kernel&m=104979788626334&w=2

Bye,
Patrick