Re: [Bugme-new] [Bug 8054] New: tipc_ref_discard tipc_deleteport locking dependency
From: Andrew Morton <akpm@linux-foundation.org>
Date: 2007-02-22 21:16:50
On Thu, 22 Feb 2007 04:20:10 -0800 bugme-daemon@bugzilla.kernel.org wrote: http://bugzilla.kernel.org/show_bug.cgi?id=8054 Summary: tipc_ref_discard tipc_deleteport locking dependency Kernel Version: 2.6.21-rc1 Status: NEW Severity: normal Owner: acme@conectiva.com.br Submitter: snakebyte@gmx.de Most recent kernel where this bug did *NOT* occur: - also seen this on 2.6.20 Distribution: Gentoo Hardware Environment: AMD-K6 Software Environment: Problem Description: [ 273.837978] ======================================================= [ 273.838209] [ INFO: possible circular locking dependency detected ] [ 273.838324] 2.6.21-rc1 #5 [ 273.838427] ------------------------------------------------------- [ 273.838535] sfuzz/3950 is trying to acquire lock: [ 273.838642] (ref_table_lock){-+..}, at: [<c0531389>] tipc_ref_discard+0x29/0xe0 [ 273.839116] [ 273.839121] but task is already holding lock: [ 273.839297] (&table[i].lock){-+..}, at: [<c052fae0>] tipc_deleteport+0x40/0x180 [ 273.839728] [ 273.839733] which lock already depends on the new lock. [ 273.839741] [ 273.839998] [ 273.840003] the existing dependency chain (in reverse order) is: [ 273.840182] [ 273.840187] -> #1 (&table[i].lock){-+..}: [ 273.840581] [<c0138994>] __lock_acquire+0xeb4/0x1020 [ 273.841352] [<c0138b69>] lock_acquire+0x69/0xa0 [ 273.842041] [<c0537f80>] _spin_lock_bh+0x40/0x60 [ 273.842801] [<c05314ab>] tipc_ref_acquire+0x6b/0xe0 [ 273.844131] [<c052ef93>] tipc_createport_raw+0x33/0x260 [ 273.844823] [<c0530121>] tipc_createport+0x41/0x120 [ 273.845528] [<c0529b4c>] tipc_subscr_start+0xcc/0x120 [ 273.846217] [<c0521d04>] process_signal_queue+0x44/0x80 [ 273.846926] [<c011df38>] tasklet_action+0x38/0x80 [ 273.847626] [<c011e1db>] __do_softirq+0x5b/0xc0 [ 273.848374] [<c0105e48>] do_softirq+0x88/0xe0 [ 273.849630] [<ffffffff>] 0xffffffff [ 273.850315] [ 273.850319] -> #0 (ref_table_lock){-+..}: [ 273.850713] [<c013878d>] __lock_acquire+0xcad/0x1020 [ 273.851416] [<c0138b69>] lock_acquire+0x69/0xa0 [ 273.852127] [<c0538040>] _write_lock_bh+0x40/0x60 [ 273.852835] [<c0531389>] tipc_ref_discard+0x29/0xe0 [ 273.853552] [<c052fafa>] tipc_deleteport+0x5a/0x180 [ 273.854832] [<c05317d8>] tipc_create+0x58/0x160 [ 273.855530] [<c04603d2>] __sock_create+0x112/0x280 [ 273.856237] [<c046057a>] sock_create+0x1a/0x20 [ 273.856931] [<c04605a3>] sys_socketpair+0x23/0x1a0 [ 273.857642] [<c0461257>] sys_socketcall+0x137/0x260 [ 273.858343] [<c0102cb0>] syscall_call+0x7/0xb [ 273.859052] [<ffffffff>] 0xffffffff [ 273.860333] [ 273.860338] other info that might help us debug this: [ 273.860345] [ 273.860596] 1 lock held by sfuzz/3950: [ 273.860697] #0: (&table[i].lock){-+..}, at: [<c052fae0>] tipc_deleteport+0x40/0x180 [ 273.861208] [ 273.861213] stack backtrace: [ 273.861391] [<c01042ba>] show_trace_log_lvl+0x1a/0x40 [ 273.861591] [<c0104a92>] show_trace+0x12/0x20 [ 273.861779] [<c0104b99>] dump_stack+0x19/0x20 [ 273.861968] [<c013686e>] print_circular_bug_tail+0x6e/0x80 [ 273.862162] [<c013878d>] __lock_acquire+0xcad/0x1020 [ 273.862354] [<c0138b69>] lock_acquire+0x69/0xa0 [ 273.862544] [<c0538040>] _write_lock_bh+0x40/0x60 [ 273.862740] [<c0531389>] tipc_ref_discard+0x29/0xe0 [ 273.862958] [<c052fafa>] tipc_deleteport+0x5a/0x180 [ 273.863153] [<c05317d8>] tipc_create+0x58/0x160 [ 273.863347] [<c04603d2>] __sock_create+0x112/0x280 [ 273.863541] [<c046057a>] sock_create+0x1a/0x20 [ 273.863731] [<c04605a3>] sys_socketpair+0x23/0x1a0 [ 273.863922] [<c0461257>] sys_socketcall+0x137/0x260 [ 273.864113] [<c0102cb0>] syscall_call+0x7/0xb [ 273.864299] ======================= Steps to reproduce: enable tipc, recompile, reboot, run sfuzz (http://www.digitaldwarf.be/products/sfuzz.c) for some time ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.