Messages in this thread | | | Date | Wed, 31 Jul 2013 16:54:03 +0530 | Subject | ipv4: crash at leaf_walk_rcu | From | vinayak menon <> |
| |
Hi,
A crash was seen on 3.4.5 kernel during some random wlan operations.
CPU: Single core ARM Cortex A9.
fib_route_seq_next was called with second argument (void *v) as 0xd6e3e360 which is a "freed" object of the "ip_fib_trie" cache. I confirmed that the object was freed with crash utility.
Sequence: fib_route_seq_next->trie_nextleaf->leaf_walk_rcu
As "v" was a freed object, inside trie_nextleaf(), node_parent_rcu() returned an invalid tnode. But as I had enabled slab poisoning and the object was already freed, the tnode was 0x6b6b6b6b. And this was passed to leaf_walk_rcu and resulted in the crash.
fib_route_seq_start, takes rcu_read_lock(), but free_leaf calls call_rcu_bh. Can this be the problem ? Should rcu_read_lock() in fib_route_seq_start be changed to rcu_read_lock_bh() ?
---------------------------------------------------------------------------- PC is at leaf_walk_rcu+0x10/0xa0 LR is at fib_route_seq_next+0x58/0x74 pc : [<c0500e5c>] lr : [<c050108c>] psr: a0000013 sp : c150bee0 ip : 00000000 fp : 00000000 r10: 00000400 r9 : 53701020 r8 : c32345c0 r7 : 00000000 r6 : 00000001 r5 : 00000000 r4 : 00000002 r3 : 6b6b6b6b r2 : 00000001 r1 : d6e3e360 r0 : 6b6b6b6a Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 10c53c7d Table: 835dc059 DAC: 00000015
Backtrace: [<c0500e5c>] (leaf_walk_rcu+0x10/0xa0) from [<c050108c>] (fib_route_seq_next+0x58/0x74) [<c050108c>] (fib_route_seq_next+0x58/0x74) from [<c011c06c>] (seq_read+0x2cc/0x438) [<c011c06c>] (seq_read+0x2cc/0x438) from [<c0145734>] (proc_reg_read+0xb0/0xcc) [<c0145734>] (proc_reg_read+0xb0/0xcc) from [<c0100798>] (vfs_read+0xac/0x124) [<c0100798>] (vfs_read+0xac/0x124) from [<c0100848>] (sys_read+0x38/0x64) [<c0100848>] (sys_read+0x38/0x64) from [<c000e100>] (ret_fast_syscall+0x0/0x48)
Thanks, Vinayak
| |