lkml.org 
[lkml]   [2012]   [Feb]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectTTY: NULL dereference when closing a pty file
From
Date
Hi all,

I got the following BUG() when running trinity on the KVM tool:

[ 665.738774] BUG: unable to handle kernel NULL pointer dereference at 0000000000000028
[ 665.739651] IP: [<ffffffff81257e2b>] devpts_pty_kill+0x1b/0xa0
[ 665.739651] PGD 22eba067 PUD 22eaf067 PMD 0
[ 665.739651] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
[ 665.739651] CPU 5
[ 665.739651] Pid: 3061, comm: trinity Not tainted 3.3.0-rc3-next-20120217-sasha-00001-gfa56acb #18
[ 665.739651] RIP: 0010:[<ffffffff81257e2b>] [<ffffffff81257e2b>] devpts_pty_kill+0x1b/0xa0
[ 665.739651] RSP: 0018:ffff880022ecfd88 EFLAGS: 00010286
[ 665.739651] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[ 665.739651] RDX: 0000000000000001 RSI: 0000000000000001 RDI: ffff880022921148
[ 665.739651] RBP: ffff880022ecfda8 R08: 0000000000000000 R09: 0000000000000001
[ 665.739651] R10: 0000000000000001 R11: 0000000000000001 R12: ffff880022921148
[ 665.739651] R13: ffff880022934840 R14: ffff880026831970 R15: ffff880026831970
[ 665.739651] FS: 00007f29c344e700(0000) GS:ffff88002a400000(0000) knlGS:0000000000000000
[ 665.739651] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 665.739651] CR2: 0000000000000028 CR3: 0000000022e1c000 CR4: 00000000000406e0
[ 665.739651] DR0: ffffffff810adc50 DR1: 0000000000000000 DR2: 0000000000000000
[ 665.739651] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000600
[ 665.739651] Process trinity (pid: 3061, threadinfo ffff880022ece000, task ffff880022d88000)
[ 665.739651] Stack:
[ 665.739651] ffff880022ecfd98 ffff8800229267b0 ffff880022921148 ffff880022934840
[ 665.739651] ffff880022ecfdc8 ffffffff819407f1 ffff880022934840 ffff8800229267b0
[ 665.739651] ffff880022ecfeb8 ffffffff819374a6 ffff880022ecfe18 ffffffff81078e76
[ 665.739651] Call Trace:
[ 665.739651] [<ffffffff819407f1>] pty_close+0x121/0x140
[ 665.739651] [<ffffffff819374a6>] tty_release+0x186/0x610
[ 665.739651] [<ffffffff81078e76>] ? kvm_clock_read+0x46/0x80
[ 665.739651] [<ffffffff81054973>] ? sched_clock+0x13/0x20
[ 665.739651] [<ffffffff811d9b16>] fput+0xf6/0x330
[ 665.739651] [<ffffffff811d68b4>] filp_close+0x64/0x90
[ 665.739651] [<ffffffff811d699b>] sys_close+0xbb/0x1b0
[ 665.739651] [<ffffffff8267c079>] system_call_fastpath+0x16/0x1b
[ 665.739651] Code: 4c 8b a3 08 04 00 00 eb df 0f 0b 0f 1f 44 00 00 55 48 89 e5 48 83 ec 20 48 89 5d e8 4c 89 65 f0 4c 89 6d f8 48 8b 9f 28 04 00 00 <48> 8b 43 28 48 81 78 58 d1 1c 00 00 74 0b 48 8b 05 f0 17 26 03
[ 665.739651] RIP [<ffffffff81257e2b>] devpts_pty_kill+0x1b/0xa0
[ 665.739651] RSP <ffff880022ecfd88>
[ 665.739651] CR2: 0000000000000028
[ 665.782570] ---[ end trace 5b128b9a8217de35 ]---

Looking further, it looks like devpts_pty_kill was called with tty->link==NULL.

I've bisected it down to d3bda529 ("TTY: get rid of BTM around devpts_*"), which has moved devpts_pty_kill() out of the tty_lock() protection.



\
 
 \ /
  Last update: 2012-02-19 21:39    [W:0.082 / U:0.908 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site