Re: run_timer_softirq gpf. [smc]

[Ursula Braun]

> 
> From:   Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> To:     Dave Jones <davej@xxxxxxxxxxxxxxxxx>
> Cc:     Linux Kernel <linux-kernel@xxxxxxxxxxxxxxx>, Steven Rostedt 
> <rostedt@xxxxxxxxxxx>, Ursula Braun <ubraun@xxxxxxxxxxxxxxxxxx>, 
> netdev@xxxxxxxxxxxxxxx
> Date:   21.03.2017 22:46
> Subject:        Re: run_timer_softirq gpf. [smc]
> Sent by:        netdev-owner@xxxxxxxxxxxxxxx
> 
> 
> 
> On Tue, 21 Mar 2017, Dave Jones wrote:
>> On Tue, Mar 21, 2017 at 08:25:39PM +0100, Thomas Gleixner wrote:
>>
>>  > > I just hit this while fuzzing..
>>  > > 
>>  > > general protection fault: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
>>  > > CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.11.0-rc2-think+ #1 
>>  > > task: ffff88017f0ed440 task.stack: ffffc90000094000
>>  > > RIP: 0010:run_timer_softirq+0x15f/0x700
>>  > > RSP: 0018:ffff880507c03ec8 EFLAGS: 00010086
>>  > > RAX: dead000000000200 RBX: ffff880507dd0d00 RCX: 0000000000000002
>>  > > RDX: ffff880507c03ed0 RSI: 00000000ffffffff RDI: ffffffff8204b3a0
>>  > > RBP: ffff880507c03f48 R08: ffff880507dd12d0 R09: ffff880507c03ed8
>>  > > R10: ffff880507dd0db0 R11: 0000000000000000 R12: ffffffff8215cc38
>>  > > R13: ffff880507c03ed0 R14: ffffffff82005188 R15: ffff8804b55491a8
>>  > > FS:  0000000000000000(0000) GS:ffff880507c00000(0000) 
> knlGS:0000000000000000
>>  > > CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>  > > CR2: 0000000000000004 CR3: 0000000005011000 CR4: 00000000001406e0
>>  > > Call Trace:
>>  > >  <IRQ>
>>  > >  ? clockevents_program_event+0x47/0x120
>>  > >  __do_softirq+0xbf/0x5b1
>>  > >  irq_exit+0xb5/0xc0
>>  > >  smp_apic_timer_interrupt+0x3d/0x50
>>  > >  apic_timer_interrupt+0x97/0xa0
>>  > > RIP: 0010:cpuidle_enter_state+0x12e/0x400
>>  > > RSP: 0018:ffffc90000097e40 EFLAGS: 00000202
>>  > > [CONT START]  ORIG_RAX: ffffffffffffff10
>>  > > RAX: ffff88017f0ed440 RBX: ffffe8ffffa03cc8 RCX: 0000000000000001
>>  > > RDX: 20c49ba5e353f7cf RSI: 0000000000000001 RDI: ffff88017f0ed440
>>  > > RBP: ffffc90000097e80 R08: 00000000ffffffff R09: 0000000000000008
>>  > > R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000005
>>  > > R13: ffffffff820b9338 R14: 0000000000000005 R15: ffffffff820b9320
>>  > >  </IRQ>
>>  > >  cpuidle_enter+0x17/0x20
>>  > >  call_cpuidle+0x23/0x40
>>  > >  do_idle+0xfb/0x200
>>  > >  cpu_startup_entry+0x71/0x80
>>  > >  start_secondary+0x16a/0x210
>>  > >  start_cpu+0x14/0x14
>>  > > Code: 8b 05 ce 1b ef 7e 83 f8 03 0f 87 4e 01 00 00 89 c0 49 0f a3 
> 04 24 0f 82 0a 01 00 00 49 8b 07 49 8b 57 08 48 85 c0 48 89 02 74 04 <48> 
> 89 50 08 41 f6 47 2a 20 49 c7 47 08 00 00 00 00 48 89 df 48 
>>  > 
>>  > The timer which expires has timer->entry.next == POISON2 !
>>  > 
>>  > it's a classic list corruption.  The
>>  > bad news is that there is no trace of the culprit because that 
> happens when
>>  > some other timer expires after some random amount of time.
>>  > 
>>  > If that is reproducible, then please enable debugobjects. That should
>>  > pinpoint the culprit.
>>
>> It's net/smc.  This recently had a similar bug with workqueues. 
>> (https://marc.info/?l=linux-kernel&m=148821582909541) fixed by
>> 637fdbae60d6cb9f6e963c1079d7e0445c86ff7d
> 
> Fixed? It's not fixed by that commit. The workqueue code merily got a new
> WARN_ON_ONCE(). But the underlying problem is still unfixed in net/smc
> 
>> so it's probably unsurprising that there are similar issues.
> 
> That one is related to workqueues:
> 
>> WARNING: CPU: 0 PID: 2430 at lib/debugobjects.c:289 
> debug_print_object+0x87/0xb0
>> ODEBUG: free active (active state 0) object type: timer_list hint: 
> delayed_work_timer_fn+0x0/0x20
> 
> delayed_work_timer_fn() is what queues the work once the timer expires.
> 
>> CPU: 0 PID: 2430 Comm: trinity-c4 Not tainted 4.11.0-rc3-think+ #3 
>> Call Trace:
>>  dump_stack+0x68/0x93
>>  __warn+0xcb/0xf0
>>  warn_slowpath_fmt+0x5f/0x80
>>  ? debug_check_no_obj_freed+0xd9/0x260
>>  debug_print_object+0x87/0xb0
>>  ? work_on_cpu+0xd0/0xd0
>>  debug_check_no_obj_freed+0x219/0x260
>>  ? __sk_destruct+0x10d/0x1c0
>>  kmem_cache_free+0x9f/0x370
>>  __sk_destruct+0x10d/0x1c0
>>  sk_destruct+0x20/0x30
>>  __sk_free+0x43/0xa0
>>  sk_free+0x18/0x20
> 
> smc_release does at the end of the function:
> 
>         if (smc->use_fallback) {
>                 schedule_delayed_work(&smc->sock_put_work, 
> TCP_TIMEWAIT_LEN);
>         } else if (sk->sk_state == SMC_CLOSED) {
>                 smc_conn_free(&smc->conn);
>                 schedule_delayed_work(&smc->sock_put_work,
>                                       SMC_CLOSE_SOCK_PUT_DELAY);
>         }
>         sk->sk_prot->unhash(sk);
>         release_sock(sk);
> 
>         sock_put(sk);
> 
> sock_put(sk)
> {
>         if (atomic_dec_and_test(&sk->sk_refcnt))
>                 sk_free(sk);
> }
> 
> That means either smc_release() queued delayed work or it was already
> queued.
> 
> But in neither case it holds an extra refcount on sk. Otherwise sock_put()
> would not end up in sk_free().
We are working on a fix for the smc socket closing code right now.
> 
> Thanks,
> 
>                  tglx