On Thu, Aug 22, 2019 at 09:59:53AM -0400, Joel Fernandes wrote:
> On Wed, Aug 21, 2019 at 06:19:06PM -0500, Scott Wood wrote:
> > Besides restoring behavior that used to be default on RT, this avoids
> > a deadlock on scheduler locks:
> >
> > [ 136.894657] 039: ============================================
> > [ 136.900401] 039: WARNING: possible recursive locking detected
> > [ 136.906146] 039: 5.2.9-rt3.dbg+ #174 Tainted: G E
> > [ 136.912152] 039: --------------------------------------------
> > [ 136.917896] 039: rcu_torture_rea/13474 is trying to acquire lock:
> > [ 136.923990] 039: 000000005f25146d
> > [ 136.927310] 039: (
> > [ 136.929414] 039: &p->pi_lock
> > [ 136.932303] 039: ){-...}
> > [ 136.934840] 039: , at: try_to_wake_up+0x39/0x920
> > [ 136.939461] 039:
> > but task is already holding lock:
> > [ 136.944425] 039: 000000005f25146d
> > [ 136.947745] 039: (
> > [ 136.949852] 039: &p->pi_lock
> > [ 136.952738] 039: ){-...}
> > [ 136.955274] 039: , at: try_to_wake_up+0x39/0x920
> > [ 136.959895] 039:
> > other info that might help us debug this:
> > [ 136.965555] 039: Possible unsafe locking scenario:
> >
> > [ 136.970608] 039: CPU0
> > [ 136.973493] 039: ----
> > [ 136.976378] 039: lock(
> > [ 136.978918] 039: &p->pi_lock
> > [ 136.981806] 039: );
> > [ 136.983911] 039: lock(
> > [ 136.986451] 039: &p->pi_lock
> > [ 136.989336] 039: );
> > [ 136.991444] 039:
> > *** DEADLOCK ***
> >
> > [ 136.995194] 039: May be due to missing lock nesting notation
> >
> > [ 137.001115] 039: 3 locks held by rcu_torture_rea/13474:
> > [ 137.006341] 039: #0:
> > [ 137.008707] 039: 000000005f25146d
> > [ 137.012024] 039: (
> > [ 137.014131] 039: &p->pi_lock
> > [ 137.017015] 039: ){-...}
> > [ 137.019558] 039: , at: try_to_wake_up+0x39/0x920
> > [ 137.024175] 039: #1:
> > [ 137.026540] 039: 0000000011c8e51d
> > [ 137.029859] 039: (
> > [ 137.031966] 039: &rq->lock
> > [ 137.034679] 039: ){-...}
> > [ 137.037217] 039: , at: try_to_wake_up+0x241/0x920
> > [ 137.041924] 039: #2:
> > [ 137.044291] 039: 00000000098649b9
> > [ 137.047610] 039: (
> > [ 137.049714] 039: rcu_read_lock
> > [ 137.052774] 039: ){....}
> > [ 137.055314] 039: , at: cpuacct_charge+0x33/0x1e0
> > [ 137.059934] 039:
> > stack backtrace:
> > [ 137.063425] 039: CPU: 39 PID: 13474 Comm: rcu_torture_rea Kdump: loaded Tainted: G E 5.2.9-rt3.dbg+ #174
> > [ 137.074197] 039: Hardware name: Intel Corporation S2600BT/S2600BT, BIOS SE5C620.86B.01.00.0763.022420181017 02/24/2018
> > [ 137.084886] 039: Call Trace:
> > [ 137.087773] 039: <IRQ>
> > [ 137.090226] 039: dump_stack+0x5e/0x8b
> > [ 137.093997] 039: __lock_acquire+0x725/0x1100
> > [ 137.098358] 039: lock_acquire+0xc0/0x240
> > [ 137.102374] 039: ? try_to_wake_up+0x39/0x920
> > [ 137.106737] 039: _raw_spin_lock_irqsave+0x47/0x90
> > [ 137.111534] 039: ? try_to_wake_up+0x39/0x920
> > [ 137.115910] 039: try_to_wake_up+0x39/0x920
> > [ 137.120098] 039: rcu_read_unlock_special+0x65/0xb0
> > [ 137.124977] 039: __rcu_read_unlock+0x5d/0x70
> > [ 137.129337] 039: cpuacct_charge+0xd9/0x1e0
> > [ 137.133522] 039: ? cpuacct_charge+0x33/0x1e0
> > [ 137.137880] 039: update_curr+0x14b/0x420
> > [ 137.141894] 039: enqueue_entity+0x42/0x370
> > [ 137.146080] 039: enqueue_task_fair+0xa9/0x490
> > [ 137.150528] 039: activate_task+0x5a/0xf0
> > [ 137.154539] 039: ttwu_do_activate+0x4e/0x90
> > [ 137.158813] 039: try_to_wake_up+0x277/0x920
> > [ 137.163086] 039: irq_exit+0xb6/0xf0
> > [ 137.166661] 039: smp_apic_timer_interrupt+0xe3/0x3a0
> > [ 137.171714] 039: apic_timer_interrupt+0xf/0x20
> > [ 137.176249] 039: </IRQ>
> > [ 137.178785] 039: RIP: 0010:__schedule+0x0/0x8e0
> > [ 137.183319] 039: Code: 00 02 48 89 43 20 e8 0f 5a 00 00 48 8d 7b 28 e8 86 f2 fd ff 31 c0 5b 5d 41 5c c3 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 <55> 48 89 e5 41 57 41 56 49 c7 c6 c0 ca 1e 00 41 55 41 89 fd 41 54
> > [ 137.202498] 039: RSP: 0018:ffffc9005835fbc0 EFLAGS: 00000246
> > [ 137.208158] 039: ORIG_RAX: ffffffffffffff13
> > [ 137.212428] 039: RAX: 0000000000000000 RBX: ffff8897c3e1bb00 RCX: 0000000000000001
> > [ 137.219994] 039: RDX: 0000000080004008 RSI: 0000000000000006 RDI: 0000000000000001
> > [ 137.227560] 039: RBP: ffff8897c3e1bb00 R08: 0000000000000000 R09: 0000000000000000
> > [ 137.235126] 039: R10: 0000000000000001 R11: 0000000000000001 R12: ffffffff81001fd1
> > [ 137.242694] 039: R13: 0000000000000044 R14: 0000000000000000 R15: ffffc9005835fcac
> > [ 137.250259] 039: ? ___preempt_schedule+0x16/0x18
> > [ 137.254969] 039: preempt_schedule_common+0x32/0x80
> > [ 137.259846] 039: ___preempt_schedule+0x16/0x18
> > [ 137.264379] 039: rcutorture_one_extend+0x33a/0x510 [rcutorture]
> > [ 137.270397] 039: rcu_torture_one_read+0x18c/0x450 [rcutorture]
> > [ 137.276334] 039: rcu_torture_reader+0xac/0x1f0 [rcutorture]
> > [ 137.281998] 039: ? rcu_torture_reader+0x1f0/0x1f0 [rcutorture]
> > [ 137.287920] 039: kthread+0x106/0x140
> > [ 137.291591] 039: ? rcu_torture_one_read+0x450/0x450 [rcutorture]
> > [ 137.297681] 039: ? kthread_bind+0x10/0x10
> > [ 137.301783] 039: ret_from_fork+0x3a/0x50
> >
> > Signed-off-by: Scott Wood <swood@xxxxxxxxxx>
> > ---
> > I think the prohibition on use_softirq can be dropped once RT gets the
> > latest RCU code, but the question of what use_softirq should default
> > to on PREEMPT_RT remains.
>
> Independent of the question of what use_softirq should default to, could we
> test RT with latest RCU code now to check if the deadlock goes away? That
> way, maybe we can find any issues in current RCU that cause scheduler
> deadlocks in the situation you pointed. The reason I am asking is because
> recently additional commits [1] try to prevent deadlock and it'd be nice to
> ensure that other conditions are not lingering (I don't think they are but
> it'd be nice to be sure).
>
> I am happy to do such testing myself if you want, however what does it take
> to apply the RT patchset to the latest mainline? Is it an achievable feat?
I suggest just using the -rt git tree, which I believe lives here:
https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git
I would guess that branch linux-5.2.y-rt is the one you want, but I
would ask Scott instead of blindly trusting my guess. ;-)
Thanx, Paul
