On 8/17/2021 3:32 AM, David Chen wrote:
-----Original Message-----
From: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Sent: Monday, August 16, 2021 12:31 PM
To: David Chen <david.chen@xxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx; Paul E. McKenney
<paulmck@xxxxxxxxxxxxxxxxxx>; neeraju@xxxxxxxxxxxxxx
Subject: Re: Request for backport fd6bc19d7676 to 4.14 and 4.19 branch
On Mon, Aug 16, 2021 at 07:19:34PM +0000, David Chen wrote:
Hi Greg,
We recently hit a hung task timeout issue in synchronize_rcu_expedited on
4.14 branch.
The issue seems to be identical to the one described in `fd6bc19d7676
rcu: Fix missed wakeup of exp_wq waiters` Can we backport it to 4.14 and
4.19 branch?
The patch doesn't apply cleanly, but it should be trivial to resolve,
just do this
- wake_up_all(&rnp->exp_wq[rcu_seq_ctr(rsp-
expedited_sequence) & 0x3]);
+ wake_up_all(&rnp->exp_wq[rcu_seq_ctr(s) & 0x3]);
I don't know if we should do it for 4.9, because the handling of sequence
number is a bit different.
Please provide a working backport, me hand-editing patches does not scale,
and this way you get the proper credit for backporting it (after testing it).
Sure, appended at the end.
You have tested, this, right?
I don't have a good repro for the original issue, so I only ran rcutorture and
some basic work load test to see if anything obvious went wrong.
thanks,
greg k-h
--------
From 307a212335fe143027e3a9f7a9d548beead7ba33 Mon Sep 17 00:00:00 2001
From: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx>
Date: Tue, 19 Nov 2019 03:17:07 +0000
Subject: [PATCH] rcu: Fix missed wakeup of exp_wq waiters
[ Upstream commit fd6bc19d7676a060a171d1cf3dcbf6fd797eb05f ]
Tasks waiting within exp_funnel_lock() for an expedited grace period to
elapse can be starved due to the following sequence of events:
1. Tasks A and B both attempt to start an expedited grace
period at about the same time. This grace period will have
completed when the lower four bits of the rcu_state structure's
->expedited_sequence field are 0b'0100', for example, when the
initial value of this counter is zero. Task A wins, and thus
does the actual work of starting the grace period, including
acquiring the rcu_state structure's .exp_mutex and sets the
counter to 0b'0001'.
2. Because task B lost the race to start the grace period, it
waits on ->expedited_sequence to reach 0b'0100' inside of
exp_funnel_lock(). This task therefore blocks on the rcu_node
structure's ->exp_wq[1] field, keeping in mind that the
end-of-grace-period value of ->expedited_sequence (0b'0100')
is shifted down two bits before indexing the ->exp_wq[] field.
3. Task C attempts to start another expedited grace period,
but blocks on ->exp_mutex, which is still held by Task A.
4. The aforementioned expedited grace period completes, so that
->expedited_sequence now has the value 0b'0100'. A kworker task
therefore acquires the rcu_state structure's ->exp_wake_mutex
and starts awakening any tasks waiting for this grace period.
5. One of the first tasks awakened happens to be Task A. Task A
therefore releases the rcu_state structure's ->exp_mutex,
which allows Task C to start the next expedited grace period,
which causes the lower four bits of the rcu_state structure's
->expedited_sequence field to become 0b'0101'.
6. Task C's expedited grace period completes, so that the lower four
bits of the rcu_state structure's ->expedited_sequence field now
become 0b'1000'.
7. The kworker task from step 4 above continues its wakeups.
Unfortunately, the wake_up_all() refetches the rcu_state
structure's .expedited_sequence field:
wake_up_all(&rnp->exp_wq[rcu_seq_ctr(rcu_state.expedited_sequence) & 0x3]);
Minor: On these kernel versions, we had rsp pointer, per RCU flavor,
whereas post 4.20 kernel versions, we have a single rcu_state. So, the
commit message can be corrected here. The functionality is mostly
unchanged and same fix is applicable.
This results in the wakeup being applied to the rcu_node
structure's ->exp_wq[2] field, which is unfortunate given that
Task B is instead waiting on ->exp_wq[1].
On a busy system, no harm is done (or at least no permanent harm is done).
Some later expedited grace period will redo the wakeup. But on a quiet
system, such as many embedded systems, it might be a good long time before
there was another expedited grace period. On such embedded systems,
this situation could therefore result in a system hang.
This issue manifested as DPM device timeout during suspend (which
usually qualifies as a quiet time) due to a SCSI device being stuck in
_synchronize_rcu_expedited(), with the following stack trace:
schedule()
synchronize_rcu_expedited()
synchronize_rcu()
scsi_device_quiesce()
scsi_bus_suspend()
dpm_run_callback()
__device_suspend()
This commit therefore prevents such delays, timeouts, and hangs by
making rcu_exp_wait_wake() use its "s" argument consistently instead of
refetching from rcu_state.expedited_sequence.
Fixes: 3b5f668e715b ("rcu: Overlap wakeups with next expedited grace period")
Signed-off-by: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx>
Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
Signed-off-by: David Chen <david.chen@xxxxxxxxxxx>
---
kernel/rcu/tree_exp.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index 46d61b597731..f90d10c1c3c8 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -534,7 +534,7 @@ static void rcu_exp_wait_wake(struct rcu_state *rsp, unsigned long s)
spin_unlock(&rnp->exp_lock);
}
smp_mb(); /* All above changes before wakeup. */
- wake_up_all(&rnp->exp_wq[rcu_seq_ctr(rsp->expedited_sequence) & 0x3]);
+ wake_up_all(&rnp->exp_wq[rcu_seq_ctr(s) & 0x3]);
}
trace_rcu_exp_grace_period(rsp->name, s, TPS("endwake"));
mutex_unlock(&rsp->exp_wake_mutex);
Acked-by: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx>
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of the Code Aurora Forum, hosted by The Linux Foundation
