Re: [PATCH rcu 3/3] rcu: Allow expedited RCU grace periods on incoming CPUs

From: Mukesh Ojha
Date: Tue Feb 15 2022 - 09:28:23 EST



On 2/14/2022 10:14 PM, Paul E. McKenney wrote:
On Thu, Feb 10, 2022 at 12:38:11AM +0100, Frederic Weisbecker wrote:
On Fri, Feb 04, 2022 at 02:55:07PM -0800, Paul E. McKenney wrote:
Although it is usually safe to invoke synchronize_rcu_expedited() from a
preemption-enabled CPU-hotplug notifier, if it is invoked from a notifier
between CPUHP_AP_RCUTREE_ONLINE and CPUHP_AP_ACTIVE, its attempts to
invoke a workqueue handler will hang due to RCU waiting on a CPU that
the scheduler is not paying attention to. This commit therefore expands
use of the existing workqueue-independent synchronize_rcu_expedited()
from early boot to also include CPUs that are being hotplugged.

Link: https://lore.kernel.org/lkml/7359f994-8aaf-3cea-f5cf-c0d3929689d6@xxxxxxxxxxx/
Reported-by: Mukesh Ojha <quic_mojha@xxxxxxxxxxx>
Cc: Tejun Heo <tj@xxxxxxxxxx>
Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
I'm surprised by this scheduler behaviour.

Since sched_cpu_activate() hasn't been called yet,
rq->balance_callback = balance_push_callback. As a result, balance_push() should
be called at the end of schedule() when the workqueue is picked as the next task.
Then eventually the workqueue should be immediately preempted by the stop task to
be migrated elsewhere.

So I must be missing something. For the fun, I booted the following and it
didn't produce any issue:

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 80faf2273ce9..b1e74a508881 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -4234,6 +4234,8 @@ int rcutree_online_cpu(unsigned int cpu)
// Stop-machine done, so allow nohz_full to disable tick.
tick_dep_clear(TICK_DEP_BIT_RCU);
+ if (cpu != 0)
+ synchronize_rcu_expedited();
return 0;
}
That does seem compelling. And others have argued that the workqueue
system's handling of offline CPUs should deal with this.

Mukesh, was this a theoretical bug, or did you actually make it happen?
If you made it happen, as seems to have been the case given your original
email [1], could you please post your reproducer?

Thanx, Paul


No, it was not theoretical one. We saw this issue only once in our testing and i don't think it is easy to reproduce otherwise
it would been fixed by now.

When one of thread calling synchronize_expedite_rcu with timer of 20s but it did not get the exp funnel
lock for 20s and there we crash it with panic() on timeout.

The other thread cpuhp which was having the lock got stuck at the point mentioned at the below link.

[1] https://lore.kernel.org/lkml/7359f994-8aaf-3cea-f5cf-c0d3929689d6@xxxxxxxxxxx/

Below sample test in combination of many other tests

:loop
adb shell "echo 0 > /sys/devices/system/cpu/cpu0/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu0/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu1/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu1/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu2/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu2/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu3/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu3/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu4/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu4/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu5/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu5/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu6/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu6/online"
adb shell "echo 0 > /sys/devices/system/cpu/cpu7/online"
adb shell "echo 1 > /sys/devices/system/cpu/cpu7/online"
goto loop

-Mukesh