rcu: Don't disable CPU hotplug during OOM notifiers
authorPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tue, 14 Jul 2015 23:24:14 +0000 (16:24 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Wed, 22 Jul 2015 22:27:43 +0000 (15:27 -0700)
commit9a54f98e341d09793247a6e598012edefb5ae7cb
tree4a30b36698274e22b17651a0cad70500af2cef21
parent3ad81779ad35cd2e2ab34ba3f47f6f485696a105
rcu: Don't disable CPU hotplug during OOM notifiers

RCU's rcu_oom_notify() disables CPU hotplug in order to stabilize the
list of online CPUs, which it traverses.  However, this is completely
pointless because smp_call_function_single() will quietly fail if invoked
on an offline CPU.  Because the count of requests is incremented in the
rcu_oom_notify_cpu() function that is remotely invoked, everything works
nicely even in the face of concurrent CPU-hotplug operations.

Furthermore, in recent kernels, invoking get_online_cpus() from an OOM
notifier can result in deadlock.  This commit therefore removes the
call to get_online_cpus() and put_online_cpus() from rcu_oom_notify().

Reported-by: Marcin Ślusarz <marcin.slusarz@gmail.com>
Reported-by: David Rientjes <rientjes@google.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: David Rientjes <rientjes@google.com>
Tested-by: Marcin Ślusarz <marcin.slusarz@gmail.com>
kernel/rcu/tree_plugin.h