sparc, sysfs: Fix CPU hotplug callback registration
authorSrivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Mon, 10 Mar 2014 20:36:09 +0000 (02:06 +0530)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Thu, 20 Mar 2014 12:43:41 +0000 (13:43 +0100)
commit1e0b0c4c6b9241b66c6032e14962a149f852f23a
tree853d28f3239cc68ac038924b6a3a8a835ac7c33f
parentf4edbcd5d1f350b90583c8f3e0e7112b28f65e62
sparc, sysfs: Fix CPU hotplug callback registration

Subsystems that want to register CPU hotplug callbacks, as well as perform
initialization for the CPUs that are already online, often do it as shown
below:

get_online_cpus();

for_each_online_cpu(cpu)
init_cpu(cpu);

register_cpu_notifier(&foobar_cpu_notifier);

put_online_cpus();

This is wrong, since it is prone to ABBA deadlocks involving the
cpu_add_remove_lock and the cpu_hotplug.lock (when running concurrently
with CPU hotplug operations).

Instead, the correct and race-free way of performing the callback
registration is:

cpu_notifier_register_begin();

for_each_online_cpu(cpu)
init_cpu(cpu);

/* Note the use of the double underscored version of the API */
__register_cpu_notifier(&foobar_cpu_notifier);

cpu_notifier_register_done();

Fix the sysfs code in sparc by using this latter form of callback
registration.

Cc: Ingo Molnar <mingo@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
arch/sparc/kernel/sysfs.c
This page took 0.024729 seconds and 5 git commands to generate.