On Tuesday, August 14, 2012, Jon Medhurst (Tixy) wrote:
> When a kernel is built to support multiple hardware types it's possible
> that CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED is set but the hardware the
> kernel is run on doesn't support cpuidle and therefore doesn't load a
> driver for it. In this case, when the system is shut down,
> cpuidle_coupled_cpu_notify() gets called with cpuidle_devices set to
> NULL. There are quite possibly other circumstances where this
> situation can also occur and we should check for it.
>
> Signed-off-by: Jon Medhurst <tixy@xxxxxxxxxx>
Applied to the linux-next branch of the linux-pm.git tree as v3.6 material.
Thanks,
Rafael
> ---
>
> Would a better fix be to delay registering the notifier until
> cpuidle_coupled_register_device() is called? Though following through
> with that logic would require unregistering the notifier when cpuidle
> driver is unregistered and it still seems possible that there would
> still be a window of opportunity for things to go wrong.
>
>
> drivers/cpuidle/coupled.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cpuidle/coupled.c b/drivers/cpuidle/coupled.c
> index 2c9bf26..c78a5ff 100644
> --- a/drivers/cpuidle/coupled.c
> +++ b/drivers/cpuidle/coupled.c
> @@ -681,7 +681,7 @@ static int cpuidle_coupled_cpu_notify(struct notifier_block *nb,
> mutex_lock(&cpuidle_lock);
>
> dev = per_cpu(cpuidle_devices, cpu);
> - if (!dev->coupled)
> + if (!dev || !dev->coupled)
> goto out;
>
> switch (action & ~CPU_TASKS_FROZEN) {
>
