Re: [PATCH] PM: Fix bug in the error handling of async suspend

From: Rafael J. Wysocki
Date: Wed Dec 07 2016 - 19:22:44 EST


On Wednesday, December 07, 2016 08:10:32 PM Sahitya Tummala wrote:
> If async_suspend is enabled for parent and child devices, then
> PM framework has to ensure that parent's async suspend gets called
> only after child's async suspend is done. In case if child's async
> suspend fails with error, then parent's async suspend must not be
> invoked. The current code uses async_error to ensure this but there
> is a problem with it in __device_suspend(). This function notifies
> the completion of child's async suspend before updating its error
> via async_error variable. As a result, parent's async suspend gets
> invoked even though it's child suspend has failed. Fix this bug by
> updating the async_error before notifying the child's completion.
>
> Signed-off-by: Sahitya Tummala <stummala@xxxxxxxxxxxxxx>
> ---
> drivers/base/power/main.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
> index e44944f4be77..f733f50f8321 100644
> --- a/drivers/base/power/main.c
> +++ b/drivers/base/power/main.c
> @@ -1460,9 +1460,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
> dpm_watchdog_clear(&wd);
>
> Complete:
> - complete_all(&dev->power.completion);
> if (error)
> async_error = error;
> + complete_all(&dev->power.completion);
>
> TRACE_SUSPEND(error);
> return error;

Good catch!

I'm queuing this one up (with some whitespace rearrangements).

Thanks,
Rafael