Hi,
On Thu, Dec 07, 2017 at 12:22:13AM +0900, Takuo Koguchi wrote:
> mclk and hclk need to be disabled explicitly since pm_runtime_disable does
> not disable the clocks.
>
> Found by Linux Driver Verification project (linuxtesting.org).
>
> Signed-off-by: Takuo Koguchi <takuo.koguchi.sw@xxxxxxxxxxx>
> ---
> drivers/spi/spi-sun4i.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/spi/spi-sun4i.c b/drivers/spi/spi-sun4i.c
> index c5cd635..9011c09 100644
> --- a/drivers/spi/spi-sun4i.c
> +++ b/drivers/spi/spi-sun4i.c
> @@ -525,6 +525,11 @@ static int sun4i_spi_probe(struct platform_device *pdev)
>
> static int sun4i_spi_remove(struct platform_device *pdev)
> {
> + struct spi_master *master = dev_get_drvdata(&pdev->dev);
> + struct sun4i_spi *sspi = spi_master_get_devdata(master);
> +
> + clk_disable_unprepare(sspi->mclk);
> + clk_disable_unprepare(sspi->hclk);
And that will in turn create an unbalanced disable call if suspend was
called before the remove, which is very likely.
Why are you not calling pm_runtime_force_suspend? How did you test
that patch?
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com
Attachment:
signature.asc
Description: PGP signature
