On Mon, Jul 15, 2013 at 08:47:30PM +0200, Gerhard Sittig wrote:
> clocks need to get prepared before they can get enabled,
> fix the MPC512x PSC SPI master's initialization

> Signed-off-by: Gerhard Sittig <gsi-ynQEQJNshbs@public.gmane.org>
> ---
>  drivers/spi/spi-mpc512x-psc.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/spi/spi-mpc512x-psc.c b/drivers/spi/spi-mpc512x-psc.c
> index 29fce6a..76b20ea 100644
> --- a/drivers/spi/spi-mpc512x-psc.c
> +++ b/drivers/spi/spi-mpc512x-psc.c
> @@ -395,7 +395,7 @@ static int mpc512x_psc_spi_port_config(struct spi_master *master,
>  
>  	sprintf(name, "psc%d_mclk", master->bus_num);
>  	spiclk = clk_get(&master->dev, name);
> -	clk_enable(spiclk);
> +	clk_prepare_enable(spiclk);
>  	mps->mclk = clk_get_rate(spiclk);
>  	clk_put(spiclk);

This code is *clearly* buggy and should be fixed rather than papered
over.  Not only is there no matching put the driver is also dropping the
reference to the clock rather than holding on to it while it's in use.