2017-06-06 09:44:35

by Arvind Yadav

Subject: [PATCH] spi: pxa2xx: Handle return value of clk_prepare_enable

clk_prepare_enable() can fail here and we must check its return value.

Signed-off-by: Arvind Yadav <[email protected]>
---
drivers/spi/spi-pxa2xx.c | 18 +++++++++++++-----
1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index 47b65d7..e9c0fe5 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -1676,7 +1676,11 @@ static int pxa2xx_spi_probe(struct platform_device *pdev)
}

/* Enable SOC clock */
- clk_prepare_enable(ssp->clk);
+ status = clk_prepare_enable(ssp->clk);
+ if (status) {
+ dev_err(&pdev->dev, "Failed to prepare clock\n");
+ goto out_error_master_alloc;
+ }

master->max_speed_hz = clk_get_rate(ssp->clk);

@@ -1855,8 +1859,13 @@ static int pxa2xx_spi_resume(struct device *dev)
int status;

/* Enable the SSP clock */
- if (!pm_runtime_suspended(dev))
- clk_prepare_enable(ssp->clk);
+ if (!pm_runtime_suspended(dev)) {
+ status = clk_prepare_enable(ssp->clk);
+ if (status) {
+ dev_err(dev, "Failed to prepare clock\n");
+ return status;
+ }
+ }

/* Restore LPSS private register bits */
if (is_lpss_ssp(drv_data))
@@ -1886,8 +1895,7 @@ static int pxa2xx_spi_runtime_resume(struct device *dev)
{
struct driver_data *drv_data = dev_get_drvdata(dev);

- clk_prepare_enable(drv_data->ssp->clk);
- return 0;
+ return clk_prepare_enable(drv_data->ssp->clk);
}
#endif

--
1.9.1

2017-06-06 10:54:47

by Andy Shevchenko

Subject: Re: [PATCH] spi: pxa2xx: Handle return value of clk_prepare_enable

On Tue, Jun 6, 2017 at 12:44 PM, Arvind Yadav <[email protected]> wrote:
> clk_prepare_enable() can fail here and we must check its return value.

> @@ -1676,7 +1676,11 @@ static int pxa2xx_spi_probe(struct platform_device *pdev)

> - clk_prepare_enable(ssp->clk);
> + status = clk_prepare_enable(ssp->clk);

This one looks fine.

> @@ -1855,8 +1859,13 @@ static int pxa2xx_spi_resume(struct device *dev)

> /* Enable the SSP clock */
> - if (!pm_runtime_suspended(dev))
> - clk_prepare_enable(ssp->clk);
> + if (!pm_runtime_suspended(dev)) {
> + status = clk_prepare_enable(ssp->clk);
> + if (status) {
> + dev_err(dev, "Failed to prepare clock\n");
> + return status;
> + }

This...

> @@ -1886,8 +1895,7 @@ static int pxa2xx_spi_runtime_resume(struct device *dev)
> {
> struct driver_data *drv_data = dev_get_drvdata(dev);
>
> - clk_prepare_enable(drv_data->ssp->clk);
> - return 0;
> + return clk_prepare_enable(drv_data->ssp->clk);

...and especially this should be carefully checked since there are
differences in behaviour how system or driver will be resumed.

So, the question is how did you test it?

--
With Best Regards,
Andy Shevchenko

2017-06-06 11:34:01

by Arvind Yadav

Subject: Re: [PATCH] spi: pxa2xx: Handle return value of clk_prepare_enable

On Tuesday 06 June 2017 04:24 PM, Andy Shevchenko wrote:
> On Tue, Jun 6, 2017 at 12:44 PM, Arvind Yadav <[email protected]> wrote:
>> clk_prepare_enable() can fail here and we must check its return value.
>> @@ -1676,7 +1676,11 @@ static int pxa2xx_spi_probe(struct platform_device *pdev)
>> - clk_prepare_enable(ssp->clk);
>> + status = clk_prepare_enable(ssp->clk);
> This one looks fine.
>
>> @@ -1855,8 +1859,13 @@ static int pxa2xx_spi_resume(struct device *dev)
>> /* Enable the SSP clock */
>> - if (!pm_runtime_suspended(dev))
>> - clk_prepare_enable(ssp->clk);
>> + if (!pm_runtime_suspended(dev)) {
>> + status = clk_prepare_enable(ssp->clk);
>> + if (status) {
>> + dev_err(dev, "Failed to prepare clock\n");
>> + return status;
>> + }
> This...
>
>> @@ -1886,8 +1895,7 @@ static int pxa2xx_spi_runtime_resume(struct device *dev)
>> {
>> struct driver_data *drv_data = dev_get_drvdata(dev);
>>
>> - clk_prepare_enable(drv_data->ssp->clk);
>> - return 0;
>> + return clk_prepare_enable(drv_data->ssp->clk);
> ...and especially this should be carefully checked since there are
> differences in behaviour how system or driver will be resumed.
yes true, here clk_prepare_enable will return 0 on successful attempt.
what do you suggest here, we should not return like this.?
>
> So, the question is how did you test it?
It can fail, I am not able to produce clock failure issue. If you have
any suggestion.
please let me know.
>
-arvind