On Sat, 12 Sep 2015, Fabio Estevam wrote:

On Fri, Sep 11, 2015 at 5:08 PM, Lee Jones [off-list ref] wrote:

quoted

+       res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+       base = devm_ioremap_resource(&pdev->dev, res);
+       if (IS_ERR(base))
+               return PTR_ERR(base);
+
+       clk = devm_clk_get(&pdev->dev, NULL);
+       if (!clk)
+               return -EINVAL;

This should be:

    if (IS_ERR(clk))
        return PTR_ERR(clk);

You're right.  Will fix.

quoted

+
+       clk_prepare_enable(clk);

This may fail, so better check its return value and propagate it on error.

Looks like the jury is out on this one.

$ git grep clk_prepare_enable | grep '= clk\|if (' | wc -l
659
$ git grep clk_prepare_enable | grep -v '= clk\|if (' | wc -l
569

... but it's not a problem to fix up.  Will do.

quoted

+       ddata->ops.priv = (unsigned long)ddata;
+       ddata->ops.read = st_rng_read;
+       ddata->ops.name = pdev->name;
+       ddata->base     = base;
+       ddata->clk      = clk;
+
+       dev_set_drvdata(&pdev->dev, ddata);
+
+       ret = hwrng_register(&ddata->ops);
+       if (ret) {
+               dev_err(&pdev->dev, "Failed to register HW RNG\n");
+               return ret;
+       }
+
+       dev_info(&pdev->dev, "Successfully registered HW RNG\n");

No need to put this info.

Ah, you caught me!

I know these types of prints are usually deemed not useful; however,
unless there is a failure both this driver and the core driver are
silent, so the user doesn't know that 1 or more of these devices are
available.  I personally like to see an entry in the bootlog for this
kind of functionality.

-- 
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org ? Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog