On Wed, Feb 5, 2014 at 5:02 PM, Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> On Wed, Feb 5, 2014 at 7:52 AM, Magnus Damm <magnus.damm@xxxxxxxxx> wrote:
>> +static int rcar_pci_probe(struct platform_device *pdev)
>> {
>> struct resource *cfg_res, *mem_res;
>> struct rcar_pci_priv *priv;
>> void __iomem *reg;
>> + struct hw_pci hw;
>>
>> cfg_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
>> reg = devm_ioremap_resource(&pdev->dev, cfg_res);
>> @@ -308,31 +271,26 @@ static int __init rcar_pci_probe(struct
>> priv->reg = reg;
>> priv->dev = &pdev->dev;
>>
>> - return rcar_pci_add_controller(priv);
>> + memset(&hw, 0, sizeof(hw));
>> + hw.nr_controllers = 1;
>> + hw.private_data = (void **)&priv;
>
> This raised a red flag: taking the address of a variable on the stack.
> I know it's correct, as hw.private_data is an array of pointers copied
> by pci_common_init_dev() below.
> But perhaps it's a good idea to turn priv into an array with one element,
> to make this clearer, and avoid the ugly cast?
I simply followed the same style as pci-tegra.c, but I agree that it
can be made prettier. Also, there may be some better way how to
register independent instances, not sure.
Thanks,
/ magnus
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
