PATCH Add support for the Iomega IX2-200 Device Tree

[alan butler]

On 16 August 2012 00:12, alan butler <alanbutty12 at gmail.com> wrote:
> On 16 August 2012 00:02, alan butler <alanbutty12 at gmail.com> wrote:
>> On 15 August 2012 23:16, Andrew Lunn <andrew at lunn.ch> wrote:
>>> On Wed, Aug 15, 2012 at 10:57:27PM +0100, alan butler wrote:
>>>> On 15 August 2012 22:52, Andrew Lunn <andrew at lunn.ch> wrote:
>>>> >> That would be great i just need to recreate the patch with those
>>>> >> changes but i was talking with jason cooper i believe and he said to
>>>> >> put the buttons for the device in the dts file and im working on that
>>>> >> but the kernel panics
>>>> >
>>>> > Please show us the panic.
>>>> >
>>>> >        Andrew
>>>>
>>>> i sent it to jason a lil bit earlier but i guess i must of forgotten
>>>> to cc the list on the last message or 2 but here is the panic again:
>>>>
>>>> [    5.976074] Key type dns_resolver registered
>>>> [    5.981782] Unable to handle kernel NULL pointer dereference at
>>>> virtual address 00000004
>>>> [    5.989932] pgd = c0004000
>>>> [    5.992649] [00000004] *pgd=00000000
>>>> [    5.996247] Internal error: Oops: 805 [#1] ARM
>>>> [    6.000707] Modules linked in:
>>>> [    6.003780] CPU: 0    Not tainted  (3.6.0-rc1-next-20120814 #13)
>>>> [    6.009822] PC is at gpio_keys_probe+0x110/0x6d4
>>>
>>> You might want to scatter some printk() in gpio_keys_probe() and see
>>> where it panics. It will probably then be clear what is wrong.
>>>
>>>       Andrew
>>
>> where do i find the gpio_keys_probe() ?
>
> i found the gpio_keys_probe() and it seems that it is failing in this
> piece of code i put a printk() before and after it and it didnt reach
> the printk after it:
>
> if (!pdata) {
>                 pdata = gpio_keys_get_devtree_pdata(dev);
>                 if (IS_ERR(pdata))
>                         return PTR_ERR(pdata);
>         }
>
>         ddata = kzalloc(sizeof(struct gpio_keys_drvdata) +
>                         pdata->nbuttons * sizeof(struct gpio_button_data),
>                         GFP_KERNEL);
>         input = input_allocate_device();
>         if (!ddata || !input) {
>                 dev_err(dev, "failed to allocate state\n");
>                 error = -ENOMEM;
>                 goto fail1;
>         }

actualy it seems to not be getting past this piece of code in what i
posted before:

if (!pdata) {
		pdata = gpio_keys_get_devtree_pdata(dev);
		if (IS_ERR(pdata))
			return PTR_ERR(pdata);
	}