Re: [PATCH v3 3/4] memory: omap-gpmc: Use a compatible match table when checking for NAND controller

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 20/12/2021 12:51, Roger Quadros wrote:
> 
> 
> On 20/12/2021 13:05, Krzysztof Kozlowski wrote:
>> On 20/12/2021 11:53, Roger Quadros wrote:
>>>
>>>
>>> On 17/12/2021 17:21, Krzysztof Kozlowski wrote:
>>>> On 17/12/2021 11:29, Roger Quadros wrote:
>>>>> As more compatibles can be added to the GPMC NAND controller driver
>>>>> use a compatible match table.
>>>>>
>>>>> Signed-off-by: Roger Quadros <rogerq@xxxxxxxxxx>
>>>>> ---
>>>>>  drivers/memory/omap-gpmc.c                   | 8 +++++++-
>>>>>  drivers/mtd/nand/raw/omap2.c                 | 2 +-
>>>>>  include/linux/platform_data/mtd-nand-omap2.h | 5 +++++
>>>>>  3 files changed, 13 insertions(+), 2 deletions(-)
>>>>>
>>>>> diff --git a/drivers/memory/omap-gpmc.c b/drivers/memory/omap-gpmc.c
>>>>> index 624153048182..814ddb45c13d 100644
>>>>> --- a/drivers/memory/omap-gpmc.c
>>>>> +++ b/drivers/memory/omap-gpmc.c
>>>>> @@ -2091,6 +2091,7 @@ static int gpmc_probe_generic_child(struct platform_device *pdev,
>>>>>  	u32 val;
>>>>>  	struct gpio_desc *waitpin_desc = NULL;
>>>>>  	struct gpmc_device *gpmc = platform_get_drvdata(pdev);
>>>>> +	bool is_nand = false;
>>>>>  
>>>>>  	if (of_property_read_u32(child, "reg", &cs) < 0) {
>>>>>  		dev_err(&pdev->dev, "%pOF has no 'reg' property\n",
>>>>> @@ -2183,7 +2184,12 @@ static int gpmc_probe_generic_child(struct platform_device *pdev,
>>>>>  		}
>>>>>  	}
>>>>>  
>>>>> -	if (of_device_is_compatible(child, "ti,omap2-nand")) {
>>>>> +#if defined(CONFIG_MTD_NAND_OMAP2)
>>>>
>>>> if (IS_ENABLED()) is preferred. If needed, you could make omap_nand_ids
>>>> symbol visible always (so without ifdef around it), because extern
>>>> structure should not have impact when not defined (if I recall
>>>> correctly...).
>>>>
>>>>> +	if (of_match_node(omap_nand_ids, child))
>>>>> +		is_nand = true;
>>>>> +#endif
>>>>> +
>>>>> +	if (is_nand) {
>>>>>  		/* NAND specific setup */
>>>>>  		val = 8;
>>>>>  		of_property_read_u32(child, "nand-bus-width", &val);
>>>>> diff --git a/drivers/mtd/nand/raw/omap2.c b/drivers/mtd/nand/raw/omap2.c
>>>>> index b26d4947af02..fff834ee726f 100644
>>>>> --- a/drivers/mtd/nand/raw/omap2.c
>>>>> +++ b/drivers/mtd/nand/raw/omap2.c
>>>>> @@ -2352,7 +2352,7 @@ static int omap_nand_remove(struct platform_device *pdev)
>>>>>  	return ret;
>>>>>  }
>>>>>  
>>>>> -static const struct of_device_id omap_nand_ids[] = {
>>>>> +const struct of_device_id omap_nand_ids[] = {
>>>>>  	{ .compatible = "ti,omap2-nand", },
>>>>>  	{},
>>>>>  };
>>>>
>>>> I think OMAP2 NAND driver can be a module, so this should have
>>>> EXPORT_SYMBOL.
>>>
>>> To make it work in all combinations (e.g. omap_gpmc built in and
>>> nand/raw/omap2.c as module) I had to define omap_nand_ids table as static
>>> in the linux/platform_data/mtd-nand-omap2.h header.
>>>
>>> EXPORT_SYMBOL will of course be not required there. ;)
>>>
>> Which case exactly does it require to be static in the header?
> 
> Maybe there is a better way to do it.
> e.g. If omap_gpmc.c is built-in and nand/raw/omap2.c is not built or built as
> a module, what better way we can use?

Ah, you are right, that is the tricky configuration. It could be a
separate built-in source file which is being selected by OMAP_GPMC and
MTD_NAND_OMAP2, but it would be also an overkill for one-item array.

I guess your solution with header is the easiest.


Best regards,
Krzysztof



[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]


  Powered by Linux