[PATCH] mtd: gpmi: add gpmi_devdata{} to simplify the code

[Gupta, Pekon]

>From: Huang Shijie [mailto:b32955 at freescale.com]
>于 2014年03月21日 18:29, Gupta, Pekon 写道:
>> Ezequiel has got approval for 2 new generic MTD bindings (refer l2-mtd.git)
>> 	commit 8dd49165ef5f46b5ad9ba296c559ccff315f9421
>> 	mtd: nand: Add a devicetree binding for ECC strength and ECC step size
>> So, I think<1>  can be replaced by "nand-ecc-strength".
>>
>>
>"nand-ecc-strength" is add the ECC strength for the NAND, but my patch
>is add
>the maximum ECC strength the BCH controller can support.
>
>they are different things.
>
Ok then it's also tied to your IP / SoC revision right ?

>>> >
>>> >    [2] add the gpmi_devdata_imx{23|28|6q} to replace the gpmi_ids.
>>> >
>> Can you use "compatible" string in DT for this ?
>>
>>
>sorry, could you please give me an example?
>
>I do not know what's your meaning. :(
>

Can you use something like this, instead of populating static for each chip.

if (of_device_is_compatible(of_node, "imx23") || of_device_is_compatible(child, "imx28")) {
	bch_max_ecc_strength = 20;
	max_chain_delay = 16;
} elseif (of_device_is_compatible(of_node, "imx6q") {
	bch_max_ecc_strength = 40;
	max_chain_delay = 12;
} elseif (of_device_is_compatible(of_node, "imx6sx") {
	bch_max_ecc_strength = 40;
	max_chain_delay = 0; /* whatever is here */
} else {
	bch_max_ecc_strength = 1; /* whatever is default */
	max_chain_delay = 0; /* whatever is default */
}


with regards, pekon