[PATCH 1/3] imx-esdhc: update devices registration

Uwe Kleine-König u.kleine-koenig at pengutronix.de
Tue Oct 12 09:04:35 EDT 2010


Salut Eric,
On Tue, Oct 12, 2010 at 02:57:46PM +0200, Eric Bénard wrote:
> Hi Uwe,
>
> Le 12/10/2010 14:33, Uwe Kleine-König a écrit :
>>> +#ifdef CONFIG_ARCH_MX51
>>> +const struct imx_esdhc_imx_data imx51_esdhc_data[] __initconst = {
>>> +#define imx51_esdhc_data_entry(_id, _hwid)				\
>>> +	imx_esdhc_imx_data_entry(MX51, MMC_SDHC, "sdhci-esdhc-imx", _id, _hwid, SZ_16K)
>>> +	imx51_esdhc_data_entry(0, 1),
>>> +	imx51_esdhc_data_entry(1, 2),
>>> +	imx51_esdhc_data_entry(2, 3),
>>> +	imx51_esdhc_data_entry(3, 4),
>>> +};
>>> +#endif /* ifdef CONFIG_ARCH_MX51 */
>> Hmmm, I think it's sensible to use the config struct only for data that
>> is actually different for the different socs.  That is MMC_SDHC,
>> "sdhci-esdhc-imx" and SZ_16K doesn't need to be part of the data.
>>
> keeping it like this will allow to handle the case of future i.MX socs  
> which may need special (like for spi for example) for new revisions of  
> the esdhc IP.
OK, the "danger" is real as we're talking about Freescale here.  Still I
think if that happens we only have to touch a single file and so I
prefer to have it as simple as possible as long as possible.

>>> +#define MX51_SSI3_BASE_ADDR		(MX51_AIPS2_BASE_ADDR + 0xe8000)
>> This is unrelated, no?
>>
> fixed in V2.
Yep, just saw that after I hit the send button :-)

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |



More information about the linux-arm-kernel mailing list