Re: [PATCH 1/1] at24: add i2c_driver struct member for at24_driver, and reduce the priority of at24_init

[Wolfram Sang <w.sang-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1504 bytes --]


>    Yes, we can't cover all the eeprom types, but I think people need
>    one example with all the function can be used. If somebody need use
>    at24.c for another eeprom, he only need to do the change
>    "s/at24/eeprom type", which maybe helpful to others. Anyway,
>    current code aren't work because of the lack of these members.

The instantiation of devices is described here:

Documentation/i2c/instantiating-devices

I'd think that would do to help people?

>      > 2. Since the at24 driver are tightly coupled with i2c bus
>      > driver, make � �sure it's init priority is lower than i2c bus.
> 
>      Can you describe a scenario where the current method fails?
> 
>    In my board, at24_init() is called before i2c-bus driver, because
>    the bus driver is initialized with module_init. But some i2c bus
>    used subsys_initcall to replace module_init, those i2c bus should
>    be ok without the change. I think only do one modification is
>    better than change all the i2c bus driver.

But they should still match, even if at24 comes first, no? Otherwise
there is a bug somewhere. (BTW changing i2c-bus-drivers to
subsys_initcall() looks better to me, too, because if there is something
like a PMIC or GPIOs connected to the bus, you usually want them early).

Regards,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]