twl4030 module rewrite

twl4030 module rewrite

[Felipe Balbi]

Hello all,

twl4030 is the power management chip for omap3-based boards, its
driver can be found in [1].

My question is:

twl4030 provides 4 addresses for different functions using the same
i2c bus. Which means that one client driver would have to handle 4
clients.

Putting all the addresses in i2c_board_info would be enough for that
driver to be called 4 times or should I rely in the driver been called
only one and iterate to initialize all 4 clients ?

[1] http://git.kernel.org/?p=linux/kernel/git/tmlind/linux-omap-2.6.git;a=blob;f=drivers/i2c/chips/twl4030-core.c;h=adc45d4a98b97254999d9cc3bbc739962d727a27;hb=7786cd7a00ae0b18923185789380a88052f4eee7

-- 
Best Regards,

Felipe Balbi
felipebalbi-Rn4VEauK+AKRv+LV9MX5uipxlwaOVQ5f@public.gmane.org

_______________________________________________
i2c mailing list
i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
http://lists.lm-sensors.org/mailman/listinfo/i2c

Re: twl4030 module rewrite

[Jean Delvare]

Hi Felipe,

On Fri, 11 Jul 2008 12:16:23 +0300, Felipe Balbi wrote:
> Hello all,
> 
> twl4030 is the power management chip for omap3-based boards, its
> driver can be found in [1].
> 
> My question is:
> 
> twl4030 provides 4 addresses for different functions using the same
> i2c bus. Which means that one client driver would have to handle 4
> clients.
> 
> Putting all the addresses in i2c_board_info would be enough for that
> driver to be called 4 times or should I rely in the driver been called
> only one and iterate to initialize all 4 clients ?
> 
> [1] http://git.kernel.org/?p=linux/kernel/git/tmlind/linux-omap-2.6.git;a=blob;f=drivers/i2c/chips/twl4030-core.c;h=adc45d4a98b97254999d9cc3bbc739962d727a27;hb=7786cd7a00ae0b18923185789380a88052f4eee7

If you want the same driver to handle all 4 addresses, and you want
them to show as a single device, then the usual way to handle this
would be to have a single address declared in i2c_board_info, and have
the driver itself call i2c_new_dummy() 3 times to request the 3
additional addresses during device probe.

The alternative, if the 4 addresses are really used for different,
unrelated functions, is to declare each one separately in
i2c_board_info, leading to separate devices and possibly separate
drivers. It really depends on how related the functions are, and
whether each function may also be available separately in a different
chip. Without knowing the details of how the chip works, I can't tell.

-- 
Jean Delvare

_______________________________________________
i2c mailing list
i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
http://lists.lm-sensors.org/mailman/listinfo/i2c

Re: twl4030 module rewrite

[Felipe Balbi]

Hi Jean,

On 7/11/08, Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org> wrote:

> If you want the same driver to handle all 4 addresses, and you want
>  them to show as a single device, then the usual way to handle this
>  would be to have a single address declared in i2c_board_info, and have
>  the driver itself call i2c_new_dummy() 3 times to request the 3
>  additional addresses during device probe.

Thanks for this pointer. So to clear my mind, in this time I'd have to
keep a list of all 4 i2c_client pointers because they'd have different
addresses (although it's the same adapter).

-- 
Best Regards,

Felipe Balbi
felipebalbi-Rn4VEauK+AKRv+LV9MX5uipxlwaOVQ5f@public.gmane.org

_______________________________________________
i2c mailing list
i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
http://lists.lm-sensors.org/mailman/listinfo/i2c