Re: [PATCH] i2c-i801: Don't depend on other kernel driver config options

2011-05-13 Thread Woodhouse, David 
On Fri, 2011-05-13 at 11:01 +0100, Jean Delvare wrote:
> Don't let other driver config options influence us, as it makes the
> code more complex and fragile for a small benefit. There's nothing
> wrong with instantiating I2C devices even if they don't have a driver.
> And we're talking about 835 extra bytes in the binary on x86-64,
> that's hardly worth arguing about. 

Looks sane to me; thanks. Should it be CONFIG_DMI instead of (or in
addition to) CONFIG_X86?
-- 
David WoodhouseOpen Source Technology Centre
david.woodho...@intel.com  Intel Corporation


smime.p7s
Description: S/MIME cryptographic signature

Re: [PATCH] i2c-i801: Don't probe for slaves on IDF channels

2011-05-10 Thread Woodhouse, David 
On Tue, 2011-05-10 at 15:08 +0100, Jean Delvare wrote:
> I don't know if Fujitsu is ever going to produce Patsburg-based
> machines, but if they do, I'd rather not probe the secondary (IDF)
> SMBus channels. At least not until we have a good reason for doing so.

Seems reasonable in principle.

> +#if defined CONFIG_INPUT_APANEL || defined CONFIG_INPUT_APANEL_MODULE

> +#if defined CONFIG_SENSORS_FSCHMD || defined CONFIG_SENSORS_FSCHMD_MODULE

This kind of ifdef is horrid though, and should be avoided at all costs.
A setting of CONFIG_xxx_MODULE should not affect the *static* kernel
image. You should be able to enable an additional module, build it and
use it without having to rebuild/reboot the kernel.

I know there are *some* cases where we violate that rule, but that
doesn't make it any nicer.

-- 
David WoodhouseOpen Source Technology Centre
david.woodho...@intel.com  Intel Corporation


smime.p7s
Description: S/MIME cryptographic signature