Re: [lm-sensors] [PATCH] hwmon: (KConfig) avoid building drivers

[Dean Nelson <dnelson@redhat.com>]

Hello Jean,

On 08/16/2011 09:27 AM, Jean Delvare wrote:
> Hi Dean,
>
> On Tue, 26 Jul 2011 09:35:35 -0500, Dean Nelson wrote:
>> On 07/26/2011 04:25 AM, Jean Delvare wrote:
>>>> @@ -933,7 +938,7 @@ config SENSORS_S3C_RAW
>>>>
>>>>    config SENSORS_SIS5595
>>>>    	tristate "Silicon Integrated Systems Corp. SiS5595"
>>>> -	depends on PCI
>>>> +	depends on PCI&&   !PPC
>>>>    	help
>>>>    	  If you say yes here you get support for the integrated sensors in
>>>>    	  SiS5595 South Bridges.
>>>
>>> This one is a PCI driver, it doesn't do random I/O, so it should be
>>> left as is. You did not experience a crash on PowerPC by loading this
>>> driver, did you?
>>
>> Okay. Thanks. And no, I didn't experience an Oops.
>>
>> When sm_sis5595_init() gets called at module load time, it calls
>> pci_register_driver() to register sis5595_pci_probe(), which never
>> gets called.
>>
>> Had sis5595_pci_probe() been called, it could have called
>> platform_driver_register() to register sis5595_probe(), which is
>> where we could potentially read/write the ISA address.
>
> This is really a PCI I/O address and not an ISA address, by
> construction. Any system with this PCI device would presumably properly
> support said I/O operations.

Thanks for the clarification.


>> I'd added the "&&  !PPC" to this driver because it was in the list
>> of drivers you mentioned that I hadn't, when I first introduced this
>> issue. Sorry for misunderstanding. It's been removed from v2 of this
>> patch.
>
> Oops, this is my fault then, sorry for the confusion. For my excuse,
> this was before my pretty needed vacation ;)

My lack of understanding certainly played its part in all of this. :)
(Aren't vacations wonderful.)


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors