Re: [PATCH] irq: fix a Kconfig dependency

["Kamble, Nitin A" <nitin.a.kamble@intel.com>]

On 1/31/2014 2:27 AM, David Rientjes wrote:
> On Thu, 30 Jan 2014, nitin.a.kamble@intel.com wrote:
>
>> From: Nitin A Kamble <nitin.a.kamble@intel.com>
>>
>> The generic_chip.c uses interfaces from irq_domain.c which is controlled by
>> the IRQ_DOMAIN config option.
>>
>> Add a select statement in the Kconfig to reflect this requirement.
>>
>> Without this fix, the generic_chip.c compilation fails like this:
>>
>> linux/kernel/irq/generic-chip.c:400:11:
>>    error: 'irq_domain_xlate_onetwocell' undeclared here (not in a function)
>>
> You're not specifying which config option you have enabled that selects
> CONFIG_GENERIC_IRQ_CHIP and does not select CONFIG_IRQ_DOMAIN, but it does
> look like there is a dependency here.
It is enabled by GPIO_PCH as seen here:

config GPIO_PCH
         tristate "Intel EG20T PCH/LAPIS Semiconductor 
IOH(ML7223/ML7831) GPIO"
         depends on PCI && X86
         select GENERIC_IRQ_CHIP
         help
           This driver is for PCH(Platform controller Hub) GPIO of Intel 
Topcliff
           which is an IOH(Input/Output Hub) for x86 embedded processor.
           This driver can access PCH GPIO device.

>
> So, if this dependency is reasonable, it begs the question of why we have
> CONFIG_GENERIC_IRQ_CHIP at all and shouldn't just make generic-chip.c
> depend on CONFIG_IRQ_DOMAIN.
Looking at the code there are cases where IRQ_DOMAIN functionality is used
without using the GENERIC_IRQ_CHIP. So making generic-chip.c depend
on CONFIG_IRQ_DOMAIN in the makefile will reduce the flexibility of using
these config options separately.

Nitin