Re: [tip:x86/platform] x86: Unify current 3 similar ways of saving IRQ info

[Yinghai Lu <yinghai@kernel.org>]

On 12/07/2010 05:43 PM, Yinghai Lu wrote:
> On 12/07/2010 02:32 PM, Thomas Gleixner wrote:
>> On Tue, 7 Dec 2010, Yinghai Lu wrote:
>>> On 12/07/2010 07:56 AM, Thomas Gleixner wrote:
>>>>
>>>>   Same here. No caller is interested in the detailed return value.
>>>>
>>>
>>
>>> ok, after look at code and git log, i knew the reasons for those
>>> strange duplicated functions.
>>>
>>> Alexy changed
>>>
>>>  /* MP IRQ source entries */
>>> -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
>>> +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
>>>
>>> to make it decouple that with mpstable definition. so mptable will
>>> have mpc_intsrc, and acpi etc will use mp_config_intsrc
>>>
>>> and he planed to add more member to mp_config_intsrc .... never happened.
>>>
>>> then I added those functions with different parameters.
>>>
>>> one year later JSR changed all back to mpc_instr...
>>>
>>> so now have some duplicated functions....
>>
>>> So yes, we should use simple versions in
>>> arch/x86/kernel/acpi/boot.c. ( assume functions mrst.c are copied
>>> from there)
>>>
>>> and should kill print_mp_irq_info or print_MP_intsrc_info...
>>
>> Nice archaelogical detective work ! :)
>>
>> So now I have Fengs patch plus Yinghais series which seem to go nicely
>> together, but I have the feeling that there is some stuff unresolved.
>>
>> Yinghai, Feng: could you please work out the missing bits an pieces
>> (if any) and provide me a delta patch on top of the existing patches?
>>
> 
> Maybe Feng could send you updated version with simpler function?
> 

also I prefer to keep save_mp_irq() instead of changing it to mp_save_irq().

Thanks

	Yinghai