Re: [PATCH V9 1/2] ACPI, PCI, irq: remove interrupt count restriction

[Sinan Kaya <okaya@codeaurora.org>]

Hi Bjorn, Andy;

On 12/9/2015 12:14 PM, Christopher Covington wrote:
> Hi Sinan,
> 
> On 12/09/2015 12:09 PM, Sinan Kaya wrote:
>> On 12/9/2015 11:59 AM, Andy Shevchenko wrote:
>>>> +       if (trigger != ACPI_MADT_TRIGGER_LEVEL ||
>>>>> +           polarity != ACPI_MADT_POLARITY_ACTIVE_LOW)
>>>>> +               penalty = PIRQ_PENALTY_ISA_ALWAYS;
>>>>> +       else
>>>>> +               penalty = PIRQ_PENALTY_PCI_USING;
>>>>> +
>>>>> +       acpi_irq_add_penalty(irq, penalty);
>>> Why not to change in place? I think a common sense rule is not to
>>> change something existing if it doesn't add any significant value.
>>>
>> Sorry, I didn't understand what you mean. Are you asking why we are
>> changing lines like above?
>>
>> If yes, acpi_irq_penalty used to be an array of 256 entries. Now,
>> acpi_irq_penalty doesn't exist anymore as it was replaced with a linklist.
>>
>>> -               acpi_irq_penalty[irq] += PIRQ_PENALTY_PCI_USING;
>>> +              acpi_irq_add_penalty(irq, PIRQ_PENALTY_PCI_USING);
> 
> I think Andy was suggesting that you make the change without introducing
> the penalty variable.
> 
> Christopher Covington
> 

Andy,
Is Chris' interpretation correct?

BTW, I suggest you spend some time around checkpatch for contributions. I could
have caught most of the issues you are generally concerned before submitting a patch.

Bjorn,
Is there any other question you need me to address on this patch? 


-- 
Sinan Kaya
Qualcomm Technologies, Inc. on behalf of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project