Re: [PATCH V3 1/3] x86/PCI: Fix PCI root numa_node info on AMD family15h

[Suravee Suthikulanit <suravee.suthikulpanit@amd.com>]

On 5/8/2014 10:14 AM, Robert Richter wrote:
> On 08.05.14 09:39:55, Suravee Suthikulanit wrote:
>> On 5/8/2014 4:01 AM, Robert Richter wrote:
>>> On 08.05.14 10:59:05, Robert Richter wrote:
>>>> On 07.05.14 13:58:45, suravee.suthikulpanit@amd.com wrote:
>>>>> @@ -113,10 +122,17 @@ static int __init early_fill_mp_bus_info(void)
>>>>>   		info = alloc_pci_root_info(min_bus, max_bus, node, link);
>>>>>   	}
>>>>>
>>>>> +	/*
>>>>> +	 * The following code is only supported until Fam11h.
>>>>> +	 * Newer processors will depend on ACPI MCFG table instead.
>>>>> +	 */
>>>>> +	if (boot_cpu_data.x86 > 0x11)
>>>>> +		return 0;
>>>>> +
>>>>>   	/* get the default node and link for left over res */
>>>>
>>>> As this is the only substantial change of your patch, I would better
>>>> drop ther rest or at least split it in two patches. Should this change
>>>> also be for stable?
>>>
>>> Of course adding the hostbridge must be also part of the patch, didn't
>>> note this due to the other noise. See why the split would be good?
>>>
>>>>
>>>> -Robert
>>
>>
>> Robert,
>>
>> I have already added the hostbridge for family15h in this patch.
>>
>> +static struct amd_hostbridge hb_probes[] __initdata = {
>> +	{ 0, 0x18, 0x1100 }, /* K8 */
>> +	{ 0, 0x18, 0x1200 }, /* Family10h */
>> +	{ 0xff, 0, 0x1200 }, /* Family10h */
>> +	{ 0, 0x18, 0x1300 }, /* Family11h */
>> +	{ 0, 0x18, 0x1600 }, /* Family15h */ <--- HERE
>
> Yes, I noticed that, but later, thus my 2nd mail.
>
>>   };
>>
>> The rest of the changes are mostly comments, some minor renaming of
>> variables for clarity, and replace hardcode values with preprocessor macro.
>> If needed, I can split them.
>
> I just would drop it, you just need the fam15h device and the cpu mode
> check.
>
> -Robert
>
The reason I put it all these comments here is because it took us a 
while to discuss what to do with this file going forward. There were 
some confusions.  Therefore, I just want to document it here.

Also, the check for (boot_cpu_data.x86 > 0x11) was needed because it 
should not be done for family15h.

Suravee