Re: [PATCH v2 04/11] mm: Make page_mapped_in_vma conditional on CONFIG_MEMORY_FAILURE

[Jane Chu <jane.chu@oracle.com>]

On 4/8/2024 3:52 PM, Matthew Wilcox wrote:

> On Mon, Apr 08, 2024 at 03:45:37PM -0700, Jane Chu wrote:
>> On 4/8/2024 12:42 PM, Matthew Wilcox (Oracle) wrote:
>>> +#ifdef CONFIG_MEMORY_FAILURE
>>>    /**
>>>     * page_mapped_in_vma - check whether a page is really mapped in a VMA
>>>     * @page: the page to test
>>> @@ -344,3 +345,4 @@ int page_mapped_in_vma(struct page *page, struct vm_area_struct *vma)
>>>    out:
>>>    	return pvmw.address;
>>>    }
>>> +#endif
>> Should below be put in the CONFIG_MEMORY_FAILURE bracket as well ?
>>
>> include/linux/rmap.h: int page_mapped_in_vma(struct page *page, struct
>> vm_area_struct *vma);
> I don't see why we should; I see no advantage to adding that ifdef.
> What would motivate you to add it?

Just to be consistent, like, either both function definition and 
function declaration be put in the config switch bracket, or neither.  
Perhaps this is a nitpick, up to you.

Thanks,

-jane

>
>> with that,
>>
>> Reviewed-by: Jane Chu <jane.chu@oracle.com>
> Thanks!