Re: [PATCH] mm/hugetlb: Convert &folio->page to folio_page(folio, 0)

["Zi Yan" <ziy@nvidia.com>]

On Tue Apr 8, 2025 at 11:15 PM EDT, Matthew Wilcox wrote:
> On Tue, Apr 08, 2025 at 05:49:10PM -0700, nifan.cxl@gmail.com wrote:
>> From: Fan Ni <fan.ni@samsung.com>
>> 
>> Convert the use of &folio->page to folio_page(folio, 0) where struct
>> filio fits in. This is part of the efforts to move some fields out of
>> struct page to reduce its size.
>
> Thanks for sending the patch.  You've mixed together quite a few things;
> I'd suggest focusing on one API at a time.
>
<snip>

>> @@ -3403,7 +3405,7 @@ static void __split_folio_to_order(struct folio *folio, int old_order,
>>  	if (new_order)
>>  		folio_set_order(folio, new_order);
>>  	else
>> -		ClearPageCompound(&folio->page);
>> +		ClearPageCompound(folio_page(folio, 0));
>>  }
>
> I might be inclined to leave this one alone; this whole function needs
> to be rewritten as part of the folio split.

You mean __split_folio_to_order() needs a rewrite or
a folio version of ClearPageCompound()? Some thing like
__folio_clear_compound()?

>
>>  		folio_split_memcg_refs(folio, old_order, split_order);
>> -		split_page_owner(&folio->page, old_order, split_order);
>> +		split_page_owner(folio_page(folio, 0), old_order, split_order);
>>  		pgalloc_tag_split(folio, old_order, split_order);
>
> Not sure if split_folio_owner is something that should exist.  Haven't
> looked into it.

page owner info seems to be per page, but it should be OK to have
split_folio_owner() and do page level operations inside.

>
>>  		 */
>> -		free_page_and_swap_cache(&new_folio->page);
>> +		free_page_and_swap_cache(folio_page(new_folio, 0));
>>  	}
>
> free_page_and_swap_cache() should be converted to be
> free_folio_and_swap_cache().
>
>>  
>> -	return __folio_split(folio, new_order, &folio->page, page, list, true);
>> +	return __folio_split(folio, new_order, folio_page(folio, 0), page,
>> +			     list, true);
>>  }
>
> Probably right.

Yes, this is uniform split, using first page as split_at works.

>
>>  {
>> -	return __folio_split(folio, new_order, split_at, &folio->page, list,
>> -			false);
>> +	return __folio_split(folio, new_order, split_at, folio_page(folio, 0),
>> +			     list, false);
>>  }
>
> Ditto.

Right. For non-uniform split, caller holds the folio lock, so lock_at,
which tells the folio containing lock_at to keep locked when returning to caller,
is the first page.

>
>>  
>> -	return split_huge_page_to_list_to_order(&folio->page, list, ret);
>> +	return split_huge_page_to_list_to_order(folio_page(folio, 0), list,
>> +						ret);
>>  }
>
> Ditto.

Yes. And there are two additional instances in include/linux/huge_mm.h
(Yeah, the header file of mm/huge_memory.c is huge_mm.h).


-- 
Best Regards,
Yan, Zi