Re: [PATCH 01/14] mm/memory_hotplug: remove for_each_valid_pfn() usage

[David Hildenbrand (Arm)]

On 3/17/26 17:56, David Hildenbrand (Arm) wrote:
> When offlining memory, we know that the memory range has no holes.
> Checking for valid pfns is not required.
> 
> Signed-off-by: David Hildenbrand (Arm) <david@xxxxxxxxxx>
> ---
>  mm/memory_hotplug.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> index 86d3faf50453..3495d94587e7 100644
> --- a/mm/memory_hotplug.c
> +++ b/mm/memory_hotplug.c
> @@ -1746,7 +1746,7 @@ static int scan_movable_pages(unsigned long start, unsigned long end,
>  {
>  	unsigned long pfn;
>  
> -	for_each_valid_pfn(pfn, start, end) {
> +	for (pfn = start; pfn < end; pfn++) {
>  		struct page *page;
>  		struct folio *folio;
>  
> @@ -1791,7 +1791,7 @@ static void do_migrate_range(unsigned long start_pfn, unsigned long end_pfn)
>  	static DEFINE_RATELIMIT_STATE(migrate_rs, DEFAULT_RATELIMIT_INTERVAL,
>  				      DEFAULT_RATELIMIT_BURST);
>  
> -	for_each_valid_pfn(pfn, start_pfn, end_pfn) {
> +	for (pfn = start_pfn; pfn < end_pfn; pfn++) {
>  		struct page *page;
>  
>  		page = pfn_to_page(pfn);

AI review reports something rather unrelated to this patch: if the stars
align, folio_nr_pages(folio) might return questionable values.

We certainly don't want to tryget all folios here, so we might just want
to make sure that the value we get from folio_nr_pages() is something
reasonable (e.g., >= 1, power of 2). Alternatively we might snapshot the
page.

Will look into it.

-- 
Cheers,

David