Re: [PATCH 7/8] mm/khugepaged: clarify a comment regarding max_ptes_none check

[Usama Arif]

On Mon,  6 Jul 2026 09:44:54 -0600 Nico Pache <npache@xxxxxxxxxx> wrote:

> While reading collapse_scan_pmd, one may be confused on where the deferred
> max_ptes_none check is done. Expand on this current comment by explaining
> which function the max_ptes_none check is now done.
> 
> Suggested-by: David Hildenbrand <david@xxxxxxxxxx>
> Signed-off-by: Nico Pache <npache@xxxxxxxxxx>
> ---
>  mm/khugepaged.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 3c6f1254deca..388045a524a3 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -1728,7 +1728,8 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  
>  	/*
>  	 * If PMD is the only enabled order, enforce max_ptes_none, otherwise
> -	 * scan all pages to populate the bitmap for mTHP collapse.
> +	 * scan all pages to populate the bitmap for mTHP collapse. The bitmap
> +	 * is then checked again in mthp_collapse() for each attempted order.

Could be combined with patch 4?

  	 */
>  	if (enabled_orders != BIT(HPAGE_PMD_ORDER))
>  		ctx.max_ptes_none = KHUGEPAGED_MAX_PTES_LIMIT;
> -- 
> 2.54.0
> 
>