Re: [PATCH v2] mm/hugetlb: fix set_max_huge_pages() when there are surplus pages

[Jinjiang Tu <tujinjiang@huawei.com>]

在 2025/4/4 21:18, Oscar Salvador 写道:
> On Thu, Apr 03, 2025 at 04:26:36PM +0200, Oscar Salvador wrote:
>> Unfortunately, this will not fly.
> I did not spend much time, but solving the problem of hvo free pages
> becoming surplus not being accounted twice, should not the following
> fix the entire problem?

Sorry for late replay.

>   diff --git a/mm/hugetlb.c b/mm/hugetlb.c
>   index 39f92aad7bd1..6e9534a825ed 100644
>   --- a/mm/hugetlb.c
>   +++ b/mm/hugetlb.c
>   @@ -3825,6 +3825,7 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
>    static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
>    			      nodemask_t *nodes_allowed)
>    {
>   +	unsigned long substract;
>    	unsigned long min_count;
>    	unsigned long allocated;
>    	struct folio *folio;
>   @@ -3960,7 +3961,12 @@ static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
>    	 * and won't grow the pool anywhere else. Not until one of the
>    	 * sysctls are changed, or the surplus pages go out of use.
>    	 */
>   -	min_count = h->resv_huge_pages + h->nr_huge_pages - h->free_huge_pages;
>   +	if (h->free_huge_pages > persistent_huge_pages(h))
>   +		substract = h->free_huge_pages - h->surplus_huge_pages;

h->free_huge_pages may be less than h->surplus_huge_pages

>   +	else
>   +		h->free_huge_pages;
>   +
>   +	min_count = h->resv_huge_pages + persistent_huge_pages(h) - substract;
>    	min_count = max(count, min_count);
>    	try_to_free_low(h, min_count, nodes_allowed);
>
I tries the following diff, and the problem is solved.

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 39f92aad7bd1..4afcd5ce417c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3826,6 +3826,7 @@ static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
                               nodemask_t *nodes_allowed)
  {
         unsigned long min_count;
+       unsigned long persistent_free_count;
         unsigned long allocated;
         struct folio *folio;
         LIST_HEAD(page_list);
@@ -3960,7 +3961,14 @@ static int set_max_huge_pages(struct hstate *h, unsigned long count, int nid,
          * and won't grow the pool anywhere else. Not until one of the
          * sysctls are changed, or the surplus pages go out of use.
          */
-       min_count = h->resv_huge_pages + h->nr_huge_pages - h->free_huge_pages;
+       persistent_free_count = h->free_huge_pages;
+       if (h->free_huge_pages > persistent_huge_pages(h)) {
+               if (h->free_huge_pages > h->surplus_huge_pages)
+                       persistent_free_count -= h->surplus_huge_pages;
+               else
+                       persistent_free_count = 0;
+       }
+       min_count = h->resv_huge_pages + persistent_huge_pages(h) - persistent_free_count;
         min_count = max(count, min_count);
         try_to_free_low(h, min_count, nodes_allowed);

Thanks for reviewing.