Re: [PATCH v2 1/2] mm: Make alloc_contig_range handle free hugetlb pages

[Mike Kravetz <mike.kravetz@oracle.com>]

On 2/18/21 4:00 AM, Oscar Salvador wrote:
> alloc_contig_range will fail if it ever sees a HugeTLB page within the
> range we are trying to allocate, even when that page is free and can be
> easily reallocated.
> This has proofed to be problematic for some users of alloc_contic_range,
> e.g: CMA and virtio-mem, where those would fail the call even when those
> pages lay in ZONE_MOVABLE and are free.
> 
> We can do better by trying to dissolve such pages.
> 
> Free hugepages are tricky to handle so as to no userspace application
> notices disruption, we need to replace the current free hugepage with
> a new one.
> 
> In order to do that, a new function called alloc_and_dissolve_huge_page
> is introduced.
> This function will first try to get a new fresh hugepage, and if it
> succeeds, it will dissolve the old one.
> 
> If the old hugepage cannot be be dissolved, we have to dissolve the new
> hugepage we just got.
> Should that fail as well, we count is as a surplus, so the pool will be
> re-balanced when a hugepage gets free instead of enqueues again.
> 
> With regard to the allocation, we restrict it to the node the page belongs
> to with __GFP_THISNODE, meaning we do not fallback on other node's zones.
> 
> Note that gigantic hugetlb pages are fenced off since there is a cyclic
> dependency between them and alloc_contig_range.
> 
> Signed-off-by: Oscar Salvador <osalvador@suse.de>
> ---
>  include/linux/hugetlb.h |  6 ++++
>  mm/compaction.c         | 12 ++++++++
>  mm/hugetlb.c            | 75 +++++++++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 93 insertions(+)
> 
> diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
> index b5807f23caf8..72352d718829 100644
> --- a/include/linux/hugetlb.h
> +++ b/include/linux/hugetlb.h
> @@ -505,6 +505,7 @@ struct huge_bootmem_page {
>  	struct hstate *hstate;
>  };
>  
> +bool isolate_or_dissolve_huge_page(struct page *page);
>  struct page *alloc_huge_page(struct vm_area_struct *vma,
>  				unsigned long addr, int avoid_reserve);
>  struct page *alloc_huge_page_nodemask(struct hstate *h, int preferred_nid,
> @@ -775,6 +776,11 @@ void set_page_huge_active(struct page *page);
>  #else	/* CONFIG_HUGETLB_PAGE */
>  struct hstate {};
>  
> +static inline bool isolate_or_dissolve_huge_page(struct page *page)
> +{
> +	return false;
> +}
> +
>  static inline struct page *alloc_huge_page(struct vm_area_struct *vma,
>  					   unsigned long addr,
>  					   int avoid_reserve)
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 190ccdaa6c19..d52506ed9db7 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -905,6 +905,18 @@ isolate_migratepages_block(struct compact_control *cc, unsigned long low_pfn,
>  			valid_page = page;
>  		}
>  
> +		if (PageHuge(page) && cc->alloc_contig) {
> +			if (!isolate_or_dissolve_huge_page(page))
> +				goto isolate_fail;
> +
> +			/*
> +			 * Ok, the hugepage was dissolved. Now these pages are
> +			 * Buddy and cannot be re-allocated because they are
> +			 * isolated. Fall-through as the check below handles
> +			 * Buddy pages.
> +			 */
> +		}
> +
>  		/*
>  		 * Skip if free. We read page order here without zone lock
>  		 * which is generally unsafe, but the race window is small and
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 4bdb58ab14cb..a4fbbe924a55 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -2294,6 +2294,81 @@ static void restore_reserve_on_error(struct hstate *h,
>  	}
>  }
>  
> +static bool alloc_and_dissolve_huge_page(struct hstate *h, struct page *page)
> +{
> +	gfp_t gfp_mask = htlb_alloc_mask(h) | __GFP_THISNODE;
> +	int nid = page_to_nid(page);
> +	struct page *new_page;
> +	bool ret = false;
> +
> +	/*
> +	 * Before dissolving the page, we need to allocate a new one,
> +	 * so the pool remains stable.
> +	 */
> +	new_page = alloc_fresh_huge_page(h, gfp_mask, nid, NULL, NULL);
> +	if (new_page) {
> +		/*
> +		 * Free it into the hugepage allocator
> +		 */
> +		put_page(new_page);
> +

Suppose an admin does

echo 0 > \
/sys/devices/system/node/node<nid>/hugepages/hugepages-2048kB/nr_hugepages

right now and dissolves both the original and new page.

> +		/*
> +		 * Ok, we got a new free hugepage to replace this one. Try to
> +		 * dissolve the old page.
> +		 */
> +		if (!dissolve_free_huge_page(page)) {
> +			ret = true;

dissolve_free_huge_page will fail for the original page

> +		} else if (dissolve_free_huge_page(new_page)) {

and, will fail for the new page

> +			/*
> +			 * Seems the old page could not be dissolved, so try to
> +			 * dissolve the freshly allocated page. If that fails
> +			 * too, let us count the new page as a surplus. Doing so
> +			 * allows the pool to be re-balanced when pages are freed
> +			 * instead of enqueued again.
> +			 */
> +			spin_lock(&hugetlb_lock);
> +			h->surplus_huge_pages++;
> +			h->surplus_huge_pages_node[nid]++;
> +			spin_unlock(&hugetlb_lock);

Those counts will be wrong as there are no huge pages on the node.

I'll think about this more tomorrow.
Pretty sure this is an issue, but I could be wrong.  Just wanted to give
a heads up.
-- 
Mike Kravetz

> +		}
> +	}
> +
> +	return ret;
> +}
> +
> +bool isolate_or_dissolve_huge_page(struct page *page)
> +{
> +	struct hstate *h = NULL;
> +	struct page *head;
> +	bool ret = false;
> +
> +	spin_lock(&hugetlb_lock);
> +	if (PageHuge(page)) {
> +		head = compound_head(page);
> +		h = page_hstate(head);
> +	}
> +	spin_unlock(&hugetlb_lock);
> +
> +	/*
> +	 * The page might have been dissolved from under our feet.
> +	 * If that is the case, return success as if we dissolved it ourselves.
> +	 */
> +	if (!h)
> +		return true;
> +
> +	/*
> +	 * Fence off gigantic pages as there is a cyclic dependency
> +	 * between alloc_contig_range and them.
> +	 */
> +	if (hstate_is_gigantic(h))
> +		return ret;
> +
> +	if(!page_count(head) && alloc_and_dissolve_huge_page(h, head))
> +		ret = true;
> +
> +	return ret;
> +}
> +
>  struct page *alloc_huge_page(struct vm_area_struct *vma,
>  				    unsigned long addr, int avoid_reserve)
>  {
>