Re: [PATCH v4 2/3] mm: khugepaged: fix folio is used after pte_unmap_unlock()

["David Hildenbrand (Arm)" <david@kernel.org>]

On 8/28/26 07:59, Vernon Yang wrote:
> From: Vernon Yang <yanglincheng@kylinos.cn>
> 
> After the page table lock has dropped, the folio can be freed
> concurrently. The trace_mm_khugepaged_scan_pmd() is left with
> a dangling folio pointer.
> 
> So using the folio_pfn() before dropping the page table lock, closing
> use-after-free window.
> 
> And other pre-existing bug, When the `for (i = 0; i < HPAGE_PMD_NR; i++)`
> iteration to terminate and the folio operation preceding is normal, but
> pfn will be incorrect. so we really only trace the PFN if it really was
> problematic.
> 
> About calling the respective trace_xxx() functions separately on success
> and failure, refer to [1].
> 
> [1] https://lore.kernel.org/linux-mm/ao6jVbVHLUmuY2UA@gremlin/
> 
> Acked-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
> Fixes: 7d2eba0557c1 ("mm: add tracepoint for scanning pages")
> Cc: stable@vger.kernel.org
> Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
> ---
>  include/trace/events/huge_memory.h |  6 +++---
>  mm/khugepaged.c                    | 17 ++++++++++++++---
>  2 files changed, 17 insertions(+), 6 deletions(-)
> 
> diff --git a/include/trace/events/huge_memory.h b/include/trace/events/huge_memory.h
> index 7b526528f85b..fa828967e1fb 100644
> --- a/include/trace/events/huge_memory.h
> +++ b/include/trace/events/huge_memory.h
> @@ -55,10 +55,10 @@ SCAN_STATUS
>  
>  TRACE_EVENT(mm_khugepaged_scan_pmd,
>  
> -	TP_PROTO(struct mm_struct *mm, struct folio *folio,
> +	TP_PROTO(struct mm_struct *mm, unsigned long pfn,
>  		 int referenced, int none_or_zero, int status, int unmapped),
>  
> -	TP_ARGS(mm, folio, referenced, none_or_zero, status, unmapped),
> +	TP_ARGS(mm, pfn, referenced, none_or_zero, status, unmapped),
>  
>  	TP_STRUCT__entry(
>  		__field(struct mm_struct *, mm)
> @@ -71,7 +71,7 @@ TRACE_EVENT(mm_khugepaged_scan_pmd,
>  
>  	TP_fast_assign(
>  		__entry->mm = mm;
> -		__entry->pfn = folio ? folio_pfn(folio) : -1;
> +		__entry->pfn = pfn;
>  		__entry->referenced = referenced;
>  		__entry->none_or_zero = none_or_zero;
>  		__entry->status = status;
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index b597a3e68606..4d360ae87769 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -1612,6 +1612,7 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  	enum scan_result result = SCAN_FAIL;
>  	struct page *page = NULL;
>  	struct folio *folio = NULL;
> +	unsigned long failed_pfn = -1;
>  	unsigned long addr;
>  	unsigned long enabled_orders;
>  	spinlock_t *ptl;
> @@ -1706,11 +1707,13 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  		if (cc->is_khugepaged && !(vma->vm_flags & VM_DROPPABLE) &&
>  		    folio_test_lazyfree(folio) && !pte_dirty(pteval)) {
>  			result = SCAN_PAGE_LAZYFREE;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  
>  		if (!folio_test_anon(folio)) {
>  			result = SCAN_PAGE_ANON;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  
> @@ -1721,6 +1724,7 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  		if (folio_maybe_mapped_shared(folio)) {
>  			if (++shared > max_ptes_shared) {
>  				result = SCAN_EXCEED_SHARED_PTE;
> +				failed_pfn = folio_pfn(folio);
>  				count_collapse_event(HPAGE_PMD_ORDER, THP_SCAN_EXCEED_SHARED_PTE,
>  						     MTHP_STAT_COLLAPSE_EXCEED_SHARED);
>  				goto out_unmap;
> @@ -1738,15 +1742,18 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  		node = folio_nid(folio);
>  		if (collapse_scan_abort(node, cc)) {
>  			result = SCAN_SCAN_ABORT;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  		cc->node_load[node]++;
>  		if (!folio_test_lru(folio)) {
>  			result = SCAN_PAGE_LRU;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  		if (folio_test_locked(folio)) {
>  			result = SCAN_PAGE_LOCK;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  
> @@ -1759,6 +1766,7 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  		 */
>  		if (folio_expected_ref_count(folio) != folio_ref_count(folio)) {
>  			result = SCAN_PAGE_COUNT;
> +			failed_pfn = folio_pfn(folio);
>  			goto out_unmap;
>  		}
>  
> @@ -1782,10 +1790,13 @@ static enum scan_result collapse_scan_pmd(struct mm_struct *mm,
>  				       unmapped, cc, enabled_orders);
>  		/* mmap_lock was released above, set lock_dropped */
>  		*lock_dropped = true;
> -	}
> +		trace_mm_khugepaged_scan_pmd(mm, -1, referenced, none_or_zero,
> +					     SCAN_SUCCEED, unmapped);
> +	} else {
>  out:
> -	trace_mm_khugepaged_scan_pmd(mm, folio, referenced,
> -				     none_or_zero, result, unmapped);
> +		trace_mm_khugepaged_scan_pmd(mm, failed_pfn, referenced,
> +					     none_or_zero, result, unmapped);
> +	}
>  	return result;
>  }
>  

Same thought

Acked-by: David Hildenbrand (Arm) <david@kernel.org>


-- 
Cheers,

David