mm: split underused THPs

This is an attempt to mitigate the issue of running out of memory when THP
is always enabled.  During runtime whenever a THP is being faulted in
(__do_huge_pmd_anonymous_page) or collapsed by khugepaged
(collapse_huge_page), the THP is added to _deferred_list.  Whenever memory
reclaim happens in linux, the kernel runs the deferred_split shrinker
which goes through the _deferred_list.

If the folio was partially mapped, the shrinker attempts to split it.  If
the folio is not partially mapped, the shrinker checks if the THP was
underused, i.e.  how many of the base 4K pages of the entire THP were
zero-filled.  If this number goes above a certain threshold (decided by
/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none), the
shrinker will attempt to split that THP.  Then at remap time, the pages
that were zero-filled are mapped to the shared zeropage, hence saving
memory.

Link: https://lkml.kernel.org/r/20240830100438.3623486-6-usamaarif642@gmail.com
Signed-off-by: Usama Arif <usamaarif642@gmail.com>
Suggested-by: Rik van Riel <riel@surriel.com>
Co-authored-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Alexander Zhu <alexlzhu@fb.com>
Cc: Barry Song <baohua@kernel.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Domenico Cerasuolo <cerasuolodomenico@gmail.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Kairui Song <ryncsn@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Mike Rapoport <rppt@kernel.org>
Cc: Nico Pache <npache@redhat.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: Shuang Zhai <zhais@google.com>
Cc: Yu Zhao <yuzhao@google.com>
Cc: Shuang Zhai <szhai2@cs.rochester.edu>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

6 files changed, 69 insertions, 3 deletions

diff --git a/Documentation/admin-guide/mm/transhuge.rst b/Documentation/admin-guide/mm/transhuge.rst
index 56a086900651..aca0cff852b8 100644
--- a/Documentation/admin-guide/mm/transhuge.rst
+++ b/Documentation/admin-guide/mm/transhuge.rst
@@ -471,6 +471,12 @@ thp_deferred_split_page
 	splitting it would free up some memory. Pages on split queue are
 	going to be split under memory pressure.
 
+thp_underused_split_page
+	is incremented when a huge page on the split queue was split
+	because it was underused. A THP is underused if the number of
+	zero pages in the THP is above a certain threshold
+	(/sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none).
+
 thp_split_pmd
 	is incremented every time a PMD split into table of PTEs.
 	This can happen, for instance, when application calls mprotect() or
diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
index f68865e19b0b..30baae91b225 100644
--- a/include/linux/khugepaged.h
+++ b/include/linux/khugepaged.h
@@ -4,6 +4,7 @@
 
 #include <linux/sched/coredump.h> /* MMF_VM_HUGEPAGE */
 
+extern unsigned int khugepaged_max_ptes_none __read_mostly;
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 extern struct attribute_group khugepaged_attr_group;
 
diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h
index aae5c7c5cfb4..aed952d04132 100644
--- a/include/linux/vm_event_item.h
+++ b/include/linux/vm_event_item.h
@@ -105,6 +105,7 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
 		THP_SPLIT_PAGE,
 		THP_SPLIT_PAGE_FAILED,
 		THP_DEFERRED_SPLIT_PAGE,
+		THP_UNDERUSED_SPLIT_PAGE,
 		THP_SPLIT_PMD,
 		THP_SCAN_EXCEED_NONE_PTE,
 		THP_SCAN_EXCEED_SWAP_PTE,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 9bc435bef2e6..cec5bce046a0 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1187,6 +1187,7 @@ static vm_fault_t __do_huge_pmd_anonymous_page(struct vm_fault *vmf,
 		update_mmu_cache_pmd(vma, vmf->address, vmf->pmd);
 		add_mm_counter(vma->vm_mm, MM_ANONPAGES, HPAGE_PMD_NR);
 		mm_inc_nr_ptes(vma->vm_mm);
+		deferred_split_folio(folio, false);
 		spin_unlock(vmf->ptl);
 		count_vm_event(THP_FAULT_ALLOC);
 		count_mthp_stat(HPAGE_PMD_ORDER, MTHP_STAT_ANON_FAULT_ALLOC);
@@ -3608,6 +3609,39 @@ static unsigned long deferred_split_count(struct shrinker *shrink,
 	return READ_ONCE(ds_queue->split_queue_len);
 }
 
+static bool thp_underused(struct folio *folio)
+{
+	int num_zero_pages = 0, num_filled_pages = 0;
+	void *kaddr;
+	int i;
+
+	if (khugepaged_max_ptes_none == HPAGE_PMD_NR - 1)
+		return false;
+
+	for (i = 0; i < folio_nr_pages(folio); i++) {
+		kaddr = kmap_local_folio(folio, i * PAGE_SIZE);
+		if (!memchr_inv(kaddr, 0, PAGE_SIZE)) {
+			num_zero_pages++;
+			if (num_zero_pages > khugepaged_max_ptes_none) {
+				kunmap_local(kaddr);
+				return true;
+			}
+		} else {
+			/*
+			 * Another path for early exit once the number
+			 * of non-zero filled pages exceeds threshold.
+			 */
+			num_filled_pages++;
+			if (num_filled_pages >= HPAGE_PMD_NR - khugepaged_max_ptes_none) {
+				kunmap_local(kaddr);
+				return false;
+			}
+		}
+		kunmap_local(kaddr);
+	}
+	return false;
+}
+
 static unsigned long deferred_split_scan(struct shrinker *shrink,
 		struct shrink_control *sc)
 {
@@ -3645,13 +3679,35 @@ static unsigned long deferred_split_scan(struct shrinker *shrink,
 	spin_unlock_irqrestore(&ds_queue->split_queue_lock, flags);
 
 	list_for_each_entry_safe(folio, next, &list, _deferred_list) {
+		bool did_split = false;
+		bool underused = false;
+
+		if (!folio_test_partially_mapped(folio)) {
+			underused = thp_underused(folio);
+			if (!underused)
+				goto next;
+		}
 		if (!folio_trylock(folio))
 			goto next;
-		/* split_huge_page() removes page from list on success */
-		if (!split_folio(folio))
+		if (!split_folio(folio)) {
+			did_split = true;
+			if (underused)
+				count_vm_event(THP_UNDERUSED_SPLIT_PAGE);
 			split++;
+		}
 		folio_unlock(folio);
 next:
+		/*
+		 * split_folio() removes folio from list on success.
+		 * Only add back to the queue if folio is partially mapped.
+		 * If thp_underused returns false, or if split_folio fails
+		 * in the case it was underused, then consider it used and
+		 * don't add it back to split_queue.
+		 */
+		if (!did_split && !folio_test_partially_mapped(folio)) {
+			list_del_init(&folio->_deferred_list);
+			ds_queue->split_queue_len--;
+		}
 		folio_put(folio);
 	}
 
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index ab646018ce25..32100041aef3 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -85,7 +85,7 @@ static DECLARE_WAIT_QUEUE_HEAD(khugepaged_wait);
  *
  * Note that these are only respected if collapse was initiated by khugepaged.
  */
-static unsigned int khugepaged_max_ptes_none __read_mostly;
+unsigned int khugepaged_max_ptes_none __read_mostly;
 static unsigned int khugepaged_max_ptes_swap __read_mostly;
 static unsigned int khugepaged_max_ptes_shared __read_mostly;
 
@@ -1237,6 +1237,7 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
 	pgtable_trans_huge_deposit(mm, pmd, pgtable);
 	set_pmd_at(mm, address, pmd, _pmd);
 	update_mmu_cache_pmd(vma, address, pmd);
+	deferred_split_folio(folio, false);
 	spin_unlock(pmd_ptl);
 
 	folio = NULL;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index aea58e9fce60..b5a4cea423e1 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1385,6 +1385,7 @@ const char * const vmstat_text[] = {
 	"thp_split_page",
 	"thp_split_page_failed",
 	"thp_deferred_split_page",
+	"thp_underused_split_page",
 	"thp_split_pmd",
 	"thp_scan_exceed_none_pte",
 	"thp_scan_exceed_swap_pte",