[PATCH mm-unstable v4 0/5] mm/mglru: Kconfig cleanup

[PATCH mm-unstable v4 0/5] mm/mglru: Kconfig cleanup

[Kinsey Ho]

This series is the result of the following discussion:
https://lore.kernel.org/47066176-bd93-55dd-c2fa-002299d9e034@linux.ibm.com/

It mainly avoids building the code that walks page tables on CPUs that
use it, i.e., those don't support hardware accessed bit. Specifically,
it introduces a new Kconfig to guard some of functions added by
commit bd74fdaea146 ("mm: multi-gen LRU: support page table walks")
on CPUs like POWER9, on which the series was tested.
---
v4: fix kernel build bot error by adding dummy pmd_dirty()

v3: https://lore.kernel.org/20231220040037.883811-1-kinseyho@google.com/


Kinsey Ho (5):
  mm/mglru: add CONFIG_ARCH_HAS_HW_PTE_YOUNG
  mm/mglru: add CONFIG_LRU_GEN_WALKS_MMU
  mm/mglru: remove CONFIG_MEMCG
  mm/mglru: add dummy pmd_dirty()
  mm/mglru: remove CONFIG_TRANSPARENT_HUGEPAGE

 arch/Kconfig                         |   8 +
 arch/arm64/Kconfig                   |   1 +
 arch/loongarch/include/asm/pgtable.h |   1 +
 arch/mips/include/asm/pgtable.h      |   1 +
 arch/riscv/include/asm/pgtable.h     |   1 +
 arch/s390/include/asm/pgtable.h      |   1 +
 arch/sparc/include/asm/pgtable_64.h  |   1 +
 arch/x86/Kconfig                     |   1 +
 arch/x86/include/asm/pgtable.h       |   7 +-
 include/linux/memcontrol.h           |   2 +-
 include/linux/mm_types.h             |  16 +-
 include/linux/mmzone.h               |  28 +--
 include/linux/pgtable.h              |   9 +-
 kernel/fork.c                        |   2 +-
 mm/Kconfig                           |   4 +
 mm/vmscan.c                          | 271 ++++++++++++++-------------
 16 files changed, 187 insertions(+), 167 deletions(-)

-- 
2.43.0.472.g3155946c3a-goog

[PATCH mm-unstable v4 1/5] mm/mglru: add CONFIG_ARCH_HAS_HW_PTE_YOUNG

[Kinsey Ho]

Some architectures are able to set the accessed bit in PTEs when PTEs
are used as part of linear address translations.

Add CONFIG_ARCH_HAS_HW_PTE_YOUNG for such architectures to be able to
override arch_has_hw_pte_young().

Signed-off-by: Kinsey Ho <kinseyho@google.com>
Co-developed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Tested-by: Donet Tom <donettom@linux.vnet.ibm.com>
Acked-by: Yu Zhao <yuzhao@google.com>
---
 arch/Kconfig                   | 8 ++++++++
 arch/arm64/Kconfig             | 1 +
 arch/x86/Kconfig               | 1 +
 arch/x86/include/asm/pgtable.h | 6 ------
 include/linux/pgtable.h        | 2 +-
 5 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/arch/Kconfig b/arch/Kconfig
index f4b210ab0612..8c8901f80586 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -1470,6 +1470,14 @@ config DYNAMIC_SIGFRAME
 config HAVE_ARCH_NODE_DEV_GROUP
 	bool
 
+config ARCH_HAS_HW_PTE_YOUNG
+	bool
+	help
+	  Architectures that select this option are capable of setting the
+	  accessed bit in PTE entries when using them as part of linear address
+	  translations. Architectures that require runtime check should select
+	  this option and override arch_has_hw_pte_young().
+
 config ARCH_HAS_NONLEAF_PMD_YOUNG
 	bool
 	help
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 7b071a00425d..12d611f3da5d 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -36,6 +36,7 @@ config ARM64
 	select ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
 	select ARCH_HAS_PTE_DEVMAP
 	select ARCH_HAS_PTE_SPECIAL
+	select ARCH_HAS_HW_PTE_YOUNG
 	select ARCH_HAS_SETUP_DMA_OPS
 	select ARCH_HAS_SET_DIRECT_MAP
 	select ARCH_HAS_SET_MEMORY
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 1566748f16c4..04941a1ffc0a 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -88,6 +88,7 @@ config X86
 	select ARCH_HAS_PMEM_API		if X86_64
 	select ARCH_HAS_PTE_DEVMAP		if X86_64
 	select ARCH_HAS_PTE_SPECIAL
+	select ARCH_HAS_HW_PTE_YOUNG
 	select ARCH_HAS_NONLEAF_PMD_YOUNG	if PGTABLE_LEVELS > 2
 	select ARCH_HAS_UACCESS_FLUSHCACHE	if X86_64
 	select ARCH_HAS_COPY_MC			if X86_64
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 57bab91bbf50..08b5cb22d9a6 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -1679,12 +1679,6 @@ static inline bool arch_has_pfn_modify_check(void)
 	return boot_cpu_has_bug(X86_BUG_L1TF);
 }
 
-#define arch_has_hw_pte_young arch_has_hw_pte_young
-static inline bool arch_has_hw_pte_young(void)
-{
-	return true;
-}
-
 #define arch_check_zapped_pte arch_check_zapped_pte
 void arch_check_zapped_pte(struct vm_area_struct *vma, pte_t pte);
 
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index af7639c3b0a3..9ecc20fa6269 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -375,7 +375,7 @@ static inline bool arch_has_hw_nonleaf_pmd_young(void)
  */
 static inline bool arch_has_hw_pte_young(void)
 {
-	return false;
+	return IS_ENABLED(CONFIG_ARCH_HAS_HW_PTE_YOUNG);
 }
 #endif
 
-- 
2.43.0.472.g3155946c3a-goog

[PATCH mm-unstable v4 2/5] mm/mglru: add CONFIG_LRU_GEN_WALKS_MMU

[Kinsey Ho]

Add CONFIG_LRU_GEN_WALKS_MMU such that if disabled, the code that
walks page tables to promote pages into the youngest generation will
not be built.

Also improves code readability by adding two helper functions
get_mm_state() and get_next_mm().

Signed-off-by: Kinsey Ho <kinseyho@google.com>
Co-developed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Tested-by: Donet Tom <donettom@linux.vnet.ibm.com>
Acked-by: Yu Zhao <yuzhao@google.com>
---
 include/linux/memcontrol.h |   2 +-
 include/linux/mm_types.h   |  12 ++-
 include/linux/mmzone.h     |   2 +
 kernel/fork.c              |   2 +-
 mm/Kconfig                 |   4 +
 mm/vmscan.c                | 192 ++++++++++++++++++++++++-------------
 6 files changed, 139 insertions(+), 75 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 5de775e6cdd9..20ff87f8e001 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -330,7 +330,7 @@ struct mem_cgroup {
 	struct deferred_split deferred_split_queue;
 #endif
 
-#ifdef CONFIG_LRU_GEN
+#ifdef CONFIG_LRU_GEN_WALKS_MMU
 	/* per-memcg mm_struct list */
 	struct lru_gen_mm_list mm_list;
 #endif
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index a66534c78c4d..552fa2d11c57 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -958,7 +958,7 @@ struct mm_struct {
 		 */
 		unsigned long ksm_zero_pages;
 #endif /* CONFIG_KSM */
-#ifdef CONFIG_LRU_GEN
+#ifdef CONFIG_LRU_GEN_WALKS_MMU
 		struct {
 			/* this mm_struct is on lru_gen_mm_list */
 			struct list_head list;
@@ -973,7 +973,7 @@ struct mm_struct {
 			struct mem_cgroup *memcg;
 #endif
 		} lru_gen;
-#endif /* CONFIG_LRU_GEN */
+#endif /* CONFIG_LRU_GEN_WALKS_MMU */
 	} __randomize_layout;
 
 	/*
@@ -1011,6 +1011,10 @@ struct lru_gen_mm_list {
 	spinlock_t lock;
 };
 
+#endif /* CONFIG_LRU_GEN */
+
+#ifdef CONFIG_LRU_GEN_WALKS_MMU
+
 void lru_gen_add_mm(struct mm_struct *mm);
 void lru_gen_del_mm(struct mm_struct *mm);
 #ifdef CONFIG_MEMCG
@@ -1036,7 +1040,7 @@ static inline void lru_gen_use_mm(struct mm_struct *mm)
 	WRITE_ONCE(mm->lru_gen.bitmap, -1);
 }
 
-#else /* !CONFIG_LRU_GEN */
+#else /* !CONFIG_LRU_GEN_WALKS_MMU */
 
 static inline void lru_gen_add_mm(struct mm_struct *mm)
 {
@@ -1060,7 +1064,7 @@ static inline void lru_gen_use_mm(struct mm_struct *mm)
 {
 }
 
-#endif /* CONFIG_LRU_GEN */
+#endif /* CONFIG_LRU_GEN_WALKS_MMU */
 
 struct vma_iterator {
 	struct ma_state mas;
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 2efd3be484fd..bc3f63ec4291 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -640,9 +640,11 @@ struct lruvec {
 #ifdef CONFIG_LRU_GEN
 	/* evictable pages divided into generations */
 	struct lru_gen_folio		lrugen;
+#ifdef CONFIG_LRU_GEN_WALKS_MMU
 	/* to concurrently iterate lru_gen_mm_list */
 	struct lru_gen_mm_state		mm_state;
 #endif
+#endif /* CONFIG_LRU_GEN */
 #ifdef CONFIG_MEMCG
 	struct pglist_data *pgdat;
 #endif
diff --git a/kernel/fork.c b/kernel/fork.c
index 93924392a5c3..56cf276432c8 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -2946,7 +2946,7 @@ pid_t kernel_clone(struct kernel_clone_args *args)
 		get_task_struct(p);
 	}
 
-	if (IS_ENABLED(CONFIG_LRU_GEN) && !(clone_flags & CLONE_VM)) {
+	if (IS_ENABLED(CONFIG_LRU_GEN_WALKS_MMU) && !(clone_flags & CLONE_VM)) {
 		/* lock the task to synchronize with memcg migration */
 		task_lock(p);
 		lru_gen_add_mm(p->mm);
diff --git a/mm/Kconfig b/mm/Kconfig
index 8f8b02e9c136..c98076dec5fb 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -1274,6 +1274,10 @@ config LRU_GEN_STATS
 	  from evicted generations for debugging purpose.
 
 	  This option has a per-memcg and per-node memory overhead.
+
+config LRU_GEN_WALKS_MMU
+	def_bool y
+	depends on LRU_GEN && ARCH_HAS_HW_PTE_YOUNG
 # }
 
 config ARCH_SUPPORTS_PER_VMA_LOCK
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b4ca3563bcf4..aa7ea09ffb4c 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2671,13 +2671,14 @@ static void get_item_key(void *item, int *key)
 	key[1] = hash >> BLOOM_FILTER_SHIFT;
 }
 
-static bool test_bloom_filter(struct lruvec *lruvec, unsigned long seq, void *item)
+static bool test_bloom_filter(struct lru_gen_mm_state *mm_state, unsigned long seq,
+			      void *item)
 {
 	int key[2];
 	unsigned long *filter;
 	int gen = filter_gen_from_seq(seq);
 
-	filter = READ_ONCE(lruvec->mm_state.filters[gen]);
+	filter = READ_ONCE(mm_state->filters[gen]);
 	if (!filter)
 		return true;
 
@@ -2686,13 +2687,14 @@ static bool test_bloom_filter(struct lruvec *lruvec, unsigned long seq, void *it
 	return test_bit(key[0], filter) && test_bit(key[1], filter);
 }
 
-static void update_bloom_filter(struct lruvec *lruvec, unsigned long seq, void *item)
+static void update_bloom_filter(struct lru_gen_mm_state *mm_state, unsigned long seq,
+				void *item)
 {
 	int key[2];
 	unsigned long *filter;
 	int gen = filter_gen_from_seq(seq);
 
-	filter = READ_ONCE(lruvec->mm_state.filters[gen]);
+	filter = READ_ONCE(mm_state->filters[gen]);
 	if (!filter)
 		return;
 
@@ -2704,12 +2706,12 @@ static void update_bloom_filter(struct lruvec *lruvec, unsigned long seq, void *
 		set_bit(key[1], filter);
 }
 
-static void reset_bloom_filter(struct lruvec *lruvec, unsigned long seq)
+static void reset_bloom_filter(struct lru_gen_mm_state *mm_state, unsigned long seq)
 {
 	unsigned long *filter;
 	int gen = filter_gen_from_seq(seq);
 
-	filter = lruvec->mm_state.filters[gen];
+	filter = mm_state->filters[gen];
 	if (filter) {
 		bitmap_clear(filter, 0, BIT(BLOOM_FILTER_SHIFT));
 		return;
@@ -2717,13 +2719,15 @@ static void reset_bloom_filter(struct lruvec *lruvec, unsigned long seq)
 
 	filter = bitmap_zalloc(BIT(BLOOM_FILTER_SHIFT),
 			       __GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN);
-	WRITE_ONCE(lruvec->mm_state.filters[gen], filter);
+	WRITE_ONCE(mm_state->filters[gen], filter);
 }
 
 /******************************************************************************
  *                          mm_struct list
  ******************************************************************************/
 
+#ifdef CONFIG_LRU_GEN_WALKS_MMU
+
 static struct lru_gen_mm_list *get_mm_list(struct mem_cgroup *memcg)
 {
 	static struct lru_gen_mm_list mm_list = {
@@ -2740,6 +2744,29 @@ static struct lru_gen_mm_list *get_mm_list(struct mem_cgroup *memcg)
 	return &mm_list;
 }
 
+static struct lru_gen_mm_state *get_mm_state(struct lruvec *lruvec)
+{
+	return &lruvec->mm_state;
+}
+
+static struct mm_struct *get_next_mm(struct lru_gen_mm_walk *walk)
+{
+	int key;
+	struct mm_struct *mm;
+	struct pglist_data *pgdat = lruvec_pgdat(walk->lruvec);
+	struct lru_gen_mm_state *mm_state = get_mm_state(walk->lruvec);
+
+	mm = list_entry(mm_state->head, struct mm_struct, lru_gen.list);
+	key = pgdat->node_id % BITS_PER_TYPE(mm->lru_gen.bitmap);
+
+	if (!walk->force_scan && !test_bit(key, &mm->lru_gen.bitmap))
+		return NULL;
+
+	clear_bit(key, &mm->lru_gen.bitmap);
+
+	return mmget_not_zero(mm) ? mm : NULL;
+}
+
 void lru_gen_add_mm(struct mm_struct *mm)
 {
 	int nid;
@@ -2755,10 +2782,11 @@ void lru_gen_add_mm(struct mm_struct *mm)
 
 	for_each_node_state(nid, N_MEMORY) {
 		struct lruvec *lruvec = get_lruvec(memcg, nid);
+		struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 		/* the first addition since the last iteration */
-		if (lruvec->mm_state.tail == &mm_list->fifo)
-			lruvec->mm_state.tail = &mm->lru_gen.list;
+		if (mm_state->tail == &mm_list->fifo)
+			mm_state->tail = &mm->lru_gen.list;
 	}
 
 	list_add_tail(&mm->lru_gen.list, &mm_list->fifo);
@@ -2784,14 +2812,15 @@ void lru_gen_del_mm(struct mm_struct *mm)
 
 	for_each_node(nid) {
 		struct lruvec *lruvec = get_lruvec(memcg, nid);
+		struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 		/* where the current iteration continues after */
-		if (lruvec->mm_state.head == &mm->lru_gen.list)
-			lruvec->mm_state.head = lruvec->mm_state.head->prev;
+		if (mm_state->head == &mm->lru_gen.list)
+			mm_state->head = mm_state->head->prev;
 
 		/* where the last iteration ended before */
-		if (lruvec->mm_state.tail == &mm->lru_gen.list)
-			lruvec->mm_state.tail = lruvec->mm_state.tail->next;
+		if (mm_state->tail == &mm->lru_gen.list)
+			mm_state->tail = mm_state->tail->next;
 	}
 
 	list_del_init(&mm->lru_gen.list);
@@ -2834,10 +2863,30 @@ void lru_gen_migrate_mm(struct mm_struct *mm)
 }
 #endif
 
+#else /* !CONFIG_LRU_GEN_WALKS_MMU */
+
+static struct lru_gen_mm_list *get_mm_list(struct mem_cgroup *memcg)
+{
+	return NULL;
+}
+
+static struct lru_gen_mm_state *get_mm_state(struct lruvec *lruvec)
+{
+	return NULL;
+}
+
+static struct mm_struct *get_next_mm(struct lru_gen_mm_walk *walk)
+{
+	return NULL;
+}
+
+#endif
+
 static void reset_mm_stats(struct lruvec *lruvec, struct lru_gen_mm_walk *walk, bool last)
 {
 	int i;
 	int hist;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	lockdep_assert_held(&get_mm_list(lruvec_memcg(lruvec))->lock);
 
@@ -2845,44 +2894,20 @@ static void reset_mm_stats(struct lruvec *lruvec, struct lru_gen_mm_walk *walk,
 		hist = lru_hist_from_seq(walk->max_seq);
 
 		for (i = 0; i < NR_MM_STATS; i++) {
-			WRITE_ONCE(lruvec->mm_state.stats[hist][i],
-				   lruvec->mm_state.stats[hist][i] + walk->mm_stats[i]);
+			WRITE_ONCE(mm_state->stats[hist][i],
+				   mm_state->stats[hist][i] + walk->mm_stats[i]);
 			walk->mm_stats[i] = 0;
 		}
 	}
 
 	if (NR_HIST_GENS > 1 && last) {
-		hist = lru_hist_from_seq(lruvec->mm_state.seq + 1);
+		hist = lru_hist_from_seq(mm_state->seq + 1);
 
 		for (i = 0; i < NR_MM_STATS; i++)
-			WRITE_ONCE(lruvec->mm_state.stats[hist][i], 0);
+			WRITE_ONCE(mm_state->stats[hist][i], 0);
 	}
 }
 
-static bool should_skip_mm(struct mm_struct *mm, struct lru_gen_mm_walk *walk)
-{
-	int type;
-	unsigned long size = 0;
-	struct pglist_data *pgdat = lruvec_pgdat(walk->lruvec);
-	int key = pgdat->node_id % BITS_PER_TYPE(mm->lru_gen.bitmap);
-
-	if (!walk->force_scan && !test_bit(key, &mm->lru_gen.bitmap))
-		return true;
-
-	clear_bit(key, &mm->lru_gen.bitmap);
-
-	for (type = !walk->can_swap; type < ANON_AND_FILE; type++) {
-		size += type ? get_mm_counter(mm, MM_FILEPAGES) :
-			       get_mm_counter(mm, MM_ANONPAGES) +
-			       get_mm_counter(mm, MM_SHMEMPAGES);
-	}
-
-	if (size < MIN_LRU_BATCH)
-		return true;
-
-	return !mmget_not_zero(mm);
-}
-
 static bool iterate_mm_list(struct lruvec *lruvec, struct lru_gen_mm_walk *walk,
 			    struct mm_struct **iter)
 {
@@ -2891,7 +2916,7 @@ static bool iterate_mm_list(struct lruvec *lruvec, struct lru_gen_mm_walk *walk,
 	struct mm_struct *mm = NULL;
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	struct lru_gen_mm_list *mm_list = get_mm_list(memcg);
-	struct lru_gen_mm_state *mm_state = &lruvec->mm_state;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	/*
 	 * mm_state->seq is incremented after each iteration of mm_list. There
@@ -2929,11 +2954,7 @@ static bool iterate_mm_list(struct lruvec *lruvec, struct lru_gen_mm_walk *walk,
 			mm_state->tail = mm_state->head->next;
 			walk->force_scan = true;
 		}
-
-		mm = list_entry(mm_state->head, struct mm_struct, lru_gen.list);
-		if (should_skip_mm(mm, walk))
-			mm = NULL;
-	} while (!mm);
+	} while (!(mm = get_next_mm(walk)));
 done:
 	if (*iter || last)
 		reset_mm_stats(lruvec, walk, last);
@@ -2941,7 +2962,7 @@ static bool iterate_mm_list(struct lruvec *lruvec, struct lru_gen_mm_walk *walk,
 	spin_unlock(&mm_list->lock);
 
 	if (mm && first)
-		reset_bloom_filter(lruvec, walk->max_seq + 1);
+		reset_bloom_filter(mm_state, walk->max_seq + 1);
 
 	if (*iter)
 		mmput_async(*iter);
@@ -2956,7 +2977,7 @@ static bool iterate_mm_list_nowalk(struct lruvec *lruvec, unsigned long max_seq)
 	bool success = false;
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	struct lru_gen_mm_list *mm_list = get_mm_list(memcg);
-	struct lru_gen_mm_state *mm_state = &lruvec->mm_state;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	spin_lock(&mm_list->lock);
 
@@ -3469,6 +3490,7 @@ static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 	DECLARE_BITMAP(bitmap, MIN_LRU_BATCH);
 	unsigned long first = -1;
 	struct lru_gen_mm_walk *walk = args->private;
+	struct lru_gen_mm_state *mm_state = get_mm_state(walk->lruvec);
 
 	VM_WARN_ON_ONCE(pud_leaf(*pud));
 
@@ -3520,7 +3542,7 @@ static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 			walk_pmd_range_locked(pud, addr, vma, args, bitmap, &first);
 		}
 
-		if (!walk->force_scan && !test_bloom_filter(walk->lruvec, walk->max_seq, pmd + i))
+		if (!walk->force_scan && !test_bloom_filter(mm_state, walk->max_seq, pmd + i))
 			continue;
 
 		walk->mm_stats[MM_NONLEAF_FOUND]++;
@@ -3531,7 +3553,7 @@ static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 		walk->mm_stats[MM_NONLEAF_ADDED]++;
 
 		/* carry over to the next generation */
-		update_bloom_filter(walk->lruvec, walk->max_seq + 1, pmd + i);
+		update_bloom_filter(mm_state, walk->max_seq + 1, pmd + i);
 	}
 
 	walk_pmd_range_locked(pud, -1, vma, args, bitmap, &first);
@@ -3738,16 +3760,25 @@ static bool try_to_inc_min_seq(struct lruvec *lruvec, bool can_swap)
 	return success;
 }
 
-static void inc_max_seq(struct lruvec *lruvec, bool can_swap, bool force_scan)
+static bool inc_max_seq(struct lruvec *lruvec, unsigned long max_seq,
+			bool can_swap, bool force_scan)
 {
+	bool success;
 	int prev, next;
 	int type, zone;
 	struct lru_gen_folio *lrugen = &lruvec->lrugen;
 restart:
+	if (max_seq < READ_ONCE(lrugen->max_seq))
+		return false;
+
 	spin_lock_irq(&lruvec->lru_lock);
 
 	VM_WARN_ON_ONCE(!seq_is_valid(lruvec));
 
+	success = max_seq == lrugen->max_seq;
+	if (!success)
+		goto unlock;
+
 	for (type = ANON_AND_FILE - 1; type >= 0; type--) {
 		if (get_nr_gens(lruvec, type) != MAX_NR_GENS)
 			continue;
@@ -3791,8 +3822,10 @@ static void inc_max_seq(struct lruvec *lruvec, bool can_swap, bool force_scan)
 	WRITE_ONCE(lrugen->timestamps[next], jiffies);
 	/* make sure preceding modifications appear */
 	smp_store_release(&lrugen->max_seq, lrugen->max_seq + 1);
-
+unlock:
 	spin_unlock_irq(&lruvec->lru_lock);
+
+	return success;
 }
 
 static bool try_to_inc_max_seq(struct lruvec *lruvec, unsigned long max_seq,
@@ -3802,14 +3835,16 @@ static bool try_to_inc_max_seq(struct lruvec *lruvec, unsigned long max_seq,
 	struct lru_gen_mm_walk *walk;
 	struct mm_struct *mm = NULL;
 	struct lru_gen_folio *lrugen = &lruvec->lrugen;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	VM_WARN_ON_ONCE(max_seq > READ_ONCE(lrugen->max_seq));
 
+	if (!mm_state)
+		return inc_max_seq(lruvec, max_seq, can_swap, force_scan);
+
 	/* see the comment in iterate_mm_list() */
-	if (max_seq <= READ_ONCE(lruvec->mm_state.seq)) {
-		success = false;
-		goto done;
-	}
+	if (max_seq <= READ_ONCE(mm_state->seq))
+		return false;
 
 	/*
 	 * If the hardware doesn't automatically set the accessed bit, fallback
@@ -3839,8 +3874,10 @@ static bool try_to_inc_max_seq(struct lruvec *lruvec, unsigned long max_seq,
 			walk_mm(lruvec, mm, walk);
 	} while (mm);
 done:
-	if (success)
-		inc_max_seq(lruvec, can_swap, force_scan);
+	if (success) {
+		success = inc_max_seq(lruvec, max_seq, can_swap, force_scan);
+		WARN_ON_ONCE(!success);
+	}
 
 	return success;
 }
@@ -3964,6 +4001,7 @@ void lru_gen_look_around(struct page_vma_mapped_walk *pvmw)
 	struct mem_cgroup *memcg = folio_memcg(folio);
 	struct pglist_data *pgdat = folio_pgdat(folio);
 	struct lruvec *lruvec = mem_cgroup_lruvec(memcg, pgdat);
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 	DEFINE_MAX_SEQ(lruvec);
 	int old_gen, new_gen = lru_gen_from_seq(max_seq);
 
@@ -4042,8 +4080,8 @@ void lru_gen_look_around(struct page_vma_mapped_walk *pvmw)
 	mem_cgroup_unlock_pages();
 
 	/* feedback from rmap walkers to page table walkers */
-	if (suitable_to_scan(i, young))
-		update_bloom_filter(lruvec, max_seq, pvmw->pmd);
+	if (mm_state && suitable_to_scan(i, young))
+		update_bloom_filter(mm_state, max_seq, pvmw->pmd);
 }
 
 /******************************************************************************
@@ -5219,6 +5257,7 @@ static void lru_gen_seq_show_full(struct seq_file *m, struct lruvec *lruvec,
 	int type, tier;
 	int hist = lru_hist_from_seq(seq);
 	struct lru_gen_folio *lrugen = &lruvec->lrugen;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	for (tier = 0; tier < MAX_NR_TIERS; tier++) {
 		seq_printf(m, "            %10d", tier);
@@ -5244,6 +5283,9 @@ static void lru_gen_seq_show_full(struct seq_file *m, struct lruvec *lruvec,
 		seq_putc(m, '\n');
 	}
 
+	if (!mm_state)
+		return;
+
 	seq_puts(m, "                      ");
 	for (i = 0; i < NR_MM_STATS; i++) {
 		const char *s = "      ";
@@ -5251,10 +5293,10 @@ static void lru_gen_seq_show_full(struct seq_file *m, struct lruvec *lruvec,
 
 		if (seq == max_seq && NR_HIST_GENS == 1) {
 			s = "LOYNFA";
-			n = READ_ONCE(lruvec->mm_state.stats[hist][i]);
+			n = READ_ONCE(mm_state->stats[hist][i]);
 		} else if (seq != max_seq && NR_HIST_GENS > 1) {
 			s = "loynfa";
-			n = READ_ONCE(lruvec->mm_state.stats[hist][i]);
+			n = READ_ONCE(mm_state->stats[hist][i]);
 		}
 
 		seq_printf(m, " %10lu%c", n, s[i]);
@@ -5523,6 +5565,7 @@ void lru_gen_init_lruvec(struct lruvec *lruvec)
 	int i;
 	int gen, type, zone;
 	struct lru_gen_folio *lrugen = &lruvec->lrugen;
+	struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 	lrugen->max_seq = MIN_NR_GENS + 1;
 	lrugen->enabled = lru_gen_enabled();
@@ -5533,7 +5576,8 @@ void lru_gen_init_lruvec(struct lruvec *lruvec)
 	for_each_gen_type_zone(gen, type, zone)
 		INIT_LIST_HEAD(&lrugen->folios[gen][type][zone]);
 
-	lruvec->mm_state.seq = MIN_NR_GENS;
+	if (mm_state)
+		mm_state->seq = MIN_NR_GENS;
 }
 
 #ifdef CONFIG_MEMCG
@@ -5552,28 +5596,38 @@ void lru_gen_init_pgdat(struct pglist_data *pgdat)
 
 void lru_gen_init_memcg(struct mem_cgroup *memcg)
 {
-	INIT_LIST_HEAD(&memcg->mm_list.fifo);
-	spin_lock_init(&memcg->mm_list.lock);
+	struct lru_gen_mm_list *mm_list = get_mm_list(memcg);
+
+	if (!mm_list)
+		return;
+
+	INIT_LIST_HEAD(&mm_list->fifo);
+	spin_lock_init(&mm_list->lock);
 }
 
 void lru_gen_exit_memcg(struct mem_cgroup *memcg)
 {
 	int i;
 	int nid;
+	struct lru_gen_mm_list *mm_list = get_mm_list(memcg);
 
-	VM_WARN_ON_ONCE(!list_empty(&memcg->mm_list.fifo));
+	VM_WARN_ON_ONCE(mm_list && !list_empty(&mm_list->fifo));
 
 	for_each_node(nid) {
 		struct lruvec *lruvec = get_lruvec(memcg, nid);
+		struct lru_gen_mm_state *mm_state = get_mm_state(lruvec);
 
 		VM_WARN_ON_ONCE(memchr_inv(lruvec->lrugen.nr_pages, 0,
 					   sizeof(lruvec->lrugen.nr_pages)));
 
 		lruvec->lrugen.list.next = LIST_POISON1;
 
+		if (!mm_state)
+			continue;
+
 		for (i = 0; i < NR_BLOOM_FILTERS; i++) {
-			bitmap_free(lruvec->mm_state.filters[i]);
-			lruvec->mm_state.filters[i] = NULL;
+			bitmap_free(mm_state->filters[i]);
+			mm_state->filters[i] = NULL;
 		}
 	}
 }
-- 
2.43.0.472.g3155946c3a-goog

[PATCH mm-unstable v4 3/5] mm/mglru: remove CONFIG_MEMCG

[Kinsey Ho]

Remove CONFIG_MEMCG in a refactoring to improve code readability at
the cost of a few bytes in struct lru_gen_folio per node when
CONFIG_MEMCG=n.

Signed-off-by: Kinsey Ho <kinseyho@google.com>
Co-developed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Tested-by: Donet Tom <donettom@linux.vnet.ibm.com>
Acked-by: Yu Zhao <yuzhao@google.com>
---
 include/linux/mm_types.h |  4 ---
 include/linux/mmzone.h   | 26 ++--------------
 mm/vmscan.c              | 67 +++++++++++++---------------------------
 3 files changed, 23 insertions(+), 74 deletions(-)

diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 552fa2d11c57..55b7121809ff 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -1017,9 +1017,7 @@ struct lru_gen_mm_list {
 
 void lru_gen_add_mm(struct mm_struct *mm);
 void lru_gen_del_mm(struct mm_struct *mm);
-#ifdef CONFIG_MEMCG
 void lru_gen_migrate_mm(struct mm_struct *mm);
-#endif
 
 static inline void lru_gen_init_mm(struct mm_struct *mm)
 {
@@ -1050,11 +1048,9 @@ static inline void lru_gen_del_mm(struct mm_struct *mm)
 {
 }
 
-#ifdef CONFIG_MEMCG
 static inline void lru_gen_migrate_mm(struct mm_struct *mm)
 {
 }
-#endif
 
 static inline void lru_gen_init_mm(struct mm_struct *mm)
 {
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index bc3f63ec4291..28665e1b8475 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -440,14 +440,12 @@ struct lru_gen_folio {
 	atomic_long_t refaulted[NR_HIST_GENS][ANON_AND_FILE][MAX_NR_TIERS];
 	/* whether the multi-gen LRU is enabled */
 	bool enabled;
-#ifdef CONFIG_MEMCG
 	/* the memcg generation this lru_gen_folio belongs to */
 	u8 gen;
 	/* the list segment this lru_gen_folio belongs to */
 	u8 seg;
 	/* per-node lru_gen_folio list for global reclaim */
 	struct hlist_nulls_node list;
-#endif
 };
 
 enum {
@@ -493,11 +491,6 @@ struct lru_gen_mm_walk {
 	bool force_scan;
 };
 
-void lru_gen_init_lruvec(struct lruvec *lruvec);
-void lru_gen_look_around(struct page_vma_mapped_walk *pvmw);
-
-#ifdef CONFIG_MEMCG
-
 /*
  * For each node, memcgs are divided into two generations: the old and the
  * young. For each generation, memcgs are randomly sharded into multiple bins
@@ -555,6 +548,8 @@ struct lru_gen_memcg {
 };
 
 void lru_gen_init_pgdat(struct pglist_data *pgdat);
+void lru_gen_init_lruvec(struct lruvec *lruvec);
+void lru_gen_look_around(struct page_vma_mapped_walk *pvmw);
 
 void lru_gen_init_memcg(struct mem_cgroup *memcg);
 void lru_gen_exit_memcg(struct mem_cgroup *memcg);
@@ -563,19 +558,6 @@ void lru_gen_offline_memcg(struct mem_cgroup *memcg);
 void lru_gen_release_memcg(struct mem_cgroup *memcg);
 void lru_gen_soft_reclaim(struct mem_cgroup *memcg, int nid);
 
-#else /* !CONFIG_MEMCG */
-
-#define MEMCG_NR_GENS	1
-
-struct lru_gen_memcg {
-};
-
-static inline void lru_gen_init_pgdat(struct pglist_data *pgdat)
-{
-}
-
-#endif /* CONFIG_MEMCG */
-
 #else /* !CONFIG_LRU_GEN */
 
 static inline void lru_gen_init_pgdat(struct pglist_data *pgdat)
@@ -590,8 +572,6 @@ static inline void lru_gen_look_around(struct page_vma_mapped_walk *pvmw)
 {
 }
 
-#ifdef CONFIG_MEMCG
-
 static inline void lru_gen_init_memcg(struct mem_cgroup *memcg)
 {
 }
@@ -616,8 +596,6 @@ static inline void lru_gen_soft_reclaim(struct mem_cgroup *memcg, int nid)
 {
 }
 
-#endif /* CONFIG_MEMCG */
-
 #endif /* CONFIG_LRU_GEN */
 
 struct lruvec {
diff --git a/mm/vmscan.c b/mm/vmscan.c
index aa7ea09ffb4c..351a0b5043c0 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -4097,13 +4097,6 @@ enum {
 	MEMCG_LRU_YOUNG,
 };
 
-#ifdef CONFIG_MEMCG
-
-static int lru_gen_memcg_seg(struct lruvec *lruvec)
-{
-	return READ_ONCE(lruvec->lrugen.seg);
-}
-
 static void lru_gen_rotate_memcg(struct lruvec *lruvec, int op)
 {
 	int seg;
@@ -4150,6 +4143,8 @@ static void lru_gen_rotate_memcg(struct lruvec *lruvec, int op)
 	spin_unlock_irqrestore(&pgdat->memcg_lru.lock, flags);
 }
 
+#ifdef CONFIG_MEMCG
+
 void lru_gen_online_memcg(struct mem_cgroup *memcg)
 {
 	int gen;
@@ -4217,18 +4212,11 @@ void lru_gen_soft_reclaim(struct mem_cgroup *memcg, int nid)
 	struct lruvec *lruvec = get_lruvec(memcg, nid);
 
 	/* see the comment on MEMCG_NR_GENS */
-	if (lru_gen_memcg_seg(lruvec) != MEMCG_LRU_HEAD)
+	if (READ_ONCE(lruvec->lrugen.seg) != MEMCG_LRU_HEAD)
 		lru_gen_rotate_memcg(lruvec, MEMCG_LRU_HEAD);
 }
 
-#else /* !CONFIG_MEMCG */
-
-static int lru_gen_memcg_seg(struct lruvec *lruvec)
-{
-	return 0;
-}
-
-#endif
+#endif /* CONFIG_MEMCG */
 
 /******************************************************************************
  *                          the eviction
@@ -4776,7 +4764,7 @@ static int shrink_one(struct lruvec *lruvec, struct scan_control *sc)
 
 	if (mem_cgroup_below_low(NULL, memcg)) {
 		/* see the comment on MEMCG_NR_GENS */
-		if (lru_gen_memcg_seg(lruvec) != MEMCG_LRU_TAIL)
+		if (READ_ONCE(lruvec->lrugen.seg) != MEMCG_LRU_TAIL)
 			return MEMCG_LRU_TAIL;
 
 		memcg_memory_event(memcg, MEMCG_LOW);
@@ -4799,12 +4787,10 @@ static int shrink_one(struct lruvec *lruvec, struct scan_control *sc)
 		return 0;
 
 	/* one retry if offlined or too small */
-	return lru_gen_memcg_seg(lruvec) != MEMCG_LRU_TAIL ?
+	return READ_ONCE(lruvec->lrugen.seg) != MEMCG_LRU_TAIL ?
 	       MEMCG_LRU_TAIL : MEMCG_LRU_YOUNG;
 }
 
-#ifdef CONFIG_MEMCG
-
 static void shrink_many(struct pglist_data *pgdat, struct scan_control *sc)
 {
 	int op;
@@ -4896,20 +4882,6 @@ static void lru_gen_shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc
 	blk_finish_plug(&plug);
 }
 
-#else /* !CONFIG_MEMCG */
-
-static void shrink_many(struct pglist_data *pgdat, struct scan_control *sc)
-{
-	BUILD_BUG();
-}
-
-static void lru_gen_shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
-{
-	BUILD_BUG();
-}
-
-#endif
-
 static void set_initial_priority(struct pglist_data *pgdat, struct scan_control *sc)
 {
 	int priority;
@@ -5560,6 +5532,18 @@ static const struct file_operations lru_gen_ro_fops = {
  *                          initialization
  ******************************************************************************/
 
+void lru_gen_init_pgdat(struct pglist_data *pgdat)
+{
+	int i, j;
+
+	spin_lock_init(&pgdat->memcg_lru.lock);
+
+	for (i = 0; i < MEMCG_NR_GENS; i++) {
+		for (j = 0; j < MEMCG_NR_BINS; j++)
+			INIT_HLIST_NULLS_HEAD(&pgdat->memcg_lru.fifo[i][j], i);
+	}
+}
+
 void lru_gen_init_lruvec(struct lruvec *lruvec)
 {
 	int i;
@@ -5582,18 +5566,6 @@ void lru_gen_init_lruvec(struct lruvec *lruvec)
 
 #ifdef CONFIG_MEMCG
 
-void lru_gen_init_pgdat(struct pglist_data *pgdat)
-{
-	int i, j;
-
-	spin_lock_init(&pgdat->memcg_lru.lock);
-
-	for (i = 0; i < MEMCG_NR_GENS; i++) {
-		for (j = 0; j < MEMCG_NR_BINS; j++)
-			INIT_HLIST_NULLS_HEAD(&pgdat->memcg_lru.fifo[i][j], i);
-	}
-}
-
 void lru_gen_init_memcg(struct mem_cgroup *memcg)
 {
 	struct lru_gen_mm_list *mm_list = get_mm_list(memcg);
@@ -5653,14 +5625,17 @@ late_initcall(init_lru_gen);
 
 static void lru_gen_age_node(struct pglist_data *pgdat, struct scan_control *sc)
 {
+	BUILD_BUG();
 }
 
 static void lru_gen_shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
 {
+	BUILD_BUG();
 }
 
 static void lru_gen_shrink_node(struct pglist_data *pgdat, struct scan_control *sc)
 {
+	BUILD_BUG();
 }
 
 #endif /* CONFIG_LRU_GEN */
-- 
2.43.0.472.g3155946c3a-goog

[PATCH mm-unstable v4 4/5] mm/mglru: add dummy pmd_dirty()

[Kinsey Ho]

Add dummy pmd_dirty() for architectures that don't provide it.
This is similar to commit 6617da8fb565 ("mm: add dummy pmd_young()
for architectures not having it").

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202312210606.1Etqz3M4-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202312210042.xQEiqlEh-lkp@intel.com/
Signed-off-by: Kinsey Ho <kinseyho@google.com>
Suggested-by: Yu Zhao <yuzhao@google.com>
---
 arch/loongarch/include/asm/pgtable.h | 1 +
 arch/mips/include/asm/pgtable.h      | 1 +
 arch/riscv/include/asm/pgtable.h     | 1 +
 arch/s390/include/asm/pgtable.h      | 1 +
 arch/sparc/include/asm/pgtable_64.h  | 1 +
 arch/x86/include/asm/pgtable.h       | 1 +
 include/linux/pgtable.h              | 7 +++++++
 7 files changed, 13 insertions(+)

diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h
index 29d9b12298bc..8b5df1bbf9e9 100644
--- a/arch/loongarch/include/asm/pgtable.h
+++ b/arch/loongarch/include/asm/pgtable.h
@@ -523,6 +523,7 @@ static inline pmd_t pmd_wrprotect(pmd_t pmd)
 	return pmd;
 }
 
+#define pmd_dirty pmd_dirty
 static inline int pmd_dirty(pmd_t pmd)
 {
 	return !!(pmd_val(pmd) & (_PAGE_DIRTY | _PAGE_MODIFIED));
diff --git a/arch/mips/include/asm/pgtable.h b/arch/mips/include/asm/pgtable.h
index 430b208c0130..e27a4c83c548 100644
--- a/arch/mips/include/asm/pgtable.h
+++ b/arch/mips/include/asm/pgtable.h
@@ -655,6 +655,7 @@ static inline pmd_t pmd_mkwrite_novma(pmd_t pmd)
 	return pmd;
 }
 
+#define pmd_dirty pmd_dirty
 static inline int pmd_dirty(pmd_t pmd)
 {
 	return !!(pmd_val(pmd) & _PAGE_MODIFIED);
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index ab00235b018f..7b4287f36054 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -673,6 +673,7 @@ static inline int pmd_write(pmd_t pmd)
 	return pte_write(pmd_pte(pmd));
 }
 
+#define pmd_dirty pmd_dirty
 static inline int pmd_dirty(pmd_t pmd)
 {
 	return pte_dirty(pmd_pte(pmd));
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 601e87fa8a9a..1299b56e43f6 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -770,6 +770,7 @@ static inline int pud_write(pud_t pud)
 	return (pud_val(pud) & _REGION3_ENTRY_WRITE) != 0;
 }
 
+#define pmd_dirty pmd_dirty
 static inline int pmd_dirty(pmd_t pmd)
 {
 	return (pmd_val(pmd) & _SEGMENT_ENTRY_DIRTY) != 0;
diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index 5e41033bf4ca..a8c871b7d786 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -706,6 +706,7 @@ static inline unsigned long pmd_write(pmd_t pmd)
 #define pud_write(pud)	pte_write(__pte(pud_val(pud)))
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+#define pmd_dirty pmd_dirty
 static inline unsigned long pmd_dirty(pmd_t pmd)
 {
 	pte_t pte = __pte(pmd_val(pmd));
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 08b5cb22d9a6..9d077bca6a10 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -141,6 +141,7 @@ static inline int pte_young(pte_t pte)
 	return pte_flags(pte) & _PAGE_ACCESSED;
 }
 
+#define pmd_dirty pmd_dirty
 static inline bool pmd_dirty(pmd_t pmd)
 {
 	return pmd_flags(pmd) & _PAGE_DIRTY_BITS;
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index 9ecc20fa6269..466cf477551a 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -184,6 +184,13 @@ static inline int pmd_young(pmd_t pmd)
 }
 #endif
 
+#ifndef pmd_dirty
+static inline int pmd_dirty(pmd_t pmd)
+{
+	return 0;
+}
+#endif
+
 /*
  * A facility to provide lazy MMU batching.  This allows PTE updates and
  * page invalidations to be delayed until a call to leave lazy MMU mode
-- 
2.43.0.472.g3155946c3a-goog

[PATCH mm-unstable v4 5/5] mm/mglru: remove CONFIG_TRANSPARENT_HUGEPAGE

[Kinsey Ho]

Improve code readability by removing CONFIG_TRANSPARENT_HUGEPAGE,
since the compiler should be able to automatically optimize out the
code that promotes THPs during page table walks.

No functional changes.

Signed-off-by: Kinsey Ho <kinseyho@google.com>
Co-developed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Tested-by: Donet Tom <donettom@linux.vnet.ibm.com>
Acked-by: Yu Zhao <yuzhao@google.com>
---
 mm/vmscan.c | 12 +-----------
 1 file changed, 1 insertion(+), 11 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 351a0b5043c0..ceba905e5630 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -3273,7 +3273,6 @@ static unsigned long get_pte_pfn(pte_t pte, struct vm_area_struct *vma, unsigned
 	return pfn;
 }
 
-#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG)
 static unsigned long get_pmd_pfn(pmd_t pmd, struct vm_area_struct *vma, unsigned long addr)
 {
 	unsigned long pfn = pmd_pfn(pmd);
@@ -3291,7 +3290,6 @@ static unsigned long get_pmd_pfn(pmd_t pmd, struct vm_area_struct *vma, unsigned
 
 	return pfn;
 }
-#endif
 
 static struct folio *get_pfn_folio(unsigned long pfn, struct mem_cgroup *memcg,
 				   struct pglist_data *pgdat, bool can_swap)
@@ -3394,7 +3392,6 @@ static bool walk_pte_range(pmd_t *pmd, unsigned long start, unsigned long end,
 	return suitable_to_scan(total, young);
 }
 
-#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG)
 static void walk_pmd_range_locked(pud_t *pud, unsigned long addr, struct vm_area_struct *vma,
 				  struct mm_walk *args, unsigned long *bitmap, unsigned long *first)
 {
@@ -3472,12 +3469,6 @@ static void walk_pmd_range_locked(pud_t *pud, unsigned long addr, struct vm_area
 done:
 	*first = -1;
 }
-#else
-static void walk_pmd_range_locked(pud_t *pud, unsigned long addr, struct vm_area_struct *vma,
-				  struct mm_walk *args, unsigned long *bitmap, unsigned long *first)
-{
-}
-#endif
 
 static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 			   struct mm_walk *args)
@@ -3513,7 +3504,6 @@ static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 			continue;
 		}
 
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
 		if (pmd_trans_huge(val)) {
 			unsigned long pfn = pmd_pfn(val);
 			struct pglist_data *pgdat = lruvec_pgdat(walk->lruvec);
@@ -3532,7 +3522,7 @@ static void walk_pmd_range(pud_t *pud, unsigned long start, unsigned long end,
 			walk_pmd_range_locked(pud, addr, vma, args, bitmap, &first);
 			continue;
 		}
-#endif
+
 		walk->mm_stats[MM_NONLEAF_TOTAL]++;
 
 		if (should_clear_pmd_young()) {
-- 
2.43.0.472.g3155946c3a-goog