Merge branch 'slab/for-6.19/sheaves_cleanups' into slab/for-next

Merge series "slab: preparatory cleanups before adding sheaves to all
caches" [1]

Cleanups that were written as part of the full sheaves conversion, which
is not fully ready yet, but they are useful on their own.

Link: https://lore.kernel.org/all/20251105-sheaves-cleanups-v1-0-b8218e1ac7ef@suse.cz/ [1]

3 files changed, 173 insertions, 166 deletions

diff --git a/include/linux/gfp_types.h b/include/linux/gfp_types.h
index 65db9349f905..3de43b12209e 100644
--- a/include/linux/gfp_types.h
+++ b/include/linux/gfp_types.h
@@ -55,9 +55,7 @@ enum {
 #ifdef CONFIG_LOCKDEP
 	___GFP_NOLOCKDEP_BIT,
 #endif
-#ifdef CONFIG_SLAB_OBJ_EXT
 	___GFP_NO_OBJ_EXT_BIT,
-#endif
 	___GFP_LAST_BIT
 };
 
@@ -98,11 +96,7 @@ enum {
 #else
 #define ___GFP_NOLOCKDEP	0
 #endif
-#ifdef CONFIG_SLAB_OBJ_EXT
 #define ___GFP_NO_OBJ_EXT       BIT(___GFP_NO_OBJ_EXT_BIT)
-#else
-#define ___GFP_NO_OBJ_EXT       0
-#endif
 
 /*
  * Physical address zone modifiers (see linux/mmzone.h - low four bits)
diff --git a/mm/slab.h b/mm/slab.h
index 078daecc7cf5..f7b8df56727d 100644
--- a/mm/slab.h
+++ b/mm/slab.h
@@ -236,10 +236,8 @@ struct kmem_cache_order_objects {
  * Slab cache management.
  */
 struct kmem_cache {
-#ifndef CONFIG_SLUB_TINY
 	struct kmem_cache_cpu __percpu *cpu_slab;
 	struct lock_class_key lock_key;
-#endif
 	struct slub_percpu_sheaves __percpu *cpu_sheaves;
 	/* Used for retrieving partial slabs, etc. */
 	slab_flags_t flags;
diff --git a/mm/slub.c b/mm/slub.c
index 3095f10e0fe4..16d723c2dc36 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -410,7 +410,6 @@ enum stat_item {
 	NR_SLUB_STAT_ITEMS
 };
 
-#ifndef CONFIG_SLUB_TINY
 /*
  * When changing the layout, make sure freelist and tid are still compatible
  * with this_cpu_cmpxchg_double() alignment requirements.
@@ -432,7 +431,6 @@ struct kmem_cache_cpu {
 	unsigned int stat[NR_SLUB_STAT_ITEMS];
 #endif
 };
-#endif /* CONFIG_SLUB_TINY */
 
 static inline void stat(const struct kmem_cache *s, enum stat_item si)
 {
@@ -469,7 +467,10 @@ struct slab_sheaf {
 		struct rcu_head rcu_head;
 		struct list_head barn_list;
 		/* only used for prefilled sheafs */
-		unsigned int capacity;
+		struct {
+			unsigned int capacity;
+			bool pfmemalloc;
+		};
 	};
 	struct kmem_cache *cache;
 	unsigned int size;
@@ -594,12 +595,10 @@ static inline void *get_freepointer(struct kmem_cache *s, void *object)
 	return freelist_ptr_decode(s, p, ptr_addr);
 }
 
-#ifndef CONFIG_SLUB_TINY
 static void prefetch_freepointer(const struct kmem_cache *s, void *object)
 {
 	prefetchw(object + s->offset);
 }
-#endif
 
 /*
  * When running under KMSAN, get_freepointer_safe() may return an uninitialized
@@ -711,10 +710,12 @@ static inline unsigned int slub_get_cpu_partial(struct kmem_cache *s)
 	return s->cpu_partial_slabs;
 }
 #else
+#ifdef SLAB_SUPPORTS_SYSFS
 static inline void
 slub_set_cpu_partial(struct kmem_cache *s, unsigned int nr_objects)
 {
 }
+#endif
 
 static inline unsigned int slub_get_cpu_partial(struct kmem_cache *s)
 {
@@ -2027,15 +2028,21 @@ static inline void inc_slabs_node(struct kmem_cache *s, int node,
 							int objects) {}
 static inline void dec_slabs_node(struct kmem_cache *s, int node,
 							int objects) {}
-#ifndef CONFIG_SLUB_TINY
 static bool freelist_corrupted(struct kmem_cache *s, struct slab *slab,
 			       void **freelist, void *nextfree)
 {
 	return false;
 }
-#endif
 #endif /* CONFIG_SLUB_DEBUG */
 
+/*
+ * The allocated objcg pointers array is not accounted directly.
+ * Moreover, it should not come from DMA buffer and is not readily
+ * reclaimable. So those GFP bits should be masked off.
+ */
+#define OBJCGS_CLEAR_MASK	(__GFP_DMA | __GFP_RECLAIMABLE | \
+				__GFP_ACCOUNT | __GFP_NOFAIL)
+
 #ifdef CONFIG_SLAB_OBJ_EXT
 
 #ifdef CONFIG_MEM_ALLOC_PROFILING_DEBUG
@@ -2086,14 +2093,6 @@ static inline void handle_failed_objexts_alloc(unsigned long obj_exts,
 
 #endif /* CONFIG_MEM_ALLOC_PROFILING_DEBUG */
 
-/*
- * The allocated objcg pointers array is not accounted directly.
- * Moreover, it should not come from DMA buffer and is not readily
- * reclaimable. So those GFP bits should be masked off.
- */
-#define OBJCGS_CLEAR_MASK	(__GFP_DMA | __GFP_RECLAIMABLE | \
-				__GFP_ACCOUNT | __GFP_NOFAIL)
-
 static inline void init_slab_obj_exts(struct slab *slab)
 {
 	slab->obj_exts = 0;
@@ -2601,8 +2600,24 @@ static void *setup_object(struct kmem_cache *s, void *object)
 
 static struct slab_sheaf *alloc_empty_sheaf(struct kmem_cache *s, gfp_t gfp)
 {
-	struct slab_sheaf *sheaf = kzalloc(struct_size(sheaf, objects,
-					s->sheaf_capacity), gfp);
+	struct slab_sheaf *sheaf;
+	size_t sheaf_size;
+
+	if (gfp & __GFP_NO_OBJ_EXT)
+		return NULL;
+
+	gfp &= ~OBJCGS_CLEAR_MASK;
+
+	/*
+	 * Prevent recursion to the same cache, or a deep stack of kmallocs of
+	 * varying sizes (sheaf capacity might differ for each kmalloc size
+	 * bucket)
+	 */
+	if (s->flags & SLAB_KMALLOC)
+		gfp |= __GFP_NO_OBJ_EXT;
+
+	sheaf_size = struct_size(sheaf, objects, s->sheaf_capacity);
+	sheaf = kzalloc(sheaf_size, gfp);
 
 	if (unlikely(!sheaf))
 		return NULL;
@@ -2655,7 +2670,7 @@ static struct slab_sheaf *alloc_full_sheaf(struct kmem_cache *s, gfp_t gfp)
 	if (!sheaf)
 		return NULL;
 
-	if (refill_sheaf(s, sheaf, gfp)) {
+	if (refill_sheaf(s, sheaf, gfp | __GFP_NOMEMALLOC)) {
 		free_empty_sheaf(s, sheaf);
 		return NULL;
 	}
@@ -2733,12 +2748,13 @@ static void sheaf_flush_unused(struct kmem_cache *s, struct slab_sheaf *sheaf)
 	sheaf->size = 0;
 }
 
-static void __rcu_free_sheaf_prepare(struct kmem_cache *s,
+static bool __rcu_free_sheaf_prepare(struct kmem_cache *s,
 				     struct slab_sheaf *sheaf)
 {
 	bool init = slab_want_init_on_free(s);
 	void **p = &sheaf->objects[0];
 	unsigned int i = 0;
+	bool pfmemalloc = false;
 
 	while (i < sheaf->size) {
 		struct slab *slab = virt_to_slab(p[i]);
@@ -2751,8 +2767,13 @@ static void __rcu_free_sheaf_prepare(struct kmem_cache *s,
 			continue;
 		}
 
+		if (slab_test_pfmemalloc(slab))
+			pfmemalloc = true;
+
 		i++;
 	}
+
+	return pfmemalloc;
 }
 
 static void rcu_free_sheaf_nobarn(struct rcu_head *head)
@@ -3015,14 +3036,11 @@ static void barn_init(struct node_barn *barn)
 
 static void barn_shrink(struct kmem_cache *s, struct node_barn *barn)
 {
-	struct list_head empty_list;
-	struct list_head full_list;
+	LIST_HEAD(empty_list);
+	LIST_HEAD(full_list);
 	struct slab_sheaf *sheaf, *sheaf2;
 	unsigned long flags;
 
-	INIT_LIST_HEAD(&empty_list);
-	INIT_LIST_HEAD(&full_list);
-
 	spin_lock_irqsave(&barn->lock, flags);
 
 	list_splice_init(&barn->sheaves_full, &full_list);
@@ -3618,8 +3636,6 @@ static struct slab *get_partial(struct kmem_cache *s, int node,
 	return get_any_partial(s, pc);
 }
 
-#ifndef CONFIG_SLUB_TINY
-
 #ifdef CONFIG_PREEMPTION
 /*
  * Calculate the next globally unique transaction for disambiguation
@@ -4019,12 +4035,6 @@ static bool has_cpu_slab(int cpu, struct kmem_cache *s)
 	return c->slab || slub_percpu_partial(c);
 }
 
-#else /* CONFIG_SLUB_TINY */
-static inline void __flush_cpu_slab(struct kmem_cache *s, int cpu) { }
-static inline bool has_cpu_slab(int cpu, struct kmem_cache *s) { return false; }
-static inline void flush_this_cpu_slab(struct kmem_cache *s) { }
-#endif /* CONFIG_SLUB_TINY */
-
 static bool has_pcs_used(int cpu, struct kmem_cache *s)
 {
 	struct slub_percpu_sheaves *pcs;
@@ -4365,7 +4375,6 @@ static inline bool pfmemalloc_match(struct slab *slab, gfp_t gfpflags)
 	return true;
 }
 
-#ifndef CONFIG_SLUB_TINY
 static inline bool
 __update_cpu_freelist_fast(struct kmem_cache *s,
 			   void *freelist_old, void *freelist_new,
@@ -4629,7 +4638,7 @@ new_objects:
 	pc.orig_size = orig_size;
 	slab = get_partial(s, node, &pc);
 	if (slab) {
-		if (kmem_cache_debug(s)) {
+		if (IS_ENABLED(CONFIG_SLUB_TINY) || kmem_cache_debug(s)) {
 			freelist = pc.object;
 			/*
 			 * For debug caches here we had to go through
@@ -4667,11 +4676,15 @@ new_objects:
 
 	stat(s, ALLOC_SLAB);
 
-	if (kmem_cache_debug(s)) {
+	if (IS_ENABLED(CONFIG_SLUB_TINY) || kmem_cache_debug(s)) {
 		freelist = alloc_single_from_new_slab(s, slab, orig_size, gfpflags);
 
-		if (unlikely(!freelist))
+		if (unlikely(!freelist)) {
+			/* This could cause an endless loop. Fail instead. */
+			if (!allow_spin)
+				return NULL;
 			goto new_objects;
+		}
 
 		if (s->flags & SLAB_STORE_USER)
 			set_track(s, freelist, TRACK_ALLOC, addr,
@@ -4875,32 +4888,6 @@ redo:
 
 	return object;
 }
-#else /* CONFIG_SLUB_TINY */
-static void *__slab_alloc_node(struct kmem_cache *s,
-		gfp_t gfpflags, int node, unsigned long addr, size_t orig_size)
-{
-	struct partial_context pc;
-	struct slab *slab;
-	void *object;
-
-	pc.flags = gfpflags;
-	pc.orig_size = orig_size;
-	slab = get_partial(s, node, &pc);
-
-	if (slab)
-		return pc.object;
-
-	slab = new_slab(s, gfpflags, node);
-	if (unlikely(!slab)) {
-		slab_out_of_memory(s, gfpflags, node);
-		return NULL;
-	}
-
-	object = alloc_single_from_new_slab(s, slab, orig_size, gfpflags);
-
-	return object;
-}
-#endif /* CONFIG_SLUB_TINY */
 
 /*
  * If the object has been wiped upon free, make sure it's fully initialized by
@@ -5041,7 +5028,7 @@ __pcs_replace_empty_main(struct kmem_cache *s, struct slub_percpu_sheaves *pcs,
 		return NULL;
 
 	if (empty) {
-		if (!refill_sheaf(s, empty, gfp)) {
+		if (!refill_sheaf(s, empty, gfp | __GFP_NOMEMALLOC)) {
 			full = empty;
 		} else {
 			/*
@@ -5341,6 +5328,26 @@ void *kmem_cache_alloc_node_noprof(struct kmem_cache *s, gfp_t gfpflags, int nod
 }
 EXPORT_SYMBOL(kmem_cache_alloc_node_noprof);
 
+static int __prefill_sheaf_pfmemalloc(struct kmem_cache *s,
+				      struct slab_sheaf *sheaf, gfp_t gfp)
+{
+	int ret = 0;
+
+	ret = refill_sheaf(s, sheaf, gfp | __GFP_NOMEMALLOC);
+
+	if (likely(!ret || !gfp_pfmemalloc_allowed(gfp)))
+		return ret;
+
+	/*
+	 * if we are allowed to, refill sheaf with pfmemalloc but then remember
+	 * it for when it's returned
+	 */
+	ret = refill_sheaf(s, sheaf, gfp);
+	sheaf->pfmemalloc = true;
+
+	return ret;
+}
+
 /*
  * returns a sheaf that has at least the requested size
  * when prefilling is needed, do so with given gfp flags
@@ -5375,6 +5382,10 @@ kmem_cache_prefill_sheaf(struct kmem_cache *s, gfp_t gfp, unsigned int size)
 		sheaf->cache = s;
 		sheaf->capacity = size;
 
+		/*
+		 * we do not need to care about pfmemalloc here because oversize
+		 * sheaves area always flushed and freed when returned
+		 */
 		if (!__kmem_cache_alloc_bulk(s, gfp, size,
 					     &sheaf->objects[0])) {
 			kfree(sheaf);
@@ -5411,17 +5422,18 @@ kmem_cache_prefill_sheaf(struct kmem_cache *s, gfp_t gfp, unsigned int size)
 	if (!sheaf)
 		sheaf = alloc_empty_sheaf(s, gfp);
 
-	if (sheaf && sheaf->size < size) {
-		if (refill_sheaf(s, sheaf, gfp)) {
+	if (sheaf) {
+		sheaf->capacity = s->sheaf_capacity;
+		sheaf->pfmemalloc = false;
+
+		if (sheaf->size < size &&
+		    __prefill_sheaf_pfmemalloc(s, sheaf, gfp)) {
 			sheaf_flush_unused(s, sheaf);
 			free_empty_sheaf(s, sheaf);
 			sheaf = NULL;
 		}
 	}
 
-	if (sheaf)
-		sheaf->capacity = s->sheaf_capacity;
-
 	return sheaf;
 }
 
@@ -5441,7 +5453,8 @@ void kmem_cache_return_sheaf(struct kmem_cache *s, gfp_t gfp,
 	struct slub_percpu_sheaves *pcs;
 	struct node_barn *barn;
 
-	if (unlikely(sheaf->capacity != s->sheaf_capacity)) {
+	if (unlikely((sheaf->capacity != s->sheaf_capacity)
+		     || sheaf->pfmemalloc)) {
 		sheaf_flush_unused(s, sheaf);
 		kfree(sheaf);
 		return;
@@ -5507,7 +5520,7 @@ int kmem_cache_refill_sheaf(struct kmem_cache *s, gfp_t gfp,
 
 	if (likely(sheaf->capacity >= size)) {
 		if (likely(sheaf->capacity == s->sheaf_capacity))
-			return refill_sheaf(s, sheaf, gfp);
+			return __prefill_sheaf_pfmemalloc(s, sheaf, gfp);
 
 		if (!__kmem_cache_alloc_bulk(s, gfp, sheaf->capacity - sheaf->size,
 					     &sheaf->objects[sheaf->size])) {
@@ -5540,6 +5553,9 @@ int kmem_cache_refill_sheaf(struct kmem_cache *s, gfp_t gfp,
  *
  * The gfp parameter is meant only to specify __GFP_ZERO or __GFP_ACCOUNT
  * memcg charging is forced over limit if necessary, to avoid failure.
+ *
+ * It is possible that the allocation comes from kfence and then the sheaf
+ * size is not decreased.
  */
 void *
 kmem_cache_alloc_from_sheaf_noprof(struct kmem_cache *s, gfp_t gfp,
@@ -5551,7 +5567,10 @@ kmem_cache_alloc_from_sheaf_noprof(struct kmem_cache *s, gfp_t gfp,
 	if (sheaf->size == 0)
 		goto out;
 
-	ret = sheaf->objects[--sheaf->size];
+	ret = kfence_alloc(s, s->object_size, gfp);
+
+	if (likely(!ret))
+		ret = sheaf->objects[--sheaf->size];
 
 	init = slab_want_init_on_alloc(gfp, s);
 
@@ -5719,9 +5738,7 @@ retry:
 	 * it did local_lock_irqsave(&s->cpu_slab->lock, flags).
 	 * In this case fast path with __update_cpu_freelist_fast() is not safe.
 	 */
-#ifndef CONFIG_SLUB_TINY
 	if (!in_nmi() || !local_lock_is_locked(&s->cpu_slab->lock))
-#endif
 		ret = __slab_alloc_node(s, alloc_gfp, node, _RET_IP_, size);
 
 	if (PTR_ERR(ret) == -EBUSY) {
@@ -5859,8 +5876,8 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 			unsigned long addr)
 
 {
-	void *prior;
-	int was_frozen;
+	void *old_head;
+	bool was_frozen, was_full;
 	struct slab new;
 	unsigned long counters;
 	struct kmem_cache_node *n = NULL;
@@ -5874,20 +5891,37 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 		return;
 	}
 
+	/*
+	 * It is enough to test IS_ENABLED(CONFIG_SLUB_CPU_PARTIAL) below
+	 * instead of kmem_cache_has_cpu_partial(s), because kmem_cache_debug(s)
+	 * is the only other reason it can be false, and it is already handled
+	 * above.
+	 */
+
 	do {
 		if (unlikely(n)) {
 			spin_unlock_irqrestore(&n->list_lock, flags);
 			n = NULL;
 		}
-		prior = slab->freelist;
+		old_head = slab->freelist;
 		counters = slab->counters;
-		set_freepointer(s, tail, prior);
+		set_freepointer(s, tail, old_head);
 		new.counters = counters;
-		was_frozen = new.frozen;
+		was_frozen = !!new.frozen;
+		was_full = (old_head == NULL);
 		new.inuse -= cnt;
-		if ((!new.inuse || !prior) && !was_frozen) {
-			/* Needs to be taken off a list */
-			if (!kmem_cache_has_cpu_partial(s) || prior) {
+		/*
+		 * Might need to be taken off (due to becoming empty) or added
+		 * to (due to not being full anymore) the partial list.
+		 * Unless it's frozen.
+		 */
+		if ((!new.inuse || was_full) && !was_frozen) {
+			/*
+			 * If slab becomes non-full and we have cpu partial
+			 * lists, we put it there unconditionally to avoid
+			 * taking the list_lock. Otherwise we need it.
+			 */
+			if (!(IS_ENABLED(CONFIG_SLUB_CPU_PARTIAL) && was_full)) {
 
 				n = get_node(s, slab_nid(slab));
 				/*
@@ -5905,7 +5939,7 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 		}
 
 	} while (!slab_update_freelist(s, slab,
-		prior, counters,
+		old_head, counters,
 		head, new.counters,
 		"__slab_free"));
 
@@ -5917,7 +5951,7 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 			 * activity can be necessary.
 			 */
 			stat(s, FREE_FROZEN);
-		} else if (kmem_cache_has_cpu_partial(s) && !prior) {
+		} else if (IS_ENABLED(CONFIG_SLUB_CPU_PARTIAL) && was_full) {
 			/*
 			 * If we started with a full slab then put it onto the
 			 * per cpu partial list.
@@ -5926,6 +5960,11 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 			stat(s, CPU_PARTIAL_FREE);
 		}
 
+		/*
+		 * In other cases we didn't take the list_lock because the slab
+		 * was already on the partial list and will remain there.
+		 */
+
 		return;
 	}
 
@@ -5933,19 +5972,24 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 	 * This slab was partially empty but not on the per-node partial list,
 	 * in which case we shouldn't manipulate its list, just return.
 	 */
-	if (prior && !on_node_partial) {
+	if (!was_full && !on_node_partial) {
 		spin_unlock_irqrestore(&n->list_lock, flags);
 		return;
 	}
 
+	/*
+	 * If slab became empty, should we add/keep it on the partial list or we
+	 * have enough?
+	 */
 	if (unlikely(!new.inuse && n->nr_partial >= s->min_partial))
 		goto slab_empty;
 
 	/*
 	 * Objects left in the slab. If it was not on the partial list before
-	 * then add it.
+	 * then add it. This can only happen when cache has no per cpu partial
+	 * list otherwise we would have put it there.
 	 */
-	if (!kmem_cache_has_cpu_partial(s) && unlikely(!prior)) {
+	if (!IS_ENABLED(CONFIG_SLUB_CPU_PARTIAL) && unlikely(was_full)) {
 		add_partial(n, slab, DEACTIVATE_TO_TAIL);
 		stat(s, FREE_ADD_PARTIAL);
 	}
@@ -5953,10 +5997,11 @@ static void __slab_free(struct kmem_cache *s, struct slab *slab,
 	return;
 
 slab_empty:
-	if (prior) {
-		/*
-		 * Slab on the partial list.
-		 */
+	/*
+	 * The slab could have a single object and thus go from full to empty in
+	 * a single free, but more likely it was on the partial list. Remove it.
+	 */
+	if (likely(!was_full)) {
 		remove_partial(n, slab);
 		stat(s, FREE_REMOVE_PARTIAL);
 	}
@@ -6181,8 +6226,12 @@ static void rcu_free_sheaf(struct rcu_head *head)
 	 * handles it fine. The only downside is that sheaf will serve fewer
 	 * allocations when reused. It only happens due to debugging, which is a
 	 * performance hit anyway.
+	 *
+	 * If it returns true, there was at least one object from pfmemalloc
+	 * slab so simply flush everything.
 	 */
-	__rcu_free_sheaf_prepare(s, sheaf);
+	if (__rcu_free_sheaf_prepare(s, sheaf))
+		goto flush;
 
 	n = get_node(s, sheaf->node);
 	if (!n)
@@ -6337,7 +6386,8 @@ next_remote_batch:
 			continue;
 		}
 
-		if (unlikely(IS_ENABLED(CONFIG_NUMA) && slab_nid(slab) != node)) {
+		if (unlikely((IS_ENABLED(CONFIG_NUMA) && slab_nid(slab) != node)
+			     || slab_test_pfmemalloc(slab))) {
 			remote_objects[remote_nr] = p[i];
 			p[i] = p[--size];
 			if (++remote_nr >= PCS_BATCH_MAX)
@@ -6479,14 +6529,10 @@ static void free_deferred_objects(struct irq_work *work)
 	llist_for_each_safe(pos, t, llnode) {
 		struct slab *slab = container_of(pos, struct slab, llnode);
 
-#ifdef CONFIG_SLUB_TINY
-		free_slab(slab->slab_cache, slab);
-#else
 		if (slab->frozen)
 			deactivate_slab(slab->slab_cache, slab, slab->flush_freelist);
 		else
 			free_slab(slab->slab_cache, slab);
-#endif
 	}
 }
 
@@ -6522,7 +6568,6 @@ void defer_free_barrier(void)
 		irq_work_sync(&per_cpu_ptr(&defer_free_objects, cpu)->work);
 }
 
-#ifndef CONFIG_SLUB_TINY
 /*
  * Fastpath with forced inlining to produce a kfree and kmem_cache_free that
  * can perform fastpath freeing without additional function calls.
@@ -6615,14 +6660,6 @@ redo:
 	}
 	stat_add(s, FREE_FASTPATH, cnt);
 }
-#else /* CONFIG_SLUB_TINY */
-static void do_slab_free(struct kmem_cache *s,
-				struct slab *slab, void *head, void *tail,
-				int cnt, unsigned long addr)
-{
-	__slab_free(s, slab, head, tail, cnt, addr);
-}
-#endif /* CONFIG_SLUB_TINY */
 
 static __fastpath_inline
 void slab_free(struct kmem_cache *s, struct slab *slab, void *object,
@@ -6635,7 +6672,8 @@ void slab_free(struct kmem_cache *s, struct slab *slab, void *object,
 		return;
 
 	if (s->cpu_sheaves && likely(!IS_ENABLED(CONFIG_NUMA) ||
-				     slab_nid(slab) == numa_mem_id())) {
+				     slab_nid(slab) == numa_mem_id())
+			   && likely(!slab_test_pfmemalloc(slab))) {
 		if (likely(free_to_pcs(s, object)))
 			return;
 	}
@@ -6899,11 +6937,7 @@ void kfree_nolock(const void *object)
 	 * since kasan quarantine takes locks and not supported from NMI.
 	 */
 	kasan_slab_free(s, x, false, false, /* skip quarantine */true);
-#ifndef CONFIG_SLUB_TINY
 	do_slab_free(s, slab, x, x, 0, _RET_IP_);
-#else
-	defer_free(s, x);
-#endif
 }
 EXPORT_SYMBOL_GPL(kfree_nolock);
 
@@ -7353,7 +7387,6 @@ void kmem_cache_free_bulk(struct kmem_cache *s, size_t size, void **p)
 }
 EXPORT_SYMBOL(kmem_cache_free_bulk);
 
-#ifndef CONFIG_SLUB_TINY
 static inline
 int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
 			    void **p)
@@ -7371,14 +7404,8 @@ int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
 	local_lock_irqsave(&s->cpu_slab->lock, irqflags);
 
 	for (i = 0; i < size; i++) {
-		void *object = kfence_alloc(s, s->object_size, flags);
+		void *object = c->freelist;
 
-		if (unlikely(object)) {
-			p[i] = object;
-			continue;
-		}
-
-		object = c->freelist;
 		if (unlikely(!object)) {
 			/*
 			 * We may have removed an object from c->freelist using
@@ -7424,41 +7451,13 @@ error:
 	return 0;
 
 }
-#else /* CONFIG_SLUB_TINY */
-static int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags,
-				   size_t size, void **p)
-{
-	int i;
-
-	for (i = 0; i < size; i++) {
-		void *object = kfence_alloc(s, s->object_size, flags);
-
-		if (unlikely(object)) {
-			p[i] = object;
-			continue;
-		}
-
-		p[i] = __slab_alloc_node(s, flags, NUMA_NO_NODE,
-					 _RET_IP_, s->object_size);
-		if (unlikely(!p[i]))
-			goto error;
-
-		maybe_wipe_obj_freeptr(s, p[i]);
-	}
-
-	return i;
-
-error:
-	__kmem_cache_free_bulk(s, i, p);
-	return 0;
-}
-#endif /* CONFIG_SLUB_TINY */
 
 /* Note that interrupts must be enabled when calling this function. */
 int kmem_cache_alloc_bulk_noprof(struct kmem_cache *s, gfp_t flags, size_t size,
 				 void **p)
 {
 	unsigned int i = 0;
+	void *kfence_obj;
 
 	if (!size)
 		return 0;
@@ -7467,6 +7466,20 @@ int kmem_cache_alloc_bulk_noprof(struct kmem_cache *s, gfp_t flags, size_t size,
 	if (unlikely(!s))
 		return 0;
 
+	/*
+	 * to make things simpler, only assume at most once kfence allocated
+	 * object per bulk allocation and choose its index randomly
+	 */
+	kfence_obj = kfence_alloc(s, s->object_size, flags);
+
+	if (unlikely(kfence_obj)) {
+		if (unlikely(size == 1)) {
+			p[0] = kfence_obj;
+			goto out;
+		}
+		size--;
+	}
+
 	if (s->cpu_sheaves)
 		i = alloc_from_pcs_bulk(s, size, p);
 
@@ -7478,10 +7491,23 @@ int kmem_cache_alloc_bulk_noprof(struct kmem_cache *s, gfp_t flags, size_t size,
 		if (unlikely(__kmem_cache_alloc_bulk(s, flags, size - i, p + i) == 0)) {
 			if (i > 0)
 				__kmem_cache_free_bulk(s, i, p);
+			if (kfence_obj)
+				__kfence_free(kfence_obj);
 			return 0;
 		}
 	}
 
+	if (unlikely(kfence_obj)) {
+		int idx = get_random_u32_below(size + 1);
+
+		if (idx != size)
+			p[size] = p[idx];
+		p[idx] = kfence_obj;
+
+		size++;
+	}
+
+out:
 	/*
 	 * memcg and kmem_cache debug support and memory initialization.
 	 * Done outside of the IRQ disabled fastpath loop.
@@ -7643,7 +7669,6 @@ init_kmem_cache_node(struct kmem_cache_node *n, struct node_barn *barn)
 		barn_init(barn);
 }
 
-#ifndef CONFIG_SLUB_TINY
 static inline int alloc_kmem_cache_cpus(struct kmem_cache *s)
 {
 	BUILD_BUG_ON(PERCPU_DYNAMIC_EARLY_SIZE <
@@ -7664,12 +7689,6 @@ static inline int alloc_kmem_cache_cpus(struct kmem_cache *s)
 
 	return 1;
 }
-#else
-static inline int alloc_kmem_cache_cpus(struct kmem_cache *s)
-{
-	return 1;
-}
-#endif /* CONFIG_SLUB_TINY */
 
 static int init_percpu_sheaves(struct kmem_cache *s)
 {
@@ -7759,13 +7778,11 @@ void __kmem_cache_release(struct kmem_cache *s)
 	cache_random_seq_destroy(s);
 	if (s->cpu_sheaves)
 		pcs_destroy(s);
-#ifndef CONFIG_SLUB_TINY
 #ifdef CONFIG_PREEMPT_RT
 	if (s->cpu_slab)
 		lockdep_unregister_key(&s->lock_key);
 #endif
 	free_percpu(s->cpu_slab);
-#endif
 	free_kmem_cache_nodes(s);
 }
 
@@ -8519,10 +8536,8 @@ void __init kmem_cache_init(void)
 
 void __init kmem_cache_init_late(void)
 {
-#ifndef CONFIG_SLUB_TINY
 	flushwq = alloc_workqueue("slub_flushwq", WQ_MEM_RECLAIM, 0);
 	WARN_ON(!flushwq);
-#endif
 }
 
 struct kmem_cache *