Merge tag 'slab-for-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab

Pull slab updates from Vlastimil Babka:

 - SLOB deprecation and SLUB_TINY

   The SLOB allocator adds maintenance burden and stands in the way of
   API improvements [1]. Deprecate it by renaming the config option (to
   make users notice) to CONFIG_SLOB_DEPRECATED with updated help text.
   SLUB should be used instead as SLAB will be the next on the removal
   list.

   Based on reports from a riscv k210 board with 8MB RAM, add a
   CONFIG_SLUB_TINY option to minimize SLUB's memory usage at the
   expense of scalability. This has resolved the k210 regression [2] so
   in case there are no others (that wouldn't be resolvable by further
   tweaks to SLUB_TINY) plan is to remove SLOB in a few cycles.

   Existing defconfigs with CONFIG_SLOB are converted to
   CONFIG_SLUB_TINY.

 - kmalloc() slub_debug redzone improvements

   A series from Feng Tang that builds on the tracking or requested size
   for kmalloc() allocations (for caches with debugging enabled) added
   in 6.1, to make redzone checks consider the requested size and not
   the rounded up one, in order to catch more subtle buffer overruns.
   Includes new slub_kunit test.

 - struct slab fields reordering to accomodate larger rcu_head

   RCU folks would like to grow rcu_head with debugging options, which
   breaks current struct slab layout's assumptions, so reorganize it to
   make this possible.

 - Miscellaneous improvements/fixes:
     - __alloc_size checking compiler workaround (Kees Cook)
     - Optimize and cleanup SLUB's sysfs init (Rasmus Villemoes)
     - Make SLAB compatible with PROVE_RAW_LOCK_NESTING (Jiri Kosina)
     - Correct SLUB's percpu allocation estimates (Baoquan He)
     - Re-enableS LUB's run-time failslab sysfs control (Alexander Atanasov)
     - Make tools/vm/slabinfo more user friendly when not run as root (Rong Tao)
     - Dead code removal in SLUB (Hyeonggon Yoo)

* tag 'slab-for-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab: (31 commits)
  mm, slob: rename CONFIG_SLOB to CONFIG_SLOB_DEPRECATED
  mm, slub: don't aggressively inline with CONFIG_SLUB_TINY
  mm, slub: remove percpu slabs with CONFIG_SLUB_TINY
  mm, slub: split out allocations from pre/post hooks
  mm/slub, kunit: Add a test case for kmalloc redzone check
  mm/slub, kunit: add SLAB_SKIP_KFENCE flag for cache creation
  mm, slub: refactor free debug processing
  mm, slab: ignore SLAB_RECLAIM_ACCOUNT with CONFIG_SLUB_TINY
  mm, slub: don't create kmalloc-rcl caches with CONFIG_SLUB_TINY
  mm, slub: lower the default slub_max_order with CONFIG_SLUB_TINY
  mm, slub: retain no free slabs on partial list with CONFIG_SLUB_TINY
  mm, slub: disable SYSFS support with CONFIG_SLUB_TINY
  mm, slub: add CONFIG_SLUB_TINY
  mm, slab: ignore hardened usercopy parameters when disabled
  slab: Remove special-casing of const 0 size allocations
  slab: Clean up SLOB vs kmalloc() definition
  mm/sl[au]b: rearrange struct slab fields to allow larger rcu_head
  mm/migrate: make isolate_movable_page() skip slab pages
  mm/slab: move and adjust kernel-doc for kmem_cache_alloc
  mm/slub, percpu: correct the calculation of early percpu allocation size
  ...

1 files changed, 42 insertions, 22 deletions

diff --git a/include/linux/slab.h b/include/linux/slab.h
index e7534a7b2daf..45af70315a94 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -140,7 +140,11 @@
 
 /* The following flags affect the page allocator grouping pages by mobility */
 /* Objects are reclaimable */
+#ifndef CONFIG_SLUB_TINY
 #define SLAB_RECLAIM_ACCOUNT	((slab_flags_t __force)0x00020000U)
+#else
+#define SLAB_RECLAIM_ACCOUNT	((slab_flags_t __force)0)
+#endif
 #define SLAB_TEMPORARY		SLAB_RECLAIM_ACCOUNT	/* Objects are short-lived */
 
 /*
@@ -347,13 +351,18 @@ enum kmalloc_cache_type {
 #endif
 #ifndef CONFIG_MEMCG_KMEM
 	KMALLOC_CGROUP = KMALLOC_NORMAL,
-#else
-	KMALLOC_CGROUP,
 #endif
+#ifdef CONFIG_SLUB_TINY
+	KMALLOC_RECLAIM = KMALLOC_NORMAL,
+#else
 	KMALLOC_RECLAIM,
+#endif
 #ifdef CONFIG_ZONE_DMA
 	KMALLOC_DMA,
 #endif
+#ifdef CONFIG_MEMCG_KMEM
+	KMALLOC_CGROUP,
+#endif
 	NR_KMALLOC_TYPES
 };
 
@@ -452,7 +461,18 @@ static_assert(PAGE_SHIFT <= 20);
 #endif /* !CONFIG_SLOB */
 
 void *__kmalloc(size_t size, gfp_t flags) __assume_kmalloc_alignment __alloc_size(1);
-void *kmem_cache_alloc(struct kmem_cache *s, gfp_t flags) __assume_slab_alignment __malloc;
+
+/**
+ * kmem_cache_alloc - Allocate an object
+ * @cachep: The cache to allocate from.
+ * @flags: See kmalloc().
+ *
+ * Allocate an object from this cache.
+ * See kmem_cache_zalloc() for a shortcut of adding __GFP_ZERO to flags.
+ *
+ * Return: pointer to the new object or %NULL in case of error
+ */
+void *kmem_cache_alloc(struct kmem_cache *cachep, gfp_t flags) __assume_slab_alignment __malloc;
 void *kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
 			   gfp_t gfpflags) __assume_slab_alignment __malloc;
 void kmem_cache_free(struct kmem_cache *s, void *objp);
@@ -494,9 +514,9 @@ void *kmalloc_large_node(size_t size, gfp_t flags, int node) __assume_page_align
 							     __alloc_size(1);
 
 /**
- * kmalloc - allocate memory
+ * kmalloc - allocate kernel memory
  * @size: how many bytes of memory are required.
- * @flags: the type of memory to allocate.
+ * @flags: describe the allocation context
  *
  * kmalloc is the normal method of allocating memory
  * for objects smaller than page size in the kernel.
@@ -523,12 +543,12 @@ void *kmalloc_large_node(size_t size, gfp_t flags, int node) __assume_page_align
  * %GFP_ATOMIC
  *	Allocation will not sleep.  May use emergency pools.
  *
- * %GFP_HIGHUSER
- *	Allocate memory from high memory on behalf of user.
- *
  * Also it is possible to set different flags by OR'ing
  * in one or more of the following additional @flags:
  *
+ * %__GFP_ZERO
+ *	Zero the allocated memory before returning. Also see kzalloc().
+ *
  * %__GFP_HIGH
  *	This allocation has high priority and may use emergency pools.
  *
@@ -547,42 +567,42 @@ void *kmalloc_large_node(size_t size, gfp_t flags, int node) __assume_page_align
  *	Try really hard to succeed the allocation but fail
  *	eventually.
  */
+#ifndef CONFIG_SLOB
 static __always_inline __alloc_size(1) void *kmalloc(size_t size, gfp_t flags)
 {
-	if (__builtin_constant_p(size)) {
-#ifndef CONFIG_SLOB
+	if (__builtin_constant_p(size) && size) {
 		unsigned int index;
-#endif
+
 		if (size > KMALLOC_MAX_CACHE_SIZE)
 			return kmalloc_large(size, flags);
-#ifndef CONFIG_SLOB
-		index = kmalloc_index(size);
-
-		if (!index)
-			return ZERO_SIZE_PTR;
 
+		index = kmalloc_index(size);
 		return kmalloc_trace(
 				kmalloc_caches[kmalloc_type(flags)][index],
 				flags, size);
-#endif
 	}
 	return __kmalloc(size, flags);
 }
+#else
+static __always_inline __alloc_size(1) void *kmalloc(size_t size, gfp_t flags)
+{
+	if (__builtin_constant_p(size) && size > KMALLOC_MAX_CACHE_SIZE)
+		return kmalloc_large(size, flags);
+
+	return __kmalloc(size, flags);
+}
+#endif
 
 #ifndef CONFIG_SLOB
 static __always_inline __alloc_size(1) void *kmalloc_node(size_t size, gfp_t flags, int node)
 {
-	if (__builtin_constant_p(size)) {
+	if (__builtin_constant_p(size) && size) {
 		unsigned int index;
 
 		if (size > KMALLOC_MAX_CACHE_SIZE)
 			return kmalloc_large_node(size, flags, node);
 
 		index = kmalloc_index(size);
-
-		if (!index)
-			return ZERO_SIZE_PTR;
-
 		return kmalloc_node_trace(
 				kmalloc_caches[kmalloc_type(flags)][index],
 				flags, node, size);