Commit d6a71648 authored by Hyeonggon Yoo's avatar Hyeonggon Yoo Committed by Vlastimil Babka
Browse files

mm/slab: kmalloc: pass requests larger than order-1 page to page allocator 



There is not much benefit for serving large objects in kmalloc().
Let's pass large requests to page allocator like SLUB for better
maintenance of common code.

Signed-off-by: default avatarHyeonggon Yoo <42.hyeyoo@gmail.com>
Reviewed-by: default avatarVlastimil Babka <vbabka@suse.cz>
Signed-off-by: default avatarVlastimil Babka <vbabka@suse.cz>
parent c4cab557

include/linux/slab.h

+5 −18
Original line number Diff line number Diff line
@@ -243,27 +243,17 @@ static inline unsigned int arch_slab_minalign(void) 


#ifdef CONFIG_SLAB

/*

 * The largest kmalloc size supported by the SLAB allocators is

 * 32 megabyte (2^25) or the maximum allocatable page order if that is

 * less than 32 MB.

 *

 * WARNING: Its not easy to increase this value since the allocators have

 * to do various tricks to work around compiler limitations in order to

 * ensure proper constant folding.

 * SLAB and SLUB directly allocates requests fitting in to an order-1 page

 * (PAGE_SIZE*2).  Larger requests are passed to the page allocator.

 */

#define KMALLOC_SHIFT_HIGH	((MAX_ORDER + PAGE_SHIFT - 1) <= 25 ? \

				(MAX_ORDER + PAGE_SHIFT - 1) : 25)

#define KMALLOC_SHIFT_MAX	KMALLOC_SHIFT_HIGH

#define KMALLOC_SHIFT_HIGH	(PAGE_SHIFT + 1)

#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)

#ifndef KMALLOC_SHIFT_LOW

#define KMALLOC_SHIFT_LOW	5

#endif

#endif



#ifdef CONFIG_SLUB

/*

 * SLUB directly allocates requests fitting in to an order-1 page

 * (PAGE_SIZE*2).  Larger requests are passed to the page allocator.

 */

#define KMALLOC_SHIFT_HIGH	(PAGE_SHIFT + 1)

#define KMALLOC_SHIFT_MAX	(MAX_ORDER + PAGE_SHIFT - 1)

#ifndef KMALLOC_SHIFT_LOW
@@ -415,10 +405,6 @@ static __always_inline unsigned int __kmalloc_index(size_t size, 
	if (size <= 512 * 1024) return 19;

	if (size <= 1024 * 1024) return 20;

	if (size <=  2 * 1024 * 1024) return 21;

	if (size <=  4 * 1024 * 1024) return 22;

	if (size <=  8 * 1024 * 1024) return 23;

	if (size <=  16 * 1024 * 1024) return 24;

	if (size <=  32 * 1024 * 1024) return 25;



	if (!IS_ENABLED(CONFIG_PROFILE_ALL_BRANCHES) && size_is_constant)

		BUILD_BUG_ON_MSG(1, "unexpected size in kmalloc_index()");
@@ -428,6 +414,7 @@ static __always_inline unsigned int __kmalloc_index(size_t size, 
	/* Will never be reached. Needed because the compiler may complain */

	return -1;

}

static_assert(PAGE_SHIFT <= 20);

#define kmalloc_index(s) __kmalloc_index(s, true)

#endif /* !CONFIG_SLOB */

mm/slab.c

+42 −18
Original line number Diff line number Diff line
@@ -3585,11 +3585,19 @@ __do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller) 
	struct kmem_cache *cachep;

	void *ret;



	if (unlikely(size > KMALLOC_MAX_CACHE_SIZE))

		return NULL;

	if (unlikely(size > KMALLOC_MAX_CACHE_SIZE)) {

		ret = kmalloc_large_node_notrace(size, flags, node);



		trace_kmalloc_node(caller, ret, NULL, size,

				   PAGE_SIZE << get_order(size),

				   flags, node);

		return ret;

	}



	cachep = kmalloc_slab(size, flags);

	if (unlikely(ZERO_OR_NULL_PTR(cachep)))

		return cachep;



	ret = kmem_cache_alloc_node_trace(cachep, flags, node, size);

	ret = kasan_kmalloc(cachep, ret, size, flags);
@@ -3664,17 +3672,27 @@ EXPORT_SYMBOL(kmem_cache_free); 


void kmem_cache_free_bulk(struct kmem_cache *orig_s, size_t size, void **p)

{

	struct kmem_cache *s;

	size_t i;



	local_irq_disable();

	for (i = 0; i < size; i++) {

	for (int i = 0; i < size; i++) {

		void *objp = p[i];

		struct kmem_cache *s;



		if (!orig_s) /* called via kfree_bulk */

			s = virt_to_cache(objp);

		else

		if (!orig_s) {

			struct folio *folio = virt_to_folio(objp);



			/* called via kfree_bulk */

			if (!folio_test_slab(folio)) {

				local_irq_enable();

				free_large_kmalloc(folio, objp);

				local_irq_disable();

				continue;

			}

			s = folio_slab(folio)->slab_cache;

		} else {

			s = cache_from_obj(orig_s, objp);

		}



		if (!s)

			continue;
@@ -3703,20 +3721,24 @@ void kfree(const void *objp) 
{

	struct kmem_cache *c;

	unsigned long flags;

	struct folio *folio;



	trace_kfree(_RET_IP_, objp);



	if (unlikely(ZERO_OR_NULL_PTR(objp)))

		return;

	local_irq_save(flags);

	kfree_debugcheck(objp);

	c = virt_to_cache(objp);

	if (!c) {

		local_irq_restore(flags);



	folio = virt_to_folio(objp);

	if (!folio_test_slab(folio)) {

		free_large_kmalloc(folio, (void *)objp);

		return;

	}

	debug_check_no_locks_freed(objp, c->object_size);



	c = folio_slab(folio)->slab_cache;



	local_irq_save(flags);

	kfree_debugcheck(objp);

	debug_check_no_locks_freed(objp, c->object_size);

	debug_check_no_obj_freed(objp, c->object_size);

	__cache_free(c, (void *)objp, _RET_IP_);

	local_irq_restore(flags);
@@ -4138,15 +4160,17 @@ void __check_heap_object(const void *ptr, unsigned long n, 
size_t __ksize(const void *objp)

{

	struct kmem_cache *c;

	size_t size;

	struct folio *folio;



	BUG_ON(!objp);

	if (unlikely(objp == ZERO_SIZE_PTR))

		return 0;



	c = virt_to_cache(objp);

	size = c ? c->object_size : 0;

	folio = virt_to_folio(objp);

	if (!folio_test_slab(folio))

		return folio_size(folio);



	return size;

	c = folio_slab(folio)->slab_cache;

	return c->object_size;

}

EXPORT_SYMBOL(__ksize);

mm/slab.h

+3 −0
Original line number Diff line number Diff line
@@ -660,6 +660,9 @@ static inline struct kmem_cache *cache_from_obj(struct kmem_cache *s, void *x) 
		print_tracking(cachep, x);

	return cachep;

}



void free_large_kmalloc(struct folio *folio, void *object);



#endif /* CONFIG_SLOB */



static inline size_t slab_ksize(const struct kmem_cache *s)

mm/slab_common.c

+18 −7
Original line number Diff line number Diff line
@@ -744,8 +744,8 @@ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) 


/*

 * kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time.

 * kmalloc_index() supports up to 2^25=32MB, so the final entry of the table is

 * kmalloc-32M.

 * kmalloc_index() supports up to 2^21=2MB, so the final entry of the table is

 * kmalloc-2M.

 */

const struct kmalloc_info_struct kmalloc_info[] __initconst = {

	INIT_KMALLOC_INFO(0, 0),
@@ -769,11 +769,7 @@ const struct kmalloc_info_struct kmalloc_info[] __initconst = { 
	INIT_KMALLOC_INFO(262144, 256k),

	INIT_KMALLOC_INFO(524288, 512k),

	INIT_KMALLOC_INFO(1048576, 1M),

	INIT_KMALLOC_INFO(2097152, 2M),

	INIT_KMALLOC_INFO(4194304, 4M),

	INIT_KMALLOC_INFO(8388608, 8M),

	INIT_KMALLOC_INFO(16777216, 16M),

	INIT_KMALLOC_INFO(33554432, 32M)

	INIT_KMALLOC_INFO(2097152, 2M)

};



/*
@@ -886,6 +882,21 @@ void __init create_kmalloc_caches(slab_flags_t flags) 
	/* Kmalloc array is now usable */

	slab_state = UP;

}



void free_large_kmalloc(struct folio *folio, void *object)

{

	unsigned int order = folio_order(folio);



	if (WARN_ON_ONCE(order == 0))

		pr_warn_once("object pointer: 0x%p\n", object);



	kmemleak_free(object);

	kasan_kfree_large(object);



	mod_lruvec_page_state(folio_page(folio, 0), NR_SLAB_UNRECLAIMABLE_B,

			      -(PAGE_SIZE << order));

	__free_pages(folio_page(folio, 0), order);

}

#endif /* !CONFIG_SLOB */



gfp_t kmalloc_fix_flags(gfp_t flags)

mm/slub.c

+0 −19
Original line number Diff line number Diff line
@@ -1704,12 +1704,6 @@ static bool freelist_corrupted(struct kmem_cache *s, struct slab *slab, 
 * Hooks for other subsystems that check memory allocations. In a typical

 * production configuration these hooks all should produce no code at all.

 */

static __always_inline void kfree_hook(void *x)

{

	kmemleak_free(x);

	kasan_kfree_large(x);

}



static __always_inline bool slab_free_hook(struct kmem_cache *s,

						void *x, bool init)

{
@@ -3550,19 +3544,6 @@ struct detached_freelist { 
	struct kmem_cache *s;

};



static inline void free_large_kmalloc(struct folio *folio, void *object)

{

	unsigned int order = folio_order(folio);



	if (WARN_ON_ONCE(order == 0))

		pr_warn_once("object pointer: 0x%p\n", object);



	kfree_hook(object);

	mod_lruvec_page_state(folio_page(folio, 0), NR_SLAB_UNRECLAIMABLE_B,

			      -(PAGE_SIZE << order));

	__free_pages(folio_page(folio, 0), order);

}



/*

 * This function progressively scans the array with free objects (with

 * a limited look ahead) and extract objects belonging to the same