Merge branch 'akpm' (patches from Andrew)

Merge misc updates from Andrew Morton:

 - a few misc things

 - the rest of MM

-  remove flex_arrays, replace with new simple radix-tree implementation

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (38 commits)
  Drop flex_arrays
  sctp: convert to genradix
  proc: commit to genradix
  generic radix trees
  selinux: convert to kvmalloc
  md: convert to kvmalloc
  openvswitch: convert to kvmalloc
  of: fix kmemleak crash caused by imbalance in early memory reservation
  mm: memblock: update comments and kernel-doc
  memblock: split checks whether a region should be skipped to a helper function
  memblock: remove memblock_{set,clear}_region_flags
  memblock: drop memblock_alloc_*_nopanic() variants
  memblock: memblock_alloc_try_nid: don't panic
  treewide: add checks for the return value of memblock_alloc*()
  swiotlb: add checks for the return value of memblock_alloc*()
  init/main: add checks for the return value of memblock_alloc*()
  mm/percpu: add checks for the return value of memblock_alloc*()
  sparc: add checks for the return value of memblock_alloc*()
  ia64: add checks for the return value of memblock_alloc*()
  arch: don't memset(0) memory returned by memblock_alloc()
  ...

4 files changed, 223 insertions, 400 deletions

diff --git a/lib/Makefile b/lib/Makefile
index 647517940b29..4e066120a0d6 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -35,10 +35,11 @@ obj-y	+= lockref.o
 
 obj-y += bcd.o div64.o sort.o parser.o debug_locks.o random32.o \
 	 bust_spinlocks.o kasprintf.o bitmap.o scatterlist.o \
-	 gcd.o lcm.o list_sort.o uuid.o flex_array.o iov_iter.o clz_ctz.o \
+	 gcd.o lcm.o list_sort.o uuid.o iov_iter.o clz_ctz.o \
 	 bsearch.o find_bit.o llist.o memweight.o kfifo.o \
 	 percpu-refcount.o rhashtable.o reciprocal_div.o \
-	 once.o refcount.o usercopy.o errseq.o bucket_locks.o
+	 once.o refcount.o usercopy.o errseq.o bucket_locks.o \
+	 generic-radix-tree.o
 obj-$(CONFIG_STRING_SELFTEST) += test_string.o
 obj-y += string_helpers.o
 obj-$(CONFIG_TEST_STRING_HELPERS) += test-string_helpers.o
diff --git a/lib/cpumask.c b/lib/cpumask.c
index 087a3e9a0202..0cb672eb107c 100644
--- a/lib/cpumask.c
+++ b/lib/cpumask.c
@@ -165,6 +165,9 @@ EXPORT_SYMBOL(zalloc_cpumask_var);
 void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
 	*mask = memblock_alloc(cpumask_size(), SMP_CACHE_BYTES);
+	if (!*mask)
+		panic("%s: Failed to allocate %u bytes\n", __func__,
+		      cpumask_size());
 }
 
 /**
diff --git a/lib/flex_array.c b/lib/flex_array.c
deleted file mode 100644
index 2eed22fa507c..000000000000
--- a/lib/flex_array.c
+++ /dev/null
@@ -1,398 +0,0 @@
-/*
- * Flexible array managed in PAGE_SIZE parts
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
- *
- * Copyright IBM Corporation, 2009
- *
- * Author: Dave Hansen <dave@linux.vnet.ibm.com>
- */
-
-#include <linux/flex_array.h>
-#include <linux/slab.h>
-#include <linux/stddef.h>
-#include <linux/export.h>
-#include <linux/reciprocal_div.h>
-
-struct flex_array_part {
-	char elements[FLEX_ARRAY_PART_SIZE];
-};
-
-/*
- * If a user requests an allocation which is small
- * enough, we may simply use the space in the
- * flex_array->parts[] array to store the user
- * data.
- */
-static inline int elements_fit_in_base(struct flex_array *fa)
-{
-	int data_size = fa->element_size * fa->total_nr_elements;
-	if (data_size <= FLEX_ARRAY_BASE_BYTES_LEFT)
-		return 1;
-	return 0;
-}
-
-/**
- * flex_array_alloc - allocate a new flexible array
- * @element_size:	the size of individual elements in the array
- * @total:		total number of elements that this should hold
- * @flags:		page allocation flags to use for base array
- *
- * Note: all locking must be provided by the caller.
- *
- * @total is used to size internal structures.  If the user ever
- * accesses any array indexes >=@total, it will produce errors.
- *
- * The maximum number of elements is defined as: the number of
- * elements that can be stored in a page times the number of
- * page pointers that we can fit in the base structure or (using
- * integer math):
- *
- * 	(PAGE_SIZE/element_size) * (PAGE_SIZE-8)/sizeof(void *)
- *
- * Here's a table showing example capacities.  Note that the maximum
- * index that the get/put() functions is just nr_objects-1.   This
- * basically means that you get 4MB of storage on 32-bit and 2MB on
- * 64-bit.
- *
- *
- * Element size | Objects | Objects |
- * PAGE_SIZE=4k |  32-bit |  64-bit |
- * ---------------------------------|
- *      1 bytes | 4177920 | 2088960 |
- *      2 bytes | 2088960 | 1044480 |
- *      3 bytes | 1392300 |  696150 |
- *      4 bytes | 1044480 |  522240 |
- *     32 bytes |  130560 |   65408 |
- *     33 bytes |  126480 |   63240 |
- *   2048 bytes |    2040 |    1020 |
- *   2049 bytes |    1020 |     510 |
- *       void * | 1044480 |  261120 |
- *
- * Since 64-bit pointers are twice the size, we lose half the
- * capacity in the base structure.  Also note that no effort is made
- * to efficiently pack objects across page boundaries.
- */
-struct flex_array *flex_array_alloc(int element_size, unsigned int total,
-					gfp_t flags)
-{
-	struct flex_array *ret;
-	int elems_per_part = 0;
-	int max_size = 0;
-	struct reciprocal_value reciprocal_elems = { 0 };
-
-	if (element_size) {
-		elems_per_part = FLEX_ARRAY_ELEMENTS_PER_PART(element_size);
-		reciprocal_elems = reciprocal_value(elems_per_part);
-		max_size = FLEX_ARRAY_NR_BASE_PTRS * elems_per_part;
-	}
-
-	/* max_size will end up 0 if element_size > PAGE_SIZE */
-	if (total > max_size)
-		return NULL;
-	ret = kzalloc(sizeof(struct flex_array), flags);
-	if (!ret)
-		return NULL;
-	ret->element_size = element_size;
-	ret->total_nr_elements = total;
-	ret->elems_per_part = elems_per_part;
-	ret->reciprocal_elems = reciprocal_elems;
-	if (elements_fit_in_base(ret) && !(flags & __GFP_ZERO))
-		memset(&ret->parts[0], FLEX_ARRAY_FREE,
-						FLEX_ARRAY_BASE_BYTES_LEFT);
-	return ret;
-}
-EXPORT_SYMBOL(flex_array_alloc);
-
-static int fa_element_to_part_nr(struct flex_array *fa,
-					unsigned int element_nr)
-{
-	/*
-	 * if element_size == 0 we don't get here, so we never touch
-	 * the zeroed fa->reciprocal_elems, which would yield invalid
-	 * results
-	 */
-	return reciprocal_divide(element_nr, fa->reciprocal_elems);
-}
-
-/**
- * flex_array_free_parts - just free the second-level pages
- * @fa:		the flex array from which to free parts
- *
- * This is to be used in cases where the base 'struct flex_array'
- * has been statically allocated and should not be free.
- */
-void flex_array_free_parts(struct flex_array *fa)
-{
-	int part_nr;
-
-	if (elements_fit_in_base(fa))
-		return;
-	for (part_nr = 0; part_nr < FLEX_ARRAY_NR_BASE_PTRS; part_nr++)
-		kfree(fa->parts[part_nr]);
-}
-EXPORT_SYMBOL(flex_array_free_parts);
-
-void flex_array_free(struct flex_array *fa)
-{
-	flex_array_free_parts(fa);
-	kfree(fa);
-}
-EXPORT_SYMBOL(flex_array_free);
-
-static unsigned int index_inside_part(struct flex_array *fa,
-					unsigned int element_nr,
-					unsigned int part_nr)
-{
-	unsigned int part_offset;
-
-	part_offset = element_nr - part_nr * fa->elems_per_part;
-	return part_offset * fa->element_size;
-}
-
-static struct flex_array_part *
-__fa_get_part(struct flex_array *fa, int part_nr, gfp_t flags)
-{
-	struct flex_array_part *part = fa->parts[part_nr];
-	if (!part) {
-		part = kmalloc(sizeof(struct flex_array_part), flags);
-		if (!part)
-			return NULL;
-		if (!(flags & __GFP_ZERO))
-			memset(part, FLEX_ARRAY_FREE,
-				sizeof(struct flex_array_part));
-		fa->parts[part_nr] = part;
-	}
-	return part;
-}
-
-/**
- * flex_array_put - copy data into the array at @element_nr
- * @fa:		the flex array to copy data into
- * @element_nr:	index of the position in which to insert
- * 		the new element.
- * @src:	address of data to copy into the array
- * @flags:	page allocation flags to use for array expansion
- *
- *
- * Note that this *copies* the contents of @src into
- * the array.  If you are trying to store an array of
- * pointers, make sure to pass in &ptr instead of ptr.
- * You may instead wish to use the flex_array_put_ptr()
- * helper function.
- *
- * Locking must be provided by the caller.
- */
-int flex_array_put(struct flex_array *fa, unsigned int element_nr, void *src,
-			gfp_t flags)
-{
-	int part_nr = 0;
-	struct flex_array_part *part;
-	void *dst;
-
-	if (element_nr >= fa->total_nr_elements)
-		return -ENOSPC;
-	if (!fa->element_size)
-		return 0;
-	if (elements_fit_in_base(fa))
-		part = (struct flex_array_part *)&fa->parts[0];
-	else {
-		part_nr = fa_element_to_part_nr(fa, element_nr);
-		part = __fa_get_part(fa, part_nr, flags);
-		if (!part)
-			return -ENOMEM;
-	}
-	dst = &part->elements[index_inside_part(fa, element_nr, part_nr)];
-	memcpy(dst, src, fa->element_size);
-	return 0;
-}
-EXPORT_SYMBOL(flex_array_put);
-
-/**
- * flex_array_clear - clear element in array at @element_nr
- * @fa:		the flex array of the element.
- * @element_nr:	index of the position to clear.
- *
- * Locking must be provided by the caller.
- */
-int flex_array_clear(struct flex_array *fa, unsigned int element_nr)
-{
-	int part_nr = 0;
-	struct flex_array_part *part;
-	void *dst;
-
-	if (element_nr >= fa->total_nr_elements)
-		return -ENOSPC;
-	if (!fa->element_size)
-		return 0;
-	if (elements_fit_in_base(fa))
-		part = (struct flex_array_part *)&fa->parts[0];
-	else {
-		part_nr = fa_element_to_part_nr(fa, element_nr);
-		part = fa->parts[part_nr];
-		if (!part)
-			return -EINVAL;
-	}
-	dst = &part->elements[index_inside_part(fa, element_nr, part_nr)];
-	memset(dst, FLEX_ARRAY_FREE, fa->element_size);
-	return 0;
-}
-EXPORT_SYMBOL(flex_array_clear);
-
-/**
- * flex_array_prealloc - guarantee that array space exists
- * @fa:			the flex array for which to preallocate parts
- * @start:		index of first array element for which space is allocated
- * @nr_elements:	number of elements for which space is allocated
- * @flags:		page allocation flags
- *
- * This will guarantee that no future calls to flex_array_put()
- * will allocate memory.  It can be used if you are expecting to
- * be holding a lock or in some atomic context while writing
- * data into the array.
- *
- * Locking must be provided by the caller.
- */
-int flex_array_prealloc(struct flex_array *fa, unsigned int start,
-			unsigned int nr_elements, gfp_t flags)
-{
-	int start_part;
-	int end_part;
-	int part_nr;
-	unsigned int end;
-	struct flex_array_part *part;
-
-	if (!start && !nr_elements)
-		return 0;
-	if (start >= fa->total_nr_elements)
-		return -ENOSPC;
-	if (!nr_elements)
-		return 0;
-
-	end = start + nr_elements - 1;
-
-	if (end >= fa->total_nr_elements)
-		return -ENOSPC;
-	if (!fa->element_size)
-		return 0;
-	if (elements_fit_in_base(fa))
-		return 0;
-	start_part = fa_element_to_part_nr(fa, start);
-	end_part = fa_element_to_part_nr(fa, end);
-	for (part_nr = start_part; part_nr <= end_part; part_nr++) {
-		part = __fa_get_part(fa, part_nr, flags);
-		if (!part)
-			return -ENOMEM;
-	}
-	return 0;
-}
-EXPORT_SYMBOL(flex_array_prealloc);
-
-/**
- * flex_array_get - pull data back out of the array
- * @fa:		the flex array from which to extract data
- * @element_nr:	index of the element to fetch from the array
- *
- * Returns a pointer to the data at index @element_nr.  Note
- * that this is a copy of the data that was passed in.  If you
- * are using this to store pointers, you'll get back &ptr.  You
- * may instead wish to use the flex_array_get_ptr helper.
- *
- * Locking must be provided by the caller.
- */
-void *flex_array_get(struct flex_array *fa, unsigned int element_nr)
-{
-	int part_nr = 0;
-	struct flex_array_part *part;
-
-	if (!fa->element_size)
-		return NULL;
-	if (element_nr >= fa->total_nr_elements)
-		return NULL;
-	if (elements_fit_in_base(fa))
-		part = (struct flex_array_part *)&fa->parts[0];
-	else {
-		part_nr = fa_element_to_part_nr(fa, element_nr);
-		part = fa->parts[part_nr];
-		if (!part)
-			return NULL;
-	}
-	return &part->elements[index_inside_part(fa, element_nr, part_nr)];
-}
-EXPORT_SYMBOL(flex_array_get);
-
-/**
- * flex_array_get_ptr - pull a ptr back out of the array
- * @fa:		the flex array from which to extract data
- * @element_nr:	index of the element to fetch from the array
- *
- * Returns the pointer placed in the flex array at element_nr using
- * flex_array_put_ptr().  This function should not be called if the
- * element in question was not set using the _put_ptr() helper.
- */
-void *flex_array_get_ptr(struct flex_array *fa, unsigned int element_nr)
-{
-	void **tmp;
-
-	tmp = flex_array_get(fa, element_nr);
-	if (!tmp)
-		return NULL;
-
-	return *tmp;
-}
-EXPORT_SYMBOL(flex_array_get_ptr);
-
-static int part_is_free(struct flex_array_part *part)
-{
-	int i;
-
-	for (i = 0; i < sizeof(struct flex_array_part); i++)
-		if (part->elements[i] != FLEX_ARRAY_FREE)
-			return 0;
-	return 1;
-}
-
-/**
- * flex_array_shrink - free unused second-level pages
- * @fa:		the flex array to shrink
- *
- * Frees all second-level pages that consist solely of unused
- * elements.  Returns the number of pages freed.
- *
- * Locking must be provided by the caller.
- */
-int flex_array_shrink(struct flex_array *fa)
-{
-	struct flex_array_part *part;
-	int part_nr;
-	int ret = 0;
-
-	if (!fa->total_nr_elements || !fa->element_size)
-		return 0;
-	if (elements_fit_in_base(fa))
-		return ret;
-	for (part_nr = 0; part_nr < FLEX_ARRAY_NR_BASE_PTRS; part_nr++) {
-		part = fa->parts[part_nr];
-		if (!part)
-			continue;
-		if (part_is_free(part)) {
-			fa->parts[part_nr] = NULL;
-			kfree(part);
-			ret++;
-		}
-	}
-	return ret;
-}
-EXPORT_SYMBOL(flex_array_shrink);
diff --git a/lib/generic-radix-tree.c b/lib/generic-radix-tree.c
new file mode 100644
index 000000000000..a7bafc413730
--- /dev/null
+++ b/lib/generic-radix-tree.c
@@ -0,0 +1,217 @@
+
+#include <linux/export.h>
+#include <linux/generic-radix-tree.h>
+#include <linux/gfp.h>
+
+#define GENRADIX_ARY		(PAGE_SIZE / sizeof(struct genradix_node *))
+#define GENRADIX_ARY_SHIFT	ilog2(GENRADIX_ARY)
+
+struct genradix_node {
+	union {
+		/* Interior node: */
+		struct genradix_node	*children[GENRADIX_ARY];
+
+		/* Leaf: */
+		u8			data[PAGE_SIZE];
+	};
+};
+
+static inline int genradix_depth_shift(unsigned depth)
+{
+	return PAGE_SHIFT + GENRADIX_ARY_SHIFT * depth;
+}
+
+/*
+ * Returns size (of data, in bytes) that a tree of a given depth holds:
+ */
+static inline size_t genradix_depth_size(unsigned depth)
+{
+	return 1UL << genradix_depth_shift(depth);
+}
+
+/* depth that's needed for a genradix that can address up to ULONG_MAX: */
+#define GENRADIX_MAX_DEPTH	\
+	DIV_ROUND_UP(BITS_PER_LONG - PAGE_SHIFT, GENRADIX_ARY_SHIFT)
+
+#define GENRADIX_DEPTH_MASK				\
+	((unsigned long) (roundup_pow_of_two(GENRADIX_MAX_DEPTH + 1) - 1))
+
+unsigned genradix_root_to_depth(struct genradix_root *r)
+{
+	return (unsigned long) r & GENRADIX_DEPTH_MASK;
+}
+
+struct genradix_node *genradix_root_to_node(struct genradix_root *r)
+{
+	return (void *) ((unsigned long) r & ~GENRADIX_DEPTH_MASK);
+}
+
+/*
+ * Returns pointer to the specified byte @offset within @radix, or NULL if not
+ * allocated
+ */
+void *__genradix_ptr(struct __genradix *radix, size_t offset)
+{
+	struct genradix_root *r = READ_ONCE(radix->root);
+	struct genradix_node *n = genradix_root_to_node(r);
+	unsigned level		= genradix_root_to_depth(r);
+
+	if (ilog2(offset) >= genradix_depth_shift(level))
+		return NULL;
+
+	while (1) {
+		if (!n)
+			return NULL;
+		if (!level)
+			break;
+
+		level--;
+
+		n = n->children[offset >> genradix_depth_shift(level)];
+		offset &= genradix_depth_size(level) - 1;
+	}
+
+	return &n->data[offset];
+}
+EXPORT_SYMBOL(__genradix_ptr);
+
+/*
+ * Returns pointer to the specified byte @offset within @radix, allocating it if
+ * necessary - newly allocated slots are always zeroed out:
+ */
+void *__genradix_ptr_alloc(struct __genradix *radix, size_t offset,
+			   gfp_t gfp_mask)
+{
+	struct genradix_root *v = READ_ONCE(radix->root);
+	struct genradix_node *n, *new_node = NULL;
+	unsigned level;
+
+	/* Increase tree depth if necessary: */
+	while (1) {
+		struct genradix_root *r = v, *new_root;
+
+		n	= genradix_root_to_node(r);
+		level	= genradix_root_to_depth(r);
+
+		if (n && ilog2(offset) < genradix_depth_shift(level))
+			break;
+
+		if (!new_node) {
+			new_node = (void *)
+				__get_free_page(gfp_mask|__GFP_ZERO);
+			if (!new_node)
+				return NULL;
+		}
+
+		new_node->children[0] = n;
+		new_root = ((struct genradix_root *)
+			    ((unsigned long) new_node | (n ? level + 1 : 0)));
+
+		if ((v = cmpxchg_release(&radix->root, r, new_root)) == r) {
+			v = new_root;
+			new_node = NULL;
+		}
+	}
+
+	while (level--) {
+		struct genradix_node **p =
+			&n->children[offset >> genradix_depth_shift(level)];
+		offset &= genradix_depth_size(level) - 1;
+
+		n = READ_ONCE(*p);
+		if (!n) {
+			if (!new_node) {
+				new_node = (void *)
+					__get_free_page(gfp_mask|__GFP_ZERO);
+				if (!new_node)
+					return NULL;
+			}
+
+			if (!(n = cmpxchg_release(p, NULL, new_node)))
+				swap(n, new_node);
+		}
+	}
+
+	if (new_node)
+		free_page((unsigned long) new_node);
+
+	return &n->data[offset];
+}
+EXPORT_SYMBOL(__genradix_ptr_alloc);
+
+void *__genradix_iter_peek(struct genradix_iter *iter,
+			   struct __genradix *radix,
+			   size_t objs_per_page)
+{
+	struct genradix_root *r;
+	struct genradix_node *n;
+	unsigned level, i;
+restart:
+	r = READ_ONCE(radix->root);
+	if (!r)
+		return NULL;
+
+	n	= genradix_root_to_node(r);
+	level	= genradix_root_to_depth(r);
+
+	if (ilog2(iter->offset) >= genradix_depth_shift(level))
+		return NULL;
+
+	while (level) {
+		level--;
+
+		i = (iter->offset >> genradix_depth_shift(level)) &
+			(GENRADIX_ARY - 1);
+
+		while (!n->children[i]) {
+			i++;
+			iter->offset = round_down(iter->offset +
+					   genradix_depth_size(level),
+					   genradix_depth_size(level));
+			iter->pos = (iter->offset >> PAGE_SHIFT) *
+				objs_per_page;
+			if (i == GENRADIX_ARY)
+				goto restart;
+		}
+
+		n = n->children[i];
+	}
+
+	return &n->data[iter->offset & (PAGE_SIZE - 1)];
+}
+EXPORT_SYMBOL(__genradix_iter_peek);
+
+static void genradix_free_recurse(struct genradix_node *n, unsigned level)
+{
+	if (level) {
+		unsigned i;
+
+		for (i = 0; i < GENRADIX_ARY; i++)
+			if (n->children[i])
+				genradix_free_recurse(n->children[i], level - 1);
+	}
+
+	free_page((unsigned long) n);
+}
+
+int __genradix_prealloc(struct __genradix *radix, size_t size,
+			gfp_t gfp_mask)
+{
+	size_t offset;
+
+	for (offset = 0; offset < size; offset += PAGE_SIZE)
+		if (!__genradix_ptr_alloc(radix, offset, gfp_mask))
+			return -ENOMEM;
+
+	return 0;
+}
+EXPORT_SYMBOL(__genradix_prealloc);
+
+void __genradix_free(struct __genradix *radix)
+{
+	struct genradix_root *r = xchg(&radix->root, NULL);
+
+	genradix_free_recurse(genradix_root_to_node(r),
+			      genradix_root_to_depth(r));
+}
+EXPORT_SYMBOL(__genradix_free);