* - define CONFIG_HAVE_DYNAMIC_PER_CPU_AREA
*
* - define __addr_to_pcpu_ptr() and __pcpu_ptr_to_addr() to translate
- * regular address to percpu pointer and back
+ * regular address to percpu pointer and back if they need to be
+ * different from the default
*
- * - use pcpu_setup_static() during percpu area initialization to
- * setup kernel static percpu area
+ * - use pcpu_setup_first_chunk() during percpu area initialization to
+ * setup the first chunk containing the kernel static percpu area
*/
#include <linux/bitmap.h>
#include <linux/pfn.h>
#include <linux/rbtree.h>
#include <linux/slab.h>
+#include <linux/spinlock.h>
#include <linux/vmalloc.h>
+#include <linux/workqueue.h>
#include <asm/cacheflush.h>
+#include <asm/sections.h>
#include <asm/tlbflush.h>
-#define PCPU_MIN_UNIT_PAGES_SHIFT 4 /* also max alloc size */
#define PCPU_SLOT_BASE_SHIFT 5 /* 1-31 shares the same slot */
#define PCPU_DFL_MAP_ALLOC 16 /* start a map with 16 ents */
+/* default addr <-> pcpu_ptr mapping, override in asm/percpu.h if necessary */
+#ifndef __addr_to_pcpu_ptr
+#define __addr_to_pcpu_ptr(addr) \
+ (void *)((unsigned long)(addr) - (unsigned long)pcpu_base_addr \
+ + (unsigned long)__per_cpu_start)
+#endif
+#ifndef __pcpu_ptr_to_addr
+#define __pcpu_ptr_to_addr(ptr) \
+ (void *)((unsigned long)(ptr) + (unsigned long)pcpu_base_addr \
+ - (unsigned long)__per_cpu_start)
+#endif
+
struct pcpu_chunk {
struct list_head list; /* linked to pcpu_slot lists */
struct rb_node rb_node; /* key is chunk->vm->addr */
int map_used; /* # of map entries used */
int map_alloc; /* # of map entries allocated */
int *map; /* allocation map */
- struct page *page[]; /* #cpus * UNIT_PAGES */
+ bool immutable; /* no [de]population allowed */
+ struct page **page; /* points to page array */
+ struct page *page_ar[]; /* #cpus * UNIT_PAGES */
};
-static int pcpu_unit_pages_shift;
-static int pcpu_unit_pages;
-static int pcpu_unit_shift;
-static int pcpu_unit_size;
-static int pcpu_chunk_size;
-static int pcpu_nr_slots;
-static size_t pcpu_chunk_struct_size;
+static int pcpu_unit_pages __read_mostly;
+static int pcpu_unit_size __read_mostly;
+static int pcpu_chunk_size __read_mostly;
+static int pcpu_nr_slots __read_mostly;
+static size_t pcpu_chunk_struct_size __read_mostly;
/* the address of the first chunk which starts with the kernel static area */
-void *pcpu_base_addr;
+void *pcpu_base_addr __read_mostly;
EXPORT_SYMBOL_GPL(pcpu_base_addr);
-/* the size of kernel static area */
-static int pcpu_static_size;
+/* optional reserved chunk, only accessible for reserved allocations */
+static struct pcpu_chunk *pcpu_reserved_chunk;
+/* offset limit of the reserved chunk */
+static int pcpu_reserved_chunk_limit;
/*
- * One mutex to rule them all.
- *
- * The following mutex is grabbed in the outermost public alloc/free
- * interface functions and released only when the operation is
- * complete. As such, every function in this file other than the
- * outermost functions are called under pcpu_mutex.
- *
- * It can easily be switched to use spinlock such that only the area
- * allocation and page population commit are protected with it doing
- * actual [de]allocation without holding any lock. However, given
- * what this allocator does, I think it's better to let them run
- * sequentially.
+ * Synchronization rules.
+ *
+ * There are two locks - pcpu_alloc_mutex and pcpu_lock. The former
+ * protects allocation/reclaim paths, chunks and chunk->page arrays.
+ * The latter is a spinlock and protects the index data structures -
+ * chunk slots, rbtree, chunks and area maps in chunks.
+ *
+ * During allocation, pcpu_alloc_mutex is kept locked all the time and
+ * pcpu_lock is grabbed and released as necessary. All actual memory
+ * allocations are done using GFP_KERNEL with pcpu_lock released.
+ *
+ * Free path accesses and alters only the index data structures, so it
+ * can be safely called from atomic context. When memory needs to be
+ * returned to the system, free path schedules reclaim_work which
+ * grabs both pcpu_alloc_mutex and pcpu_lock, unlinks chunks to be
+ * reclaimed, release both locks and frees the chunks. Note that it's
+ * necessary to grab both locks to remove a chunk from circulation as
+ * allocation path might be referencing the chunk with only
+ * pcpu_alloc_mutex locked.
*/
-static DEFINE_MUTEX(pcpu_mutex);
+static DEFINE_MUTEX(pcpu_alloc_mutex); /* protects whole alloc and reclaim */
+static DEFINE_SPINLOCK(pcpu_lock); /* protects index data structures */
-static struct list_head *pcpu_slot; /* chunk list slots */
+static struct list_head *pcpu_slot __read_mostly; /* chunk list slots */
static struct rb_root pcpu_addr_root = RB_ROOT; /* chunks by address */
-static int pcpu_size_to_slot(int size)
+/* reclaim work to release fully free chunks, scheduled from free path */
+static void pcpu_reclaim(struct work_struct *work);
+static DECLARE_WORK(pcpu_reclaim_work, pcpu_reclaim);
+
+static int __pcpu_size_to_slot(int size)
{
int highbit = fls(size); /* size is in bytes */
return max(highbit - PCPU_SLOT_BASE_SHIFT + 2, 1);
}
+static int pcpu_size_to_slot(int size)
+{
+ if (size == pcpu_unit_size)
+ return pcpu_nr_slots - 1;
+ return __pcpu_size_to_slot(size);
+}
+
static int pcpu_chunk_slot(const struct pcpu_chunk *chunk)
{
if (chunk->free_size < sizeof(int) || chunk->contig_hint < sizeof(int))
static int pcpu_page_idx(unsigned int cpu, int page_idx)
{
- return (cpu << pcpu_unit_pages_shift) + page_idx;
+ return cpu * pcpu_unit_pages + page_idx;
}
static struct page **pcpu_chunk_pagep(struct pcpu_chunk *chunk,
}
/**
- * pcpu_realloc - versatile realloc
- * @p: the current pointer (can be NULL for new allocations)
- * @size: the current size in bytes (can be 0 for new allocations)
- * @new_size: the wanted new size in bytes (can be 0 for free)
+ * pcpu_mem_alloc - allocate memory
+ * @size: bytes to allocate
+ *
+ * Allocate @size bytes. If @size is smaller than PAGE_SIZE,
+ * kzalloc() is used; otherwise, vmalloc() is used. The returned
+ * memory is always zeroed.
*
- * More robust realloc which can be used to allocate, resize or free a
- * memory area of arbitrary size. If the needed size goes over
- * PAGE_SIZE, kernel VM is used.
+ * CONTEXT:
+ * Does GFP_KERNEL allocation.
*
* RETURNS:
- * The new pointer on success, NULL on failure.
+ * Pointer to the allocated area on success, NULL on failure.
*/
-static void *pcpu_realloc(void *p, size_t size, size_t new_size)
+static void *pcpu_mem_alloc(size_t size)
{
- void *new;
-
- if (new_size <= PAGE_SIZE)
- new = kmalloc(new_size, GFP_KERNEL);
- else
- new = vmalloc(new_size);
- if (new_size && !new)
- return NULL;
-
- memcpy(new, p, min(size, new_size));
- if (new_size > size)
- memset(new + size, 0, new_size - size);
+ if (size <= PAGE_SIZE)
+ return kzalloc(size, GFP_KERNEL);
+ else {
+ void *ptr = vmalloc(size);
+ if (ptr)
+ memset(ptr, 0, size);
+ return ptr;
+ }
+}
+/**
+ * pcpu_mem_free - free memory
+ * @ptr: memory to free
+ * @size: size of the area
+ *
+ * Free @ptr. @ptr should have been allocated using pcpu_mem_alloc().
+ */
+static void pcpu_mem_free(void *ptr, size_t size)
+{
if (size <= PAGE_SIZE)
- kfree(p);
+ kfree(ptr);
else
- vfree(p);
-
- return new;
+ vfree(ptr);
}
/**
*
* This function is called after an allocation or free changed @chunk.
* New slot according to the changed state is determined and @chunk is
- * moved to the slot.
+ * moved to the slot. Note that the reserved chunk is never put on
+ * chunk slots.
+ *
+ * CONTEXT:
+ * pcpu_lock.
*/
static void pcpu_chunk_relocate(struct pcpu_chunk *chunk, int oslot)
{
int nslot = pcpu_chunk_slot(chunk);
- if (oslot != nslot) {
+ if (chunk != pcpu_reserved_chunk && oslot != nslot) {
if (oslot < nslot)
list_move(&chunk->list, &pcpu_slot[nslot]);
else
* searchs for the chunk with the highest start address which isn't
* beyond @addr.
*
+ * CONTEXT:
+ * pcpu_lock.
+ *
* RETURNS:
* The address of the found chunk.
*/
struct rb_node *n, *parent;
struct pcpu_chunk *chunk;
+ /* is it in the reserved chunk? */
+ if (pcpu_reserved_chunk) {
+ void *start = pcpu_reserved_chunk->vm->addr;
+
+ if (addr >= start && addr < start + pcpu_reserved_chunk_limit)
+ return pcpu_reserved_chunk;
+ }
+
+ /* nah... search the regular ones */
n = *pcpu_chunk_rb_search(addr, &parent);
if (!n) {
/* no exactly matching chunk, the parent is the closest */
* @new: chunk to insert
*
* Insert @new into address rb tree.
+ *
+ * CONTEXT:
+ * pcpu_lock.
*/
static void pcpu_chunk_addr_insert(struct pcpu_chunk *new)
{
}
/**
+ * pcpu_extend_area_map - extend area map for allocation
+ * @chunk: target chunk
+ *
+ * Extend area map of @chunk so that it can accomodate an allocation.
+ * A single allocation can split an area into three areas, so this
+ * function makes sure that @chunk->map has at least two extra slots.
+ *
+ * CONTEXT:
+ * pcpu_alloc_mutex, pcpu_lock. pcpu_lock is released and reacquired
+ * if area map is extended.
+ *
+ * RETURNS:
+ * 0 if noop, 1 if successfully extended, -errno on failure.
+ */
+static int pcpu_extend_area_map(struct pcpu_chunk *chunk)
+{
+ int new_alloc;
+ int *new;
+ size_t size;
+
+ /* has enough? */
+ if (chunk->map_alloc >= chunk->map_used + 2)
+ return 0;
+
+ spin_unlock_irq(&pcpu_lock);
+
+ new_alloc = PCPU_DFL_MAP_ALLOC;
+ while (new_alloc < chunk->map_used + 2)
+ new_alloc *= 2;
+
+ new = pcpu_mem_alloc(new_alloc * sizeof(new[0]));
+ if (!new) {
+ spin_lock_irq(&pcpu_lock);
+ return -ENOMEM;
+ }
+
+ /*
+ * Acquire pcpu_lock and switch to new area map. Only free
+ * could have happened inbetween, so map_used couldn't have
+ * grown.
+ */
+ spin_lock_irq(&pcpu_lock);
+ BUG_ON(new_alloc < chunk->map_used + 2);
+
+ size = chunk->map_alloc * sizeof(chunk->map[0]);
+ memcpy(new, chunk->map, size);
+
+ /*
+ * map_alloc < PCPU_DFL_MAP_ALLOC indicates that the chunk is
+ * one of the first chunks and still using static map.
+ */
+ if (chunk->map_alloc >= PCPU_DFL_MAP_ALLOC)
+ pcpu_mem_free(chunk->map, size);
+
+ chunk->map_alloc = new_alloc;
+ chunk->map = new;
+ return 0;
+}
+
+/**
* pcpu_split_block - split a map block
* @chunk: chunk of interest
* @i: index of map block to split
* depending on @head, is reduced by @tail bytes and @tail byte block
* is inserted after the target block.
*
- * RETURNS:
- * 0 on success, -errno on failure.
+ * @chunk->map must have enough free slots to accomodate the split.
+ *
+ * CONTEXT:
+ * pcpu_lock.
*/
-static int pcpu_split_block(struct pcpu_chunk *chunk, int i, int head, int tail)
+static void pcpu_split_block(struct pcpu_chunk *chunk, int i,
+ int head, int tail)
{
int nr_extra = !!head + !!tail;
- int target = chunk->map_used + nr_extra;
- /* reallocation required? */
- if (chunk->map_alloc < target) {
- int new_alloc = chunk->map_alloc;
- int *new;
+ BUG_ON(chunk->map_alloc < chunk->map_used + nr_extra);
- while (new_alloc < target)
- new_alloc *= 2;
-
- new = pcpu_realloc(chunk->map,
- chunk->map_alloc * sizeof(new[0]),
- new_alloc * sizeof(new[0]));
- if (!new)
- return -ENOMEM;
-
- chunk->map_alloc = new_alloc;
- chunk->map = new;
- }
-
- /* insert a new subblock */
+ /* insert new subblocks */
memmove(&chunk->map[i + nr_extra], &chunk->map[i],
sizeof(chunk->map[0]) * (chunk->map_used - i));
chunk->map_used += nr_extra;
chunk->map[i++] -= tail;
chunk->map[i] = tail;
}
- return 0;
}
/**
* Note that this function only allocates the offset. It doesn't
* populate or map the area.
*
+ * @chunk->map must have at least two free slots.
+ *
+ * CONTEXT:
+ * pcpu_lock.
+ *
* RETURNS:
- * Allocated offset in @chunk on success, -errno on failure.
+ * Allocated offset in @chunk on success, -1 if no matching area is
+ * found.
*/
static int pcpu_alloc_area(struct pcpu_chunk *chunk, int size, int align)
{
int max_contig = 0;
int i, off;
- /*
- * The static chunk initially doesn't have map attached
- * because kmalloc wasn't available during init. Give it one.
- */
- if (unlikely(!chunk->map)) {
- chunk->map = pcpu_realloc(NULL, 0,
- PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
- if (!chunk->map)
- return -ENOMEM;
-
- chunk->map_alloc = PCPU_DFL_MAP_ALLOC;
- chunk->map[chunk->map_used++] = -pcpu_static_size;
- if (chunk->free_size)
- chunk->map[chunk->map_used++] = chunk->free_size;
- }
-
for (i = 0, off = 0; i < chunk->map_used; off += abs(chunk->map[i++])) {
bool is_last = i + 1 == chunk->map_used;
int head, tail;
/* split if warranted */
if (head || tail) {
- if (pcpu_split_block(chunk, i, head, tail))
- return -ENOMEM;
+ pcpu_split_block(chunk, i, head, tail);
if (head) {
i++;
off += head;
chunk->contig_hint = max_contig; /* fully scanned */
pcpu_chunk_relocate(chunk, oslot);
- /*
- * Tell the upper layer that this chunk has no area left.
- * Note that this is not an error condition but a notification
- * to upper layer that it needs to look at other chunks.
- * -ENOSPC is chosen as it isn't used in memory subsystem and
- * matches the meaning in a way.
- */
- return -ENOSPC;
+ /* tell the upper layer that this chunk has no matching area */
+ return -1;
}
/**
* Free area starting from @freeme to @chunk. Note that this function
* only modifies the allocation map. It doesn't depopulate or unmap
* the area.
+ *
+ * CONTEXT:
+ * pcpu_lock.
*/
static void pcpu_free_area(struct pcpu_chunk *chunk, int freeme)
{
unsigned int last = num_possible_cpus() - 1;
unsigned int cpu;
+ /* unmap must not be done on immutable chunk */
+ WARN_ON(chunk->immutable);
+
/*
* Each flushing trial can be very expensive, issue flush on
* the whole region at once rather than doing it for each cpu.
* For each cpu, depopulate and unmap pages [@page_start,@page_end)
* from @chunk. If @flush is true, vcache is flushed before unmapping
* and tlb after.
+ *
+ * CONTEXT:
+ * pcpu_alloc_mutex.
*/
static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk, int off, int size,
bool flush)
unsigned int cpu;
int err;
+ /* map must not be done on immutable chunk */
+ WARN_ON(chunk->immutable);
+
for_each_possible_cpu(cpu) {
err = map_kernel_range_noflush(
pcpu_chunk_addr(chunk, cpu, page_start),
*
* For each cpu, populate and map pages [@page_start,@page_end) into
* @chunk. The area is cleared on return.
+ *
+ * CONTEXT:
+ * pcpu_alloc_mutex, does GFP_KERNEL allocation.
*/
static int pcpu_populate_chunk(struct pcpu_chunk *chunk, int off, int size)
{
int page_start = PFN_DOWN(off);
int page_end = PFN_UP(off + size);
int map_start = -1;
- int map_end;
+ int uninitialized_var(map_end);
unsigned int cpu;
int i;
goto err;
for_each_possible_cpu(cpu)
- memset(chunk->vm->addr + (cpu << pcpu_unit_shift) + off, 0,
+ memset(chunk->vm->addr + cpu * pcpu_unit_size + off, 0,
size);
return 0;
return;
if (chunk->vm)
free_vm_area(chunk->vm);
- pcpu_realloc(chunk->map, chunk->map_alloc * sizeof(chunk->map[0]), 0);
+ pcpu_mem_free(chunk->map, chunk->map_alloc * sizeof(chunk->map[0]));
kfree(chunk);
}
if (!chunk)
return NULL;
- chunk->map = pcpu_realloc(NULL, 0,
- PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
+ chunk->map = pcpu_mem_alloc(PCPU_DFL_MAP_ALLOC * sizeof(chunk->map[0]));
chunk->map_alloc = PCPU_DFL_MAP_ALLOC;
chunk->map[chunk->map_used++] = pcpu_unit_size;
+ chunk->page = chunk->page_ar;
chunk->vm = get_vm_area(pcpu_chunk_size, GFP_KERNEL);
if (!chunk->vm) {
}
/**
- * __alloc_percpu - allocate percpu area
+ * pcpu_alloc - the percpu allocator
* @size: size of area to allocate in bytes
* @align: alignment of area (max PAGE_SIZE)
+ * @reserved: allocate from the reserved chunk if available
*
- * Allocate percpu area of @size bytes aligned at @align. Might
- * sleep. Might trigger writeouts.
+ * Allocate percpu area of @size bytes aligned at @align.
+ *
+ * CONTEXT:
+ * Does GFP_KERNEL allocation.
*
* RETURNS:
* Percpu pointer to the allocated area on success, NULL on failure.
*/
-void *__alloc_percpu(size_t size, size_t align)
+static void *pcpu_alloc(size_t size, size_t align, bool reserved)
{
- void *ptr = NULL;
struct pcpu_chunk *chunk;
int slot, off;
- if (unlikely(!size || size > PAGE_SIZE << PCPU_MIN_UNIT_PAGES_SHIFT ||
- align > PAGE_SIZE)) {
+ if (unlikely(!size || size > PCPU_MIN_UNIT_SIZE || align > PAGE_SIZE)) {
WARN(true, "illegal size (%zu) or align (%zu) for "
"percpu allocation\n", size, align);
return NULL;
}
- mutex_lock(&pcpu_mutex);
+ mutex_lock(&pcpu_alloc_mutex);
+ spin_lock_irq(&pcpu_lock);
+
+ /* serve reserved allocations from the reserved chunk if available */
+ if (reserved && pcpu_reserved_chunk) {
+ chunk = pcpu_reserved_chunk;
+ if (size > chunk->contig_hint ||
+ pcpu_extend_area_map(chunk) < 0)
+ goto fail_unlock;
+ off = pcpu_alloc_area(chunk, size, align);
+ if (off >= 0)
+ goto area_found;
+ goto fail_unlock;
+ }
- /* allocate area */
+restart:
+ /* search through normal chunks */
for (slot = pcpu_size_to_slot(size); slot < pcpu_nr_slots; slot++) {
list_for_each_entry(chunk, &pcpu_slot[slot], list) {
if (size > chunk->contig_hint)
continue;
+
+ switch (pcpu_extend_area_map(chunk)) {
+ case 0:
+ break;
+ case 1:
+ goto restart; /* pcpu_lock dropped, restart */
+ default:
+ goto fail_unlock;
+ }
+
off = pcpu_alloc_area(chunk, size, align);
if (off >= 0)
goto area_found;
- if (off != -ENOSPC)
- goto out_unlock;
}
}
/* hmmm... no space left, create a new chunk */
+ spin_unlock_irq(&pcpu_lock);
+
chunk = alloc_pcpu_chunk();
if (!chunk)
- goto out_unlock;
+ goto fail_unlock_mutex;
+
+ spin_lock_irq(&pcpu_lock);
pcpu_chunk_relocate(chunk, -1);
pcpu_chunk_addr_insert(chunk);
-
- off = pcpu_alloc_area(chunk, size, align);
- if (off < 0)
- goto out_unlock;
+ goto restart;
area_found:
+ spin_unlock_irq(&pcpu_lock);
+
/* populate, map and clear the area */
if (pcpu_populate_chunk(chunk, off, size)) {
+ spin_lock_irq(&pcpu_lock);
pcpu_free_area(chunk, off);
- goto out_unlock;
+ goto fail_unlock;
}
- ptr = __addr_to_pcpu_ptr(chunk->vm->addr + off);
-out_unlock:
- mutex_unlock(&pcpu_mutex);
- return ptr;
+ mutex_unlock(&pcpu_alloc_mutex);
+
+ return __addr_to_pcpu_ptr(chunk->vm->addr + off);
+
+fail_unlock:
+ spin_unlock_irq(&pcpu_lock);
+fail_unlock_mutex:
+ mutex_unlock(&pcpu_alloc_mutex);
+ return NULL;
+}
+
+/**
+ * __alloc_percpu - allocate dynamic percpu area
+ * @size: size of area to allocate in bytes
+ * @align: alignment of area (max PAGE_SIZE)
+ *
+ * Allocate percpu area of @size bytes aligned at @align. Might
+ * sleep. Might trigger writeouts.
+ *
+ * CONTEXT:
+ * Does GFP_KERNEL allocation.
+ *
+ * RETURNS:
+ * Percpu pointer to the allocated area on success, NULL on failure.
+ */
+void *__alloc_percpu(size_t size, size_t align)
+{
+ return pcpu_alloc(size, align, false);
}
EXPORT_SYMBOL_GPL(__alloc_percpu);
-static void pcpu_kill_chunk(struct pcpu_chunk *chunk)
+/**
+ * __alloc_reserved_percpu - allocate reserved percpu area
+ * @size: size of area to allocate in bytes
+ * @align: alignment of area (max PAGE_SIZE)
+ *
+ * Allocate percpu area of @size bytes aligned at @align from reserved
+ * percpu area if arch has set it up; otherwise, allocation is served
+ * from the same dynamic area. Might sleep. Might trigger writeouts.
+ *
+ * CONTEXT:
+ * Does GFP_KERNEL allocation.
+ *
+ * RETURNS:
+ * Percpu pointer to the allocated area on success, NULL on failure.
+ */
+void *__alloc_reserved_percpu(size_t size, size_t align)
+{
+ return pcpu_alloc(size, align, true);
+}
+
+/**
+ * pcpu_reclaim - reclaim fully free chunks, workqueue function
+ * @work: unused
+ *
+ * Reclaim all fully free chunks except for the first one.
+ *
+ * CONTEXT:
+ * workqueue context.
+ */
+static void pcpu_reclaim(struct work_struct *work)
{
- pcpu_depopulate_chunk(chunk, 0, pcpu_unit_size, false);
- list_del(&chunk->list);
- rb_erase(&chunk->rb_node, &pcpu_addr_root);
- free_pcpu_chunk(chunk);
+ LIST_HEAD(todo);
+ struct list_head *head = &pcpu_slot[pcpu_nr_slots - 1];
+ struct pcpu_chunk *chunk, *next;
+
+ mutex_lock(&pcpu_alloc_mutex);
+ spin_lock_irq(&pcpu_lock);
+
+ list_for_each_entry_safe(chunk, next, head, list) {
+ WARN_ON(chunk->immutable);
+
+ /* spare the first one */
+ if (chunk == list_first_entry(head, struct pcpu_chunk, list))
+ continue;
+
+ rb_erase(&chunk->rb_node, &pcpu_addr_root);
+ list_move(&chunk->list, &todo);
+ }
+
+ spin_unlock_irq(&pcpu_lock);
+ mutex_unlock(&pcpu_alloc_mutex);
+
+ list_for_each_entry_safe(chunk, next, &todo, list) {
+ pcpu_depopulate_chunk(chunk, 0, pcpu_unit_size, false);
+ free_pcpu_chunk(chunk);
+ }
}
/**
* free_percpu - free percpu area
* @ptr: pointer to area to free
*
- * Free percpu area @ptr. Might sleep.
+ * Free percpu area @ptr.
+ *
+ * CONTEXT:
+ * Can be called from atomic context.
*/
void free_percpu(void *ptr)
{
void *addr = __pcpu_ptr_to_addr(ptr);
struct pcpu_chunk *chunk;
+ unsigned long flags;
int off;
if (!ptr)
return;
- mutex_lock(&pcpu_mutex);
+ spin_lock_irqsave(&pcpu_lock, flags);
chunk = pcpu_chunk_addr_search(addr);
off = addr - chunk->vm->addr;
pcpu_free_area(chunk, off);
- /* the chunk became fully free, kill one if there are other free ones */
+ /* if there are more than one fully free chunks, wake up grim reaper */
if (chunk->free_size == pcpu_unit_size) {
struct pcpu_chunk *pos;
- list_for_each_entry(pos,
- &pcpu_slot[pcpu_chunk_slot(chunk)], list)
+ list_for_each_entry(pos, &pcpu_slot[pcpu_nr_slots - 1], list)
if (pos != chunk) {
- pcpu_kill_chunk(pos);
+ schedule_work(&pcpu_reclaim_work);
break;
}
}
- mutex_unlock(&pcpu_mutex);
+ spin_unlock_irqrestore(&pcpu_lock, flags);
}
EXPORT_SYMBOL_GPL(free_percpu);
/**
- * pcpu_setup_static - initialize kernel static percpu area
- * @populate_pte_fn: callback to allocate pagetable
- * @pages: num_possible_cpus() * PFN_UP(cpu_size) pages
- * @cpu_size: the size of static percpu area in bytes
+ * pcpu_setup_first_chunk - initialize the first percpu chunk
+ * @get_page_fn: callback to fetch page pointer
+ * @static_size: the size of static percpu area in bytes
+ * @reserved_size: the size of reserved percpu area in bytes
+ * @dyn_size: free size for dynamic allocation in bytes, -1 for auto
+ * @unit_size: unit size in bytes, must be multiple of PAGE_SIZE, -1 for auto
+ * @base_addr: mapped address, NULL for auto
+ * @populate_pte_fn: callback to allocate pagetable, NULL if unnecessary
+ *
+ * Initialize the first percpu chunk which contains the kernel static
+ * perpcu area. This function is to be called from arch percpu area
+ * setup path. The first two parameters are mandatory. The rest are
+ * optional.
+ *
+ * @get_page_fn() should return pointer to percpu page given cpu
+ * number and page number. It should at least return enough pages to
+ * cover the static area. The returned pages for static area should
+ * have been initialized with valid data. If @unit_size is specified,
+ * it can also return pages after the static area. NULL return
+ * indicates end of pages for the cpu. Note that @get_page_fn() must
+ * return the same number of pages for all cpus.
+ *
+ * @reserved_size, if non-zero, specifies the amount of bytes to
+ * reserve after the static area in the first chunk. This reserves
+ * the first chunk such that it's available only through reserved
+ * percpu allocation. This is primarily used to serve module percpu
+ * static areas on architectures where the addressing model has
+ * limited offset range for symbol relocations to guarantee module
+ * percpu symbols fall inside the relocatable range.
*
- * Initialize kernel static percpu area. The caller should allocate
- * all the necessary pages and pass them in @pages.
- * @populate_pte_fn() is called on each page to be used for percpu
- * mapping and is responsible for making sure all the necessary page
- * tables for the page is allocated.
+ * @dyn_size, if non-negative, determines the number of bytes
+ * available for dynamic allocation in the first chunk. Specifying
+ * non-negative value makes percpu leave alone the area beyond
+ * @static_size + @reserved_size + @dyn_size.
+ *
+ * @unit_size, if non-negative, specifies unit size and must be
+ * aligned to PAGE_SIZE and equal to or larger than @static_size +
+ * @reserved_size + if non-negative, @dyn_size.
+ *
+ * Non-null @base_addr means that the caller already allocated virtual
+ * region for the first chunk and mapped it. percpu must not mess
+ * with the chunk. Note that @base_addr with 0 @unit_size or non-NULL
+ * @populate_pte_fn doesn't make any sense.
+ *
+ * @populate_pte_fn is used to populate the pagetable. NULL means the
+ * caller already populated the pagetable.
+ *
+ * If the first chunk ends up with both reserved and dynamic areas, it
+ * is served by two chunks - one to serve the core static and reserved
+ * areas and the other for the dynamic area. They share the same vm
+ * and page map but uses different area allocation map to stay away
+ * from each other. The latter chunk is circulated in the chunk slots
+ * and available for dynamic allocation like any other chunks.
*
* RETURNS:
* The determined pcpu_unit_size which can be used to initialize
* percpu access.
*/
-size_t __init pcpu_setup_static(pcpu_populate_pte_fn_t populate_pte_fn,
- struct page **pages, size_t cpu_size)
+size_t __init pcpu_setup_first_chunk(pcpu_get_page_fn_t get_page_fn,
+ size_t static_size, size_t reserved_size,
+ ssize_t dyn_size, ssize_t unit_size,
+ void *base_addr,
+ pcpu_populate_pte_fn_t populate_pte_fn)
{
- static struct vm_struct static_vm;
- struct pcpu_chunk *static_chunk;
- int nr_cpu_pages = DIV_ROUND_UP(cpu_size, PAGE_SIZE);
+ static struct vm_struct first_vm;
+ static int smap[2], dmap[2];
+ size_t size_sum = static_size + reserved_size +
+ (dyn_size >= 0 ? dyn_size : 0);
+ struct pcpu_chunk *schunk, *dchunk = NULL;
unsigned int cpu;
+ int nr_pages;
int err, i;
- pcpu_unit_pages_shift = max_t(int, PCPU_MIN_UNIT_PAGES_SHIFT,
- order_base_2(cpu_size) - PAGE_SHIFT);
+ /* santiy checks */
+ BUILD_BUG_ON(ARRAY_SIZE(smap) >= PCPU_DFL_MAP_ALLOC ||
+ ARRAY_SIZE(dmap) >= PCPU_DFL_MAP_ALLOC);
+ BUG_ON(!static_size);
+ if (unit_size >= 0) {
+ BUG_ON(unit_size < size_sum);
+ BUG_ON(unit_size & ~PAGE_MASK);
+ BUG_ON(unit_size < PCPU_MIN_UNIT_SIZE);
+ } else
+ BUG_ON(base_addr);
+ BUG_ON(base_addr && populate_pte_fn);
+
+ if (unit_size >= 0)
+ pcpu_unit_pages = unit_size >> PAGE_SHIFT;
+ else
+ pcpu_unit_pages = max_t(int, PCPU_MIN_UNIT_SIZE >> PAGE_SHIFT,
+ PFN_UP(size_sum));
- pcpu_static_size = cpu_size;
- pcpu_unit_pages = 1 << pcpu_unit_pages_shift;
- pcpu_unit_shift = PAGE_SHIFT + pcpu_unit_pages_shift;
- pcpu_unit_size = 1 << pcpu_unit_shift;
+ pcpu_unit_size = pcpu_unit_pages << PAGE_SHIFT;
pcpu_chunk_size = num_possible_cpus() * pcpu_unit_size;
- pcpu_nr_slots = pcpu_size_to_slot(pcpu_unit_size) + 1;
pcpu_chunk_struct_size = sizeof(struct pcpu_chunk)
- + (1 << pcpu_unit_pages_shift) * sizeof(struct page *);
+ + num_possible_cpus() * pcpu_unit_pages * sizeof(struct page *);
- /* allocate chunk slots */
+ if (dyn_size < 0)
+ dyn_size = pcpu_unit_size - static_size - reserved_size;
+
+ /*
+ * Allocate chunk slots. The additional last slot is for
+ * empty chunks.
+ */
+ pcpu_nr_slots = __pcpu_size_to_slot(pcpu_unit_size) + 2;
pcpu_slot = alloc_bootmem(pcpu_nr_slots * sizeof(pcpu_slot[0]));
for (i = 0; i < pcpu_nr_slots; i++)
INIT_LIST_HEAD(&pcpu_slot[i]);
- /* init and register vm area */
- static_vm.flags = VM_ALLOC;
- static_vm.size = pcpu_chunk_size;
- vm_area_register_early(&static_vm);
+ /*
+ * Initialize static chunk. If reserved_size is zero, the
+ * static chunk covers static area + dynamic allocation area
+ * in the first chunk. If reserved_size is not zero, it
+ * covers static area + reserved area (mostly used for module
+ * static percpu allocation).
+ */
+ schunk = alloc_bootmem(pcpu_chunk_struct_size);
+ INIT_LIST_HEAD(&schunk->list);
+ schunk->vm = &first_vm;
+ schunk->map = smap;
+ schunk->map_alloc = ARRAY_SIZE(smap);
+ schunk->page = schunk->page_ar;
+
+ if (reserved_size) {
+ schunk->free_size = reserved_size;
+ pcpu_reserved_chunk = schunk; /* not for dynamic alloc */
+ } else {
+ schunk->free_size = dyn_size;
+ dyn_size = 0; /* dynamic area covered */
+ }
+ schunk->contig_hint = schunk->free_size;
+
+ schunk->map[schunk->map_used++] = -static_size;
+ if (schunk->free_size)
+ schunk->map[schunk->map_used++] = schunk->free_size;
+
+ pcpu_reserved_chunk_limit = static_size + schunk->free_size;
+
+ /* init dynamic chunk if necessary */
+ if (dyn_size) {
+ dchunk = alloc_bootmem(sizeof(struct pcpu_chunk));
+ INIT_LIST_HEAD(&dchunk->list);
+ dchunk->vm = &first_vm;
+ dchunk->map = dmap;
+ dchunk->map_alloc = ARRAY_SIZE(dmap);
+ dchunk->page = schunk->page_ar; /* share page map with schunk */
+
+ dchunk->contig_hint = dchunk->free_size = dyn_size;
+ dchunk->map[dchunk->map_used++] = -pcpu_reserved_chunk_limit;
+ dchunk->map[dchunk->map_used++] = dchunk->free_size;
+ }
+
+ /* allocate vm address */
+ first_vm.flags = VM_ALLOC;
+ first_vm.size = pcpu_chunk_size;
- /* init static_chunk */
- static_chunk = alloc_bootmem(pcpu_chunk_struct_size);
- INIT_LIST_HEAD(&static_chunk->list);
- static_chunk->vm = &static_vm;
- static_chunk->free_size = pcpu_unit_size - pcpu_static_size;
- static_chunk->contig_hint = static_chunk->free_size;
+ if (!base_addr)
+ vm_area_register_early(&first_vm, PAGE_SIZE);
+ else {
+ /*
+ * Pages already mapped. No need to remap into
+ * vmalloc area. In this case the first chunks can't
+ * be mapped or unmapped by percpu and are marked
+ * immutable.
+ */
+ first_vm.addr = base_addr;
+ schunk->immutable = true;
+ if (dchunk)
+ dchunk->immutable = true;
+ }
- /* assign pages and map them */
+ /* assign pages */
+ nr_pages = -1;
for_each_possible_cpu(cpu) {
- for (i = 0; i < nr_cpu_pages; i++) {
- *pcpu_chunk_pagep(static_chunk, cpu, i) = *pages++;
- populate_pte_fn(pcpu_chunk_addr(static_chunk, cpu, i));
+ for (i = 0; i < pcpu_unit_pages; i++) {
+ struct page *page = get_page_fn(cpu, i);
+
+ if (!page)
+ break;
+ *pcpu_chunk_pagep(schunk, cpu, i) = page;
}
+
+ BUG_ON(i < PFN_UP(static_size));
+
+ if (nr_pages < 0)
+ nr_pages = i;
+ else
+ BUG_ON(nr_pages != i);
}
- err = pcpu_map(static_chunk, 0, nr_cpu_pages);
- if (err)
- panic("failed to setup static percpu area, err=%d\n", err);
+ /* map them */
+ if (populate_pte_fn) {
+ for_each_possible_cpu(cpu)
+ for (i = 0; i < nr_pages; i++)
+ populate_pte_fn(pcpu_chunk_addr(schunk,
+ cpu, i));
+
+ err = pcpu_map(schunk, 0, nr_pages);
+ if (err)
+ panic("failed to setup static percpu area, err=%d\n",
+ err);
+ }
- /* link static_chunk in */
- pcpu_chunk_relocate(static_chunk, -1);
- pcpu_chunk_addr_insert(static_chunk);
+ /* link the first chunk in */
+ if (!dchunk) {
+ pcpu_chunk_relocate(schunk, -1);
+ pcpu_chunk_addr_insert(schunk);
+ } else {
+ pcpu_chunk_relocate(dchunk, -1);
+ pcpu_chunk_addr_insert(dchunk);
+ }
/* we're done */
- pcpu_base_addr = (void *)pcpu_chunk_addr(static_chunk, 0, 0);
+ pcpu_base_addr = (void *)pcpu_chunk_addr(schunk, 0, 0);
return pcpu_unit_size;
}
+
+/*
+ * Embedding first chunk setup helper.
+ */
+static void *pcpue_ptr __initdata;
+static size_t pcpue_size __initdata;
+static size_t pcpue_unit_size __initdata;
+
+static struct page * __init pcpue_get_page(unsigned int cpu, int pageno)
+{
+ size_t off = (size_t)pageno << PAGE_SHIFT;
+
+ if (off >= pcpue_size)
+ return NULL;
+
+ return virt_to_page(pcpue_ptr + cpu * pcpue_unit_size + off);
+}
+
+/**
+ * pcpu_embed_first_chunk - embed the first percpu chunk into bootmem
+ * @static_size: the size of static percpu area in bytes
+ * @reserved_size: the size of reserved percpu area in bytes
+ * @dyn_size: free size for dynamic allocation in bytes, -1 for auto
+ * @unit_size: unit size in bytes, must be multiple of PAGE_SIZE, -1 for auto
+ *
+ * This is a helper to ease setting up embedded first percpu chunk and
+ * can be called where pcpu_setup_first_chunk() is expected.
+ *
+ * If this function is used to setup the first chunk, it is allocated
+ * as a contiguous area using bootmem allocator and used as-is without
+ * being mapped into vmalloc area. This enables the first chunk to
+ * piggy back on the linear physical mapping which often uses larger
+ * page size.
+ *
+ * When @dyn_size is positive, dynamic area might be larger than
+ * specified to fill page alignment. Also, when @dyn_size is auto,
+ * @dyn_size does not fill the whole first chunk but only what's
+ * necessary for page alignment after static and reserved areas.
+ *
+ * If the needed size is smaller than the minimum or specified unit
+ * size, the leftover is returned to the bootmem allocator.
+ *
+ * RETURNS:
+ * The determined pcpu_unit_size which can be used to initialize
+ * percpu access on success, -errno on failure.
+ */
+ssize_t __init pcpu_embed_first_chunk(size_t static_size, size_t reserved_size,
+ ssize_t dyn_size, ssize_t unit_size)
+{
+ unsigned int cpu;
+
+ /* determine parameters and allocate */
+ pcpue_size = PFN_ALIGN(static_size + reserved_size +
+ (dyn_size >= 0 ? dyn_size : 0));
+ if (dyn_size != 0)
+ dyn_size = pcpue_size - static_size - reserved_size;
+
+ if (unit_size >= 0) {
+ BUG_ON(unit_size < pcpue_size);
+ pcpue_unit_size = unit_size;
+ } else
+ pcpue_unit_size = max_t(size_t, pcpue_size, PCPU_MIN_UNIT_SIZE);
+
+ pcpue_ptr = __alloc_bootmem_nopanic(
+ num_possible_cpus() * pcpue_unit_size,
+ PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
+ if (!pcpue_ptr)
+ return -ENOMEM;
+
+ /* return the leftover and copy */
+ for_each_possible_cpu(cpu) {
+ void *ptr = pcpue_ptr + cpu * pcpue_unit_size;
+
+ free_bootmem(__pa(ptr + pcpue_size),
+ pcpue_unit_size - pcpue_size);
+ memcpy(ptr, __per_cpu_load, static_size);
+ }
+
+ /* we're ready, commit */
+ pr_info("PERCPU: Embedded %zu pages at %p, static data %zu bytes\n",
+ pcpue_size >> PAGE_SHIFT, pcpue_ptr, static_size);
+
+ return pcpu_setup_first_chunk(pcpue_get_page, static_size,
+ reserved_size, dyn_size,
+ pcpue_unit_size, pcpue_ptr, NULL);
+}
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff