kernel中的mempool机制

mempool 为了防止某些情况下分配内存不能失败而采用的技术。实现思路是会预留一定数量大小相等的内存块来备用。其源码在/mm/mempool.c中.
在kernel中通过
typedef struct mempool_s {
	spinlock_t lock;
	int min_nr;		/* nr of elements at *elements */
	int curr_nr;		/* Current nr of elements at *elements */
	void **elements;

	void *pool_data;
	mempool_alloc_t *alloc;
	mempool_free_t *free;
	wait_queue_head_t wait;
} mempool_t;
来表示一个mempool。使用mempool的时候必须先调用mempool_create来create 一个memory pool
mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
				mempool_free_t *free_fn, void *pool_data)
{
	return mempool_create_node(min_nr,alloc_fn,free_fn, pool_data,
				   GFP_KERNEL, NUMA_NO_NODE);
}

mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn,
			       mempool_free_t *free_fn, void *pool_data,
			       gfp_t gfp_mask, int node_id)
{
	mempool_t *pool;
	pool = kzalloc_node(sizeof(*pool), gfp_mask, node_id);
	if (!pool)
		return NULL;
//可以看到这里预留了min_nr个sizeof(void *)的size，简单举例下如果是64 位系统的话sizeof(void *) 是8个byte，假如min_nr等于1的话，在create的时候就预留了8个byte
	pool->elements = kmalloc_node(min_nr * sizeof(void *),
				      gfp_mask, node_id);
	if (!pool->elements) {
		kfree(pool);
		return NULL;
	}
	spin_lock_init(&pool->lock);
	pool->min_nr = min_nr;
	pool->pool_data = pool_data;
	init_waitqueue_head(&pool->wait);
	pool->alloc = alloc_fn;
	pool->free = free_fn;

	/*
	 * First pre-allocate the guaranteed number of buffers.
	 */
//如果pool->curr_nr < pool->min_nr 的话，会再预留一部分 
	while (pool->curr_nr < pool->min_nr) {
		void *element;

		element = pool->alloc(gfp_mask, pool->pool_data);
		if (unlikely(!element)) {
			mempool_destroy(pool);
			return NULL;
		}
//通过add_element 将pool 添加到list中去
		add_element(pool, element);
	}
	return pool;
}
可见在新建memory pool的时候，就会预留一定的memory，其中预留的memory是以byte为单位的。预留的内存块都是等长的，其中长度是sizeof(void *)。这里还要注意在新建memory pool的时候，还要提供两个函数指针用于申请和释放memory。这两个函数可以自己写，也可以用kernel提供的线程的。
其中kernel提供的如下：
/*
 * A commonly used alloc and free fn.
 */
void *mempool_alloc_slab(gfp_t gfp_mask, void *pool_data)
{
	struct kmem_cache *mem = pool_data;
	VM_BUG_ON(mem->ctor);
	return kmem_cache_alloc(mem, gfp_mask);
}
EXPORT_SYMBOL(mempool_alloc_slab);

void mempool_free_slab(void *element, void *pool_data)
{
	struct kmem_cache *mem = pool_data;
	kmem_cache_free(mem, element);
}
EXPORT_SYMBOL(mempool_free_slab);

/*
 * A commonly used alloc and free fn that kmalloc/kfrees the amount of memory
 * specified by pool_data
 */
void *mempool_kmalloc(gfp_t gfp_mask, void *pool_data)
{
	size_t size = (size_t)pool_data;
	return kmalloc(size, gfp_mask);
}
EXPORT_SYMBOL(mempool_kmalloc);

void mempool_kfree(void *element, void *pool_data)
{
	kfree(element);
}
EXPORT_SYMBOL(mempool_kfree);

/*
 * A simple mempool-backed page allocator that allocates pages
 * of the order specified by pool_data.
 */
void *mempool_alloc_pages(gfp_t gfp_mask, void *pool_data)
{
	int order = (int)(long)pool_data;
	return alloc_pages(gfp_mask, order);
}
EXPORT_SYMBOL(mempool_alloc_pages);

void mempool_free_pages(void *element, void *pool_data)
{
	int order = (int)(long)pool_data;
	__free_pages(element, order);
}
EXPORT_SYMBOL(mempool_free_pages);

新建memory pool后，如果要从memory pool中申请memory的话，需要调用mempool_alloc
void *mempool_alloc(mempool_t *pool, gfp_t gfp_mask)
{
	void *element;
	unsigned long flags;
	wait_queue_entry_t wait;
	gfp_t gfp_temp;

	VM_WARN_ON_ONCE(gfp_mask & __GFP_ZERO);
	might_sleep_if(gfp_mask & __GFP_DIRECT_RECLAIM);

	gfp_mask |= __GFP_NOMEMALLOC;	/* don't allocate emergency reserves */
	gfp_mask |= __GFP_NORETRY;	/* don't loop in __alloc_pages */
	gfp_mask |= __GFP_NOWARN;	/* failures are OK */

	gfp_temp = gfp_mask & ~(__GFP_DIRECT_RECLAIM|__GFP_IO);

repeat_alloc:
//首先通过pool的alloc函数申请，假如我们在新建pool的时候，提供的pool函数是mempool_alloc_slab的话，则这里其实是从个mempool_alloc_slab 来向kernel申请memory。如果这里没有失败的话，则mempool就和普通的kmalloc 函数没啥区别。如果这里失败的话，则会向新建pool时候预留的8 byte(本例子的举例数据)来申请
	element = pool->alloc(gfp_temp, pool->pool_data);
	if (likely(element != NULL))
		return element;

	spin_lock_irqsave(&pool->lock, flags);
	if (likely(pool->curr_nr)) {
从mempool的list中去掉一个，然后返回给调用者
		element = remove_element(pool, gfp_temp);
		spin_unlock_irqrestore(&pool->lock, flags);
		/* paired with rmb in mempool_free(), read comment there */
		smp_wmb();
		/*
		 * Update the allocation stack trace as this is more useful
		 * for debugging.
		 */
		kmemleak_update_trace(element);
		return element;
	}

	/*
	 * We use gfp mask w/o direct reclaim or IO for the first round.  If
	 * alloc failed with that and @pool was empty, retry immediately.
	 */
	if (gfp_temp != gfp_mask) {
		spin_unlock_irqrestore(&pool->lock, flags);
		gfp_temp = gfp_mask;
		goto repeat_alloc;
	}

	/* We must not sleep if !__GFP_DIRECT_RECLAIM */
	if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) {
		spin_unlock_irqrestore(&pool->lock, flags);
		return NULL;
	}

	/* Let's wait for someone else to return an element to @pool */
	init_wait(&wait);
	prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);

	spin_unlock_irqrestore(&pool->lock, flags);

	/*
	 * FIXME: this should be io_schedule().  The timeout is there as a
	 * workaround for some DM problems in 2.6.18.
	 */
	io_schedule_timeout(5*HZ);

	finish_wait(&pool->wait, &wait);
	goto repeat_alloc;
}