Re: [PATCH] block: avoid unconditionally freeing previously allocatedrequest_queue

[Kiyoshi Ueda]

Hi Mike,

On 05/26/2010 01:34 AM +0900, Mike Snitzer wrote:
> Mike Snitzer <snitzer@xxxxxxxxxx> wrote:
>> Kiyoshi Ueda <k-ueda@xxxxxxxxxxxxx> wrote:
>>>> +/*
>>>> + * Fully initialize a request-based queue (->elevator, ->request_fn, etc).
>>>> + */
>>>> +static int dm_init_request_based_queue(struct mapped_device *md)
>>>> +{
>>>> +	struct request_queue *q = NULL;
>>>> +
>>>> +	/* Avoid re-initializing the queue if already fully initialized */
>>>> +	if (!md->queue->elevator) {
>>>> +		/* Fully initialize the queue */
>>>> +		q = blk_init_allocated_queue(md->queue, dm_request_fn, NULL);
>>>> +		if (!q)
>>>> +			return 0;
>>>
>>> When blk_init_allocated_queue() fails, the block-layer seems not to
>>> guarantee that the queue is still available.
>>
>> Ouch, yes this portion of blk_init_allocated_queue_node() is certainly
>> problematic:
>>
>>         if (blk_init_free_list(q)) {
>>                 kmem_cache_free(blk_requestq_cachep, q);
>>                 return NULL;
>>         }

Not only that.  The blk_put_queue() in blk_init_allocated_queue_node()
will also free the queue:

	if (!elevator_init(q, NULL)) {
		blk_queue_congestion_threshold(q);
		return q;
	}

	blk_put_queue(q);
	return NULL;

Thanks,
Kiyoshi Ueda
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/