On Wed, Sep 15, 2021 at 04:15:37PM +0800, Yu Kuai wrote:
> There is a problem that nbd_handle_reply() might access freed request:
>
> 1) At first, a normal io is submitted and completed with scheduler:
>
> internel_tag = blk_mq_get_tag -> get tag from sched_tags
> blk_mq_rq_ctx_init
> sched_tags->rq[internel_tag] = sched_tag->static_rq[internel_tag]
> ...
> blk_mq_get_driver_tag
> __blk_mq_get_driver_tag -> get tag from tags
> tags->rq[tag] = sched_tag->static_rq[internel_tag]
>
> So, both tags->rq[tag] and sched_tags->rq[internel_tag] are pointing
> to the request: sched_tags->static_rq[internal_tag]. Even if the
> io is finished.
>
> 2) nbd server send a reply with random tag directly:
>
> recv_work
> nbd_handle_reply
> blk_mq_tag_to_rq(tags, tag)
> rq = tags->rq[tag]
>
> 3) if the sched_tags->static_rq is freed:
>
> blk_mq_sched_free_requests
> blk_mq_free_rqs(q->tag_set, hctx->sched_tags, i)
> -> step 2) access rq before clearing rq mapping
> blk_mq_clear_rq_mapping(set, tags, hctx_idx);
> __free_pages() -> rq is freed here
>
> 4) Then, nbd continue to use the freed request in nbd_handle_reply
>
> Fix the problem by get 'q_usage_counter' before blk_mq_tag_to_rq(),
> thus request is ensured not to be freed because 'q_usage_counter' is
> not zero.
>
> Signed-off-by: Yu Kuai <yukuai3@xxxxxxxxxx>
> ---
> block/blk-core.c | 1 +
> drivers/block/nbd.c | 19 ++++++++++++++++++-
> 2 files changed, 19 insertions(+), 1 deletion(-)
>
> diff --git a/block/blk-core.c b/block/blk-core.c
> index 5454db2fa263..2008e6903166 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -489,6 +489,7 @@ void blk_queue_exit(struct request_queue *q)
> {
> percpu_ref_put(&q->q_usage_counter);
> }
> +EXPORT_SYMBOL(blk_queue_exit);
>
> static void blk_queue_usage_counter_release(struct percpu_ref *ref)
> {
> diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
> index 9a7bbf8ebe74..f065afcc7586 100644
> --- a/drivers/block/nbd.c
> +++ b/drivers/block/nbd.c
> @@ -824,6 +824,7 @@ static void recv_work(struct work_struct *work)
> work);
> struct nbd_device *nbd = args->nbd;
> struct nbd_config *config = nbd->config;
> + struct request_queue *q = nbd->disk->queue;
> struct nbd_sock *nsock;
> struct nbd_cmd *cmd;
> struct request *rq;
> @@ -834,13 +835,29 @@ static void recv_work(struct work_struct *work)
> if (nbd_read_reply(nbd, args->index, &reply))
> break;
>
> + /*
> + * Get q_usage_counter can prevent accessing freed request
> + * through blk_mq_tag_to_rq() in nbd_handle_reply(). If
> + * q_usage_counter is zero, then no request is inflight, which
> + * means something is wrong since we expect to find a request to
> + * complete here.
> + */
> + if (!percpu_ref_tryget(&q->q_usage_counter)) {
> + dev_err(disk_to_dev(nbd->disk), "%s: no io inflight\n",
> + __func__);
> + break;
> + }
> +
> cmd = nbd_handle_reply(nbd, args->index, &reply);
> - if (IS_ERR(cmd))
> + if (IS_ERR(cmd)) {
> + blk_queue_exit(q);
> break;
> + }
>
> rq = blk_mq_rq_from_pdu(cmd);
> if (likely(!blk_should_fake_timeout(rq->q)))
> blk_mq_complete_request(rq);
> + blk_queue_exit(q);
You can simply call percpu_ref_put() directly just like what scsi_end_request()
is doing.
--
Ming
