NVMe: Fix memory leak on retried commands

Resources are reallocated for requeued commands, so unmap and release
the iod for the failed command.

It's a pretty bad memory leak and causes a kernel hang if you remove a
drive because of a busy dma pool. You'll get messages spewing like this:

  nvme 0000:xx:xx.x: dma_pool_destroy prp list 256, ffff880420dec000 busy

and lock up pci and the driver since removal never completes while
holding a lock.

Cc: stable@vger.kernel.org
Cc: <stable@vger.kernel.org> # 4.0.x-
Signed-off-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>

1 files changed, 6 insertions, 2 deletions

diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 84e4a8088386..ccc0c1f93daa 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -604,6 +604,7 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx,
         struct request *req = iod_get_private(iod);
         struct nvme_cmd_info *cmd_rq = blk_mq_rq_to_pdu(req);
         u16 status = le16_to_cpup(&cqe->status) >> 1;
+        bool requeue = false;
         int error = 0;
 
         if (unlikely(status)) {
@@ -611,12 +612,13 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx,
                     && (jiffies - req->start_time) < req->timeout) {
                         unsigned long flags;
 
+                        requeue = true;
                         blk_mq_requeue_request(req);
                         spin_lock_irqsave(req->q->queue_lock, flags);
                         if (!blk_queue_stopped(req->q))
                                 blk_mq_kick_requeue_list(req->q);
                         spin_unlock_irqrestore(req->q->queue_lock, flags);
-                        return;
+                        goto release_iod;
                 }
 
                 if (req->cmd_type == REQ_TYPE_DRV_PRIV) {
@@ -639,6 +641,7 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx,
                         "completing aborted command with status:%04x\n",
                         error);
 
+release_iod:
         if (iod->nents) {
                 dma_unmap_sg(nvmeq->dev->dev, iod->sg, iod->nents,
                         rq_data_dir(req) ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
@@ -651,7 +654,8 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx,
         }
         nvme_free_iod(nvmeq->dev, iod);
 
-        blk_mq_complete_request(req, error);
+        if (likely(!requeue))
+                blk_mq_complete_request(req, error);
 }
 
 /* length is in bytes.  gfp flags indicates whether we may sleep. */