nvme-pci: fix NULL pointer reference in nvme_alloc_ns

When the io queues setup or tagset allocation failed, ctrl.tagset is
NULL.  But the scan work will still be queued and executed, then panic
comes up due to NULL pointer reference of ctrl.tagset.

To fix this, add a new ctrl state NVME_CTRL_ADMIN_ONLY to inidcate only
admin queue is live. When non io queues or tagset allocation failed, ctrl
enters into this state, scan work will not be started.  But async event
work and nvme dev ioctl will be still available.  This will be helpful to
do further investigation and recovery.

Suggested-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

1 files changed, 21 insertions, 9 deletions

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index add7b18d825d..62119078c2bf 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -2035,13 +2035,12 @@ static void nvme_disable_io_queues(struct nvme_dev *dev, int queues)
 }
 
 /*
- * Return: error value if an error occurred setting up the queues or calling
- * Identify Device.  0 if these succeeded, even if adding some of the
- * namespaces failed.  At the moment, these failures are silent.  TBD which
- * failures should be reported.
+ * return error value only when tagset allocation failed
  */
 static int nvme_dev_add(struct nvme_dev *dev)
 {
+	int ret;
+
 	if (!dev->ctrl.tagset) {
 		dev->tagset.ops = &nvme_mq_ops;
 		dev->tagset.nr_hw_queues = dev->online_queues - 1;
@@ -2057,8 +2056,12 @@ static int nvme_dev_add(struct nvme_dev *dev)
 		dev->tagset.flags = BLK_MQ_F_SHOULD_MERGE;
 		dev->tagset.driver_data = dev;
 
-		if (blk_mq_alloc_tag_set(&dev->tagset))
-			return 0;
+		ret = blk_mq_alloc_tag_set(&dev->tagset);
+		if (ret) {
+			dev_warn(dev->ctrl.device,
+				"IO queues tagset allocation failed %d\n", ret);
+			return ret;
+		}
 		dev->ctrl.tagset = &dev->tagset;
 
 		nvme_dbbuf_set(dev);
@@ -2291,6 +2294,7 @@ static void nvme_reset_work(struct work_struct *work)
 		container_of(work, struct nvme_dev, ctrl.reset_work);
 	bool was_suspend = !!(dev->ctrl.ctrl_config & NVME_CC_SHN_NORMAL);
 	int result = -ENODEV;
+	enum nvme_ctrl_state new_state = NVME_CTRL_LIVE;
 
 	if (WARN_ON(dev->ctrl.state != NVME_CTRL_RESETTING))
 		goto out;
@@ -2354,15 +2358,23 @@ static void nvme_reset_work(struct work_struct *work)
 		dev_warn(dev->ctrl.device, "IO queues not created\n");
 		nvme_kill_queues(&dev->ctrl);
 		nvme_remove_namespaces(&dev->ctrl);
+		new_state = NVME_CTRL_ADMIN_ONLY;
 	} else {
 		nvme_start_queues(&dev->ctrl);
 		nvme_wait_freeze(&dev->ctrl);
-		nvme_dev_add(dev);
+		/* hit this only when allocate tagset fails */
+		if (nvme_dev_add(dev))
+			new_state = NVME_CTRL_ADMIN_ONLY;
 		nvme_unfreeze(&dev->ctrl);
 	}
 
-	if (!nvme_change_ctrl_state(&dev->ctrl, NVME_CTRL_LIVE)) {
-		dev_warn(dev->ctrl.device, "failed to mark controller live\n");
+	/*
+	 * If only admin queue live, keep it to do further investigation or
+	 * recovery.
+	 */
+	if (!nvme_change_ctrl_state(&dev->ctrl, new_state)) {
+		dev_warn(dev->ctrl.device,
+			"failed to mark controller state %d\n", new_state);
 		goto out;
 	}