scsi: mpi3mr: Add support for recovering controller

Detection of firmware fault or any kind of unresponsiveness in the
controller (any admin command which times out) results in resetting the
controller. The primary reset mechanisms used are either soft reset or diag
fault reset. A reset is performed if the host sets the ResetAction field in
the HostDiagnostic register to either 001b (soft reset) or 007b (diag fault
reset). After successfully resetting the controller the driver
reinitializes the controller by going through start of the day
initialization procedure. Pending I/Os during the reset are returned back
to the SCSI midlayer for retry.

Link: https://lore.kernel.org/r/20210520152545.2710479-10-kashyap.desai@broadcom.com
Cc: sathya.prakash@broadcom.co
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Tomas Henzl <thenzl@redhat.com>
Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

1 files changed, 86 insertions, 4 deletions

diff --git a/drivers/scsi/mpi3mr/mpi3mr_os.c b/drivers/scsi/mpi3mr/mpi3mr_os.c
index 97b6497498d4..9a54cc143dc5 100644
--- a/drivers/scsi/mpi3mr/mpi3mr_os.c
+++ b/drivers/scsi/mpi3mr/mpi3mr_os.c
@@ -311,6 +311,86 @@ void mpi3mr_cleanup_fwevt_list(struct mpi3mr_ioc *mrioc)
 }
 
 /**
+ * mpi3mr_invalidate_devhandles -Invalidate device handles
+ * @mrioc: Adapter instance reference
+ *
+ * Invalidate the device handles in the target device structures
+ * . Called post reset prior to reinitializing the controller.
+ *
+ * Return: Nothing.
+ */
+void mpi3mr_invalidate_devhandles(struct mpi3mr_ioc *mrioc)
+{
+	struct mpi3mr_tgt_dev *tgtdev;
+	struct mpi3mr_stgt_priv_data *tgt_priv;
+
+	list_for_each_entry(tgtdev, &mrioc->tgtdev_list, list) {
+		tgtdev->dev_handle = MPI3MR_INVALID_DEV_HANDLE;
+		if (tgtdev->starget && tgtdev->starget->hostdata) {
+			tgt_priv = tgtdev->starget->hostdata;
+			tgt_priv->dev_handle = MPI3MR_INVALID_DEV_HANDLE;
+		}
+	}
+}
+
+/**
+ * mpi3mr_flush_scmd - Flush individual SCSI command
+ * @rq: Block request
+ * @data: Adapter instance reference
+ *
+ * Return the SCSI command to the upper layers if it is in LLD
+ * scope.
+ *
+ * Return: true always.
+ */
+
+static bool mpi3mr_flush_scmd(struct request *rq,
+	void *data, bool reserved)
+{
+	struct mpi3mr_ioc *mrioc = (struct mpi3mr_ioc *)data;
+	struct scsi_cmnd *scmd = blk_mq_rq_to_pdu(rq);
+	struct scmd_priv *priv = NULL;
+
+	if (scmd) {
+		priv = scsi_cmd_priv(scmd);
+		if (!priv->in_lld_scope)
+			goto out;
+
+		mpi3mr_clear_scmd_priv(mrioc, scmd);
+		scsi_dma_unmap(scmd);
+		scmd->result = DID_RESET << 16;
+		scsi_print_command(scmd);
+		scmd->scsi_done(scmd);
+		mrioc->flush_io_count++;
+	}
+
+out:
+	return(true);
+}
+
+/**
+ * mpi3mr_flush_host_io -  Flush host I/Os
+ * @mrioc: Adapter instance reference
+ *
+ * Flush all of the pending I/Os by calling
+ * blk_mq_tagset_busy_iter() for each possible tag. This is
+ * executed post controller reset
+ *
+ * Return: Nothing.
+ */
+void mpi3mr_flush_host_io(struct mpi3mr_ioc *mrioc)
+{
+	struct Scsi_Host *shost = mrioc->shost;
+
+	mrioc->flush_io_count = 0;
+	ioc_info(mrioc, "%s :Flushing Host I/O cmds post reset\n", __func__);
+	blk_mq_tagset_busy_iter(&shost->tag_set,
+	    mpi3mr_flush_scmd, (void *)mrioc);
+	ioc_info(mrioc, "%s :Flushed %d Host I/O cmds\n", __func__,
+	    mrioc->flush_io_count);
+}
+
+/**
  * mpi3mr_alloc_tgtdev - target device allocator
  *
  * Allocate target device instance and initialize the reference
@@ -2495,6 +2575,7 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	INIT_LIST_HEAD(&mrioc->tgtdev_list);
 	INIT_LIST_HEAD(&mrioc->delayed_rmhs_list);
 
+	mutex_init(&mrioc->reset_mutex);
 	mpi3mr_init_drv_cmd(&mrioc->init_cmds, MPI3MR_HOSTTAG_INITCMDS);
 
 	for (i = 0; i < MPI3MR_NUM_DEVRMCMD; i++)
@@ -2504,6 +2585,7 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (pdev->revision)
 		mrioc->enable_segqueue = true;
 
+	init_waitqueue_head(&mrioc->reset_waitq);
 	mrioc->logging_level = logging_level;
 	mrioc->shost = shost;
 	mrioc->pdev = pdev;
@@ -2528,7 +2610,7 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	mrioc->is_driver_loading = 1;
-	if (mpi3mr_init_ioc(mrioc)) {
+	if (mpi3mr_init_ioc(mrioc, 0)) {
 		ioc_err(mrioc, "failure at %s:%d/%s()!\n",
 		    __FILE__, __LINE__, __func__);
 		retval = -ENODEV;
@@ -2551,7 +2633,7 @@ mpi3mr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	return retval;
 
 addhost_failed:
-	mpi3mr_cleanup_ioc(mrioc);
+	mpi3mr_cleanup_ioc(mrioc, 0);
 out_iocinit_failed:
 	destroy_workqueue(mrioc->fwevt_worker_thread);
 out_fwevtthread_failed:
@@ -2600,7 +2682,7 @@ static void mpi3mr_remove(struct pci_dev *pdev)
 		mpi3mr_tgtdev_del_from_list(mrioc, tgtdev);
 		mpi3mr_tgtdev_put(tgtdev);
 	}
-	mpi3mr_cleanup_ioc(mrioc);
+	mpi3mr_cleanup_ioc(mrioc, 0);
 
 	spin_lock(&mrioc_list_lock);
 	list_del(&mrioc->list);
@@ -2640,7 +2722,7 @@ static void mpi3mr_shutdown(struct pci_dev *pdev)
 	spin_unlock_irqrestore(&mrioc->fwevt_lock, flags);
 	if (wq)
 		destroy_workqueue(wq);
-	mpi3mr_cleanup_ioc(mrioc);
+	mpi3mr_cleanup_ioc(mrioc, 0);
 }
 
 static const struct pci_device_id mpi3mr_pci_id_table[] = {