scsi: megaraid_sas: Early detection of VD deletion through RaidMap update

Consider the case where a VD is deleted and the targetID of that VD is
assigned to a newly created VD. If the sequence of deletion/addition of VD
happens very quickly there is a possibility that second event (VD add)
occurs even before the driver processes the first event (VD delete).  As
event processing is done in deferred context the device list remains the
same (but targetID is re-used) so driver will not learn the VD
deletion/additon. I/Os meant for the older VD will be directed to new VD
which may lead to data corruption.

Make driver detect the deleted VD as soon as possible based on the RaidMap
update and block further I/O to that device.

Link: https://lore.kernel.org/r/20210528131307.25683-4-chandrakanth.patil@broadcom.comReported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Chandrakanth Patil <chandrakanth.patil@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

drivers/scsi/megaraid/megaraid_sas.h

@@ -2262,6 +2262,15 @@ enum MR_PERF_MODE {
 		 (mode) == MR_LATENCY_PERF_MODE ? "Latency" : \
 		 "Unknown")
 
+enum MEGASAS_LD_TARGET_ID_STATUS {
+	LD_TARGET_ID_INITIAL,
+	LD_TARGET_ID_ACTIVE,
+	LD_TARGET_ID_DELETED,
+};
+
+#define MEGASAS_TARGET_ID(sdev)						\
+	(((sdev->channel % 2) * MEGASAS_MAX_DEV_PER_CHANNEL) + sdev->id)
+
 struct megasas_instance {
 
 	unsigned int *reply_map;
@@ -2326,6 +2335,9 @@ struct megasas_instance {
 	struct megasas_pd_list          pd_list[MEGASAS_MAX_PD];
 	struct megasas_pd_list          local_pd_list[MEGASAS_MAX_PD];
 	u8 ld_ids[MEGASAS_MAX_LD_IDS];
+	u8 ld_tgtid_status[MEGASAS_MAX_LD_IDS];
+	u8 ld_ids_prev[MEGASAS_MAX_LD_IDS];
+	u8 ld_ids_from_raidmap[MEGASAS_MAX_LD_IDS];
 	s8 init_id;
 
 	u16 max_num_sge;

drivers/scsi/megaraid/megaraid_sas_base.c

@@ -141,6 +141,8 @@ static int megasas_register_aen(struct megasas_instance *instance,
 				u32 seq_num, u32 class_locale_word);
 static void megasas_get_pd_info(struct megasas_instance *instance,
 				struct scsi_device *sdev);
+static void
+megasas_set_ld_removed_by_fw(struct megasas_instance *instance);
 
 /*
  * PCI ID table for all supported controllers
@@ -436,6 +438,12 @@ megasas_decode_evt(struct megasas_instance *instance)
 			(class_locale.members.locale),
 			format_class(class_locale.members.class),
 			evt_detail->description);
+
+	if (megasas_dbg_lvl & LD_PD_DEBUG)
+		dev_info(&instance->pdev->dev,
+			 "evt_detail.args.ld.target_id/index %d/%d\n",
+			 evt_detail->args.ld.target_id, evt_detail->args.ld.ld_index);
+
 }
 
 /*
@@ -1779,6 +1787,7 @@ megasas_queue_command(struct Scsi_Host *shost, struct scsi_cmnd *scmd)
 {
 	struct megasas_instance *instance;
 	struct MR_PRIV_DEVICE *mr_device_priv_data;
+	u32 ld_tgt_id;
 
 	instance = (struct megasas_instance *)
 	    scmd->device->host->hostdata;
@@ -1805,18 +1814,22 @@ megasas_queue_command(struct Scsi_Host *shost, struct scsi_cmnd *scmd)
 		}
 	}
 
-	if (atomic_read(&instance->adprecovery) == MEGASAS_HW_CRITICAL_ERROR) {
+	mr_device_priv_data = scmd->device->hostdata;
+	if (!mr_device_priv_data ||
+	    (atomic_read(&instance->adprecovery) == MEGASAS_HW_CRITICAL_ERROR)) {
 		scmd->result = DID_NO_CONNECT << 16;
 		scmd->scsi_done(scmd);
 		return 0;
 	}
 
-	mr_device_priv_data = scmd->device->hostdata;
-	if (!mr_device_priv_data) {
+	if (MEGASAS_IS_LOGICAL(scmd->device)) {
+		ld_tgt_id = MEGASAS_TARGET_ID(scmd->device);
+		if (instance->ld_tgtid_status[ld_tgt_id] == LD_TARGET_ID_DELETED) {
 			scmd->result = DID_NO_CONNECT << 16;
 			scmd->scsi_done(scmd);
 			return 0;
 		}
+	}
 
 	if (atomic_read(&instance->adprecovery) != MEGASAS_HBA_OPERATIONAL)
 		return SCSI_MLQUEUE_HOST_BUSY;
@@ -2095,7 +2108,7 @@ static int megasas_slave_configure(struct scsi_device *sdev)
 
 static int megasas_slave_alloc(struct scsi_device *sdev)
 {
-	u16 pd_index = 0;
+	u16 pd_index = 0, ld_tgt_id;
 	struct megasas_instance *instance ;
 	struct MR_PRIV_DEVICE *mr_device_priv_data;
 
@@ -2120,6 +2133,14 @@ static int megasas_slave_alloc(struct scsi_device *sdev)
 					GFP_KERNEL);
 	if (!mr_device_priv_data)
 		return -ENOMEM;
+
+	if (MEGASAS_IS_LOGICAL(sdev)) {
+		ld_tgt_id = MEGASAS_TARGET_ID(sdev);
+		instance->ld_tgtid_status[ld_tgt_id] = LD_TARGET_ID_ACTIVE;
+		if (megasas_dbg_lvl & LD_PD_DEBUG)
+			sdev_printk(KERN_INFO, sdev, "LD target ID %d created.\n", ld_tgt_id);
+	}
+
 	sdev->hostdata = mr_device_priv_data;
 
 	atomic_set(&mr_device_priv_data->r1_ldio_hint,
@@ -2129,6 +2150,19 @@ static int megasas_slave_alloc(struct scsi_device *sdev)
 
 static void megasas_slave_destroy(struct scsi_device *sdev)
 {
+	u16 ld_tgt_id;
+	struct megasas_instance *instance;
+
+	instance = megasas_lookup_instance(sdev->host->host_no);
+
+	if (MEGASAS_IS_LOGICAL(sdev)) {
+		ld_tgt_id = MEGASAS_TARGET_ID(sdev);
+		instance->ld_tgtid_status[ld_tgt_id] = LD_TARGET_ID_DELETED;
+		if (megasas_dbg_lvl & LD_PD_DEBUG)
+			sdev_printk(KERN_INFO, sdev,
+				    "LD target ID %d removed from OS stack\n", ld_tgt_id);
+	}
+
 	kfree(sdev->hostdata);
 	sdev->hostdata = NULL;
 }
@@ -3525,6 +3559,22 @@ megasas_complete_abort(struct megasas_instance *instance,
 	}
 }
 
+static void
+megasas_set_ld_removed_by_fw(struct megasas_instance *instance)
+{
+	uint i;
+
+	for (i = 0; (i < MEGASAS_MAX_LD_IDS); i++) {
+		if (instance->ld_ids_prev[i] != 0xff &&
+		    instance->ld_ids_from_raidmap[i] == 0xff) {
+			if (megasas_dbg_lvl & LD_PD_DEBUG)
+				dev_info(&instance->pdev->dev,
+					 "LD target ID %d removed from RAID map\n", i);
+			instance->ld_tgtid_status[i] = LD_TARGET_ID_DELETED;
+		}
+	}
+}
+
 /**
  * megasas_complete_cmd -	Completes a command
  * @instance:			Adapter soft state
@@ -3687,9 +3737,13 @@ megasas_complete_cmd(struct megasas_instance *instance, struct megasas_cmd *cmd,
 				fusion->fast_path_io = 0;
 			}
 
+			if (instance->adapter_type >= INVADER_SERIES)
+				megasas_set_ld_removed_by_fw(instance);
+
 			megasas_sync_map_info(instance);
 			spin_unlock_irqrestore(instance->host->host_lock,
 					       flags);
+
 			break;
 		}
 		if (opcode == MR_DCMD_CTRL_EVENT_GET_INFO ||
@@ -8831,8 +8885,10 @@ megasas_aen_polling(struct work_struct *work)
 	union megasas_evt_class_locale class_locale;
 	int event_type = 0;
 	u32 seq_num;
+	u16 ld_target_id;
 	int error;
 	u8  dcmd_ret = DCMD_SUCCESS;
+	struct scsi_device *sdev1;
 
 	if (!instance) {
 		printk(KERN_ERR "invalid instance!\n");
@@ -8855,12 +8911,23 @@ megasas_aen_polling(struct work_struct *work)
 			break;
 
 		case MR_EVT_LD_OFFLINE:
-		case MR_EVT_CFG_CLEARED:
 		case MR_EVT_LD_DELETED:
+			ld_target_id = instance->evt_detail->args.ld.target_id;
+			sdev1 = scsi_device_lookup(instance->host,
+						   MEGASAS_MAX_PD_CHANNELS +
+						   (ld_target_id / MEGASAS_MAX_DEV_PER_CHANNEL),
+						   (ld_target_id - MEGASAS_MAX_DEV_PER_CHANNEL),
+						   0);
+			if (sdev1)
+				megasas_remove_scsi_device(sdev1);
+
+			event_type = SCAN_VD_CHANNEL;
+			break;
 		case MR_EVT_LD_CREATED:
 			event_type = SCAN_VD_CHANNEL;
 			break;
 
+		case MR_EVT_CFG_CLEARED:
 		case MR_EVT_CTRL_HOST_BUS_SCAN_REQUESTED:
 		case MR_EVT_FOREIGN_CFG_IMPORTED:
 		case MR_EVT_LD_STATE_CHANGE:

drivers/scsi/megaraid/megaraid_sas_fp.c

@@ -349,6 +349,10 @@ u8 MR_ValidateMapInfo(struct megasas_instance *instance, u64 map_id)
 
 	num_lds = le16_to_cpu(drv_map->raidMap.ldCount);
 
+	memcpy(instance->ld_ids_prev,
+	       instance->ld_ids_from_raidmap,
+	       sizeof(instance->ld_ids_from_raidmap));
+	memset(instance->ld_ids_from_raidmap, 0xff, MEGASAS_MAX_LD_IDS);
 	/*Convert Raid capability values to CPU arch */
 	for (i = 0; (num_lds > 0) && (i < MAX_LOGICAL_DRIVES_EXT); i++) {
 		ld = MR_TargetIdToLdGet(i, drv_map);
@@ -359,7 +363,7 @@ u8 MR_ValidateMapInfo(struct megasas_instance *instance, u64 map_id)
 
 		raid = MR_LdRaidGet(ld, drv_map);
 		le32_to_cpus((u32 *)&raid->capability);
-
+		instance->ld_ids_from_raidmap[i] = i;
 		num_lds--;
 	}