Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

Pull SCSI updates from James Bottomley:
 "This update includes the usual round of major driver updates
  (hisi_sas, ufs, fnic, cxlflash, be2iscsi, ipr, stex). There's also the
  usual amount of cosmetic and spelling stuff"

* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (155 commits)
  scsi: qla4xxx: fix spelling mistake: "Tempalate" -> "Template"
  scsi: stex: make S6flag static
  scsi: mac_esp: fix to pass correct device identity to free_irq()
  scsi: aacraid: pci_alloc_consistent() failures on ARM64
  scsi: ufs: make ufshcd_get_lists_status() register operation obvious
  scsi: ufs: use MASK_EE_STATUS
  scsi: mac_esp: Replace bogus memory barrier with spinlock
  scsi: fcoe: make fcoe_e_d_tov and fcoe_r_a_tov static
  scsi: sd_zbc: Do not write lock zones for reset
  scsi: sd_zbc: Remove superfluous assignments
  scsi: sd: sd_zbc: Rename sd_zbc_setup_write_cmnd
  scsi: Improve scsi_get_sense_info_fld
  scsi: sd: Cleanup sd_done sense data handling
  scsi: sd: Improve sd_completed_bytes
  scsi: sd: Fix function descriptions
  scsi: mpt3sas: remove redundant wmb
  scsi: mpt: Move scsi_remove_host() out of mptscsih_remove_host()
  scsi: sg: reset 'res_in_use' after unlinking reserved array
  scsi: mvumi: remove code handling zero scsi_sg_count(scmd) case
  scsi: fusion: fix spelling mistake: "Persistancy" -> "Persistency"
  ...

102 files changed, 4250 insertions, 2711 deletions

diff --git a/drivers/scsi/BusLogic.c b/drivers/scsi/BusLogic.c
index c7be7bb37209..35380a58d3f0 100644
--- a/drivers/scsi/BusLogic.c
+++ b/drivers/scsi/BusLogic.c
@@ -3009,7 +3009,7 @@ static int blogic_hostreset(struct scsi_cmnd *SCpnt)
 
 	spin_lock_irq(SCpnt->device->host->host_lock);
 
-	blogic_inc_count(&stats->adatper_reset_req);
+	blogic_inc_count(&stats->adapter_reset_req);
 
 	rc = blogic_resetadapter(adapter, false);
 	spin_unlock_irq(SCpnt->device->host->host_lock);
@@ -3560,8 +3560,16 @@ Target	Requested Completed  Requested Completed  Requested Completed\n\
 		struct blogic_tgt_flags *tgt_flags = &adapter->tgt_flags[tgt];
 		if (!tgt_flags->tgt_exists)
 			continue;
-		seq_printf(m, "\
-  %2d	 %5d %5d %5d    %5d %5d %5d	   %5d %5d %5d\n", tgt, tgt_stats[tgt].aborts_request, tgt_stats[tgt].aborts_tried, tgt_stats[tgt].aborts_done, tgt_stats[tgt].bdr_request, tgt_stats[tgt].bdr_tried, tgt_stats[tgt].bdr_done, tgt_stats[tgt].adatper_reset_req, tgt_stats[tgt].adapter_reset_attempt, tgt_stats[tgt].adapter_reset_done);
+		seq_printf(m, "  %2d	 %5d %5d %5d    %5d %5d %5d	   %5d %5d %5d\n",
+			   tgt, tgt_stats[tgt].aborts_request,
+			   tgt_stats[tgt].aborts_tried,
+			   tgt_stats[tgt].aborts_done,
+			   tgt_stats[tgt].bdr_request,
+			   tgt_stats[tgt].bdr_tried,
+			   tgt_stats[tgt].bdr_done,
+			   tgt_stats[tgt].adapter_reset_req,
+			   tgt_stats[tgt].adapter_reset_attempt,
+			   tgt_stats[tgt].adapter_reset_done);
 	}
 	seq_printf(m, "\nExternal Host Adapter Resets: %d\n", adapter->ext_resets);
 	seq_printf(m, "Host Adapter Internal Errors: %d\n", adapter->adapter_intern_errors);
diff --git a/drivers/scsi/BusLogic.h b/drivers/scsi/BusLogic.h
index b53ec2f1e8cd..8d47e2c88d24 100644
--- a/drivers/scsi/BusLogic.h
+++ b/drivers/scsi/BusLogic.h
@@ -935,7 +935,7 @@ struct blogic_tgt_stats {
 	unsigned short bdr_request;
 	unsigned short bdr_tried;
 	unsigned short bdr_done;
-	unsigned short adatper_reset_req;
+	unsigned short adapter_reset_req;
 	unsigned short adapter_reset_attempt;
 	unsigned short adapter_reset_done;
 };
diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c
index e3e93def722b..43d88389e899 100644
--- a/drivers/scsi/aacraid/aachba.c
+++ b/drivers/scsi/aacraid/aachba.c
@@ -1678,8 +1678,8 @@ int aac_issue_bmic_identify(struct aac_dev *dev, u32 bus, u32 target)
 			sizeof(struct sgentry) + sizeof(struct sgentry64);
 	datasize = sizeof(struct aac_ciss_identify_pd);
 
-	identify_resp =  pci_alloc_consistent(dev->pdev, datasize, &addr);
-
+	identify_resp = dma_alloc_coherent(&dev->pdev->dev, datasize, &addr,
+					   GFP_KERNEL);
 	if (!identify_resp)
 		goto fib_free_ptr;
 
@@ -1720,7 +1720,7 @@ int aac_issue_bmic_identify(struct aac_dev *dev, u32 bus, u32 target)
 		dev->hba_map[bus][target].qd_limit =
 			identify_resp->current_queue_depth_limit;
 
-	pci_free_consistent(dev->pdev, datasize, (void *)identify_resp, addr);
+	dma_free_coherent(&dev->pdev->dev, datasize, identify_resp, addr);
 
 	aac_fib_complete(fibptr);
 
@@ -1814,9 +1814,8 @@ int aac_report_phys_luns(struct aac_dev *dev, struct fib *fibptr, int rescan)
 	datasize = sizeof(struct aac_ciss_phys_luns_resp)
 			+ (AAC_MAX_TARGETS - 1) * sizeof(struct _ciss_lun);
 
-	phys_luns = (struct aac_ciss_phys_luns_resp *) pci_alloc_consistent(
-			dev->pdev, datasize, &addr);
-
+	phys_luns = dma_alloc_coherent(&dev->pdev->dev, datasize, &addr,
+				       GFP_KERNEL);
 	if (phys_luns == NULL) {
 		rcode = -ENOMEM;
 		goto err_out;
@@ -1861,7 +1860,7 @@ int aac_report_phys_luns(struct aac_dev *dev, struct fib *fibptr, int rescan)
 		aac_update_hba_map(dev, phys_luns, rescan);
 	}
 
-	pci_free_consistent(dev->pdev, datasize, (void *) phys_luns, addr);
+	dma_free_coherent(&dev->pdev->dev, datasize, phys_luns, addr);
 err_out:
 	return rcode;
 }
diff --git a/drivers/scsi/aacraid/commctrl.c b/drivers/scsi/aacraid/commctrl.c
index f6afd50579c0..d2f8d5954840 100644
--- a/drivers/scsi/aacraid/commctrl.c
+++ b/drivers/scsi/aacraid/commctrl.c
@@ -100,7 +100,8 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 			goto cleanup;
 		}
 
-		kfib = pci_alloc_consistent(dev->pdev, size, &daddr);
+		kfib = dma_alloc_coherent(&dev->pdev->dev, size, &daddr,
+					  GFP_KERNEL);
 		if (!kfib) {
 			retval = -ENOMEM;
 			goto cleanup;
@@ -160,7 +161,8 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 		retval = -EFAULT;
 cleanup:
 	if (hw_fib) {
-		pci_free_consistent(dev->pdev, size, kfib, fibptr->hw_fib_pa);
+		dma_free_coherent(&dev->pdev->dev, size, kfib,
+				  fibptr->hw_fib_pa);
 		fibptr->hw_fib_pa = hw_fib_pa;
 		fibptr->hw_fib_va = hw_fib;
 	}
diff --git a/drivers/scsi/aacraid/comminit.c b/drivers/scsi/aacraid/comminit.c
index 35607005f7e1..1151505853cf 100644
--- a/drivers/scsi/aacraid/comminit.c
+++ b/drivers/scsi/aacraid/comminit.c
@@ -99,8 +99,7 @@ static int aac_alloc_comm(struct aac_dev *dev, void **commaddr, unsigned long co
 	size = fibsize + aac_init_size + commsize + commalign +
 			printfbufsiz + host_rrq_size;
 
-	base = pci_alloc_consistent(dev->pdev, size, &phys);
-
+	base = dma_alloc_coherent(&dev->pdev->dev, size, &phys, GFP_KERNEL);
 	if (base == NULL) {
 		printk(KERN_ERR "aacraid: unable to create mapping.\n");
 		return 0;
diff --git a/drivers/scsi/aacraid/commsup.c b/drivers/scsi/aacraid/commsup.c
index 1f4918355fdb..7a1b8a2ce658 100644
--- a/drivers/scsi/aacraid/commsup.c
+++ b/drivers/scsi/aacraid/commsup.c
@@ -73,13 +73,13 @@ static int fib_map_alloc(struct aac_dev *dev)
 	}
 
 	dprintk((KERN_INFO
-	  "allocate hardware fibs pci_alloc_consistent(%p, %d * (%d + %d), %p)\n",
-	  dev->pdev, dev->max_cmd_size, dev->scsi_host_ptr->can_queue,
+	  "allocate hardware fibs dma_alloc_coherent(%p, %d * (%d + %d), %p)\n",
+	  &dev->pdev->dev, dev->max_cmd_size, dev->scsi_host_ptr->can_queue,
 	  AAC_NUM_MGT_FIB, &dev->hw_fib_pa));
-	dev->hw_fib_va = pci_alloc_consistent(dev->pdev,
+	dev->hw_fib_va = dma_alloc_coherent(&dev->pdev->dev,
 		(dev->max_cmd_size + sizeof(struct aac_fib_xporthdr))
 		* (dev->scsi_host_ptr->can_queue + AAC_NUM_MGT_FIB) + (ALIGN32 - 1),
-		&dev->hw_fib_pa);
+		&dev->hw_fib_pa, GFP_KERNEL);
 	if (dev->hw_fib_va == NULL)
 		return -ENOMEM;
 	return 0;
@@ -106,8 +106,8 @@ void aac_fib_map_free(struct aac_dev *dev)
 	fib_size = dev->max_fib_size + sizeof(struct aac_fib_xporthdr);
 	alloc_size = fib_size * num_fibs + ALIGN32 - 1;
 
-	pci_free_consistent(dev->pdev, alloc_size, dev->hw_fib_va,
-							dev->hw_fib_pa);
+	dma_free_coherent(&dev->pdev->dev, alloc_size, dev->hw_fib_va,
+			  dev->hw_fib_pa);
 
 	dev->hw_fib_va = NULL;
 	dev->hw_fib_pa = 0;
@@ -1571,7 +1571,8 @@ static int _aac_reset_adapter(struct aac_dev *aac, int forced, u8 reset_type)
 	 * case.
 	 */
 	aac_fib_map_free(aac);
-	pci_free_consistent(aac->pdev, aac->comm_size, aac->comm_addr, aac->comm_phys);
+	dma_free_coherent(&aac->pdev->dev, aac->comm_size, aac->comm_addr,
+			  aac->comm_phys);
 	aac->comm_addr = NULL;
 	aac->comm_phys = 0;
 	kfree(aac->queues);
@@ -2319,7 +2320,8 @@ static int aac_send_wellness_command(struct aac_dev *dev, char *wellness_str,
 	if (!fibptr)
 		goto out;
 
-	dma_buf = pci_alloc_consistent(dev->pdev, datasize, &addr);
+	dma_buf = dma_alloc_coherent(&dev->pdev->dev, datasize, &addr,
+				     GFP_KERNEL);
 	if (!dma_buf)
 		goto fib_free_out;
 
@@ -2354,7 +2356,7 @@ static int aac_send_wellness_command(struct aac_dev *dev, char *wellness_str,
 	ret = aac_fib_send(ScsiPortCommand64, fibptr, sizeof(struct aac_srb),
 				FsaNormal, 1, 1, NULL, NULL);
 
-	pci_free_consistent(dev->pdev, datasize, (void *)dma_buf, addr);
+	dma_free_coherent(&dev->pdev->dev, datasize, dma_buf, addr);
 
 	/*
 	 * Do not set XferState to zero unless
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 520ada8266af..372a07533026 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -1592,8 +1592,8 @@ static int aac_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
  out_unmap:
 	aac_fib_map_free(aac);
 	if (aac->comm_addr)
-		pci_free_consistent(aac->pdev, aac->comm_size, aac->comm_addr,
-		  aac->comm_phys);
+		dma_free_coherent(&aac->pdev->dev, aac->comm_size,
+				  aac->comm_addr, aac->comm_phys);
 	kfree(aac->queues);
 	aac_adapter_ioremap(aac, 0);
 	kfree(aac->fibs);
@@ -1729,8 +1729,8 @@ static void aac_remove_one(struct pci_dev *pdev)
 
 	__aac_shutdown(aac);
 	aac_fib_map_free(aac);
-	pci_free_consistent(aac->pdev, aac->comm_size, aac->comm_addr,
-			aac->comm_phys);
+	dma_free_coherent(&aac->pdev->dev, aac->comm_size, aac->comm_addr,
+			  aac->comm_phys);
 	kfree(aac->queues);
 
 	aac_adapter_ioremap(aac, 0);
diff --git a/drivers/scsi/aacraid/rx.c b/drivers/scsi/aacraid/rx.c
index 5d19c31e3bba..93ef7c37e568 100644
--- a/drivers/scsi/aacraid/rx.c
+++ b/drivers/scsi/aacraid/rx.c
@@ -355,14 +355,16 @@ static int aac_rx_check_health(struct aac_dev *dev)
 
 		if (likely((status & 0xFF000000L) == 0xBC000000L))
 			return (status >> 16) & 0xFF;
-		buffer = pci_alloc_consistent(dev->pdev, 512, &baddr);
+		buffer = dma_alloc_coherent(&dev->pdev->dev, 512, &baddr,
+					    GFP_KERNEL);
 		ret = -2;
 		if (unlikely(buffer == NULL))
 			return ret;
-		post = pci_alloc_consistent(dev->pdev,
-		  sizeof(struct POSTSTATUS), &paddr);
+		post = dma_alloc_coherent(&dev->pdev->dev,
+					  sizeof(struct POSTSTATUS), &paddr,
+					  GFP_KERNEL);
 		if (unlikely(post == NULL)) {
-			pci_free_consistent(dev->pdev, 512, buffer, baddr);
+			dma_free_coherent(&dev->pdev->dev, 512, buffer, baddr);
 			return ret;
 		}
 		memset(buffer, 0, 512);
@@ -371,13 +373,13 @@ static int aac_rx_check_health(struct aac_dev *dev)
 		rx_writel(dev, MUnit.IMRx[0], paddr);
 		rx_sync_cmd(dev, COMMAND_POST_RESULTS, baddr, 0, 0, 0, 0, 0,
 		  NULL, NULL, NULL, NULL, NULL);
-		pci_free_consistent(dev->pdev, sizeof(struct POSTSTATUS),
-		  post, paddr);
+		dma_free_coherent(&dev->pdev->dev, sizeof(struct POSTSTATUS),
+				  post, paddr);
 		if (likely((buffer[0] == '0') && ((buffer[1] == 'x') || (buffer[1] == 'X')))) {
 			ret = (hex_to_bin(buffer[2]) << 4) +
 				hex_to_bin(buffer[3]);
 		}
-		pci_free_consistent(dev->pdev, 512, buffer, baddr);
+		dma_free_coherent(&dev->pdev->dev, 512, buffer, baddr);
 		return ret;
 	}
 	/*
diff --git a/drivers/scsi/advansys.c b/drivers/scsi/advansys.c
index 81dd0927246b..24e57e770432 100644
--- a/drivers/scsi/advansys.c
+++ b/drivers/scsi/advansys.c
@@ -6291,18 +6291,17 @@ static uchar AscGetSynPeriodIndex(ASC_DVC_VAR *asc_dvc, uchar syn_time)
 static uchar
 AscMsgOutSDTR(ASC_DVC_VAR *asc_dvc, uchar sdtr_period, uchar sdtr_offset)
 {
-	EXT_MSG sdtr_buf;
-	uchar sdtr_period_index;
-	PortAddr iop_base;
-
-	iop_base = asc_dvc->iop_base;
-	sdtr_buf.msg_type = EXTENDED_MESSAGE;
-	sdtr_buf.msg_len = MS_SDTR_LEN;
-	sdtr_buf.msg_req = EXTENDED_SDTR;
-	sdtr_buf.xfer_period = sdtr_period;
+	PortAddr iop_base = asc_dvc->iop_base;
+	uchar sdtr_period_index = AscGetSynPeriodIndex(asc_dvc, sdtr_period);
+	EXT_MSG sdtr_buf = {
+		.msg_type = EXTENDED_MESSAGE,
+		.msg_len = MS_SDTR_LEN,
+		.msg_req = EXTENDED_SDTR,
+		.xfer_period = sdtr_period,
+		.req_ack_offset = sdtr_offset,
+	};
 	sdtr_offset &= ASC_SYN_MAX_OFFSET;
-	sdtr_buf.req_ack_offset = sdtr_offset;
-	sdtr_period_index = AscGetSynPeriodIndex(asc_dvc, sdtr_period);
+
 	if (sdtr_period_index <= asc_dvc->max_sdtr_index) {
 		AscMemWordCopyPtrToLram(iop_base, ASCV_MSGOUT_BEG,
 					(uchar *)&sdtr_buf,
diff --git a/drivers/scsi/aic7xxx/aic7xxx_pci.c b/drivers/scsi/aic7xxx/aic7xxx_pci.c
index 22d5a949ec83..673e826d7adb 100644
--- a/drivers/scsi/aic7xxx/aic7xxx_pci.c
+++ b/drivers/scsi/aic7xxx/aic7xxx_pci.c
@@ -601,8 +601,8 @@ static const u_int ahc_num_pci_devs = ARRAY_SIZE(ahc_pci_ident_table);
 #define STA	0x08
 #define DPR	0x01
 
-static int ahc_9005_subdevinfo_valid(uint16_t vendor, uint16_t device,
-				     uint16_t subvendor, uint16_t subdevice);
+static int ahc_9005_subdevinfo_valid(uint16_t device, uint16_t vendor,
+				     uint16_t subdevice, uint16_t subvendor);
 static int ahc_ext_scbram_present(struct ahc_softc *ahc);
 static void ahc_scbram_config(struct ahc_softc *ahc, int enable,
 				  int pcheck, int fast, int large);
diff --git a/drivers/scsi/aic94xx/aic94xx_init.c b/drivers/scsi/aic94xx/aic94xx_init.c
index 662b2321d1b0..a14ba7a6b81e 100644
--- a/drivers/scsi/aic94xx/aic94xx_init.c
+++ b/drivers/scsi/aic94xx/aic94xx_init.c
@@ -703,7 +703,6 @@ static int asd_unregister_sas_ha(struct asd_ha_struct *asd_ha)
 {
 	int err;
 
-	scsi_remove_host(asd_ha->sas_ha.core.shost);
 	err = sas_unregister_ha(&asd_ha->sas_ha);
 
 	sas_remove_host(asd_ha->sas_ha.core.shost);
diff --git a/drivers/scsi/be2iscsi/be.h b/drivers/scsi/be2iscsi/be.h
index ca9440fb2325..55e3f8b40eb3 100644
--- a/drivers/scsi/be2iscsi/be.h
+++ b/drivers/scsi/be2iscsi/be.h
@@ -1,18 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #ifndef BEISCSI_H
@@ -154,7 +151,6 @@ struct be_ctrl_info {
 #define PAGE_SHIFT_4K 12
 #define PAGE_SIZE_4K (1 << PAGE_SHIFT_4K)
 #define mcc_timeout		120000 /* 12s timeout */
-#define BEISCSI_LOGOUT_SYNC_DELAY	250
 
 /* Returns number of pages spanned by the data starting at the given addr */
 #define PAGES_4K_SPANNED(_address, size)				\
diff --git a/drivers/scsi/be2iscsi/be_cmds.c b/drivers/scsi/be2iscsi/be_cmds.c
index 5d59e2630ce6..a79a5e72c777 100644
--- a/drivers/scsi/be2iscsi/be_cmds.c
+++ b/drivers/scsi/be2iscsi/be_cmds.c
@@ -1,18 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #include <scsi/iscsi_proto.h>
@@ -246,6 +243,12 @@ int beiscsi_mccq_compl_wait(struct beiscsi_hba *phba,
 {
 	int rc = 0;
 
+	if (!tag || tag > MAX_MCC_CMD) {
+		__beiscsi_log(phba, KERN_ERR,
+			      "BC_%d : invalid tag %u\n", tag);
+		return -EINVAL;
+	}
+
 	if (beiscsi_hba_in_error(phba)) {
 		clear_bit(MCC_TAG_STATE_RUNNING,
 			  &phba->ctrl.ptag_state[tag].tag_state);
diff --git a/drivers/scsi/be2iscsi/be_cmds.h b/drivers/scsi/be2iscsi/be_cmds.h
index 1d40e83b0790..d9b6773facdb 100644
--- a/drivers/scsi/be2iscsi/be_cmds.h
+++ b/drivers/scsi/be2iscsi/be_cmds.h
@@ -1,18 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #ifndef BEISCSI_CMDS_H
@@ -1145,24 +1142,49 @@ struct tcp_connect_and_offload_out {
 #define DB_DEF_PDU_EVENT_SHIFT		15
 #define DB_DEF_PDU_CQPROC_SHIFT		16
 
-struct dmsg_cqe {
-	u32 dw[4];
+struct be_invalidate_connection_params_in {
+	struct be_cmd_req_hdr hdr;
+	u32 session_handle;
+	u16 cid;
+	u16 unused;
+#define BE_CLEANUP_TYPE_INVALIDATE	0x8001
+#define BE_CLEANUP_TYPE_ISSUE_TCP_RST	0x8002
+	u16 cleanup_type;
+	u16 save_cfg;
+} __packed;
+
+struct be_invalidate_connection_params_out {
+	u32 session_handle;
+	u16 cid;
+	u16 unused;
+} __packed;
+
+union be_invalidate_connection_params {
+	struct be_invalidate_connection_params_in req;
+	struct be_invalidate_connection_params_out resp;
 } __packed;
 
-struct tcp_upload_params_in {
+struct be_tcp_upload_params_in {
 	struct be_cmd_req_hdr hdr;
 	u16 id;
+#define BE_UPLOAD_TYPE_GRACEFUL		1
+/* abortive upload with reset */
+#define BE_UPLOAD_TYPE_ABORT_RESET	2
+/* abortive upload without reset */
+#define BE_UPLOAD_TYPE_ABORT		3
+/* abortive upload with reset, sequence number by driver */
+#define BE_UPLOAD_TYPE_ABORT_WITH_SEQ	4
 	u16 upload_type;
 	u32 reset_seq;
 } __packed;
 
-struct tcp_upload_params_out {
+struct be_tcp_upload_params_out {
 	u32 dw[32];
 } __packed;
 
-union tcp_upload_params {
-	struct tcp_upload_params_in request;
-	struct tcp_upload_params_out response;
+union be_tcp_upload_params {
+	struct be_tcp_upload_params_in request;
+	struct be_tcp_upload_params_out response;
 } __packed;
 
 struct be_ulp_fw_cfg {
@@ -1243,10 +1265,7 @@ struct be_cmd_get_port_name {
 #define OPCODE_COMMON_WRITE_FLASH		96
 #define OPCODE_COMMON_READ_FLASH		97
 
-/* --- CMD_ISCSI_INVALIDATE_CONNECTION_TYPE --- */
 #define CMD_ISCSI_COMMAND_INVALIDATE		1
-#define CMD_ISCSI_CONNECTION_INVALIDATE		0x8001
-#define CMD_ISCSI_CONNECTION_ISSUE_TCP_RST	0x8002
 
 #define INI_WR_CMD			1	/* Initiator write command */
 #define INI_TMF_CMD			2	/* Initiator TMF command */
@@ -1269,27 +1288,6 @@ struct be_cmd_get_port_name {
 						 *  preparedby
 						 * driver should not be touched
 						 */
-/* --- CMD_CHUTE_TYPE --- */
-#define CMD_CONNECTION_CHUTE_0		1
-#define CMD_CONNECTION_CHUTE_1		2
-#define CMD_CONNECTION_CHUTE_2		3
-
-#define EQ_MAJOR_CODE_COMPLETION	0
-
-#define CMD_ISCSI_SESSION_DEL_CFG_FROM_FLASH 0
-#define CMD_ISCSI_SESSION_SAVE_CFG_ON_FLASH 1
-
-/* --- CONNECTION_UPLOAD_PARAMS --- */
-/* These parameters are used to define the type of upload desired.  */
-#define CONNECTION_UPLOAD_GRACEFUL      1	/* Graceful upload  */
-#define CONNECTION_UPLOAD_ABORT_RESET   2	/* Abortive upload with
-						 * reset
-						 */
-#define CONNECTION_UPLOAD_ABORT		3	/* Abortive upload without
-						 * reset
-						 */
-#define CONNECTION_UPLOAD_ABORT_WITH_SEQ 4	/* Abortive upload with reset,
-						 * sequence number by driver  */
 
 /* Returns the number of items in the field array. */
 #define BE_NUMBER_OF_FIELD(_type_, _field_)	\
diff --git a/drivers/scsi/be2iscsi/be_iscsi.c b/drivers/scsi/be2iscsi/be_iscsi.c
index a4844578e357..97dca4681784 100644
--- a/drivers/scsi/be2iscsi/be_iscsi.c
+++ b/drivers/scsi/be2iscsi/be_iscsi.c
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #include <scsi/libiscsi.h>
@@ -1263,31 +1258,58 @@ static void beiscsi_flush_cq(struct beiscsi_hba *phba)
 }
 
 /**
- * beiscsi_close_conn - Upload the  connection
+ * beiscsi_conn_close - Invalidate and upload connection
  * @ep: The iscsi endpoint
- * @flag: The type of connection closure
+ *
+ * Returns 0 on success,  -1 on failure.
  */
-static int beiscsi_close_conn(struct  beiscsi_endpoint *beiscsi_ep, int flag)
+static int beiscsi_conn_close(struct beiscsi_endpoint *beiscsi_ep)
 {
-	int ret = 0;
-	unsigned int tag;
 	struct beiscsi_hba *phba = beiscsi_ep->phba;
+	unsigned int tag, attempts;
+	int ret;
 
-	tag = mgmt_upload_connection(phba, beiscsi_ep->ep_cid, flag);
-	if (!tag) {
-		beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG,
-			    "BS_%d : upload failed for cid 0x%x\n",
-			    beiscsi_ep->ep_cid);
-
-		ret = -EAGAIN;
+	/**
+	 * Without successfully invalidating and uploading connection
+	 * driver can't reuse the CID so attempt more than once.
+	 */
+	attempts = 0;
+	while (attempts++ < 3) {
+		tag = beiscsi_invalidate_cxn(phba, beiscsi_ep);
+		if (tag) {
+			ret = beiscsi_mccq_compl_wait(phba, tag, NULL, NULL);
+			if (!ret)
+				break;
+			beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG,
+				    "BS_%d : invalidate conn failed cid %d\n",
+				    beiscsi_ep->ep_cid);
+		}
 	}
 
-	ret = beiscsi_mccq_compl_wait(phba, tag, NULL, NULL);
-
-	/* Flush the CQ entries */
+	/* wait for all completions to arrive, then process them */
+	msleep(250);
+	/* flush CQ entries */
 	beiscsi_flush_cq(phba);
 
-	return ret;
+	if (attempts > 3)
+		return -1;
+
+	attempts = 0;
+	while (attempts++ < 3) {
+		tag = beiscsi_upload_cxn(phba, beiscsi_ep);
+		if (tag) {
+			ret = beiscsi_mccq_compl_wait(phba, tag, NULL, NULL);
+			if (!ret)
+				break;
+			beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG,
+				    "BS_%d : upload conn failed cid %d\n",
+				    beiscsi_ep->ep_cid);
+		}
+	}
+	if (attempts > 3)
+		return -1;
+
+	return 0;
 }
 
 /**
@@ -1298,12 +1320,9 @@ static int beiscsi_close_conn(struct  beiscsi_endpoint *beiscsi_ep, int flag)
  */
 void beiscsi_ep_disconnect(struct iscsi_endpoint *ep)
 {
-	struct beiscsi_conn *beiscsi_conn;
 	struct beiscsi_endpoint *beiscsi_ep;
+	struct beiscsi_conn *beiscsi_conn;
 	struct beiscsi_hba *phba;
-	unsigned int tag;
-	uint8_t mgmt_invalidate_flag, tcp_upload_flag;
-	unsigned short savecfg_flag = CMD_ISCSI_SESSION_SAVE_CFG_ON_FLASH;
 	uint16_t cri_index;
 
 	beiscsi_ep = ep->dd_data;
@@ -1324,39 +1343,27 @@ void beiscsi_ep_disconnect(struct iscsi_endpoint *ep)
 	if (beiscsi_ep->conn) {
 		beiscsi_conn = beiscsi_ep->conn;
 		iscsi_suspend_queue(beiscsi_conn->conn);
-		mgmt_invalidate_flag = ~BEISCSI_NO_RST_ISSUE;
-		tcp_upload_flag = CONNECTION_UPLOAD_GRACEFUL;
-	} else {
-		mgmt_invalidate_flag = BEISCSI_NO_RST_ISSUE;
-		tcp_upload_flag = CONNECTION_UPLOAD_ABORT;
 	}
 
 	if (!beiscsi_hba_is_online(phba)) {
 		beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG,
 			    "BS_%d : HBA in error 0x%lx\n", phba->state);
-		goto free_ep;
-	}
-
-	tag = mgmt_invalidate_connection(phba, beiscsi_ep,
-					  beiscsi_ep->ep_cid,
-					  mgmt_invalidate_flag,
-					  savecfg_flag);
-	if (!tag) {
-		beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG,
-			    "BS_%d : mgmt_invalidate_connection Failed for cid=%d\n",
-			    beiscsi_ep->ep_cid);
+	} else {
+		/**
+		 * Make CID available even if close fails.
+		 * If not freed, FW might fail open using the CID.
+		 */
+		if (beiscsi_conn_close(beiscsi_ep) < 0)
+			__beiscsi_log(phba, KERN_ERR,
+				      "BS_%d : close conn failed cid %d\n",
+				      beiscsi_ep->ep_cid);
 	}
 
-	beiscsi_mccq_compl_wait(phba, tag, NULL, NULL);
-	beiscsi_close_conn(beiscsi_ep, tcp_upload_flag);
-free_ep:
-	msleep(BEISCSI_LOGOUT_SYNC_DELAY);
 	beiscsi_free_ep(beiscsi_ep);
 	if (!phba->conn_table[cri_index])
 		__beiscsi_log(phba, KERN_ERR,
-				"BS_%d : conn_table empty at %u: cid %u\n",
-				cri_index,
-				beiscsi_ep->ep_cid);
+			      "BS_%d : conn_table empty at %u: cid %u\n",
+			      cri_index, beiscsi_ep->ep_cid);
 	phba->conn_table[cri_index] = NULL;
 	iscsi_destroy_endpoint(beiscsi_ep->openiscsi_ep);
 }
diff --git a/drivers/scsi/be2iscsi/be_iscsi.h b/drivers/scsi/be2iscsi/be_iscsi.h
index e4d67dfea4cb..b9d459a21f25 100644
--- a/drivers/scsi/be2iscsi/be_iscsi.h
+++ b/drivers/scsi/be2iscsi/be_iscsi.h
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Avago Technologies
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #ifndef _BE_ISCSI_
diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c
index 32b2713cec93..f862332261f8 100644
--- a/drivers/scsi/be2iscsi/be_main.c
+++ b/drivers/scsi/be2iscsi/be_main.c
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #include <linux/reboot.h>
@@ -337,7 +332,7 @@ static int beiscsi_eh_device_reset(struct scsi_cmnd *sc)
 		inv_tbl->task[nents] = task;
 		nents++;
 	}
-	spin_unlock_bh(&session->back_lock);
+	spin_unlock(&session->back_lock);
 	spin_unlock_bh(&session->frwd_lock);
 
 	rc = SUCCESS;
@@ -636,7 +631,6 @@ static void beiscsi_get_params(struct beiscsi_hba *phba)
 				    (total_cid_count +
 				     BE2_TMFS + BE2_NOPOUT_REQ));
 	phba->params.cxns_per_ctrl = total_cid_count;
-	phba->params.asyncpdus_per_ctrl = total_cid_count;
 	phba->params.icds_per_ctrl = total_icd_count;
 	phba->params.num_sge_per_io = BE2_SGE;
 	phba->params.defpdu_hdr_sz = BE2_DEFPDU_HDR_SZ;
@@ -802,12 +796,12 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba)
 	struct pci_dev *pcidev = phba->pcidev;
 	struct hwi_controller *phwi_ctrlr;
 	struct hwi_context_memory *phwi_context;
-	int ret, msix_vec, i, j;
+	int ret, i, j;
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
 
-	if (phba->msix_enabled) {
+	if (pcidev->msix_enabled) {
 		for (i = 0; i < phba->num_cpus; i++) {
 			phba->msi_name[i] = kzalloc(BEISCSI_MSI_NAME,
 						    GFP_KERNEL);
@@ -818,9 +812,8 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba)
 
 			sprintf(phba->msi_name[i], "beiscsi_%02x_%02x",
 				phba->shost->host_no, i);
-			msix_vec = phba->msix_entries[i].vector;
-			ret = request_irq(msix_vec, be_isr_msix, 0,
-					  phba->msi_name[i],
+			ret = request_irq(pci_irq_vector(pcidev, i),
+					  be_isr_msix, 0, phba->msi_name[i],
 					  &phwi_context->be_eq[i]);
 			if (ret) {
 				beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT,
@@ -838,9 +831,8 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba)
 		}
 		sprintf(phba->msi_name[i], "beiscsi_mcc_%02x",
 			phba->shost->host_no);
-		msix_vec = phba->msix_entries[i].vector;
-		ret = request_irq(msix_vec, be_isr_mcc, 0, phba->msi_name[i],
-				  &phwi_context->be_eq[i]);
+		ret = request_irq(pci_irq_vector(pcidev, i), be_isr_mcc, 0,
+				  phba->msi_name[i], &phwi_context->be_eq[i]);
 		if (ret) {
 			beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT ,
 				    "BM_%d : beiscsi_init_irqs-"
@@ -862,9 +854,8 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba)
 	return 0;
 free_msix_irqs:
 	for (j = i - 1; j >= 0; j--) {
+		free_irq(pci_irq_vector(pcidev, i), &phwi_context->be_eq[j]);
 		kfree(phba->msi_name[j]);
-		msix_vec = phba->msix_entries[j].vector;
-		free_irq(msix_vec, &phwi_context->be_eq[j]);
 	}
 	return ret;
 }
@@ -1454,30 +1445,45 @@ static inline void
 beiscsi_hdl_put_handle(struct hd_async_context *pasync_ctx,
 			 struct hd_async_handle *pasync_handle)
 {
-	if (pasync_handle->is_header) {
-		list_add_tail(&pasync_handle->link,
-				&pasync_ctx->async_header.free_list);
-		pasync_ctx->async_header.free_entries++;
-	} else {
-		list_add_tail(&pasync_handle->link,
-				&pasync_ctx->async_data.free_list);
-		pasync_ctx->async_data.free_entries++;
-	}
+	pasync_handle->is_final = 0;
+	pasync_handle->buffer_len = 0;
+	pasync_handle->in_use = 0;
+	list_del_init(&pasync_handle->link);
+}
+
+static void
+beiscsi_hdl_purge_handles(struct beiscsi_hba *phba,
+			  struct hd_async_context *pasync_ctx,
+			  u16 cri)
+{
+	struct hd_async_handle *pasync_handle, *tmp_handle;
+	struct list_head *plist;
+
+	plist  = &pasync_ctx->async_entry[cri].wq.list;
+	list_for_each_entry_safe(pasync_handle, tmp_handle, plist, link)
+		beiscsi_hdl_put_handle(pasync_ctx, pasync_handle);
+
+	INIT_LIST_HEAD(&pasync_ctx->async_entry[cri].wq.list);
+	pasync_ctx->async_entry[cri].wq.hdr_len = 0;
+	pasync_ctx->async_entry[cri].wq.bytes_received = 0;
+	pasync_ctx->async_entry[cri].wq.bytes_needed = 0;
 }
 
 static struct hd_async_handle *
 beiscsi_hdl_get_handle(struct beiscsi_conn *beiscsi_conn,
 		       struct hd_async_context *pasync_ctx,
-		       struct i_t_dpdu_cqe *pdpdu_cqe)
+		       struct i_t_dpdu_cqe *pdpdu_cqe,
+		       u8 *header)
 {
 	struct beiscsi_hba *phba = beiscsi_conn->phba;
 	struct hd_async_handle *pasync_handle;
 	struct be_bus_address phys_addr;
+	u16 cid, code, ci, cri;
 	u8 final, error = 0;
-	u16 cid, code, ci;
 	u32 dpl;
 
 	cid = beiscsi_conn->beiscsi_conn_cid;
+	cri = BE_GET_ASYNC_CRI_FROM_CID(cid);
 	/**
 	 * This function is invoked to get the right async_handle structure
 	 * from a given DEF PDU CQ entry.
@@ -1516,6 +1522,7 @@ beiscsi_hdl_get_handle(struct beiscsi_conn *beiscsi_conn,
 	switch (code) {
 	case UNSOL_HDR_NOTIFY:
 		pasync_handle = pasync_ctx->async_entry[ci].header;
+		*header = 1;
 		break;
 	case UNSOL_DATA_DIGEST_ERROR_NOTIFY:
 		error = 1;
@@ -1524,15 +1531,7 @@ beiscsi_hdl_get_handle(struct beiscsi_conn *beiscsi_conn,
 		break;
 	/* called only for above codes */
 	default:
-		pasync_handle = NULL;
-		break;
-	}
-
-	if (!pasync_handle) {
-		beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_ISCSI,
-			    "BM_%d : cid %d async PDU handle not found - code %d ci %d addr %llx\n",
-			    cid, code, ci, phys_addr.u.a64.address);
-		return pasync_handle;
+		return NULL;
 	}
 
 	if (pasync_handle->pa.u.a64.address != phys_addr.u.a64.address ||
@@ -1549,47 +1548,32 @@ beiscsi_hdl_get_handle(struct beiscsi_conn *beiscsi_conn,
 	}
 
 	/**
-	 * Each CID is associated with unique CRI.
-	 * ASYNC_CRI_FROM_CID mapping and CRI_FROM_CID are totaly different.
-	 **/
-	pasync_handle->cri = BE_GET_ASYNC_CRI_FROM_CID(cid);
-	pasync_handle->is_final = final;
-	pasync_handle->buffer_len = dpl;
-	/* empty the slot */
-	if (pasync_handle->is_header)
-		pasync_ctx->async_entry[ci].header = NULL;
-	else
-		pasync_ctx->async_entry[ci].data = NULL;
-
-	/**
 	 * DEF PDU header and data buffers with errors should be simply
 	 * dropped as there are no consumers for it.
 	 */
 	if (error) {
 		beiscsi_hdl_put_handle(pasync_ctx, pasync_handle);
-		pasync_handle = NULL;
+		return NULL;
 	}
-	return pasync_handle;
-}
-
-static void
-beiscsi_hdl_purge_handles(struct beiscsi_hba *phba,
-			  struct hd_async_context *pasync_ctx,
-			  u16 cri)
-{
-	struct hd_async_handle *pasync_handle, *tmp_handle;
-	struct list_head *plist;
 
-	plist  = &pasync_ctx->async_entry[cri].wq.list;
-	list_for_each_entry_safe(pasync_handle, tmp_handle, plist, link) {
-		list_del(&pasync_handle->link);
-		beiscsi_hdl_put_handle(pasync_ctx, pasync_handle);
+	if (pasync_handle->in_use || !list_empty(&pasync_handle->link)) {
+		beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_ISCSI,
+			    "BM_%d : cid %d async PDU handle in use - code %d ci %d addr %llx\n",
+			    cid, code, ci, phys_addr.u.a64.address);
+		beiscsi_hdl_purge_handles(phba, pasync_ctx, cri);
 	}
 
-	INIT_LIST_HEAD(&pasync_ctx->async_entry[cri].wq.list);
-	pasync_ctx->async_entry[cri].wq.hdr_len = 0;
-	pasync_ctx->async_entry[cri].wq.bytes_received = 0;
-	pasync_ctx->async_entry[cri].wq.bytes_needed = 0;
+	list_del_init(&pasync_handle->link);
+	/**
+	 * Each CID is associated with unique CRI.
+	 * ASYNC_CRI_FROM_CID mapping and CRI_FROM_CID are totaly different.
+	 **/
+	pasync_handle->cri = cri;
+	pasync_handle->is_final = final;
+	pasync_handle->buffer_len = dpl;
+	pasync_handle->in_use = 1;
+
+	return pasync_handle;
 }
 
 static unsigned int
@@ -1619,6 +1603,10 @@ beiscsi_hdl_fwd_pdu(struct beiscsi_conn *beiscsi_conn,
 			dlen = pasync_handle->buffer_len;
 			continue;
 		}
+		if (!pasync_handle->buffer_len ||
+		    (dlen + pasync_handle->buffer_len) >
+		    pasync_ctx->async_data.buffer_size)
+			break;
 		memcpy(pdata + dlen, pasync_handle->pbuffer,
 		       pasync_handle->buffer_len);
 		dlen += pasync_handle->buffer_len;
@@ -1627,8 +1615,9 @@ beiscsi_hdl_fwd_pdu(struct beiscsi_conn *beiscsi_conn,
 	if (!plast_handle->is_final) {
 		/* last handle should have final PDU notification from FW */
 		beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_ISCSI,
-			    "BM_%d : cid %u %p fwd async PDU with last handle missing - HL%u:DN%u:DR%u\n",
+			    "BM_%d : cid %u %p fwd async PDU opcode %x with last handle missing - HL%u:DN%u:DR%u\n",
 			    beiscsi_conn->beiscsi_conn_cid, plast_handle,
+			    AMAP_GET_BITS(struct amap_pdu_base, opcode, phdr),
 			    pasync_ctx->async_entry[cri].wq.hdr_len,
 			    pasync_ctx->async_entry[cri].wq.bytes_needed,
 			    pasync_ctx->async_entry[cri].wq.bytes_received);
@@ -1709,85 +1698,53 @@ drop_pdu:
 
 static void
 beiscsi_hdq_post_handles(struct beiscsi_hba *phba,
-			 u8 header, u8 ulp_num)
+			 u8 header, u8 ulp_num, u16 nbuf)
 {
-	struct hd_async_handle *pasync_handle, *tmp, **slot;
+	struct hd_async_handle *pasync_handle;
 	struct hd_async_context *pasync_ctx;
 	struct hwi_controller *phwi_ctrlr;
-	struct list_head *hfree_list;
 	struct phys_addr *pasync_sge;
 	u32 ring_id, doorbell = 0;
 	u32 doorbell_offset;
-	u16 prod = 0, cons;
-	u16 index;
+	u16 prod, pi;
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	pasync_ctx = HWI_GET_ASYNC_PDU_CTX(phwi_ctrlr, ulp_num);
 	if (header) {
-		cons = pasync_ctx->async_header.free_entries;
-		hfree_list = &pasync_ctx->async_header.free_list;
+		pasync_sge = pasync_ctx->async_header.ring_base;
+		pi = pasync_ctx->async_header.pi;
 		ring_id = phwi_ctrlr->default_pdu_hdr[ulp_num].id;
 		doorbell_offset = phwi_ctrlr->default_pdu_hdr[ulp_num].
 					doorbell_offset;
 	} else {
-		cons = pasync_ctx->async_data.free_entries;
-		hfree_list = &pasync_ctx->async_data.free_list;
+		pasync_sge = pasync_ctx->async_data.ring_base;
+		pi = pasync_ctx->async_data.pi;
 		ring_id = phwi_ctrlr->default_pdu_data[ulp_num].id;
 		doorbell_offset = phwi_ctrlr->default_pdu_data[ulp_num].
 					doorbell_offset;
 	}
-	/* number of entries posted must be in multiples of 8 */
-	if (cons % 8)
-		return;
-
-	list_for_each_entry_safe(pasync_handle, tmp, hfree_list, link) {
-		list_del_init(&pasync_handle->link);
-		pasync_handle->is_final = 0;
-		pasync_handle->buffer_len = 0;
 
-		/* handles can be consumed out of order, use index in handle */
-		index = pasync_handle->index;
-		WARN_ON(pasync_handle->is_header != header);
+	for (prod = 0; prod < nbuf; prod++) {
 		if (header)
-			slot = &pasync_ctx->async_entry[index].header;
+			pasync_handle = pasync_ctx->async_entry[pi].header;
 		else
-			slot = &pasync_ctx->async_entry[index].data;
-		/**
-		 * The slot just tracks handle's hold and release, so
-		 * overwriting at the same index won't do any harm but
-		 * needs to be caught.
-		 */
-		if (*slot != NULL) {
-			beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_ISCSI,
-				    "BM_%d : async PDU %s slot at %u not empty\n",
-				    header ? "header" : "data", index);
+			pasync_handle = pasync_ctx->async_entry[pi].data;
+		WARN_ON(pasync_handle->is_header != header);
+		WARN_ON(pasync_handle->index != pi);
+		/* setup the ring only once */
+		if (nbuf == pasync_ctx->num_entries) {
+			/* note hi is lo */
+			pasync_sge[pi].hi = pasync_handle->pa.u.a32.address_lo;
+			pasync_sge[pi].lo = pasync_handle->pa.u.a32.address_hi;
 		}
-		/**
-		 * We use same freed index as in completion to post so this
-		 * operation is not required for refills. Its required only
-		 * for ring creation.
-		 */
-		if (header)
-			pasync_sge = pasync_ctx->async_header.ring_base;
-		else
-			pasync_sge = pasync_ctx->async_data.ring_base;
-		pasync_sge += index;
-		/* if its a refill then address is same; hi is lo */
-		WARN_ON(pasync_sge->hi &&
-			pasync_sge->hi != pasync_handle->pa.u.a32.address_lo);
-		WARN_ON(pasync_sge->lo &&
-			pasync_sge->lo != pasync_handle->pa.u.a32.address_hi);
-		pasync_sge->hi = pasync_handle->pa.u.a32.address_lo;
-		pasync_sge->lo = pasync_handle->pa.u.a32.address_hi;
-
-		*slot = pasync_handle;
-		if (++prod == cons)
-			break;
+		if (++pi == pasync_ctx->num_entries)
+			pi = 0;
 	}
+
 	if (header)
-		pasync_ctx->async_header.free_entries -= prod;
+		pasync_ctx->async_header.pi = pi;
 	else
-		pasync_ctx->async_data.free_entries -= prod;
+		pasync_ctx->async_data.pi = pi;
 
 	doorbell |= ring_id & DB_DEF_PDU_RING_ID_MASK;
 	doorbell |= 1 << DB_DEF_PDU_REARM_SHIFT;
@@ -1804,20 +1761,26 @@ beiscsi_hdq_process_compl(struct beiscsi_conn *beiscsi_conn,
 	struct hd_async_handle *pasync_handle = NULL;
 	struct hd_async_context *pasync_ctx;
 	struct hwi_controller *phwi_ctrlr;
+	u8 ulp_num, consumed, header = 0;
 	u16 cid_cri;
-	u8 ulp_num;
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	cid_cri = BE_GET_CRI_FROM_CID(beiscsi_conn->beiscsi_conn_cid);
 	ulp_num = BEISCSI_GET_ULP_FROM_CRI(phwi_ctrlr, cid_cri);
 	pasync_ctx = HWI_GET_ASYNC_PDU_CTX(phwi_ctrlr, ulp_num);
 	pasync_handle = beiscsi_hdl_get_handle(beiscsi_conn, pasync_ctx,
-					       pdpdu_cqe);
-	if (!pasync_handle)
-		return;
-
-	beiscsi_hdl_gather_pdu(beiscsi_conn, pasync_ctx, pasync_handle);
-	beiscsi_hdq_post_handles(phba, pasync_handle->is_header, ulp_num);
+					       pdpdu_cqe, &header);
+	if (is_chip_be2_be3r(phba))
+		consumed = AMAP_GET_BITS(struct amap_i_t_dpdu_cqe,
+					 num_cons, pdpdu_cqe);
+	else
+		consumed = AMAP_GET_BITS(struct amap_i_t_dpdu_cqe_v2,
+					 num_cons, pdpdu_cqe);
+	if (pasync_handle)
+		beiscsi_hdl_gather_pdu(beiscsi_conn, pasync_ctx, pasync_handle);
+	/* num_cons indicates number of 8 RQEs consumed */
+	if (consumed)
+		beiscsi_hdq_post_handles(phba, header, ulp_num, 8 * consumed);
 }
 
 void beiscsi_process_mcc_cq(struct beiscsi_hba *phba)
@@ -2407,22 +2370,22 @@ static void beiscsi_find_mem_req(struct beiscsi_hba *phba)
 		if (test_bit(ulp_num, &phba->fw_config.ulp_supported)) {
 
 			num_async_pdu_buf_sgl_pages =
-				PAGES_REQUIRED(BEISCSI_GET_CID_COUNT(
+				PAGES_REQUIRED(BEISCSI_ASYNC_HDQ_SIZE(
 					       phba, ulp_num) *
 					       sizeof(struct phys_addr));
 
 			num_async_pdu_buf_pages =
-				PAGES_REQUIRED(BEISCSI_GET_CID_COUNT(
+				PAGES_REQUIRED(BEISCSI_ASYNC_HDQ_SIZE(
 					       phba, ulp_num) *
 					       phba->params.defpdu_hdr_sz);
 
 			num_async_pdu_data_pages =
-				PAGES_REQUIRED(BEISCSI_GET_CID_COUNT(
+				PAGES_REQUIRED(BEISCSI_ASYNC_HDQ_SIZE(
 					       phba, ulp_num) *
 					       phba->params.defpdu_data_sz);
 
 			num_async_pdu_data_sgl_pages =
-				PAGES_REQUIRED(BEISCSI_GET_CID_COUNT(
+				PAGES_REQUIRED(BEISCSI_ASYNC_HDQ_SIZE(
 					       phba, ulp_num) *
 					       sizeof(struct phys_addr));
 
@@ -2459,21 +2422,21 @@ static void beiscsi_find_mem_req(struct beiscsi_hba *phba)
 			mem_descr_index = (HWI_MEM_ASYNC_HEADER_HANDLE_ULP0 +
 					  (ulp_num * MEM_DESCR_OFFSET));
 			phba->mem_req[mem_descr_index] =
-					  BEISCSI_GET_CID_COUNT(phba, ulp_num) *
-					  sizeof(struct hd_async_handle);
+				BEISCSI_ASYNC_HDQ_SIZE(phba, ulp_num) *
+				sizeof(struct hd_async_handle);
 
 			mem_descr_index = (HWI_MEM_ASYNC_DATA_HANDLE_ULP0 +
 					  (ulp_num * MEM_DESCR_OFFSET));
 			phba->mem_req[mem_descr_index] =
-					  BEISCSI_GET_CID_COUNT(phba, ulp_num) *
-					  sizeof(struct hd_async_handle);
+				BEISCSI_ASYNC_HDQ_SIZE(phba, ulp_num) *
+				sizeof(struct hd_async_handle);
 
 			mem_descr_index = (HWI_MEM_ASYNC_PDU_CONTEXT_ULP0 +
 					  (ulp_num * MEM_DESCR_OFFSET));
 			phba->mem_req[mem_descr_index] =
-					  sizeof(struct hd_async_context) +
-					 (BEISCSI_GET_CID_COUNT(phba, ulp_num) *
-					  sizeof(struct hd_async_entry));
+				sizeof(struct hd_async_context) +
+				(BEISCSI_ASYNC_HDQ_SIZE(phba, ulp_num) *
+				 sizeof(struct hd_async_entry));
 		}
 	}
 }
@@ -2757,7 +2720,7 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 					((long unsigned int)pasync_ctx +
 					sizeof(struct hd_async_context));
 
-			pasync_ctx->num_entries = BEISCSI_GET_CID_COUNT(phba,
+			pasync_ctx->num_entries = BEISCSI_ASYNC_HDQ_SIZE(phba,
 						  ulp_num);
 			/* setup header buffers */
 			mem_descr = (struct be_mem_descriptor *)phba->init_mem;
@@ -2776,6 +2739,7 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 					    "BM_%d : No Virtual address for ULP : %d\n",
 					    ulp_num);
 
+			pasync_ctx->async_header.pi = 0;
 			pasync_ctx->async_header.buffer_size = p->defpdu_hdr_sz;
 			pasync_ctx->async_header.va_base =
 				mem_descr->mem_array[0].virtual_address;
@@ -2823,7 +2787,6 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 
 			pasync_ctx->async_header.handle_base =
 				mem_descr->mem_array[0].virtual_address;
-			INIT_LIST_HEAD(&pasync_ctx->async_header.free_list);
 
 			/* setup data buffer sgls */
 			mem_descr = (struct be_mem_descriptor *)phba->init_mem;
@@ -2857,7 +2820,6 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 
 			pasync_ctx->async_data.handle_base =
 				mem_descr->mem_array[0].virtual_address;
-			INIT_LIST_HEAD(&pasync_ctx->async_data.free_list);
 
 			pasync_header_h =
 				(struct hd_async_handle *)
@@ -2884,6 +2846,7 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 					    ulp_num);
 
 			idx = 0;
+			pasync_ctx->async_data.pi = 0;
 			pasync_ctx->async_data.buffer_size = p->defpdu_data_sz;
 			pasync_ctx->async_data.va_base =
 				mem_descr->mem_array[idx].virtual_address;
@@ -2895,7 +2858,7 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 					phba->params.defpdu_data_sz);
 			num_per_mem = 0;
 
-			for (index = 0;	index < BEISCSI_GET_CID_COUNT
+			for (index = 0;	index < BEISCSI_ASYNC_HDQ_SIZE
 					(phba, ulp_num); index++) {
 				pasync_header_h->cri = -1;
 				pasync_header_h->is_header = 1;
@@ -2911,14 +2874,11 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 					pasync_ctx->async_header.pa_base.u.a64.
 					address + (p->defpdu_hdr_sz * index);
 
-				list_add_tail(&pasync_header_h->link,
-					      &pasync_ctx->async_header.
-					      free_list);
+				pasync_ctx->async_entry[index].header =
+					pasync_header_h;
 				pasync_header_h++;
-				pasync_ctx->async_header.free_entries++;
 				INIT_LIST_HEAD(&pasync_ctx->async_entry[index].
 						wq.list);
-				pasync_ctx->async_entry[index].header = NULL;
 
 				pasync_data_h->cri = -1;
 				pasync_data_h->is_header = 0;
@@ -2952,12 +2912,9 @@ static int hwi_init_async_pdu_ctx(struct beiscsi_hba *phba)
 				num_per_mem++;
 				num_async_data--;
 
-				list_add_tail(&pasync_data_h->link,
-					      &pasync_ctx->async_data.
-					      free_list);
+				pasync_ctx->async_entry[index].data =
+					pasync_data_h;
 				pasync_data_h++;
-				pasync_ctx->async_data.free_entries++;
-				pasync_ctx->async_entry[index].data = NULL;
 			}
 		}
 	}
@@ -3040,7 +2997,7 @@ static int beiscsi_create_eqs(struct beiscsi_hba *phba,
 	num_eq_pages = PAGES_REQUIRED(phba->params.num_eq_entries * \
 				      sizeof(struct be_eq_entry));
 
-	if (phba->msix_enabled)
+	if (phba->pcidev->msix_enabled)
 		eq_for_mcc = 1;
 	else
 		eq_for_mcc = 0;
@@ -3550,7 +3507,7 @@ static int be_mcc_queues_create(struct beiscsi_hba *phba,
 			sizeof(struct be_mcc_compl)))
 		goto err;
 	/* Ask BE to create MCC compl queue; */
-	if (phba->msix_enabled) {
+	if (phba->pcidev->msix_enabled) {
 		if (beiscsi_cmd_cq_create(ctrl, cq, &phwi_context->be_eq
 					 [phba->num_cpus].q, false, true, 0))
 		goto mcc_cq_free;
@@ -3581,42 +3538,35 @@ err:
 	return -ENOMEM;
 }
 
-/**
- * find_num_cpus()- Get the CPU online count
- * @phba: ptr to priv structure
- *
- * CPU count is used for creating EQ.
- **/
-static void find_num_cpus(struct beiscsi_hba *phba)
+static void be2iscsi_enable_msix(struct beiscsi_hba *phba)
 {
-	int  num_cpus = 0;
-
-	num_cpus = num_online_cpus();
+	int nvec = 1;
 
 	switch (phba->generation) {
 	case BE_GEN2:
 	case BE_GEN3:
-		phba->num_cpus = (num_cpus > BEISCSI_MAX_NUM_CPUS) ?
-				  BEISCSI_MAX_NUM_CPUS : num_cpus;
+		nvec = BEISCSI_MAX_NUM_CPUS + 1;
 		break;
 	case BE_GEN4:
-		/*
-		 * If eqid_count == 1 fall back to
-		 * INTX mechanism
-		 **/
-		if (phba->fw_config.eqid_count == 1) {
-			enable_msix = 0;
-			phba->num_cpus = 1;
-			return;
-		}
-
-		phba->num_cpus =
-			(num_cpus > (phba->fw_config.eqid_count - 1)) ?
-			(phba->fw_config.eqid_count - 1) : num_cpus;
+		nvec = phba->fw_config.eqid_count;
 		break;
 	default:
-		phba->num_cpus = 1;
+		nvec = 2;
+		break;
 	}
+
+	/* if eqid_count == 1 fall back to INTX */
+	if (enable_msix && nvec > 1) {
+		const struct irq_affinity desc = { .post_vectors = 1 };
+
+		if (pci_alloc_irq_vectors_affinity(phba->pcidev, 2, nvec,
+				PCI_IRQ_MSIX | PCI_IRQ_AFFINITY, &desc) < 0) {
+			phba->num_cpus = nvec - 1;
+			return;
+		}
+	}
+
+	phba->num_cpus = 1;
 }
 
 static void hwi_purge_eq(struct beiscsi_hba *phba)
@@ -3633,7 +3583,7 @@ static void hwi_purge_eq(struct beiscsi_hba *phba)
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
-	if (phba->msix_enabled)
+	if (phba->pcidev->msix_enabled)
 		eq_msix = 1;
 	else
 		eq_msix = 0;
@@ -3711,7 +3661,7 @@ static void hwi_cleanup_port(struct beiscsi_hba *phba)
 	}
 
 	be_mcc_queues_destroy(phba);
-	if (phba->msix_enabled)
+	if (phba->pcidev->msix_enabled)
 		eq_for_mcc = 1;
 	else
 		eq_for_mcc = 0;
@@ -3735,6 +3685,7 @@ static int hwi_init_port(struct beiscsi_hba *phba)
 	unsigned int def_pdu_ring_sz;
 	struct be_ctrl_info *ctrl = &phba->ctrl;
 	int status, ulp_num;
+	u16 nbufs;
 
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
@@ -3771,9 +3722,8 @@ static int hwi_init_port(struct beiscsi_hba *phba)
 
 	for (ulp_num = 0; ulp_num < BEISCSI_ULP_COUNT; ulp_num++) {
 		if (test_bit(ulp_num, &phba->fw_config.ulp_supported)) {
-			def_pdu_ring_sz =
-				BEISCSI_GET_CID_COUNT(phba, ulp_num) *
-				sizeof(struct phys_addr);
+			nbufs = phwi_context->pasync_ctx[ulp_num]->num_entries;
+			def_pdu_ring_sz = nbufs * sizeof(struct phys_addr);
 
 			status = beiscsi_create_def_hdr(phba, phwi_context,
 							phwi_ctrlr,
@@ -3801,9 +3751,9 @@ static int hwi_init_port(struct beiscsi_hba *phba)
 			 * let EP know about it.
 			 */
 			beiscsi_hdq_post_handles(phba, BEISCSI_DEFQ_HDR,
-						 ulp_num);
+						 ulp_num, nbufs);
 			beiscsi_hdq_post_handles(phba, BEISCSI_DEFQ_DATA,
-						 ulp_num);
+						 ulp_num, nbufs);
 		}
 	}
 
@@ -4157,7 +4107,7 @@ static void hwi_enable_intr(struct beiscsi_hba *phba)
 		iowrite32(reg, addr);
 	}
 
-	if (!phba->msix_enabled) {
+	if (!phba->pcidev->msix_enabled) {
 		eq = &phwi_context->be_eq[0].q;
 		beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT,
 			    "BM_%d : eq->id=%d\n", eq->id);
@@ -5280,19 +5230,6 @@ static void beiscsi_eqd_update_work(struct work_struct *work)
 			      msecs_to_jiffies(BEISCSI_EQD_UPDATE_INTERVAL));
 }
 
-static void beiscsi_msix_enable(struct beiscsi_hba *phba)
-{
-	int i, status;
-
-	for (i = 0; i <= phba->num_cpus; i++)
-		phba->msix_entries[i].entry = i;
-
-	status = pci_enable_msix_range(phba->pcidev, phba->msix_entries,
-				       phba->num_cpus + 1, phba->num_cpus + 1);
-	if (status > 0)
-		phba->msix_enabled = true;
-}
-
 static void beiscsi_hw_tpe_check(unsigned long ptr)
 {
 	struct beiscsi_hba *phba;
@@ -5360,15 +5297,7 @@ static int beiscsi_enable_port(struct beiscsi_hba *phba)
 	if (ret)
 		return ret;
 
-	if (enable_msix)
-		find_num_cpus(phba);
-	else
-		phba->num_cpus = 1;
-	if (enable_msix) {
-		beiscsi_msix_enable(phba);
-		if (!phba->msix_enabled)
-			phba->num_cpus = 1;
-	}
+	be2iscsi_enable_msix(phba);
 
 	beiscsi_get_params(phba);
 	/* Re-enable UER. If different TPE occurs then it is recoverable. */
@@ -5397,7 +5326,7 @@ static int beiscsi_enable_port(struct beiscsi_hba *phba)
 		irq_poll_init(&pbe_eq->iopoll, be_iopoll_budget, be_iopoll);
 	}
 
-	i = (phba->msix_enabled) ? i : 0;
+	i = (phba->pcidev->msix_enabled) ? i : 0;
 	/* Work item for MCC handling */
 	pbe_eq = &phwi_context->be_eq[i];
 	INIT_WORK(&pbe_eq->mcc_work, beiscsi_mcc_work);
@@ -5435,9 +5364,7 @@ cleanup_port:
 	hwi_cleanup_port(phba);
 
 disable_msix:
-	if (phba->msix_enabled)
-		pci_disable_msix(phba->pcidev);
-
+	pci_free_irq_vectors(phba->pcidev);
 	return ret;
 }
 
@@ -5454,7 +5381,7 @@ static void beiscsi_disable_port(struct beiscsi_hba *phba, int unload)
 	struct hwi_context_memory *phwi_context;
 	struct hwi_controller *phwi_ctrlr;
 	struct be_eq_obj *pbe_eq;
-	unsigned int i, msix_vec;
+	unsigned int i;
 
 	if (!test_and_clear_bit(BEISCSI_HBA_ONLINE, &phba->state))
 		return;
@@ -5462,16 +5389,16 @@ static void beiscsi_disable_port(struct beiscsi_hba *phba, int unload)
 	phwi_ctrlr = phba->phwi_ctrlr;
 	phwi_context = phwi_ctrlr->phwi_ctxt;
 	hwi_disable_intr(phba);
-	if (phba->msix_enabled) {
+	if (phba->pcidev->msix_enabled) {
 		for (i = 0; i <= phba->num_cpus; i++) {
-			msix_vec = phba->msix_entries[i].vector;
-			free_irq(msix_vec, &phwi_context->be_eq[i]);
+			free_irq(pci_irq_vector(phba->pcidev, i),
+				&phwi_context->be_eq[i]);
 			kfree(phba->msi_name[i]);
 		}
 	} else
 		if (phba->pcidev->irq)
 			free_irq(phba->pcidev->irq, phba);
-	pci_disable_msix(phba->pcidev);
+	pci_free_irq_vectors(phba->pcidev);
 
 	for (i = 0; i < phba->num_cpus; i++) {
 		pbe_eq = &phwi_context->be_eq[i];
@@ -5681,21 +5608,12 @@ static int beiscsi_dev_probe(struct pci_dev *pcidev,
 	beiscsi_get_params(phba);
 	beiscsi_set_uer_feature(phba);
 
-	if (enable_msix)
-		find_num_cpus(phba);
-	else
-		phba->num_cpus = 1;
+	be2iscsi_enable_msix(phba);
 
 	beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT,
 		    "BM_%d : num_cpus = %d\n",
 		    phba->num_cpus);
 
-	if (enable_msix) {
-		beiscsi_msix_enable(phba);
-		if (!phba->msix_enabled)
-			phba->num_cpus = 1;
-	}
-
 	phba->shost->max_id = phba->params.cxns_per_ctrl;
 	phba->shost->can_queue = phba->params.ios_per_ctrl;
 	ret = beiscsi_get_memory(phba);
@@ -5745,7 +5663,7 @@ static int beiscsi_dev_probe(struct pci_dev *pcidev,
 		irq_poll_init(&pbe_eq->iopoll, be_iopoll_budget, be_iopoll);
 	}
 
-	i = (phba->msix_enabled) ? i : 0;
+	i = (phba->pcidev->msix_enabled) ? i : 0;
 	/* Work item for MCC handling */
 	pbe_eq = &phwi_context->be_eq[i];
 	INIT_WORK(&pbe_eq->mcc_work, beiscsi_mcc_work);
@@ -5816,8 +5734,7 @@ free_port:
 			   phba->ctrl.mbox_mem_alloced.dma);
 	beiscsi_unmap_pci_function(phba);
 hba_free:
-	if (phba->msix_enabled)
-		pci_disable_msix(phba->pcidev);
+	pci_disable_msix(phba->pcidev);
 	pci_dev_put(phba->pcidev);
 	iscsi_host_free(phba->shost);
 	pci_set_drvdata(pcidev, NULL);
diff --git a/drivers/scsi/be2iscsi/be_main.h b/drivers/scsi/be2iscsi/be_main.h
index 218857926566..338dbe0800c1 100644
--- a/drivers/scsi/be2iscsi/be_main.h
+++ b/drivers/scsi/be2iscsi/be_main.h
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #ifndef _BEISCSI_MAIN_
@@ -36,7 +31,7 @@
 #include <scsi/scsi_transport_iscsi.h>
 
 #define DRV_NAME		"be2iscsi"
-#define BUILD_STR		"11.2.1.0"
+#define BUILD_STR		"11.4.0.0"
 #define BE_NAME			"Emulex OneConnect" \
 				"Open-iSCSI Driver version" BUILD_STR
 #define DRV_DESC		BE_NAME " " "Driver"
@@ -235,7 +230,6 @@ struct sgl_handle {
 struct hba_parameters {
 	unsigned int ios_per_ctrl;
 	unsigned int cxns_per_ctrl;
-	unsigned int asyncpdus_per_ctrl;
 	unsigned int icds_per_ctrl;
 	unsigned int num_sge_per_io;
 	unsigned int defpdu_hdr_sz;
@@ -323,9 +317,7 @@ struct beiscsi_hba {
 	struct pci_dev *pcidev;
 	unsigned int num_cpus;
 	unsigned int nxt_cqid;
-	struct msix_entry msix_entries[MAX_CPUS];
 	char *msi_name[MAX_CPUS];
-	bool msix_enabled;
 	struct be_mem_descriptor *init_mem;
 
 	unsigned short io_sgl_alloc_index;
@@ -597,8 +589,12 @@ struct hd_async_handle {
 	u16 cri;
 	u8 is_header;
 	u8 is_final;
+	u8 in_use;
 };
 
+#define BEISCSI_ASYNC_HDQ_SIZE(phba, ulp) \
+	(BEISCSI_GET_CID_COUNT((phba), (ulp)) * 2)
+
 /**
  * This has list of async PDUs that are waiting to be processed.
  * Buffers live in this list for a brief duration before they get
@@ -624,14 +620,8 @@ struct hd_async_buf_context {
 	void *va_base;
 	void *ring_base;
 	struct hd_async_handle *handle_base;
-	u16 free_entries;
 	u32 buffer_size;
-	/**
-	 * Once iSCSI layer finishes processing an async PDU, the
-	 * handles used for the PDU are added to this list.
-	 * They are posted back to FW in groups of 8.
-	 */
-	struct list_head free_list;
+	u16 pi;
 };
 
 /**
diff --git a/drivers/scsi/be2iscsi/be_mgmt.c b/drivers/scsi/be2iscsi/be_mgmt.c
index 2f6d5c2ac329..c73775368d09 100644
--- a/drivers/scsi/be2iscsi/be_mgmt.c
+++ b/drivers/scsi/be2iscsi/be_mgmt.c
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #include <linux/bsg-lib.h>
@@ -126,67 +121,6 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl,
 	return tag;
 }
 
-unsigned int mgmt_invalidate_connection(struct beiscsi_hba *phba,
-					 struct beiscsi_endpoint *beiscsi_ep,
-					 unsigned short cid,
-					 unsigned short issue_reset,
-					 unsigned short savecfg_flag)
-{
-	struct be_ctrl_info *ctrl = &phba->ctrl;
-	struct be_mcc_wrb *wrb;
-	struct iscsi_invalidate_connection_params_in *req;
-	unsigned int tag = 0;
-
-	mutex_lock(&ctrl->mbox_lock);
-	wrb = alloc_mcc_wrb(phba, &tag);
-	if (!wrb) {
-		mutex_unlock(&ctrl->mbox_lock);
-		return 0;
-	}
-
-	req = embedded_payload(wrb);
-	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
-	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI_INI,
-			   OPCODE_ISCSI_INI_DRIVER_INVALIDATE_CONNECTION,
-			   sizeof(*req));
-	req->session_handle = beiscsi_ep->fw_handle;
-	req->cid = cid;
-	if (issue_reset)
-		req->cleanup_type = CMD_ISCSI_CONNECTION_ISSUE_TCP_RST;
-	else
-		req->cleanup_type = CMD_ISCSI_CONNECTION_INVALIDATE;
-	req->save_cfg = savecfg_flag;
-	be_mcc_notify(phba, tag);
-	mutex_unlock(&ctrl->mbox_lock);
-	return tag;
-}
-
-unsigned int mgmt_upload_connection(struct beiscsi_hba *phba,
-				unsigned short cid, unsigned int upload_flag)
-{
-	struct be_ctrl_info *ctrl = &phba->ctrl;
-	struct be_mcc_wrb *wrb;
-	struct tcp_upload_params_in *req;
-	unsigned int tag;
-
-	mutex_lock(&ctrl->mbox_lock);
-	wrb = alloc_mcc_wrb(phba, &tag);
-	if (!wrb) {
-		mutex_unlock(&ctrl->mbox_lock);
-		return 0;
-	}
-
-	req = embedded_payload(wrb);
-	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
-	be_cmd_hdr_prepare(&req->hdr, CMD_COMMON_TCP_UPLOAD,
-			   OPCODE_COMMON_TCP_UPLOAD, sizeof(*req));
-	req->id = (unsigned short)cid;
-	req->upload_type = (unsigned char)upload_flag;
-	be_mcc_notify(phba, tag);
-	mutex_unlock(&ctrl->mbox_lock);
-	return tag;
-}
-
 /**
  * mgmt_open_connection()- Establish a TCP CXN
  * @dst_addr: Destination Address
@@ -1449,6 +1383,72 @@ void beiscsi_offload_cxn_v2(struct beiscsi_offload_params *params,
 		      exp_statsn) / 32] + 1));
 }
 
+unsigned int beiscsi_invalidate_cxn(struct beiscsi_hba *phba,
+				    struct beiscsi_endpoint *beiscsi_ep)
+{
+	struct be_invalidate_connection_params_in *req;
+	struct be_ctrl_info *ctrl = &phba->ctrl;
+	struct be_mcc_wrb *wrb;
+	unsigned int tag = 0;
+
+	mutex_lock(&ctrl->mbox_lock);
+	wrb = alloc_mcc_wrb(phba, &tag);
+	if (!wrb) {
+		mutex_unlock(&ctrl->mbox_lock);
+		return 0;
+	}
+
+	req = embedded_payload(wrb);
+	be_wrb_hdr_prepare(wrb, sizeof(union be_invalidate_connection_params),
+			   true, 0);
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI_INI,
+			   OPCODE_ISCSI_INI_DRIVER_INVALIDATE_CONNECTION,
+			   sizeof(*req));
+	req->session_handle = beiscsi_ep->fw_handle;
+	req->cid = beiscsi_ep->ep_cid;
+	if (beiscsi_ep->conn)
+		req->cleanup_type = BE_CLEANUP_TYPE_INVALIDATE;
+	else
+		req->cleanup_type = BE_CLEANUP_TYPE_ISSUE_TCP_RST;
+	/**
+	 * 0 - non-persistent targets
+	 * 1 - save session info on flash
+	 */
+	req->save_cfg = 0;
+	be_mcc_notify(phba, tag);
+	mutex_unlock(&ctrl->mbox_lock);
+	return tag;
+}
+
+unsigned int beiscsi_upload_cxn(struct beiscsi_hba *phba,
+				struct beiscsi_endpoint *beiscsi_ep)
+{
+	struct be_ctrl_info *ctrl = &phba->ctrl;
+	struct be_mcc_wrb *wrb;
+	struct be_tcp_upload_params_in *req;
+	unsigned int tag;
+
+	mutex_lock(&ctrl->mbox_lock);
+	wrb = alloc_mcc_wrb(phba, &tag);
+	if (!wrb) {
+		mutex_unlock(&ctrl->mbox_lock);
+		return 0;
+	}
+
+	req = embedded_payload(wrb);
+	be_wrb_hdr_prepare(wrb, sizeof(union be_tcp_upload_params), true, 0);
+	be_cmd_hdr_prepare(&req->hdr, CMD_COMMON_TCP_UPLOAD,
+			   OPCODE_COMMON_TCP_UPLOAD, sizeof(*req));
+	req->id = beiscsi_ep->ep_cid;
+	if (beiscsi_ep->conn)
+		req->upload_type = BE_UPLOAD_TYPE_GRACEFUL;
+	else
+		req->upload_type = BE_UPLOAD_TYPE_ABORT;
+	be_mcc_notify(phba, tag);
+	mutex_unlock(&ctrl->mbox_lock);
+	return tag;
+}
+
 int beiscsi_mgmt_invalidate_icds(struct beiscsi_hba *phba,
 				 struct invldt_cmd_tbl *inv_tbl,
 				 unsigned int nents)
diff --git a/drivers/scsi/be2iscsi/be_mgmt.h b/drivers/scsi/be2iscsi/be_mgmt.h
index 308f1472f98a..06ddc5ad6874 100644
--- a/drivers/scsi/be2iscsi/be_mgmt.h
+++ b/drivers/scsi/be2iscsi/be_mgmt.h
@@ -1,20 +1,15 @@
-/**
- * Copyright (C) 2005 - 2016 Broadcom
- * All rights reserved.
+/*
+ * Copyright 2017 Broadcom. All Rights Reserved.
+ * The term "Broadcom" refers to Broadcom Limited and/or its subsidiaries.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation.  The full GNU General
+ * as published by the Free Software Foundation. The full GNU General
  * Public License is included in this distribution in the file called COPYING.
  *
- * Written by: Jayamohan Kallickal (jayamohan.kallickal@broadcom.com)
- *
  * Contact Information:
  * linux-drivers@broadcom.com
  *
- * Emulex
- * 3333 Susan Street
- * Costa Mesa, CA 92626
  */
 
 #ifndef _BEISCSI_MGMT_
@@ -41,35 +36,11 @@ int mgmt_open_connection(struct beiscsi_hba *phba,
 			 struct beiscsi_endpoint *beiscsi_ep,
 			 struct be_dma_mem *nonemb_cmd);
 
-unsigned int mgmt_upload_connection(struct beiscsi_hba *phba,
-				     unsigned short cid,
-				     unsigned int upload_flag);
 unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl,
 					 struct beiscsi_hba *phba,
 					 struct bsg_job *job,
 					 struct be_dma_mem *nonemb_cmd);
 
-#define BEISCSI_NO_RST_ISSUE	0
-struct iscsi_invalidate_connection_params_in {
-	struct be_cmd_req_hdr hdr;
-	unsigned int session_handle;
-	unsigned short cid;
-	unsigned short unused;
-	unsigned short cleanup_type;
-	unsigned short save_cfg;
-} __packed;
-
-struct iscsi_invalidate_connection_params_out {
-	unsigned int session_handle;
-	unsigned short cid;
-	unsigned short unused;
-} __packed;
-
-union iscsi_invalidate_connection_params {
-	struct iscsi_invalidate_connection_params_in request;
-	struct iscsi_invalidate_connection_params_out response;
-} __packed;
-
 #define BE_INVLDT_CMD_TBL_SZ	128
 struct invldt_cmd_tbl {
 	unsigned short icd;
@@ -265,6 +236,12 @@ void beiscsi_offload_cxn_v2(struct beiscsi_offload_params *params,
 			     struct wrb_handle *pwrb_handle,
 			     struct hwi_wrb_context *pwrb_context);
 
+unsigned int beiscsi_invalidate_cxn(struct beiscsi_hba *phba,
+				    struct beiscsi_endpoint *beiscsi_ep);
+
+unsigned int beiscsi_upload_cxn(struct beiscsi_hba *phba,
+				struct beiscsi_endpoint *beiscsi_ep);
+
 int be_cmd_modify_eq_delay(struct beiscsi_hba *phba,
 			 struct be_set_eqd *, int num);
 
diff --git a/drivers/scsi/bfa/bfa_core.c b/drivers/scsi/bfa/bfa_core.c
index 7209afad82f7..3e1caec82554 100644
--- a/drivers/scsi/bfa/bfa_core.c
+++ b/drivers/scsi/bfa/bfa_core.c
@@ -23,22 +23,6 @@
 BFA_TRC_FILE(HAL, CORE);
 
 /*
- * BFA module list terminated by NULL
- */
-static struct bfa_module_s *hal_mods[] = {
-	&hal_mod_fcdiag,
-	&hal_mod_sgpg,
-	&hal_mod_fcport,
-	&hal_mod_fcxp,
-	&hal_mod_lps,
-	&hal_mod_uf,
-	&hal_mod_rport,
-	&hal_mod_fcp,
-	&hal_mod_dconf,
-	NULL
-};
-
-/*
  * Message handlers for various modules.
  */
 static bfa_isr_func_t  bfa_isrs[BFI_MC_MAX] = {
@@ -1191,8 +1175,13 @@ bfa_iocfc_start_submod(struct bfa_s *bfa)
 	for (i = 0; i < BFI_IOC_MAX_CQS; i++)
 		bfa_isr_rspq_ack(bfa, i, bfa_rspq_ci(bfa, i));
 
-	for (i = 0; hal_mods[i]; i++)
-		hal_mods[i]->start(bfa);
+	bfa_fcport_start(bfa);
+	bfa_uf_start(bfa);
+	/*
+	 * bfa_init() with flash read is complete. now invalidate the stale
+	 * content of lun mask like unit attention, rp tag and lp tag.
+	 */
+	bfa_ioim_lm_init(BFA_FCP_MOD(bfa)->bfa);
 
 	bfa->iocfc.submod_enabled = BFA_TRUE;
 }
@@ -1203,13 +1192,16 @@ bfa_iocfc_start_submod(struct bfa_s *bfa)
 static void
 bfa_iocfc_disable_submod(struct bfa_s *bfa)
 {
-	int		i;
-
 	if (bfa->iocfc.submod_enabled == BFA_FALSE)
 		return;
 
-	for (i = 0; hal_mods[i]; i++)
-		hal_mods[i]->iocdisable(bfa);
+	bfa_fcdiag_iocdisable(bfa);
+	bfa_fcport_iocdisable(bfa);
+	bfa_fcxp_iocdisable(bfa);
+	bfa_lps_iocdisable(bfa);
+	bfa_rport_iocdisable(bfa);
+	bfa_fcp_iocdisable(bfa);
+	bfa_dconf_iocdisable(bfa);
 
 	bfa->iocfc.submod_enabled = BFA_FALSE;
 }
@@ -1773,7 +1765,6 @@ void
 bfa_cfg_get_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo,
 		struct bfa_s *bfa)
 {
-	int		i;
 	struct bfa_mem_dma_s *port_dma = BFA_MEM_PORT_DMA(bfa);
 	struct bfa_mem_dma_s *ablk_dma = BFA_MEM_ABLK_DMA(bfa);
 	struct bfa_mem_dma_s *cee_dma = BFA_MEM_CEE_DMA(bfa);
@@ -1792,9 +1783,14 @@ bfa_cfg_get_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo,
 	INIT_LIST_HEAD(&meminfo->kva_info.qe);
 
 	bfa_iocfc_meminfo(cfg, meminfo, bfa);
-
-	for (i = 0; hal_mods[i]; i++)
-		hal_mods[i]->meminfo(cfg, meminfo, bfa);
+	bfa_sgpg_meminfo(cfg, meminfo, bfa);
+	bfa_fcport_meminfo(cfg, meminfo, bfa);
+	bfa_fcxp_meminfo(cfg, meminfo, bfa);
+	bfa_lps_meminfo(cfg, meminfo, bfa);
+	bfa_uf_meminfo(cfg, meminfo, bfa);
+	bfa_rport_meminfo(cfg, meminfo, bfa);
+	bfa_fcp_meminfo(cfg, meminfo, bfa);
+	bfa_dconf_meminfo(cfg, meminfo, bfa);
 
 	/* dma info setup */
 	bfa_mem_dma_setup(meminfo, port_dma, bfa_port_meminfo());
@@ -1840,7 +1836,6 @@ void
 bfa_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	       struct bfa_meminfo_s *meminfo, struct bfa_pcidev_s *pcidev)
 {
-	int	i;
 	struct bfa_mem_dma_s *dma_info, *dma_elem;
 	struct bfa_mem_kva_s *kva_info, *kva_elem;
 	struct list_head *dm_qe, *km_qe;
@@ -1869,10 +1864,15 @@ bfa_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	}
 
 	bfa_iocfc_attach(bfa, bfad, cfg, pcidev);
-
-	for (i = 0; hal_mods[i]; i++)
-		hal_mods[i]->attach(bfa, bfad, cfg, pcidev);
-
+	bfa_fcdiag_attach(bfa, bfad, cfg, pcidev);
+	bfa_sgpg_attach(bfa, bfad, cfg, pcidev);
+	bfa_fcport_attach(bfa, bfad, cfg, pcidev);
+	bfa_fcxp_attach(bfa, bfad, cfg, pcidev);
+	bfa_lps_attach(bfa, bfad, cfg, pcidev);
+	bfa_uf_attach(bfa, bfad, cfg, pcidev);
+	bfa_rport_attach(bfa, bfad, cfg, pcidev);
+	bfa_fcp_attach(bfa, bfad, cfg, pcidev);
+	bfa_dconf_attach(bfa, bfad, cfg);
 	bfa_com_port_attach(bfa);
 	bfa_com_ablk_attach(bfa);
 	bfa_com_cee_attach(bfa);
@@ -1899,10 +1899,6 @@ bfa_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 void
 bfa_detach(struct bfa_s *bfa)
 {
-	int	i;
-
-	for (i = 0; hal_mods[i]; i++)
-		hal_mods[i]->detach(bfa);
 	bfa_ioc_detach(&bfa->ioc);
 }
 
diff --git a/drivers/scsi/bfa/bfa_fcpim.c b/drivers/scsi/bfa/bfa_fcpim.c
index 20982e7cdd81..5f53b3276234 100644
--- a/drivers/scsi/bfa/bfa_fcpim.c
+++ b/drivers/scsi/bfa/bfa_fcpim.c
@@ -25,7 +25,6 @@ BFA_TRC_FILE(HAL, FCPIM);
  *  BFA ITNIM Related definitions
  */
 static void bfa_itnim_update_del_itn_stats(struct bfa_itnim_s *itnim);
-static void bfa_ioim_lm_init(struct bfa_s *bfa);
 
 #define BFA_ITNIM_FROM_TAG(_fcpim, _tag)                                \
 	(((_fcpim)->itnim_arr + ((_tag) & ((_fcpim)->num_itnims - 1))))
@@ -339,7 +338,7 @@ bfa_fcpim_attach(struct bfa_fcp_mod_s *fcp, void *bfad,
 	bfa_ioim_attach(fcpim);
 }
 
-static void
+void
 bfa_fcpim_iocdisable(struct bfa_fcp_mod_s *fcp)
 {
 	struct bfa_fcpim_s *fcpim = &fcp->fcpim;
@@ -2105,7 +2104,7 @@ bfa_ioim_sm_resfree(struct bfa_ioim_s *ioim, enum bfa_ioim_event event)
  * is complete by driver. now invalidate the stale content of lun mask
  * like unit attention, rp tag and lp tag.
  */
-static void
+void
 bfa_ioim_lm_init(struct bfa_s *bfa)
 {
 	struct bfa_lun_mask_s *lunm_list;
@@ -3634,11 +3633,7 @@ bfa_tskim_res_recfg(struct bfa_s *bfa, u16 num_tskim_fw)
 	}
 }
 
-/* BFA FCP module - parent module for fcpim */
-
-BFA_MODULE(fcp);
-
-static void
+void
 bfa_fcp_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -3696,7 +3691,7 @@ bfa_fcp_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 	bfa_mem_kva_setup(minfo, fcp_kva, km_len);
 }
 
-static void
+void
 bfa_fcp_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -3739,29 +3734,7 @@ bfa_fcp_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 			(fcp->num_itns * sizeof(struct bfa_itn_s)));
 }
 
-static void
-bfa_fcp_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_fcp_start(struct bfa_s *bfa)
-{
-	struct bfa_fcp_mod_s *fcp = BFA_FCP_MOD(bfa);
-
-	/*
-	 * bfa_init() with flash read is complete. now invalidate the stale
-	 * content of lun mask like unit attention, rp tag and lp tag.
-	 */
-	bfa_ioim_lm_init(fcp->bfa);
-}
-
-static void
-bfa_fcp_stop(struct bfa_s *bfa)
-{
-}
-
-static void
+void
 bfa_fcp_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_fcp_mod_s *fcp = BFA_FCP_MOD(bfa);
diff --git a/drivers/scsi/bfa/bfa_fcs_lport.c b/drivers/scsi/bfa/bfa_fcs_lport.c
index 4ddda72f60e6..638c0a2857f7 100644
--- a/drivers/scsi/bfa/bfa_fcs_lport.c
+++ b/drivers/scsi/bfa/bfa_fcs_lport.c
@@ -89,16 +89,27 @@ static struct {
 	void		(*online) (struct bfa_fcs_lport_s *port);
 	void		(*offline) (struct bfa_fcs_lport_s *port);
 } __port_action[] = {
-	{
-	bfa_fcs_lport_unknown_init, bfa_fcs_lport_unknown_online,
-			bfa_fcs_lport_unknown_offline}, {
-	bfa_fcs_lport_fab_init, bfa_fcs_lport_fab_online,
-			bfa_fcs_lport_fab_offline}, {
-	bfa_fcs_lport_n2n_init, bfa_fcs_lport_n2n_online,
-			bfa_fcs_lport_n2n_offline}, {
-	bfa_fcs_lport_loop_init, bfa_fcs_lport_loop_online,
-			bfa_fcs_lport_loop_offline},
-	};
+	[BFA_FCS_FABRIC_UNKNOWN] = {
+		.init = bfa_fcs_lport_unknown_init,
+		.online = bfa_fcs_lport_unknown_online,
+		.offline = bfa_fcs_lport_unknown_offline
+	},
+	[BFA_FCS_FABRIC_SWITCHED] = {
+		.init = bfa_fcs_lport_fab_init,
+		.online = bfa_fcs_lport_fab_online,
+		.offline = bfa_fcs_lport_fab_offline
+	},
+	[BFA_FCS_FABRIC_N2N] = {
+		.init = bfa_fcs_lport_n2n_init,
+		.online = bfa_fcs_lport_n2n_online,
+		.offline = bfa_fcs_lport_n2n_offline
+	},
+	[BFA_FCS_FABRIC_LOOP] = {
+		.init = bfa_fcs_lport_loop_init,
+		.online = bfa_fcs_lport_loop_online,
+		.offline = bfa_fcs_lport_loop_offline
+	},
+};
 
 /*
  *  fcs_port_sm FCS logical port state machine
diff --git a/drivers/scsi/bfa/bfa_ioc.c b/drivers/scsi/bfa/bfa_ioc.c
index a1ada4a31c97..256f4afaccf9 100644
--- a/drivers/scsi/bfa/bfa_ioc.c
+++ b/drivers/scsi/bfa/bfa_ioc.c
@@ -5822,12 +5822,6 @@ bfa_phy_intr(void *phyarg, struct bfi_mbmsg_s *msg)
 }
 
 /*
- *	DCONF module specific
- */
-
-BFA_MODULE(dconf);
-
-/*
  * DCONF state machine events
  */
 enum bfa_dconf_event {
@@ -6073,7 +6067,7 @@ bfa_dconf_sm_iocdown_dirty(struct bfa_dconf_mod_s *dconf,
 /*
  * Compute and return memory needed by DRV_CFG module.
  */
-static void
+void
 bfa_dconf_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo,
 		  struct bfa_s *bfa)
 {
@@ -6087,9 +6081,8 @@ bfa_dconf_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo,
 				sizeof(struct bfa_dconf_s));
 }
 
-static void
-bfa_dconf_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
-		struct bfa_pcidev_s *pcidev)
+void
+bfa_dconf_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg)
 {
 	struct bfa_dconf_mod_s *dconf = BFA_DCONF_MOD(bfa);
 
@@ -6134,33 +6127,20 @@ bfa_dconf_modinit(struct bfa_s *bfa)
 	struct bfa_dconf_mod_s *dconf = BFA_DCONF_MOD(bfa);
 	bfa_sm_send_event(dconf, BFA_DCONF_SM_INIT);
 }
-static void
-bfa_dconf_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_dconf_stop(struct bfa_s *bfa)
-{
-}
 
 static void bfa_dconf_timer(void *cbarg)
 {
 	struct bfa_dconf_mod_s *dconf = cbarg;
 	bfa_sm_send_event(dconf, BFA_DCONF_SM_TIMEOUT);
 }
-static void
+
+void
 bfa_dconf_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_dconf_mod_s *dconf = BFA_DCONF_MOD(bfa);
 	bfa_sm_send_event(dconf, BFA_DCONF_SM_IOCDISABLE);
 }
 
-static void
-bfa_dconf_detach(struct bfa_s *bfa)
-{
-}
-
 static bfa_status_t
 bfa_dconf_flash_write(struct bfa_dconf_mod_s *dconf)
 {
diff --git a/drivers/scsi/bfa/bfa_modules.h b/drivers/scsi/bfa/bfa_modules.h
index 53135f21fa0e..1c2ab395e616 100644
--- a/drivers/scsi/bfa/bfa_modules.h
+++ b/drivers/scsi/bfa/bfa_modules.h
@@ -61,54 +61,8 @@ enum {
 	BFA_TRC_HAL_IOCFC_CB	= 5,
 };
 
-/*
- * Macro to define a new BFA module
- */
-#define BFA_MODULE(__mod)						\
-	static void bfa_ ## __mod ## _meminfo(				\
-			struct bfa_iocfc_cfg_s *cfg,			\
-			struct bfa_meminfo_s *meminfo,			\
-			struct bfa_s *bfa);				\
-	static void bfa_ ## __mod ## _attach(struct bfa_s *bfa,		\
-			void *bfad, struct bfa_iocfc_cfg_s *cfg,	\
-			struct bfa_pcidev_s *pcidev);      \
-	static void bfa_ ## __mod ## _detach(struct bfa_s *bfa);      \
-	static void bfa_ ## __mod ## _start(struct bfa_s *bfa);      \
-	static void bfa_ ## __mod ## _stop(struct bfa_s *bfa);      \
-	static void bfa_ ## __mod ## _iocdisable(struct bfa_s *bfa);      \
-									\
-	extern struct bfa_module_s hal_mod_ ## __mod;			\
-	struct bfa_module_s hal_mod_ ## __mod = {			\
-		bfa_ ## __mod ## _meminfo,				\
-		bfa_ ## __mod ## _attach,				\
-		bfa_ ## __mod ## _detach,				\
-		bfa_ ## __mod ## _start,				\
-		bfa_ ## __mod ## _stop,					\
-		bfa_ ## __mod ## _iocdisable,				\
-	}
-
 #define BFA_CACHELINE_SZ	(256)
 
-/*
- * Structure used to interact between different BFA sub modules
- *
- * Each sub module needs to implement only the entry points relevant to it (and
- * can leave entry points as NULL)
- */
-struct bfa_module_s {
-	void (*meminfo) (struct bfa_iocfc_cfg_s *cfg,
-			 struct bfa_meminfo_s *meminfo,
-			 struct bfa_s *bfa);
-	void (*attach) (struct bfa_s *bfa, void *bfad,
-			struct bfa_iocfc_cfg_s *cfg,
-			struct bfa_pcidev_s *pcidev);
-	void (*detach) (struct bfa_s *bfa);
-	void (*start) (struct bfa_s *bfa);
-	void (*stop) (struct bfa_s *bfa);
-	void (*iocdisable) (struct bfa_s *bfa);
-};
-
-
 struct bfa_s {
 	void			*bfad;		/*  BFA driver instance    */
 	struct bfa_plog_s	*plog;		/*  portlog buffer	    */
@@ -127,14 +81,51 @@ struct bfa_s {
 };
 
 extern bfa_boolean_t bfa_auto_recover;
-extern struct bfa_module_s hal_mod_fcdiag;
-extern struct bfa_module_s hal_mod_sgpg;
-extern struct bfa_module_s hal_mod_fcport;
-extern struct bfa_module_s hal_mod_fcxp;
-extern struct bfa_module_s hal_mod_lps;
-extern struct bfa_module_s hal_mod_uf;
-extern struct bfa_module_s hal_mod_rport;
-extern struct bfa_module_s hal_mod_fcp;
-extern struct bfa_module_s hal_mod_dconf;
+
+void bfa_dconf_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *);
+void bfa_dconf_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		  struct bfa_s *);
+void bfa_dconf_iocdisable(struct bfa_s *);
+void bfa_fcp_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_fcp_iocdisable(struct bfa_s *bfa);
+void bfa_fcp_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_fcpim_iocdisable(struct bfa_fcp_mod_s *);
+void bfa_fcport_start(struct bfa_s *);
+void bfa_fcport_iocdisable(struct bfa_s *);
+void bfa_fcport_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		   struct bfa_s *);
+void bfa_fcport_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_fcxp_iocdisable(struct bfa_s *);
+void bfa_fcxp_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_fcxp_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_fcdiag_iocdisable(struct bfa_s *);
+void bfa_fcdiag_attach(struct bfa_s *bfa, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_ioim_lm_init(struct bfa_s *);
+void bfa_lps_iocdisable(struct bfa_s *bfa);
+void bfa_lps_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_lps_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+	struct bfa_pcidev_s *);
+void bfa_rport_iocdisable(struct bfa_s *bfa);
+void bfa_rport_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_rport_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_sgpg_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_sgpg_attach(struct bfa_s *, void *bfad, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_uf_iocdisable(struct bfa_s *);
+void bfa_uf_meminfo(struct bfa_iocfc_cfg_s *, struct bfa_meminfo_s *,
+		struct bfa_s *);
+void bfa_uf_attach(struct bfa_s *, void *, struct bfa_iocfc_cfg_s *,
+		struct bfa_pcidev_s *);
+void bfa_uf_start(struct bfa_s *);
 
 #endif /* __BFA_MODULES_H__ */
diff --git a/drivers/scsi/bfa/bfa_svc.c b/drivers/scsi/bfa/bfa_svc.c
index 12de292175ef..e640223bab3c 100644
--- a/drivers/scsi/bfa/bfa_svc.c
+++ b/drivers/scsi/bfa/bfa_svc.c
@@ -23,13 +23,6 @@
 #include "bfa_modules.h"
 
 BFA_TRC_FILE(HAL, FCXP);
-BFA_MODULE(fcdiag);
-BFA_MODULE(fcxp);
-BFA_MODULE(sgpg);
-BFA_MODULE(lps);
-BFA_MODULE(fcport);
-BFA_MODULE(rport);
-BFA_MODULE(uf);
 
 /*
  * LPS related definitions
@@ -121,15 +114,6 @@ static void	bfa_fcxp_queue(struct bfa_fcxp_s *fcxp,
 /*
  * forward declarations for LPS functions
  */
-static void bfa_lps_meminfo(struct bfa_iocfc_cfg_s *cfg,
-		struct bfa_meminfo_s *minfo, struct bfa_s *bfa);
-static void bfa_lps_attach(struct bfa_s *bfa, void *bfad,
-				struct bfa_iocfc_cfg_s *cfg,
-				struct bfa_pcidev_s *pcidev);
-static void bfa_lps_detach(struct bfa_s *bfa);
-static void bfa_lps_start(struct bfa_s *bfa);
-static void bfa_lps_stop(struct bfa_s *bfa);
-static void bfa_lps_iocdisable(struct bfa_s *bfa);
 static void bfa_lps_login_rsp(struct bfa_s *bfa,
 				struct bfi_lps_login_rsp_s *rsp);
 static void bfa_lps_no_res(struct bfa_lps_s *first_lps, u8 count);
@@ -484,7 +468,7 @@ claim_fcxps_mem(struct bfa_fcxp_mod_s *mod)
 	bfa_mem_kva_curp(mod) = (void *)fcxp;
 }
 
-static void
+void
 bfa_fcxp_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -522,7 +506,7 @@ bfa_fcxp_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		cfg->fwcfg.num_fcxp_reqs * sizeof(struct bfa_fcxp_s));
 }
 
-static void
+void
 bfa_fcxp_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -544,22 +528,7 @@ bfa_fcxp_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	claim_fcxps_mem(mod);
 }
 
-static void
-bfa_fcxp_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_fcxp_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_fcxp_stop(struct bfa_s *bfa)
-{
-}
-
-static void
+void
 bfa_fcxp_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_fcxp_mod_s *mod = BFA_FCXP_MOD(bfa);
@@ -1510,7 +1479,7 @@ bfa_lps_sm_logowait(struct bfa_lps_s *lps, enum bfa_lps_event event)
 /*
  * return memory requirement
  */
-static void
+void
 bfa_lps_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -1527,7 +1496,7 @@ bfa_lps_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 /*
  * bfa module attach at initialization time
  */
-static void
+void
 bfa_lps_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	struct bfa_pcidev_s *pcidev)
 {
@@ -1557,25 +1526,10 @@ bfa_lps_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	}
 }
 
-static void
-bfa_lps_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_lps_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_lps_stop(struct bfa_s *bfa)
-{
-}
-
 /*
  * IOC in disabled state -- consider all lps offline
  */
-static void
+void
 bfa_lps_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_lps_mod_s	*mod = BFA_LPS_MOD(bfa);
@@ -3055,7 +3009,7 @@ bfa_fcport_queue_cb(struct bfa_fcport_ln_s *ln, enum bfa_port_linkstate event)
 #define FCPORT_STATS_DMA_SZ (BFA_ROUNDUP(sizeof(union bfa_fcport_stats_u), \
 							BFA_CACHELINE_SZ))
 
-static void
+void
 bfa_fcport_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		   struct bfa_s *bfa)
 {
@@ -3086,7 +3040,7 @@ bfa_fcport_mem_claim(struct bfa_fcport_s *fcport)
 /*
  * Memory initialization.
  */
-static void
+void
 bfa_fcport_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -3131,34 +3085,16 @@ bfa_fcport_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	bfa_reqq_winit(&fcport->reqq_wait, bfa_fcport_qresume, fcport);
 }
 
-static void
-bfa_fcport_detach(struct bfa_s *bfa)
-{
-}
-
-/*
- * Called when IOC is ready.
- */
-static void
+void
 bfa_fcport_start(struct bfa_s *bfa)
 {
 	bfa_sm_send_event(BFA_FCPORT_MOD(bfa), BFA_FCPORT_SM_START);
 }
 
 /*
- * Called before IOC is stopped.
- */
-static void
-bfa_fcport_stop(struct bfa_s *bfa)
-{
-	bfa_sm_send_event(BFA_FCPORT_MOD(bfa), BFA_FCPORT_SM_STOP);
-	bfa_trunk_iocdisable(bfa);
-}
-
-/*
  * Called when IOC failure is detected.
  */
-static void
+void
 bfa_fcport_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_fcport_s *fcport = BFA_FCPORT_MOD(bfa);
@@ -4886,7 +4822,7 @@ bfa_rport_qresume(void *cbarg)
 	bfa_sm_send_event(rp, BFA_RPORT_SM_QRESUME);
 }
 
-static void
+void
 bfa_rport_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -4900,7 +4836,7 @@ bfa_rport_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		cfg->fwcfg.num_rports * sizeof(struct bfa_rport_s));
 }
 
-static void
+void
 bfa_rport_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -4940,22 +4876,7 @@ bfa_rport_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	bfa_mem_kva_curp(mod) = (u8 *) rp;
 }
 
-static void
-bfa_rport_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_rport_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_rport_stop(struct bfa_s *bfa)
-{
-}
-
-static void
+void
 bfa_rport_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_rport_mod_s *mod = BFA_RPORT_MOD(bfa);
@@ -5246,7 +5167,7 @@ bfa_rport_unset_lunmask(struct bfa_s *bfa, struct bfa_rport_s *rp)
 /*
  * Compute and return memory needed by FCP(im) module.
  */
-static void
+void
 bfa_sgpg_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -5281,7 +5202,7 @@ bfa_sgpg_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		cfg->drvcfg.num_sgpgs * sizeof(struct bfa_sgpg_s));
 }
 
-static void
+void
 bfa_sgpg_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -5344,26 +5265,6 @@ bfa_sgpg_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	bfa_mem_kva_curp(mod) = (u8 *) hsgpg;
 }
 
-static void
-bfa_sgpg_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_sgpg_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_sgpg_stop(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_sgpg_iocdisable(struct bfa_s *bfa)
-{
-}
-
 bfa_status_t
 bfa_sgpg_malloc(struct bfa_s *bfa, struct list_head *sgpg_q, int nsgpgs)
 {
@@ -5547,7 +5448,7 @@ uf_mem_claim(struct bfa_uf_mod_s *ufm)
 	claim_uf_post_msgs(ufm);
 }
 
-static void
+void
 bfa_uf_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		struct bfa_s *bfa)
 {
@@ -5575,7 +5476,7 @@ bfa_uf_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *minfo,
 		(sizeof(struct bfa_uf_s) + sizeof(struct bfi_uf_buf_post_s)));
 }
 
-static void
+void
 bfa_uf_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -5590,11 +5491,6 @@ bfa_uf_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	uf_mem_claim(ufm);
 }
 
-static void
-bfa_uf_detach(struct bfa_s *bfa)
-{
-}
-
 static struct bfa_uf_s *
 bfa_uf_get(struct bfa_uf_mod_s *uf_mod)
 {
@@ -5682,12 +5578,7 @@ uf_recv(struct bfa_s *bfa, struct bfi_uf_frm_rcvd_s *m)
 		bfa_cb_queue(bfa, &uf->hcb_qe, __bfa_cb_uf_recv, uf);
 }
 
-static void
-bfa_uf_stop(struct bfa_s *bfa)
-{
-}
-
-static void
+void
 bfa_uf_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_uf_mod_s *ufm = BFA_UF_MOD(bfa);
@@ -5704,7 +5595,7 @@ bfa_uf_iocdisable(struct bfa_s *bfa)
 	}
 }
 
-static void
+void
 bfa_uf_start(struct bfa_s *bfa)
 {
 	bfa_uf_post_all(BFA_UF_MOD(bfa));
@@ -5845,13 +5736,7 @@ bfa_fcdiag_set_busy_status(struct bfa_fcdiag_s *fcdiag)
 		fcport->diag_busy = BFA_FALSE;
 }
 
-static void
-bfa_fcdiag_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo,
-		struct bfa_s *bfa)
-{
-}
-
-static void
+void
 bfa_fcdiag_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 		struct bfa_pcidev_s *pcidev)
 {
@@ -5870,7 +5755,7 @@ bfa_fcdiag_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg,
 	memset(&dport->result, 0, sizeof(struct bfa_diag_dport_result_s));
 }
 
-static void
+void
 bfa_fcdiag_iocdisable(struct bfa_s *bfa)
 {
 	struct bfa_fcdiag_s *fcdiag = BFA_FCDIAG_MOD(bfa);
@@ -5888,21 +5773,6 @@ bfa_fcdiag_iocdisable(struct bfa_s *bfa)
 }
 
 static void
-bfa_fcdiag_detach(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_fcdiag_start(struct bfa_s *bfa)
-{
-}
-
-static void
-bfa_fcdiag_stop(struct bfa_s *bfa)
-{
-}
-
-static void
 bfa_fcdiag_queuetest_timeout(void *cbarg)
 {
 	struct bfa_fcdiag_s       *fcdiag = cbarg;
diff --git a/drivers/scsi/csiostor/csio_hw.h b/drivers/scsi/csiostor/csio_hw.h
index 029bef82c057..62758e830d3b 100644
--- a/drivers/scsi/csiostor/csio_hw.h
+++ b/drivers/scsi/csiostor/csio_hw.h
@@ -95,7 +95,6 @@ enum {
 };
 
 struct csio_msix_entries {
-	unsigned short	vector;		/* Assigned MSI-X vector */
 	void		*dev_id;	/* Priv object associated w/ this msix*/
 	char		desc[24];	/* Description of this vector */
 };
diff --git a/drivers/scsi/csiostor/csio_isr.c b/drivers/scsi/csiostor/csio_isr.c
index 2fb71c6c3b37..7c8814715711 100644
--- a/drivers/scsi/csiostor/csio_isr.c
+++ b/drivers/scsi/csiostor/csio_isr.c
@@ -383,17 +383,15 @@ csio_request_irqs(struct csio_hw *hw)
 	int rv, i, j, k = 0;
 	struct csio_msix_entries *entryp = &hw->msix_entries[0];
 	struct csio_scsi_cpu_info *info;
+	struct pci_dev *pdev = hw->pdev;
 
 	if (hw->intr_mode != CSIO_IM_MSIX) {
-		rv = request_irq(hw->pdev->irq, csio_fcoe_isr,
-					(hw->intr_mode == CSIO_IM_MSI) ?
-							0 : IRQF_SHARED,
-					KBUILD_MODNAME, hw);
+		rv = request_irq(pci_irq_vector(pdev, 0), csio_fcoe_isr,
+				hw->intr_mode == CSIO_IM_MSI ? 0 : IRQF_SHARED,
+				KBUILD_MODNAME, hw);
 		if (rv) {
-			if (hw->intr_mode == CSIO_IM_MSI)
-				pci_disable_msi(hw->pdev);
 			csio_err(hw, "Failed to allocate interrupt line.\n");
-			return -EINVAL;
+			goto out_free_irqs;
 		}
 
 		goto out;
@@ -402,22 +400,22 @@ csio_request_irqs(struct csio_hw *hw)
 	/* Add the MSIX vector descriptions */
 	csio_add_msix_desc(hw);
 
-	rv = request_irq(entryp[k].vector, csio_nondata_isr, 0,
+	rv = request_irq(pci_irq_vector(pdev, k), csio_nondata_isr, 0,
 			 entryp[k].desc, hw);
 	if (rv) {
 		csio_err(hw, "IRQ request failed for vec %d err:%d\n",
-			 entryp[k].vector, rv);
-		goto err;
+			 pci_irq_vector(pdev, k), rv);
+		goto out_free_irqs;
 	}
 
-	entryp[k++].dev_id = (void *)hw;
+	entryp[k++].dev_id = hw;
 
-	rv = request_irq(entryp[k].vector, csio_fwevt_isr, 0,
+	rv = request_irq(pci_irq_vector(pdev, k), csio_fwevt_isr, 0,
 			 entryp[k].desc, hw);
 	if (rv) {
 		csio_err(hw, "IRQ request failed for vec %d err:%d\n",
-			 entryp[k].vector, rv);
-		goto err;
+			 pci_irq_vector(pdev, k), rv);
+		goto out_free_irqs;
 	}
 
 	entryp[k++].dev_id = (void *)hw;
@@ -429,51 +427,31 @@ csio_request_irqs(struct csio_hw *hw)
 			struct csio_scsi_qset *sqset = &hw->sqset[i][j];
 			struct csio_q *q = hw->wrm.q_arr[sqset->iq_idx];
 
-			rv = request_irq(entryp[k].vector, csio_scsi_isr, 0,
+			rv = request_irq(pci_irq_vector(pdev, k), csio_scsi_isr, 0,
 					 entryp[k].desc, q);
 			if (rv) {
 				csio_err(hw,
 				       "IRQ request failed for vec %d err:%d\n",
-				       entryp[k].vector, rv);
-				goto err;
+				       pci_irq_vector(pdev, k), rv);
+				goto out_free_irqs;
 			}
 
-			entryp[k].dev_id = (void *)q;
+			entryp[k].dev_id = q;
 
 		} /* for all scsi cpus */
 	} /* for all ports */
 
 out:
 	hw->flags |= CSIO_HWF_HOST_INTR_ENABLED;
-
 	return 0;
 
-err:
-	for (i = 0; i < k; i++) {
-		entryp = &hw->msix_entries[i];
-		free_irq(entryp->vector, entryp->dev_id);
-	}
-	pci_disable_msix(hw->pdev);
-
+out_free_irqs:
+	for (i = 0; i < k; i++)
+		free_irq(pci_irq_vector(pdev, i), hw->msix_entries[i].dev_id);
+	pci_free_irq_vectors(hw->pdev);
 	return -EINVAL;
 }
 
-static void
-csio_disable_msix(struct csio_hw *hw, bool free)
-{
-	int i;
-	struct csio_msix_entries *entryp;
-	int cnt = hw->num_sqsets + CSIO_EXTRA_VECS;
-
-	if (free) {
-		for (i = 0; i < cnt; i++) {
-			entryp = &hw->msix_entries[i];
-			free_irq(entryp->vector, entryp->dev_id);
-		}
-	}
-	pci_disable_msix(hw->pdev);
-}
-
 /* Reduce per-port max possible CPUs */
 static void
 csio_reduce_sqsets(struct csio_hw *hw, int cnt)
@@ -500,10 +478,9 @@ static int
 csio_enable_msix(struct csio_hw *hw)
 {
 	int i, j, k, n, min, cnt;
-	struct csio_msix_entries *entryp;
-	struct msix_entry *entries;
 	int extra = CSIO_EXTRA_VECS;
 	struct csio_scsi_cpu_info *info;
+	struct irq_affinity desc = { .pre_vectors = 2 };
 
 	min = hw->num_pports + extra;
 	cnt = hw->num_sqsets + extra;
@@ -512,50 +489,35 @@ csio_enable_msix(struct csio_hw *hw)
 	if (hw->flags & CSIO_HWF_USING_SOFT_PARAMS || !csio_is_hw_master(hw))
 		cnt = min_t(uint8_t, hw->cfg_niq, cnt);
 
-	entries = kzalloc(sizeof(struct msix_entry) * cnt, GFP_KERNEL);
-	if (!entries)
-		return -ENOMEM;
-
-	for (i = 0; i < cnt; i++)
-		entries[i].entry = (uint16_t)i;
-
 	csio_dbg(hw, "FW supp #niq:%d, trying %d msix's\n", hw->cfg_niq, cnt);
 
-	cnt = pci_enable_msix_range(hw->pdev, entries, min, cnt);
-	if (cnt < 0) {
-		kfree(entries);
+	cnt = pci_alloc_irq_vectors_affinity(hw->pdev, min, cnt,
+			PCI_IRQ_MSIX | PCI_IRQ_AFFINITY, &desc);
+	if (cnt < 0)
 		return cnt;
-	}
 
 	if (cnt < (hw->num_sqsets + extra)) {
 		csio_dbg(hw, "Reducing sqsets to %d\n", cnt - extra);
 		csio_reduce_sqsets(hw, cnt - extra);
 	}
 
-	/* Save off vectors */
-	for (i = 0; i < cnt; i++) {
-		entryp = &hw->msix_entries[i];
-		entryp->vector = entries[i].vector;
-	}
-
 	/* Distribute vectors */
 	k = 0;
-	csio_set_nondata_intr_idx(hw, entries[k].entry);
-	csio_set_mb_intr_idx(csio_hw_to_mbm(hw), entries[k++].entry);
-	csio_set_fwevt_intr_idx(hw, entries[k++].entry);
+	csio_set_nondata_intr_idx(hw, k);
+	csio_set_mb_intr_idx(csio_hw_to_mbm(hw), k++);
+	csio_set_fwevt_intr_idx(hw, k++);
 
 	for (i = 0; i < hw->num_pports; i++) {
 		info = &hw->scsi_cpu_info[i];
 
 		for (j = 0; j < hw->num_scsi_msix_cpus; j++) {
 			n = (j % info->max_cpus) +  k;
-			hw->sqset[i][j].intr_idx = entries[n].entry;
+			hw->sqset[i][j].intr_idx = n;
 		}
 
 		k += info->max_cpus;
 	}
 
-	kfree(entries);
 	return 0;
 }
 
@@ -597,22 +559,26 @@ csio_intr_disable(struct csio_hw *hw, bool free)
 {
 	csio_hw_intr_disable(hw);
 
-	switch (hw->intr_mode) {
-	case CSIO_IM_MSIX:
-		csio_disable_msix(hw, free);
-		break;
-	case CSIO_IM_MSI:
-		if (free)
-			free_irq(hw->pdev->irq, hw);
-		pci_disable_msi(hw->pdev);
-		break;
-	case CSIO_IM_INTX:
-		if (free)
-			free_irq(hw->pdev->irq, hw);
-		break;
-	default:
-		break;
+	if (free) {
+		int i;
+
+		switch (hw->intr_mode) {
+		case CSIO_IM_MSIX:
+			for (i = 0; i < hw->num_sqsets + CSIO_EXTRA_VECS; i++) {
+				free_irq(pci_irq_vector(hw->pdev, i),
+					 hw->msix_entries[i].dev_id);
+			}
+			break;
+		case CSIO_IM_MSI:
+		case CSIO_IM_INTX:
+			free_irq(pci_irq_vector(hw->pdev, 0), hw);
+			break;
+		default:
+			break;
+		}
 	}
+
+	pci_free_irq_vectors(hw->pdev);
 	hw->intr_mode = CSIO_IM_NONE;
 	hw->flags &= ~CSIO_HWF_HOST_INTR_ENABLED;
 }
diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
index 3fb3f5708ff7..1076c1578322 100644
--- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
+++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
@@ -36,7 +36,7 @@ static unsigned int dbg_level;
 #include "../libcxgbi.h"
 
 #define	DRV_MODULE_NAME		"cxgb4i"
-#define DRV_MODULE_DESC		"Chelsio T4/T5 iSCSI Driver"
+#define DRV_MODULE_DESC		"Chelsio T4-T6 iSCSI Driver"
 #define	DRV_MODULE_VERSION	"0.9.5-ko"
 #define DRV_MODULE_RELDATE	"Apr. 2015"
 
diff --git a/drivers/scsi/cxlflash/common.h b/drivers/scsi/cxlflash/common.h
index d11dcc59ff46..256af819377d 100644
--- a/drivers/scsi/cxlflash/common.h
+++ b/drivers/scsi/cxlflash/common.h
@@ -15,6 +15,7 @@
 #ifndef _CXLFLASH_COMMON_H
 #define _CXLFLASH_COMMON_H
 
+#include <linux/irq_poll.h>
 #include <linux/list.h>
 #include <linux/rwsem.h>
 #include <linux/types.h>
@@ -24,30 +25,32 @@
 
 extern const struct file_operations cxlflash_cxl_fops;
 
-#define MAX_CONTEXT  CXLFLASH_MAX_CONTEXT       /* num contexts per afu */
+#define MAX_CONTEXT	CXLFLASH_MAX_CONTEXT	/* num contexts per afu */
+#define MAX_FC_PORTS	CXLFLASH_MAX_FC_PORTS	/* max ports per AFU */
+#define LEGACY_FC_PORTS	2			/* legacy ports per AFU */
 
-#define CXLFLASH_BLOCK_SIZE	4096	/* 4K blocks */
+#define CHAN2PORTBANK(_x)	((_x) >> ilog2(CXLFLASH_NUM_FC_PORTS_PER_BANK))
+#define CHAN2BANKPORT(_x)	((_x) & (CXLFLASH_NUM_FC_PORTS_PER_BANK - 1))
+
+#define CHAN2PORTMASK(_x)	(1 << (_x))	/* channel to port mask */
+#define PORTMASK2CHAN(_x)	(ilog2((_x)))	/* port mask to channel */
+#define PORTNUM2CHAN(_x)	((_x) - 1)	/* port number to channel */
+
+#define CXLFLASH_BLOCK_SIZE	4096		/* 4K blocks */
 #define CXLFLASH_MAX_XFER_SIZE	16777216	/* 16MB transfer */
 #define CXLFLASH_MAX_SECTORS	(CXLFLASH_MAX_XFER_SIZE/512)	/* SCSI wants
-								   max_sectors
-								   in units of
-								   512 byte
-								   sectors
-								*/
+								 * max_sectors
+								 * in units of
+								 * 512 byte
+								 * sectors
+								 */
 
 #define MAX_RHT_PER_CONTEXT (PAGE_SIZE / sizeof(struct sisl_rht_entry))
 
 /* AFU command retry limit */
-#define MC_RETRY_CNT         5	/* sufficient for SCSI check and
-				   certain AFU errors */
+#define MC_RETRY_CNT	5	/* Sufficient for SCSI and certain AFU errors */
 
 /* Command management definitions */
-#define CXLFLASH_NUM_CMDS	(2 * CXLFLASH_MAX_CMDS)	/* Must be a pow2 for
-							   alignment and more
-							   efficient array
-							   index derivation
-							 */
-
 #define CXLFLASH_MAX_CMDS               256
 #define CXLFLASH_MAX_CMDS_PER_LUN       CXLFLASH_MAX_CMDS
 
@@ -57,10 +60,16 @@ extern const struct file_operations cxlflash_cxl_fops;
 /* SQ for master issued cmds */
 #define NUM_SQ_ENTRY			CXLFLASH_MAX_CMDS
 
+/* Hardware queue definitions */
+#define CXLFLASH_DEF_HWQS		1
+#define CXLFLASH_MAX_HWQS		8
+#define PRIMARY_HWQ			0
+
 
 static inline void check_sizes(void)
 {
-	BUILD_BUG_ON_NOT_POWER_OF_2(CXLFLASH_NUM_CMDS);
+	BUILD_BUG_ON_NOT_POWER_OF_2(CXLFLASH_NUM_FC_PORTS_PER_BANK);
+	BUILD_BUG_ON_NOT_POWER_OF_2(CXLFLASH_MAX_CMDS);
 }
 
 /* AFU defines a fixed size of 4K for command buffers (borrow 4K page define) */
@@ -80,11 +89,20 @@ enum cxlflash_init_state {
 };
 
 enum cxlflash_state {
+	STATE_PROBING,	/* Initial state during probe */
+	STATE_PROBED,	/* Temporary state, probe completed but EEH occurred */
 	STATE_NORMAL,	/* Normal running state, everything good */
 	STATE_RESET,	/* Reset state, trying to reset/recover */
 	STATE_FAILTERM	/* Failed/terminating state, error out users/threads */
 };
 
+enum cxlflash_hwq_mode {
+	HWQ_MODE_RR,	/* Roundrobin (default) */
+	HWQ_MODE_TAG,	/* Distribute based on block MQ tag */
+	HWQ_MODE_CPU,	/* CPU affinity */
+	MAX_HWQ_MODE
+};
+
 /*
  * Each context has its own set of resource handles that is visible
  * only from that context.
@@ -92,11 +110,11 @@ enum cxlflash_state {
 
 struct cxlflash_cfg {
 	struct afu *afu;
-	struct cxl_context *mcctx;
 
 	struct pci_dev *dev;
 	struct pci_device_id *dev_id;
 	struct Scsi_Host *host;
+	int num_fc_ports;
 
 	ulong cxlflash_regs_pci;
 
@@ -117,7 +135,7 @@ struct cxlflash_cfg {
 	struct file_operations cxl_fops;
 
 	/* Parameters that are LUN table related */
-	int last_lun_index[CXLFLASH_NUM_FC_PORTS];
+	int last_lun_index[MAX_FC_PORTS];
 	int promote_lun_index;
 	struct list_head lluns; /* list of llun_info structs */
 
@@ -134,6 +152,8 @@ struct afu_cmd {
 	struct afu *parent;
 	struct scsi_cmnd *scp;
 	struct completion cevent;
+	struct list_head queue;
+	u32 hwq_index;
 
 	u8 cmd_tmf:1;
 
@@ -156,7 +176,7 @@ static inline struct afu_cmd *sc_to_afucz(struct scsi_cmnd *sc)
 	return afuc;
 }
 
-struct afu {
+struct hwq {
 	/* Stuff requiring alignment go first. */
 	struct sisl_ioarcb sq[NUM_SQ_ENTRY];		/* 16K SQ */
 	u64 rrq_entry[NUM_RRQ_ENTRY];			/* 2K RRQ */
@@ -164,40 +184,67 @@ struct afu {
 	/* Beware of alignment till here. Preferably introduce new
 	 * fields after this point
 	 */
-
-	int (*send_cmd)(struct afu *, struct afu_cmd *);
-	void (*context_reset)(struct afu_cmd *);
-
-	/* AFU HW */
+	struct afu *afu;
+	struct cxl_context *ctx;
 	struct cxl_ioctl_start_work work;
-	struct cxlflash_afu_map __iomem *afu_map;	/* entire MMIO map */
 	struct sisl_host_map __iomem *host_map;		/* MC host map */
 	struct sisl_ctrl_map __iomem *ctrl_map;		/* MC control map */
-
 	ctx_hndl_t ctx_hndl;	/* master's context handle */
+	u32 index;		/* Index of this hwq */
 
 	atomic_t hsq_credits;
 	spinlock_t hsq_slock;
 	struct sisl_ioarcb *hsq_start;
 	struct sisl_ioarcb *hsq_end;
 	struct sisl_ioarcb *hsq_curr;
+	spinlock_t hrrq_slock;
 	u64 *hrrq_start;
 	u64 *hrrq_end;
 	u64 *hrrq_curr;
 	bool toggle;
-	atomic_t cmds_active;	/* Number of currently active AFU commands */
+
 	s64 room;
 	spinlock_t rrin_slock; /* Lock to rrin queuing and cmd_room updates */
+
+	struct irq_poll irqpoll;
+} __aligned(cache_line_size());
+
+struct afu {
+	struct hwq hwqs[CXLFLASH_MAX_HWQS];
+	int (*send_cmd)(struct afu *, struct afu_cmd *);
+	void (*context_reset)(struct afu_cmd *);
+
+	/* AFU HW */
+	struct cxlflash_afu_map __iomem *afu_map;	/* entire MMIO map */
+
+	atomic_t cmds_active;	/* Number of currently active AFU commands */
 	u64 hb;
 	u32 internal_lun;	/* User-desired LUN mode for this AFU */
 
+	u32 num_hwqs;		/* Number of hardware queues */
+	u32 desired_hwqs;	/* Desired h/w queues, effective on AFU reset */
+	enum cxlflash_hwq_mode hwq_mode; /* Steering mode for h/w queues */
+	u32 hwq_rr_count;	/* Count to distribute traffic for roundrobin */
+
 	char version[16];
 	u64 interface_version;
 
+	u32 irqpoll_weight;
 	struct cxlflash_cfg *parent; /* Pointer back to parent cxlflash_cfg */
-
 };
 
+static inline struct hwq *get_hwq(struct afu *afu, u32 index)
+{
+	WARN_ON(index >= CXLFLASH_MAX_HWQS);
+
+	return &afu->hwqs[index];
+}
+
+static inline bool afu_is_irqpoll_enabled(struct afu *afu)
+{
+	return !!afu->irqpoll_weight;
+}
+
 static inline bool afu_is_cmd_mode(struct afu *afu, u64 cmd_mode)
 {
 	u64 afu_cap = afu->interface_version >> SISL_INTVER_CAP_SHIFT;
@@ -223,14 +270,36 @@ static inline u64 lun_to_lunid(u64 lun)
 	return be64_to_cpu(lun_id);
 }
 
-int cxlflash_afu_sync(struct afu *, ctx_hndl_t, res_hndl_t, u8);
+static inline struct fc_port_bank __iomem *get_fc_port_bank(
+					    struct cxlflash_cfg *cfg, int i)
+{
+	struct afu *afu = cfg->afu;
+
+	return &afu->afu_map->global.bank[CHAN2PORTBANK(i)];
+}
+
+static inline __be64 __iomem *get_fc_port_regs(struct cxlflash_cfg *cfg, int i)
+{
+	struct fc_port_bank __iomem *fcpb = get_fc_port_bank(cfg, i);
+
+	return &fcpb->fc_port_regs[CHAN2BANKPORT(i)][0];
+}
+
+static inline __be64 __iomem *get_fc_port_luns(struct cxlflash_cfg *cfg, int i)
+{
+	struct fc_port_bank __iomem *fcpb = get_fc_port_bank(cfg, i);
+
+	return &fcpb->fc_port_luns[CHAN2BANKPORT(i)][0];
+}
+
+int cxlflash_afu_sync(struct afu *afu, ctx_hndl_t c, res_hndl_t r, u8 mode);
 void cxlflash_list_init(void);
 void cxlflash_term_global_luns(void);
 void cxlflash_free_errpage(void);
-int cxlflash_ioctl(struct scsi_device *, int, void __user *);
-void cxlflash_stop_term_user_contexts(struct cxlflash_cfg *);
-int cxlflash_mark_contexts_error(struct cxlflash_cfg *);
-void cxlflash_term_local_luns(struct cxlflash_cfg *);
-void cxlflash_restore_luntable(struct cxlflash_cfg *);
+int cxlflash_ioctl(struct scsi_device *sdev, int cmd, void __user *arg);
+void cxlflash_stop_term_user_contexts(struct cxlflash_cfg *cfg);
+int cxlflash_mark_contexts_error(struct cxlflash_cfg *cfg);
+void cxlflash_term_local_luns(struct cxlflash_cfg *cfg);
+void cxlflash_restore_luntable(struct cxlflash_cfg *cfg);
 
 #endif /* ifndef _CXLFLASH_COMMON_H */
diff --git a/drivers/scsi/cxlflash/lunmgt.c b/drivers/scsi/cxlflash/lunmgt.c
index 0efed177cc8b..4d232e271af6 100644
--- a/drivers/scsi/cxlflash/lunmgt.c
+++ b/drivers/scsi/cxlflash/lunmgt.c
@@ -252,7 +252,7 @@ int cxlflash_manage_lun(struct scsi_device *sdev,
 		 * in unpacked, AFU-friendly format, and hang LUN reference in
 		 * the sdev.
 		 */
-		lli->port_sel |= CHAN2PORT(chan);
+		lli->port_sel |= CHAN2PORTMASK(chan);
 		lli->lun_id[chan] = lun_to_lunid(sdev->lun);
 		sdev->hostdata = lli;
 	} else if (flags & DK_CXLFLASH_MANAGE_LUN_DISABLE_SUPERPIPE) {
@@ -264,7 +264,7 @@ int cxlflash_manage_lun(struct scsi_device *sdev,
 			 * tracking when no more references exist.
 			 */
 			sdev->hostdata = NULL;
-			lli->port_sel &= ~CHAN2PORT(chan);
+			lli->port_sel &= ~CHAN2PORTMASK(chan);
 			if (lli->port_sel == 0U)
 				lli->in_table = false;
 		}
diff --git a/drivers/scsi/cxlflash/main.c b/drivers/scsi/cxlflash/main.c
index 3061d8045382..a7d57c343492 100644
--- a/drivers/scsi/cxlflash/main.c
+++ b/drivers/scsi/cxlflash/main.c
@@ -176,7 +176,6 @@ static void cmd_complete(struct afu_cmd *cmd)
 		dev_dbg_ratelimited(dev, "%s:scp=%p result=%08x ioasc=%08x\n",
 				    __func__, scp, scp->result, cmd->sa.ioasc);
 
-		scsi_dma_unmap(scp);
 		scp->scsi_done(scp);
 
 		if (cmd_is_tmf) {
@@ -224,8 +223,9 @@ static void context_reset(struct afu_cmd *cmd, __be64 __iomem *reset_reg)
 static void context_reset_ioarrin(struct afu_cmd *cmd)
 {
 	struct afu *afu = cmd->parent;
+	struct hwq *hwq = get_hwq(afu, cmd->hwq_index);
 
-	context_reset(cmd, &afu->host_map->ioarrin);
+	context_reset(cmd, &hwq->host_map->ioarrin);
 }
 
 /**
@@ -235,8 +235,9 @@ static void context_reset_ioarrin(struct afu_cmd *cmd)
 static void context_reset_sq(struct afu_cmd *cmd)
 {
 	struct afu *afu = cmd->parent;
+	struct hwq *hwq = get_hwq(afu, cmd->hwq_index);
 
-	context_reset(cmd, &afu->host_map->sq_ctx_reset);
+	context_reset(cmd, &hwq->host_map->sq_ctx_reset);
 }
 
 /**
@@ -251,6 +252,7 @@ static int send_cmd_ioarrin(struct afu *afu, struct afu_cmd *cmd)
 {
 	struct cxlflash_cfg *cfg = afu->parent;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq = get_hwq(afu, cmd->hwq_index);
 	int rc = 0;
 	s64 room;
 	ulong lock_flags;
@@ -259,23 +261,23 @@ static int send_cmd_ioarrin(struct afu *afu, struct afu_cmd *cmd)
 	 * To avoid the performance penalty of MMIO, spread the update of
 	 * 'room' over multiple commands.
 	 */
-	spin_lock_irqsave(&afu->rrin_slock, lock_flags);
-	if (--afu->room < 0) {
-		room = readq_be(&afu->host_map->cmd_room);
+	spin_lock_irqsave(&hwq->rrin_slock, lock_flags);
+	if (--hwq->room < 0) {
+		room = readq_be(&hwq->host_map->cmd_room);
 		if (room <= 0) {
 			dev_dbg_ratelimited(dev, "%s: no cmd_room to send "
 					    "0x%02X, room=0x%016llX\n",
 					    __func__, cmd->rcb.cdb[0], room);
-			afu->room = 0;
+			hwq->room = 0;
 			rc = SCSI_MLQUEUE_HOST_BUSY;
 			goto out;
 		}
-		afu->room = room - 1;
+		hwq->room = room - 1;
 	}
 
-	writeq_be((u64)&cmd->rcb, &afu->host_map->ioarrin);
+	writeq_be((u64)&cmd->rcb, &hwq->host_map->ioarrin);
 out:
-	spin_unlock_irqrestore(&afu->rrin_slock, lock_flags);
+	spin_unlock_irqrestore(&hwq->rrin_slock, lock_flags);
 	dev_dbg(dev, "%s: cmd=%p len=%u ea=%016llx rc=%d\n", __func__,
 		cmd, cmd->rcb.data_len, cmd->rcb.data_ea, rc);
 	return rc;
@@ -293,11 +295,12 @@ static int send_cmd_sq(struct afu *afu, struct afu_cmd *cmd)
 {
 	struct cxlflash_cfg *cfg = afu->parent;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq = get_hwq(afu, cmd->hwq_index);
 	int rc = 0;
 	int newval;
 	ulong lock_flags;
 
-	newval = atomic_dec_if_positive(&afu->hsq_credits);
+	newval = atomic_dec_if_positive(&hwq->hsq_credits);
 	if (newval <= 0) {
 		rc = SCSI_MLQUEUE_HOST_BUSY;
 		goto out;
@@ -305,22 +308,22 @@ static int send_cmd_sq(struct afu *afu, struct afu_cmd *cmd)
 
 	cmd->rcb.ioasa = &cmd->sa;
 
-	spin_lock_irqsave(&afu->hsq_slock, lock_flags);
+	spin_lock_irqsave(&hwq->hsq_slock, lock_flags);
 
-	*afu->hsq_curr = cmd->rcb;
-	if (afu->hsq_curr < afu->hsq_end)
-		afu->hsq_curr++;
+	*hwq->hsq_curr = cmd->rcb;
+	if (hwq->hsq_curr < hwq->hsq_end)
+		hwq->hsq_curr++;
 	else
-		afu->hsq_curr = afu->hsq_start;
-	writeq_be((u64)afu->hsq_curr, &afu->host_map->sq_tail);
+		hwq->hsq_curr = hwq->hsq_start;
+	writeq_be((u64)hwq->hsq_curr, &hwq->host_map->sq_tail);
 
-	spin_unlock_irqrestore(&afu->hsq_slock, lock_flags);
+	spin_unlock_irqrestore(&hwq->hsq_slock, lock_flags);
 out:
 	dev_dbg(dev, "%s: cmd=%p len=%u ea=%016llx ioasa=%p rc=%d curr=%p "
 	       "head=%016llx tail=%016llx\n", __func__, cmd, cmd->rcb.data_len,
-	       cmd->rcb.data_ea, cmd->rcb.ioasa, rc, afu->hsq_curr,
-	       readq_be(&afu->host_map->sq_head),
-	       readq_be(&afu->host_map->sq_tail));
+	       cmd->rcb.data_ea, cmd->rcb.ioasa, rc, hwq->hsq_curr,
+	       readq_be(&hwq->host_map->sq_head),
+	       readq_be(&hwq->host_map->sq_tail));
 	return rc;
 }
 
@@ -355,6 +358,43 @@ static int wait_resp(struct afu *afu, struct afu_cmd *cmd)
 }
 
 /**
+ * cmd_to_target_hwq() - selects a target hardware queue for a SCSI command
+ * @host:	SCSI host associated with device.
+ * @scp:	SCSI command to send.
+ * @afu:	SCSI command to send.
+ *
+ * Hashes a command based upon the hardware queue mode.
+ *
+ * Return: Trusted index of target hardware queue
+ */
+static u32 cmd_to_target_hwq(struct Scsi_Host *host, struct scsi_cmnd *scp,
+			     struct afu *afu)
+{
+	u32 tag;
+	u32 hwq = 0;
+
+	if (afu->num_hwqs == 1)
+		return 0;
+
+	switch (afu->hwq_mode) {
+	case HWQ_MODE_RR:
+		hwq = afu->hwq_rr_count++ % afu->num_hwqs;
+		break;
+	case HWQ_MODE_TAG:
+		tag = blk_mq_unique_tag(scp->request);
+		hwq = blk_mq_unique_tag_to_hwq(tag);
+		break;
+	case HWQ_MODE_CPU:
+		hwq = smp_processor_id() % afu->num_hwqs;
+		break;
+	default:
+		WARN_ON_ONCE(1);
+	}
+
+	return hwq;
+}
+
+/**
  * send_tmf() - sends a Task Management Function (TMF)
  * @afu:	AFU to checkout from.
  * @scp:	SCSI command from stack.
@@ -365,10 +405,12 @@ static int wait_resp(struct afu *afu, struct afu_cmd *cmd)
  */
 static int send_tmf(struct afu *afu, struct scsi_cmnd *scp, u64 tmfcmd)
 {
-	u32 port_sel = scp->device->channel + 1;
-	struct cxlflash_cfg *cfg = shost_priv(scp->device->host);
+	struct Scsi_Host *host = scp->device->host;
+	struct cxlflash_cfg *cfg = shost_priv(host);
 	struct afu_cmd *cmd = sc_to_afucz(scp);
 	struct device *dev = &cfg->dev->dev;
+	int hwq_index = cmd_to_target_hwq(host, scp, afu);
+	struct hwq *hwq = get_hwq(afu, hwq_index);
 	ulong lock_flags;
 	int rc = 0;
 	ulong to;
@@ -385,10 +427,11 @@ static int send_tmf(struct afu *afu, struct scsi_cmnd *scp, u64 tmfcmd)
 	cmd->scp = scp;
 	cmd->parent = afu;
 	cmd->cmd_tmf = true;
+	cmd->hwq_index = hwq_index;
 
-	cmd->rcb.ctx_id = afu->ctx_hndl;
+	cmd->rcb.ctx_id = hwq->ctx_hndl;
 	cmd->rcb.msi = SISL_MSI_RRQ_UPDATED;
-	cmd->rcb.port_sel = port_sel;
+	cmd->rcb.port_sel = CHAN2PORTMASK(scp->device->channel);
 	cmd->rcb.lun_id = lun_to_lunid(scp->device->lun);
 	cmd->rcb.req_flags = (SISL_REQ_FLAGS_PORT_LUN_ID |
 			      SISL_REQ_FLAGS_SUP_UNDERRUN |
@@ -444,10 +487,10 @@ static int cxlflash_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scp)
 	struct device *dev = &cfg->dev->dev;
 	struct afu_cmd *cmd = sc_to_afucz(scp);
 	struct scatterlist *sg = scsi_sglist(scp);
-	u32 port_sel = scp->device->channel + 1;
+	int hwq_index = cmd_to_target_hwq(host, scp, afu);
+	struct hwq *hwq = get_hwq(afu, hwq_index);
 	u16 req_flags = SISL_REQ_FLAGS_SUP_UNDERRUN;
 	ulong lock_flags;
-	int nseg = 0;
 	int rc = 0;
 
 	dev_dbg_ratelimited(dev, "%s: (scp=%p) %d/%d/%d/%llu "
@@ -472,6 +515,8 @@ static int cxlflash_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scp)
 	spin_unlock_irqrestore(&cfg->tmf_slock, lock_flags);
 
 	switch (cfg->state) {
+	case STATE_PROBING:
+	case STATE_PROBED:
 	case STATE_RESET:
 		dev_dbg_ratelimited(dev, "%s: device is in reset\n", __func__);
 		rc = SCSI_MLQUEUE_HOST_BUSY;
@@ -487,23 +532,17 @@ static int cxlflash_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scp)
 	}
 
 	if (likely(sg)) {
-		nseg = scsi_dma_map(scp);
-		if (unlikely(nseg < 0)) {
-			dev_err(dev, "%s: Fail DMA map\n", __func__);
-			rc = SCSI_MLQUEUE_HOST_BUSY;
-			goto out;
-		}
-
-		cmd->rcb.data_len = sg_dma_len(sg);
-		cmd->rcb.data_ea = sg_dma_address(sg);
+		cmd->rcb.data_len = sg->length;
+		cmd->rcb.data_ea = (uintptr_t)sg_virt(sg);
 	}
 
 	cmd->scp = scp;
 	cmd->parent = afu;
+	cmd->hwq_index = hwq_index;
 
-	cmd->rcb.ctx_id = afu->ctx_hndl;
+	cmd->rcb.ctx_id = hwq->ctx_hndl;
 	cmd->rcb.msi = SISL_MSI_RRQ_UPDATED;
-	cmd->rcb.port_sel = port_sel;
+	cmd->rcb.port_sel = CHAN2PORTMASK(scp->device->channel);
 	cmd->rcb.lun_id = lun_to_lunid(scp->device->lun);
 
 	if (scp->sc_data_direction == DMA_TO_DEVICE)
@@ -513,8 +552,6 @@ static int cxlflash_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scp)
 	memcpy(cmd->rcb.cdb, scp->cmnd, sizeof(cmd->rcb.cdb));
 
 	rc = afu->send_cmd(afu, cmd);
-	if (unlikely(rc))
-		scsi_dma_unmap(scp);
 out:
 	return rc;
 }
@@ -554,17 +591,28 @@ static void free_mem(struct cxlflash_cfg *cfg)
  * Safe to call with AFU in a partially allocated/initialized state.
  *
  * Cancels scheduled worker threads, waits for any active internal AFU
- * commands to timeout and then unmaps the MMIO space.
+ * commands to timeout, disables IRQ polling and then unmaps the MMIO space.
  */
 static void stop_afu(struct cxlflash_cfg *cfg)
 {
 	struct afu *afu = cfg->afu;
+	struct hwq *hwq;
+	int i;
 
 	cancel_work_sync(&cfg->work_q);
 
 	if (likely(afu)) {
 		while (atomic_read(&afu->cmds_active))
 			ssleep(1);
+
+		if (afu_is_irqpoll_enabled(afu)) {
+			for (i = 0; i < afu->num_hwqs; i++) {
+				hwq = get_hwq(afu, i);
+
+				irq_poll_disable(&hwq->irqpoll);
+			}
+		}
+
 		if (likely(afu->afu_map)) {
 			cxl_psa_unmap((void __iomem *)afu->afu_map);
 			afu->afu_map = NULL;
@@ -576,28 +624,40 @@ static void stop_afu(struct cxlflash_cfg *cfg)
  * term_intr() - disables all AFU interrupts
  * @cfg:	Internal structure associated with the host.
  * @level:	Depth of allocation, where to begin waterfall tear down.
+ * @index:	Index of the hardware queue.
  *
  * Safe to call with AFU/MC in partially allocated/initialized state.
  */
-static void term_intr(struct cxlflash_cfg *cfg, enum undo_level level)
+static void term_intr(struct cxlflash_cfg *cfg, enum undo_level level,
+		      u32 index)
 {
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq;
 
-	if (!afu || !cfg->mcctx) {
-		dev_err(dev, "%s: returning with NULL afu or MC\n", __func__);
+	if (!afu) {
+		dev_err(dev, "%s: returning with NULL afu\n", __func__);
+		return;
+	}
+
+	hwq = get_hwq(afu, index);
+
+	if (!hwq->ctx) {
+		dev_err(dev, "%s: returning with NULL MC\n", __func__);
 		return;
 	}
 
 	switch (level) {
 	case UNMAP_THREE:
-		cxl_unmap_afu_irq(cfg->mcctx, 3, afu);
+		/* SISL_MSI_ASYNC_ERROR is setup only for the primary HWQ */
+		if (index == PRIMARY_HWQ)
+			cxl_unmap_afu_irq(hwq->ctx, 3, hwq);
 	case UNMAP_TWO:
-		cxl_unmap_afu_irq(cfg->mcctx, 2, afu);
+		cxl_unmap_afu_irq(hwq->ctx, 2, hwq);
 	case UNMAP_ONE:
-		cxl_unmap_afu_irq(cfg->mcctx, 1, afu);
+		cxl_unmap_afu_irq(hwq->ctx, 1, hwq);
 	case FREE_IRQ:
-		cxl_free_afu_irqs(cfg->mcctx);
+		cxl_free_afu_irqs(hwq->ctx);
 		/* fall through */
 	case UNDO_NOOP:
 		/* No action required */
@@ -608,24 +668,32 @@ static void term_intr(struct cxlflash_cfg *cfg, enum undo_level level)
 /**
  * term_mc() - terminates the master context
  * @cfg:	Internal structure associated with the host.
- * @level:	Depth of allocation, where to begin waterfall tear down.
+ * @index:	Index of the hardware queue.
  *
  * Safe to call with AFU/MC in partially allocated/initialized state.
  */
-static void term_mc(struct cxlflash_cfg *cfg)
+static void term_mc(struct cxlflash_cfg *cfg, u32 index)
 {
-	int rc = 0;
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq;
 
-	if (!afu || !cfg->mcctx) {
-		dev_err(dev, "%s: returning with NULL afu or MC\n", __func__);
+	if (!afu) {
+		dev_err(dev, "%s: returning with NULL afu\n", __func__);
 		return;
 	}
 
-	rc = cxl_stop_context(cfg->mcctx);
-	WARN_ON(rc);
-	cfg->mcctx = NULL;
+	hwq = get_hwq(afu, index);
+
+	if (!hwq->ctx) {
+		dev_err(dev, "%s: returning with NULL MC\n", __func__);
+		return;
+	}
+
+	WARN_ON(cxl_stop_context(hwq->ctx));
+	if (index != PRIMARY_HWQ)
+		WARN_ON(cxl_release_context(hwq->ctx));
+	hwq->ctx = NULL;
 }
 
 /**
@@ -637,21 +705,25 @@ static void term_mc(struct cxlflash_cfg *cfg)
 static void term_afu(struct cxlflash_cfg *cfg)
 {
 	struct device *dev = &cfg->dev->dev;
+	int k;
 
 	/*
 	 * Tear down is carefully orchestrated to ensure
 	 * no interrupts can come in when the problem state
 	 * area is unmapped.
 	 *
-	 * 1) Disable all AFU interrupts
+	 * 1) Disable all AFU interrupts for each master
 	 * 2) Unmap the problem state area
-	 * 3) Stop the master context
+	 * 3) Stop each master context
 	 */
-	term_intr(cfg, UNMAP_THREE);
+	for (k = cfg->afu->num_hwqs - 1; k >= 0; k--)
+		term_intr(cfg, UNMAP_THREE, k);
+
 	if (cfg->afu)
 		stop_afu(cfg);
 
-	term_mc(cfg);
+	for (k = cfg->afu->num_hwqs - 1; k >= 0; k--)
+		term_mc(cfg, k);
 
 	dev_dbg(dev, "%s: returning\n", __func__);
 }
@@ -670,8 +742,8 @@ static void notify_shutdown(struct cxlflash_cfg *cfg, bool wait)
 {
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
-	struct sisl_global_map __iomem *global;
 	struct dev_dependent_vals *ddv;
+	__be64 __iomem *fc_port_regs;
 	u64 reg, status;
 	int i, retry_cnt = 0;
 
@@ -684,23 +756,25 @@ static void notify_shutdown(struct cxlflash_cfg *cfg, bool wait)
 		return;
 	}
 
-	global = &afu->afu_map->global;
-
 	/* Notify AFU */
-	for (i = 0; i < NUM_FC_PORTS; i++) {
-		reg = readq_be(&global->fc_regs[i][FC_CONFIG2 / 8]);
+	for (i = 0; i < cfg->num_fc_ports; i++) {
+		fc_port_regs = get_fc_port_regs(cfg, i);
+
+		reg = readq_be(&fc_port_regs[FC_CONFIG2 / 8]);
 		reg |= SISL_FC_SHUTDOWN_NORMAL;
-		writeq_be(reg, &global->fc_regs[i][FC_CONFIG2 / 8]);
+		writeq_be(reg, &fc_port_regs[FC_CONFIG2 / 8]);
 	}
 
 	if (!wait)
 		return;
 
 	/* Wait up to 1.5 seconds for shutdown processing to complete */
-	for (i = 0; i < NUM_FC_PORTS; i++) {
+	for (i = 0; i < cfg->num_fc_ports; i++) {
+		fc_port_regs = get_fc_port_regs(cfg, i);
 		retry_cnt = 0;
+
 		while (true) {
-			status = readq_be(&global->fc_regs[i][FC_STATUS / 8]);
+			status = readq_be(&fc_port_regs[FC_STATUS / 8]);
 			if (status & SISL_STATUS_SHUTDOWN_COMPLETE)
 				break;
 			if (++retry_cnt >= MC_RETRY_CNT) {
@@ -717,7 +791,8 @@ static void notify_shutdown(struct cxlflash_cfg *cfg, bool wait)
  * cxlflash_remove() - PCI entry point to tear down host
  * @pdev:	PCI device associated with the host.
  *
- * Safe to use as a cleanup in partially allocated/initialized state.
+ * Safe to use as a cleanup in partially allocated/initialized state. Note that
+ * the reset_waitq is flushed as part of the stop/termination of user contexts.
  */
 static void cxlflash_remove(struct pci_dev *pdev)
 {
@@ -750,7 +825,6 @@ static void cxlflash_remove(struct pci_dev *pdev)
 	case INIT_STATE_SCSI:
 		cxlflash_term_local_luns(cfg);
 		scsi_remove_host(cfg->host);
-		/* fall through */
 	case INIT_STATE_AFU:
 		term_afu(cfg);
 	case INIT_STATE_PCI:
@@ -789,6 +863,7 @@ static int alloc_mem(struct cxlflash_cfg *cfg)
 		goto out;
 	}
 	cfg->afu->parent = cfg;
+	cfg->afu->desired_hwqs = CXLFLASH_DEF_HWQS;
 	cfg->afu->afu_map = NULL;
 out:
 	return rc;
@@ -1024,53 +1099,16 @@ static void afu_link_reset(struct afu *afu, int port, __be64 __iomem *fc_regs)
 	dev_dbg(dev, "%s: returning port_sel=%016llx\n", __func__, port_sel);
 }
 
-/*
- * Asynchronous interrupt information table
- */
-static const struct asyc_intr_info ainfo[] = {
-	{SISL_ASTATUS_FC0_OTHER, "other error", 0, CLR_FC_ERROR | LINK_RESET},
-	{SISL_ASTATUS_FC0_LOGO, "target initiated LOGO", 0, 0},
-	{SISL_ASTATUS_FC0_CRC_T, "CRC threshold exceeded", 0, LINK_RESET},
-	{SISL_ASTATUS_FC0_LOGI_R, "login timed out, retrying", 0, LINK_RESET},
-	{SISL_ASTATUS_FC0_LOGI_F, "login failed", 0, CLR_FC_ERROR},
-	{SISL_ASTATUS_FC0_LOGI_S, "login succeeded", 0, SCAN_HOST},
-	{SISL_ASTATUS_FC0_LINK_DN, "link down", 0, 0},
-	{SISL_ASTATUS_FC0_LINK_UP, "link up", 0, 0},
-	{SISL_ASTATUS_FC1_OTHER, "other error", 1, CLR_FC_ERROR | LINK_RESET},
-	{SISL_ASTATUS_FC1_LOGO, "target initiated LOGO", 1, 0},
-	{SISL_ASTATUS_FC1_CRC_T, "CRC threshold exceeded", 1, LINK_RESET},
-	{SISL_ASTATUS_FC1_LOGI_R, "login timed out, retrying", 1, LINK_RESET},
-	{SISL_ASTATUS_FC1_LOGI_F, "login failed", 1, CLR_FC_ERROR},
-	{SISL_ASTATUS_FC1_LOGI_S, "login succeeded", 1, SCAN_HOST},
-	{SISL_ASTATUS_FC1_LINK_DN, "link down", 1, 0},
-	{SISL_ASTATUS_FC1_LINK_UP, "link up", 1, 0},
-	{0x0, "", 0, 0}		/* terminator */
-};
-
-/**
- * find_ainfo() - locates and returns asynchronous interrupt information
- * @status:	Status code set by AFU on error.
- *
- * Return: The located information or NULL when the status code is invalid.
- */
-static const struct asyc_intr_info *find_ainfo(u64 status)
-{
-	const struct asyc_intr_info *info;
-
-	for (info = &ainfo[0]; info->status; info++)
-		if (info->status == status)
-			return info;
-
-	return NULL;
-}
-
 /**
  * afu_err_intr_init() - clears and initializes the AFU for error interrupts
  * @afu:	AFU associated with the host.
  */
 static void afu_err_intr_init(struct afu *afu)
 {
+	struct cxlflash_cfg *cfg = afu->parent;
+	__be64 __iomem *fc_port_regs;
 	int i;
+	struct hwq *hwq = get_hwq(afu, PRIMARY_HWQ);
 	u64 reg;
 
 	/* global async interrupts: AFU clears afu_ctrl on context exit
@@ -1082,8 +1120,8 @@ static void afu_err_intr_init(struct afu *afu)
 
 	/* mask all */
 	writeq_be(-1ULL, &afu->afu_map->global.regs.aintr_mask);
-	/* set LISN# to send and point to master context */
-	reg = ((u64) (((afu->ctx_hndl << 8) | SISL_MSI_ASYNC_ERROR)) << 40);
+	/* set LISN# to send and point to primary master context */
+	reg = ((u64) (((hwq->ctx_hndl << 8) | SISL_MSI_ASYNC_ERROR)) << 40);
 
 	if (afu->internal_lun)
 		reg |= 1;	/* Bit 63 indicates local lun */
@@ -1098,17 +1136,19 @@ static void afu_err_intr_init(struct afu *afu)
 	writeq_be(-1ULL, &afu->afu_map->global.regs.aintr_clear);
 
 	/* Clear/Set internal lun bits */
-	reg = readq_be(&afu->afu_map->global.fc_regs[0][FC_CONFIG2 / 8]);
+	fc_port_regs = get_fc_port_regs(cfg, 0);
+	reg = readq_be(&fc_port_regs[FC_CONFIG2 / 8]);
 	reg &= SISL_FC_INTERNAL_MASK;
 	if (afu->internal_lun)
 		reg |= ((u64)(afu->internal_lun - 1) << SISL_FC_INTERNAL_SHIFT);
-	writeq_be(reg, &afu->afu_map->global.fc_regs[0][FC_CONFIG2 / 8]);
+	writeq_be(reg, &fc_port_regs[FC_CONFIG2 / 8]);
 
 	/* now clear FC errors */
-	for (i = 0; i < NUM_FC_PORTS; i++) {
-		writeq_be(0xFFFFFFFFU,
-			  &afu->afu_map->global.fc_regs[i][FC_ERROR / 8]);
-		writeq_be(0, &afu->afu_map->global.fc_regs[i][FC_ERRCAP / 8]);
+	for (i = 0; i < cfg->num_fc_ports; i++) {
+		fc_port_regs = get_fc_port_regs(cfg, i);
+
+		writeq_be(0xFFFFFFFFU, &fc_port_regs[FC_ERROR / 8]);
+		writeq_be(0, &fc_port_regs[FC_ERRCAP / 8]);
 	}
 
 	/* sync interrupts for master's IOARRIN write */
@@ -1117,8 +1157,12 @@ static void afu_err_intr_init(struct afu *afu)
 	/* IOARRIN yet), so there is nothing to clear. */
 
 	/* set LISN#, it is always sent to the context that wrote IOARRIN */
-	writeq_be(SISL_MSI_SYNC_ERROR, &afu->host_map->ctx_ctrl);
-	writeq_be(SISL_ISTATUS_MASK, &afu->host_map->intr_mask);
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
+
+		writeq_be(SISL_MSI_SYNC_ERROR, &hwq->host_map->ctx_ctrl);
+		writeq_be(SISL_ISTATUS_MASK, &hwq->host_map->intr_mask);
+	}
 }
 
 /**
@@ -1130,13 +1174,13 @@ static void afu_err_intr_init(struct afu *afu)
  */
 static irqreturn_t cxlflash_sync_err_irq(int irq, void *data)
 {
-	struct afu *afu = (struct afu *)data;
-	struct cxlflash_cfg *cfg = afu->parent;
+	struct hwq *hwq = (struct hwq *)data;
+	struct cxlflash_cfg *cfg = hwq->afu->parent;
 	struct device *dev = &cfg->dev->dev;
 	u64 reg;
 	u64 reg_unmasked;
 
-	reg = readq_be(&afu->host_map->intr_status);
+	reg = readq_be(&hwq->host_map->intr_status);
 	reg_unmasked = (reg & SISL_ISTATUS_UNMASK);
 
 	if (reg_unmasked == 0UL) {
@@ -1148,32 +1192,36 @@ static irqreturn_t cxlflash_sync_err_irq(int irq, void *data)
 	dev_err(dev, "%s: unexpected interrupt, intr_status=%016llx\n",
 		__func__, reg);
 
-	writeq_be(reg_unmasked, &afu->host_map->intr_clear);
+	writeq_be(reg_unmasked, &hwq->host_map->intr_clear);
 
 cxlflash_sync_err_irq_exit:
 	return IRQ_HANDLED;
 }
 
 /**
- * cxlflash_rrq_irq() - interrupt handler for read-response queue (normal path)
- * @irq:	Interrupt number.
- * @data:	Private data provided at interrupt registration, the AFU.
+ * process_hrrq() - process the read-response queue
+ * @afu:	AFU associated with the host.
+ * @doneq:	Queue of commands harvested from the RRQ.
+ * @budget:	Threshold of RRQ entries to process.
  *
- * Return: Always return IRQ_HANDLED.
+ * This routine must be called holding the disabled RRQ spin lock.
+ *
+ * Return: The number of entries processed.
  */
-static irqreturn_t cxlflash_rrq_irq(int irq, void *data)
+static int process_hrrq(struct hwq *hwq, struct list_head *doneq, int budget)
 {
-	struct afu *afu = (struct afu *)data;
+	struct afu *afu = hwq->afu;
 	struct afu_cmd *cmd;
 	struct sisl_ioasa *ioasa;
 	struct sisl_ioarcb *ioarcb;
-	bool toggle = afu->toggle;
+	bool toggle = hwq->toggle;
+	int num_hrrq = 0;
 	u64 entry,
-	    *hrrq_start = afu->hrrq_start,
-	    *hrrq_end = afu->hrrq_end,
-	    *hrrq_curr = afu->hrrq_curr;
+	    *hrrq_start = hwq->hrrq_start,
+	    *hrrq_end = hwq->hrrq_end,
+	    *hrrq_curr = hwq->hrrq_curr;
 
-	/* Process however many RRQ entries that are ready */
+	/* Process ready RRQ entries up to the specified budget (if any) */
 	while (true) {
 		entry = *hrrq_curr;
 
@@ -1190,7 +1238,7 @@ static irqreturn_t cxlflash_rrq_irq(int irq, void *data)
 			cmd = container_of(ioarcb, struct afu_cmd, rcb);
 		}
 
-		cmd_complete(cmd);
+		list_add_tail(&cmd->queue, doneq);
 
 		/* Advance to next entry or wrap and flip the toggle bit */
 		if (hrrq_curr < hrrq_end)
@@ -1200,15 +1248,123 @@ static irqreturn_t cxlflash_rrq_irq(int irq, void *data)
 			toggle ^= SISL_RESP_HANDLE_T_BIT;
 		}
 
-		atomic_inc(&afu->hsq_credits);
+		atomic_inc(&hwq->hsq_credits);
+		num_hrrq++;
+
+		if (budget > 0 && num_hrrq >= budget)
+			break;
+	}
+
+	hwq->hrrq_curr = hrrq_curr;
+	hwq->toggle = toggle;
+
+	return num_hrrq;
+}
+
+/**
+ * process_cmd_doneq() - process a queue of harvested RRQ commands
+ * @doneq:	Queue of completed commands.
+ *
+ * Note that upon return the queue can no longer be trusted.
+ */
+static void process_cmd_doneq(struct list_head *doneq)
+{
+	struct afu_cmd *cmd, *tmp;
+
+	WARN_ON(list_empty(doneq));
+
+	list_for_each_entry_safe(cmd, tmp, doneq, queue)
+		cmd_complete(cmd);
+}
+
+/**
+ * cxlflash_irqpoll() - process a queue of harvested RRQ commands
+ * @irqpoll:	IRQ poll structure associated with queue to poll.
+ * @budget:	Threshold of RRQ entries to process per poll.
+ *
+ * Return: The number of entries processed.
+ */
+static int cxlflash_irqpoll(struct irq_poll *irqpoll, int budget)
+{
+	struct hwq *hwq = container_of(irqpoll, struct hwq, irqpoll);
+	unsigned long hrrq_flags;
+	LIST_HEAD(doneq);
+	int num_entries = 0;
+
+	spin_lock_irqsave(&hwq->hrrq_slock, hrrq_flags);
+
+	num_entries = process_hrrq(hwq, &doneq, budget);
+	if (num_entries < budget)
+		irq_poll_complete(irqpoll);
+
+	spin_unlock_irqrestore(&hwq->hrrq_slock, hrrq_flags);
+
+	process_cmd_doneq(&doneq);
+	return num_entries;
+}
+
+/**
+ * cxlflash_rrq_irq() - interrupt handler for read-response queue (normal path)
+ * @irq:	Interrupt number.
+ * @data:	Private data provided at interrupt registration, the AFU.
+ *
+ * Return: IRQ_HANDLED or IRQ_NONE when no ready entries found.
+ */
+static irqreturn_t cxlflash_rrq_irq(int irq, void *data)
+{
+	struct hwq *hwq = (struct hwq *)data;
+	struct afu *afu = hwq->afu;
+	unsigned long hrrq_flags;
+	LIST_HEAD(doneq);
+	int num_entries = 0;
+
+	spin_lock_irqsave(&hwq->hrrq_slock, hrrq_flags);
+
+	if (afu_is_irqpoll_enabled(afu)) {
+		irq_poll_sched(&hwq->irqpoll);
+		spin_unlock_irqrestore(&hwq->hrrq_slock, hrrq_flags);
+		return IRQ_HANDLED;
 	}
 
-	afu->hrrq_curr = hrrq_curr;
-	afu->toggle = toggle;
+	num_entries = process_hrrq(hwq, &doneq, -1);
+	spin_unlock_irqrestore(&hwq->hrrq_slock, hrrq_flags);
+
+	if (num_entries == 0)
+		return IRQ_NONE;
 
+	process_cmd_doneq(&doneq);
 	return IRQ_HANDLED;
 }
 
+/*
+ * Asynchronous interrupt information table
+ *
+ * NOTE:
+ *	- Order matters here as this array is indexed by bit position.
+ *
+ *	- The checkpatch script considers the BUILD_SISL_ASTATUS_FC_PORT macro
+ *	  as complex and complains due to a lack of parentheses/braces.
+ */
+#define ASTATUS_FC(_a, _b, _c, _d)					 \
+	{ SISL_ASTATUS_FC##_a##_##_b, _c, _a, (_d) }
+
+#define BUILD_SISL_ASTATUS_FC_PORT(_a)					 \
+	ASTATUS_FC(_a, LINK_UP, "link up", 0),				 \
+	ASTATUS_FC(_a, LINK_DN, "link down", 0),			 \
+	ASTATUS_FC(_a, LOGI_S, "login succeeded", SCAN_HOST),		 \
+	ASTATUS_FC(_a, LOGI_F, "login failed", CLR_FC_ERROR),		 \
+	ASTATUS_FC(_a, LOGI_R, "login timed out, retrying", LINK_RESET), \
+	ASTATUS_FC(_a, CRC_T, "CRC threshold exceeded", LINK_RESET),	 \
+	ASTATUS_FC(_a, LOGO, "target initiated LOGO", 0),		 \
+	ASTATUS_FC(_a, OTHER, "other error", CLR_FC_ERROR | LINK_RESET)
+
+static const struct asyc_intr_info ainfo[] = {
+	BUILD_SISL_ASTATUS_FC_PORT(1),
+	BUILD_SISL_ASTATUS_FC_PORT(0),
+	BUILD_SISL_ASTATUS_FC_PORT(3),
+	BUILD_SISL_ASTATUS_FC_PORT(2)
+};
+
 /**
  * cxlflash_async_err_irq() - interrupt handler for asynchronous errors
  * @irq:	Interrupt number.
@@ -1218,20 +1374,22 @@ static irqreturn_t cxlflash_rrq_irq(int irq, void *data)
  */
 static irqreturn_t cxlflash_async_err_irq(int irq, void *data)
 {
-	struct afu *afu = (struct afu *)data;
+	struct hwq *hwq = (struct hwq *)data;
+	struct afu *afu = hwq->afu;
 	struct cxlflash_cfg *cfg = afu->parent;
 	struct device *dev = &cfg->dev->dev;
-	u64 reg_unmasked;
 	const struct asyc_intr_info *info;
 	struct sisl_global_map __iomem *global = &afu->afu_map->global;
+	__be64 __iomem *fc_port_regs;
+	u64 reg_unmasked;
 	u64 reg;
+	u64 bit;
 	u8 port;
-	int i;
 
 	reg = readq_be(&global->regs.aintr_status);
 	reg_unmasked = (reg & SISL_ASTATUS_UNMASK);
 
-	if (reg_unmasked == 0) {
+	if (unlikely(reg_unmasked == 0)) {
 		dev_err(dev, "%s: spurious interrupt, aintr_status=%016llx\n",
 			__func__, reg);
 		goto out;
@@ -1241,16 +1399,24 @@ static irqreturn_t cxlflash_async_err_irq(int irq, void *data)
 	writeq_be(reg_unmasked, &global->regs.aintr_clear);
 
 	/* Check each bit that is on */
-	for (i = 0; reg_unmasked; i++, reg_unmasked = (reg_unmasked >> 1)) {
-		info = find_ainfo(1ULL << i);
-		if (((reg_unmasked & 0x1) == 0) || !info)
+	for_each_set_bit(bit, (ulong *)&reg_unmasked, BITS_PER_LONG) {
+		if (unlikely(bit >= ARRAY_SIZE(ainfo))) {
+			WARN_ON_ONCE(1);
 			continue;
+		}
+
+		info = &ainfo[bit];
+		if (unlikely(info->status != 1ULL << bit)) {
+			WARN_ON_ONCE(1);
+			continue;
+		}
 
 		port = info->port;
+		fc_port_regs = get_fc_port_regs(cfg, port);
 
 		dev_err(dev, "%s: FC Port %d -> %s, fc_status=%016llx\n",
 			__func__, port, info->desc,
-		       readq_be(&global->fc_regs[port][FC_STATUS / 8]));
+		       readq_be(&fc_port_regs[FC_STATUS / 8]));
 
 		/*
 		 * Do link reset first, some OTHER errors will set FC_ERROR
@@ -1265,7 +1431,7 @@ static irqreturn_t cxlflash_async_err_irq(int irq, void *data)
 		}
 
 		if (info->action & CLR_FC_ERROR) {
-			reg = readq_be(&global->fc_regs[port][FC_ERROR / 8]);
+			reg = readq_be(&fc_port_regs[FC_ERROR / 8]);
 
 			/*
 			 * Since all errors are unmasked, FC_ERROR and FC_ERRCAP
@@ -1275,8 +1441,8 @@ static irqreturn_t cxlflash_async_err_irq(int irq, void *data)
 			dev_err(dev, "%s: fc %d: clearing fc_error=%016llx\n",
 				__func__, port, reg);
 
-			writeq_be(reg, &global->fc_regs[port][FC_ERROR / 8]);
-			writeq_be(0, &global->fc_regs[port][FC_ERRCAP / 8]);
+			writeq_be(reg, &fc_port_regs[FC_ERROR / 8]);
+			writeq_be(0, &fc_port_regs[FC_ERRCAP / 8]);
 		}
 
 		if (info->action & SCAN_HOST) {
@@ -1292,16 +1458,18 @@ out:
 /**
  * start_context() - starts the master context
  * @cfg:	Internal structure associated with the host.
+ * @index:	Index of the hardware queue.
  *
  * Return: A success or failure value from CXL services.
  */
-static int start_context(struct cxlflash_cfg *cfg)
+static int start_context(struct cxlflash_cfg *cfg, u32 index)
 {
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq = get_hwq(cfg->afu, index);
 	int rc = 0;
 
-	rc = cxl_start_context(cfg->mcctx,
-			       cfg->afu->work.work_element_descriptor,
+	rc = cxl_start_context(hwq->ctx,
+			       hwq->work.work_element_descriptor,
 			       NULL);
 
 	dev_dbg(dev, "%s: returning rc=%d\n", __func__, rc);
@@ -1311,7 +1479,7 @@ static int start_context(struct cxlflash_cfg *cfg)
 /**
  * read_vpd() - obtains the WWPNs from VPD
  * @cfg:	Internal structure associated with the host.
- * @wwpn:	Array of size NUM_FC_PORTS to pass back WWPNs
+ * @wwpn:	Array of size MAX_FC_PORTS to pass back WWPNs
  *
  * Return: 0 on success, -errno on failure
  */
@@ -1324,7 +1492,7 @@ static int read_vpd(struct cxlflash_cfg *cfg, u64 wwpn[])
 	ssize_t vpd_size;
 	char vpd_data[CXLFLASH_VPD_LEN];
 	char tmp_buf[WWPN_BUF_LEN] = { 0 };
-	char *wwpn_vpd_tags[NUM_FC_PORTS] = { "V5", "V6" };
+	char *wwpn_vpd_tags[MAX_FC_PORTS] = { "V5", "V6", "V7", "V8" };
 
 	/* Get the VPD data from the device */
 	vpd_size = cxl_read_adapter_vpd(pdev, vpd_data, sizeof(vpd_data));
@@ -1362,7 +1530,7 @@ static int read_vpd(struct cxlflash_cfg *cfg, u64 wwpn[])
 	 * because the conversion service requires that the ASCII
 	 * string be terminated.
 	 */
-	for (k = 0; k < NUM_FC_PORTS; k++) {
+	for (k = 0; k < cfg->num_fc_ports; k++) {
 		j = ro_size;
 		i = ro_start + PCI_VPD_LRDT_TAG_SIZE;
 
@@ -1391,6 +1559,8 @@ static int read_vpd(struct cxlflash_cfg *cfg, u64 wwpn[])
 			rc = -ENODEV;
 			goto out;
 		}
+
+		dev_dbg(dev, "%s: wwpn%d=%016llx\n", __func__, k, wwpn[k]);
 	}
 
 out:
@@ -1409,6 +1579,7 @@ static void init_pcr(struct cxlflash_cfg *cfg)
 {
 	struct afu *afu = cfg->afu;
 	struct sisl_ctrl_map __iomem *ctrl_map;
+	struct hwq *hwq;
 	int i;
 
 	for (i = 0; i < MAX_CONTEXT; i++) {
@@ -1420,13 +1591,17 @@ static void init_pcr(struct cxlflash_cfg *cfg)
 		writeq_be(0, &ctrl_map->ctx_cap);
 	}
 
-	/* Copy frequently used fields into afu */
-	afu->ctx_hndl = (u16) cxl_process_element(cfg->mcctx);
-	afu->host_map = &afu->afu_map->hosts[afu->ctx_hndl].host;
-	afu->ctrl_map = &afu->afu_map->ctrls[afu->ctx_hndl].ctrl;
+	/* Copy frequently used fields into hwq */
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
+
+		hwq->ctx_hndl = (u16) cxl_process_element(hwq->ctx);
+		hwq->host_map = &afu->afu_map->hosts[hwq->ctx_hndl].host;
+		hwq->ctrl_map = &afu->afu_map->ctrls[hwq->ctx_hndl].ctrl;
 
-	/* Program the Endian Control for the master context */
-	writeq_be(SISL_ENDIAN_CTRL, &afu->host_map->endian_ctrl);
+		/* Program the Endian Control for the master context */
+		writeq_be(SISL_ENDIAN_CTRL, &hwq->host_map->endian_ctrl);
+	}
 }
 
 /**
@@ -1437,7 +1612,10 @@ static int init_global(struct cxlflash_cfg *cfg)
 {
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
-	u64 wwpn[NUM_FC_PORTS];	/* wwpn of AFU ports */
+	struct hwq *hwq;
+	struct sisl_host_map __iomem *hmap;
+	__be64 __iomem *fc_port_regs;
+	u64 wwpn[MAX_FC_PORTS];	/* wwpn of AFU ports */
 	int i = 0, num_ports = 0;
 	int rc = 0;
 	u64 reg;
@@ -1448,16 +1626,18 @@ static int init_global(struct cxlflash_cfg *cfg)
 		goto out;
 	}
 
-	dev_dbg(dev, "%s: wwpn0=%016llx wwpn1=%016llx\n",
-		__func__, wwpn[0], wwpn[1]);
+	/* Set up RRQ and SQ in HWQ for master issued cmds */
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
+		hmap = hwq->host_map;
 
-	/* Set up RRQ and SQ in AFU for master issued cmds */
-	writeq_be((u64) afu->hrrq_start, &afu->host_map->rrq_start);
-	writeq_be((u64) afu->hrrq_end, &afu->host_map->rrq_end);
+		writeq_be((u64) hwq->hrrq_start, &hmap->rrq_start);
+		writeq_be((u64) hwq->hrrq_end, &hmap->rrq_end);
 
-	if (afu_is_sq_cmd_mode(afu)) {
-		writeq_be((u64)afu->hsq_start, &afu->host_map->sq_start);
-		writeq_be((u64)afu->hsq_end, &afu->host_map->sq_end);
+		if (afu_is_sq_cmd_mode(afu)) {
+			writeq_be((u64)hwq->hsq_start, &hmap->sq_start);
+			writeq_be((u64)hwq->hsq_end, &hmap->sq_end);
+		}
 	}
 
 	/* AFU configuration */
@@ -1473,26 +1653,25 @@ static int init_global(struct cxlflash_cfg *cfg)
 	if (afu->internal_lun) {
 		/* Only use port 0 */
 		writeq_be(PORT0, &afu->afu_map->global.regs.afu_port_sel);
-		num_ports = NUM_FC_PORTS - 1;
+		num_ports = 0;
 	} else {
-		writeq_be(BOTH_PORTS, &afu->afu_map->global.regs.afu_port_sel);
-		num_ports = NUM_FC_PORTS;
+		writeq_be(PORT_MASK(cfg->num_fc_ports),
+			  &afu->afu_map->global.regs.afu_port_sel);
+		num_ports = cfg->num_fc_ports;
 	}
 
 	for (i = 0; i < num_ports; i++) {
+		fc_port_regs = get_fc_port_regs(cfg, i);
+
 		/* Unmask all errors (but they are still masked at AFU) */
-		writeq_be(0, &afu->afu_map->global.fc_regs[i][FC_ERRMSK / 8]);
+		writeq_be(0, &fc_port_regs[FC_ERRMSK / 8]);
 		/* Clear CRC error cnt & set a threshold */
-		(void)readq_be(&afu->afu_map->global.
-			       fc_regs[i][FC_CNT_CRCERR / 8]);
-		writeq_be(MC_CRC_THRESH, &afu->afu_map->global.fc_regs[i]
-			  [FC_CRC_THRESH / 8]);
+		(void)readq_be(&fc_port_regs[FC_CNT_CRCERR / 8]);
+		writeq_be(MC_CRC_THRESH, &fc_port_regs[FC_CRC_THRESH / 8]);
 
 		/* Set WWPNs. If already programmed, wwpn[i] is 0 */
 		if (wwpn[i] != 0)
-			afu_set_wwpn(afu, i,
-				     &afu->afu_map->global.fc_regs[i][0],
-				     wwpn[i]);
+			afu_set_wwpn(afu, i, &fc_port_regs[0], wwpn[i]);
 		/* Programming WWPN back to back causes additional
 		 * offline/online transitions and a PLOGI
 		 */
@@ -1502,11 +1681,15 @@ static int init_global(struct cxlflash_cfg *cfg)
 	/* Set up master's own CTX_CAP to allow real mode, host translation */
 	/* tables, afu cmds and read/write GSCSI cmds. */
 	/* First, unlock ctx_cap write by reading mbox */
-	(void)readq_be(&afu->ctrl_map->mbox_r);	/* unlock ctx_cap */
-	writeq_be((SISL_CTX_CAP_REAL_MODE | SISL_CTX_CAP_HOST_XLATE |
-		   SISL_CTX_CAP_READ_CMD | SISL_CTX_CAP_WRITE_CMD |
-		   SISL_CTX_CAP_AFU_CMD | SISL_CTX_CAP_GSCSI_CMD),
-		  &afu->ctrl_map->ctx_cap);
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
+
+		(void)readq_be(&hwq->ctrl_map->mbox_r);	/* unlock ctx_cap */
+		writeq_be((SISL_CTX_CAP_REAL_MODE | SISL_CTX_CAP_HOST_XLATE |
+			SISL_CTX_CAP_READ_CMD | SISL_CTX_CAP_WRITE_CMD |
+			SISL_CTX_CAP_AFU_CMD | SISL_CTX_CAP_GSCSI_CMD),
+			&hwq->ctrl_map->ctx_cap);
+	}
 	/* Initialize heartbeat */
 	afu->hb = readq_be(&afu->afu_map->global.regs.afu_hb);
 out:
@@ -1521,28 +1704,42 @@ static int start_afu(struct cxlflash_cfg *cfg)
 {
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq;
 	int rc = 0;
+	int i;
 
 	init_pcr(cfg);
 
-	/* After an AFU reset, RRQ entries are stale, clear them */
-	memset(&afu->rrq_entry, 0, sizeof(afu->rrq_entry));
+	/* Initialize each HWQ */
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
 
-	/* Initialize RRQ pointers */
-	afu->hrrq_start = &afu->rrq_entry[0];
-	afu->hrrq_end = &afu->rrq_entry[NUM_RRQ_ENTRY - 1];
-	afu->hrrq_curr = afu->hrrq_start;
-	afu->toggle = 1;
+		/* After an AFU reset, RRQ entries are stale, clear them */
+		memset(&hwq->rrq_entry, 0, sizeof(hwq->rrq_entry));
 
-	/* Initialize SQ */
-	if (afu_is_sq_cmd_mode(afu)) {
-		memset(&afu->sq, 0, sizeof(afu->sq));
-		afu->hsq_start = &afu->sq[0];
-		afu->hsq_end = &afu->sq[NUM_SQ_ENTRY - 1];
-		afu->hsq_curr = afu->hsq_start;
+		/* Initialize RRQ pointers */
+		hwq->hrrq_start = &hwq->rrq_entry[0];
+		hwq->hrrq_end = &hwq->rrq_entry[NUM_RRQ_ENTRY - 1];
+		hwq->hrrq_curr = hwq->hrrq_start;
+		hwq->toggle = 1;
+		spin_lock_init(&hwq->hrrq_slock);
+
+		/* Initialize SQ */
+		if (afu_is_sq_cmd_mode(afu)) {
+			memset(&hwq->sq, 0, sizeof(hwq->sq));
+			hwq->hsq_start = &hwq->sq[0];
+			hwq->hsq_end = &hwq->sq[NUM_SQ_ENTRY - 1];
+			hwq->hsq_curr = hwq->hsq_start;
+
+			spin_lock_init(&hwq->hsq_slock);
+			atomic_set(&hwq->hsq_credits, NUM_SQ_ENTRY - 1);
+		}
+
+		/* Initialize IRQ poll */
+		if (afu_is_irqpoll_enabled(afu))
+			irq_poll_init(&hwq->irqpoll, afu->irqpoll_weight,
+				      cxlflash_irqpoll);
 
-		spin_lock_init(&afu->hsq_slock);
-		atomic_set(&afu->hsq_credits, NUM_SQ_ENTRY - 1);
 	}
 
 	rc = init_global(cfg);
@@ -1554,18 +1751,21 @@ static int start_afu(struct cxlflash_cfg *cfg)
 /**
  * init_intr() - setup interrupt handlers for the master context
  * @cfg:	Internal structure associated with the host.
+ * @hwq:	Hardware queue to initialize.
  *
  * Return: 0 on success, -errno on failure
  */
 static enum undo_level init_intr(struct cxlflash_cfg *cfg,
-				 struct cxl_context *ctx)
+				 struct hwq *hwq)
 {
-	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	struct cxl_context *ctx = hwq->ctx;
 	int rc = 0;
 	enum undo_level level = UNDO_NOOP;
+	bool is_primary_hwq = (hwq->index == PRIMARY_HWQ);
+	int num_irqs = is_primary_hwq ? 3 : 2;
 
-	rc = cxl_allocate_afu_irqs(ctx, 3);
+	rc = cxl_allocate_afu_irqs(ctx, num_irqs);
 	if (unlikely(rc)) {
 		dev_err(dev, "%s: allocate_afu_irqs failed rc=%d\n",
 			__func__, rc);
@@ -1573,7 +1773,7 @@ static enum undo_level init_intr(struct cxlflash_cfg *cfg,
 		goto out;
 	}
 
-	rc = cxl_map_afu_irq(ctx, 1, cxlflash_sync_err_irq, afu,
+	rc = cxl_map_afu_irq(ctx, 1, cxlflash_sync_err_irq, hwq,
 			     "SISL_MSI_SYNC_ERROR");
 	if (unlikely(rc <= 0)) {
 		dev_err(dev, "%s: SISL_MSI_SYNC_ERROR map failed\n", __func__);
@@ -1581,7 +1781,7 @@ static enum undo_level init_intr(struct cxlflash_cfg *cfg,
 		goto out;
 	}
 
-	rc = cxl_map_afu_irq(ctx, 2, cxlflash_rrq_irq, afu,
+	rc = cxl_map_afu_irq(ctx, 2, cxlflash_rrq_irq, hwq,
 			     "SISL_MSI_RRQ_UPDATED");
 	if (unlikely(rc <= 0)) {
 		dev_err(dev, "%s: SISL_MSI_RRQ_UPDATED map failed\n", __func__);
@@ -1589,7 +1789,11 @@ static enum undo_level init_intr(struct cxlflash_cfg *cfg,
 		goto out;
 	}
 
-	rc = cxl_map_afu_irq(ctx, 3, cxlflash_async_err_irq, afu,
+	/* SISL_MSI_ASYNC_ERROR is setup only for the primary HWQ */
+	if (!is_primary_hwq)
+		goto out;
+
+	rc = cxl_map_afu_irq(ctx, 3, cxlflash_async_err_irq, hwq,
 			     "SISL_MSI_ASYNC_ERROR");
 	if (unlikely(rc <= 0)) {
 		dev_err(dev, "%s: SISL_MSI_ASYNC_ERROR map failed\n", __func__);
@@ -1603,55 +1807,106 @@ out:
 /**
  * init_mc() - create and register as the master context
  * @cfg:	Internal structure associated with the host.
+ * index:	HWQ Index of the master context.
  *
  * Return: 0 on success, -errno on failure
  */
-static int init_mc(struct cxlflash_cfg *cfg)
+static int init_mc(struct cxlflash_cfg *cfg, u32 index)
 {
 	struct cxl_context *ctx;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq = get_hwq(cfg->afu, index);
 	int rc = 0;
 	enum undo_level level;
 
-	ctx = cxl_get_context(cfg->dev);
+	hwq->afu = cfg->afu;
+	hwq->index = index;
+
+	if (index == PRIMARY_HWQ)
+		ctx = cxl_get_context(cfg->dev);
+	else
+		ctx = cxl_dev_context_init(cfg->dev);
 	if (unlikely(!ctx)) {
 		rc = -ENOMEM;
-		goto ret;
+		goto err1;
 	}
-	cfg->mcctx = ctx;
+
+	WARN_ON(hwq->ctx);
+	hwq->ctx = ctx;
 
 	/* Set it up as a master with the CXL */
 	cxl_set_master(ctx);
 
-	/* During initialization reset the AFU to start from a clean slate */
-	rc = cxl_afu_reset(cfg->mcctx);
-	if (unlikely(rc)) {
-		dev_err(dev, "%s: AFU reset failed rc=%d\n", __func__, rc);
-		goto ret;
+	/* Reset AFU when initializing primary context */
+	if (index == PRIMARY_HWQ) {
+		rc = cxl_afu_reset(ctx);
+		if (unlikely(rc)) {
+			dev_err(dev, "%s: AFU reset failed rc=%d\n",
+				      __func__, rc);
+			goto err1;
+		}
 	}
 
-	level = init_intr(cfg, ctx);
+	level = init_intr(cfg, hwq);
 	if (unlikely(level)) {
 		dev_err(dev, "%s: interrupt init failed rc=%d\n", __func__, rc);
-		goto out;
+		goto err2;
 	}
 
 	/* This performs the equivalent of the CXL_IOCTL_START_WORK.
 	 * The CXL_IOCTL_GET_PROCESS_ELEMENT is implicit in the process
 	 * element (pe) that is embedded in the context (ctx)
 	 */
-	rc = start_context(cfg);
+	rc = start_context(cfg, index);
 	if (unlikely(rc)) {
 		dev_err(dev, "%s: start context failed rc=%d\n", __func__, rc);
 		level = UNMAP_THREE;
-		goto out;
+		goto err2;
 	}
-ret:
+
+out:
 	dev_dbg(dev, "%s: returning rc=%d\n", __func__, rc);
 	return rc;
-out:
-	term_intr(cfg, level);
-	goto ret;
+err2:
+	term_intr(cfg, level, index);
+	if (index != PRIMARY_HWQ)
+		cxl_release_context(ctx);
+err1:
+	hwq->ctx = NULL;
+	goto out;
+}
+
+/**
+ * get_num_afu_ports() - determines and configures the number of AFU ports
+ * @cfg:	Internal structure associated with the host.
+ *
+ * This routine determines the number of AFU ports by converting the global
+ * port selection mask. The converted value is only valid following an AFU
+ * reset (explicit or power-on). This routine must be invoked shortly after
+ * mapping as other routines are dependent on the number of ports during the
+ * initialization sequence.
+ *
+ * To support legacy AFUs that might not have reflected an initial global
+ * port mask (value read is 0), default to the number of ports originally
+ * supported by the cxlflash driver (2) before hardware with other port
+ * offerings was introduced.
+ */
+static void get_num_afu_ports(struct cxlflash_cfg *cfg)
+{
+	struct afu *afu = cfg->afu;
+	struct device *dev = &cfg->dev->dev;
+	u64 port_mask;
+	int num_fc_ports = LEGACY_FC_PORTS;
+
+	port_mask = readq_be(&afu->afu_map->global.regs.afu_port_sel);
+	if (port_mask != 0ULL)
+		num_fc_ports = min(ilog2(port_mask) + 1, MAX_FC_PORTS);
+
+	dev_dbg(dev, "%s: port_mask=%016llx num_fc_ports=%d\n",
+		__func__, port_mask, num_fc_ports);
+
+	cfg->num_fc_ports = num_fc_ports;
+	cfg->host->max_channel = PORTNUM2CHAN(num_fc_ports);
 }
 
 /**
@@ -1669,18 +1924,24 @@ static int init_afu(struct cxlflash_cfg *cfg)
 	int rc = 0;
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	struct hwq *hwq;
+	int i;
 
 	cxl_perst_reloads_same_image(cfg->cxl_afu, true);
 
-	rc = init_mc(cfg);
-	if (rc) {
-		dev_err(dev, "%s: init_mc failed rc=%d\n",
-			__func__, rc);
-		goto out;
+	afu->num_hwqs = afu->desired_hwqs;
+	for (i = 0; i < afu->num_hwqs; i++) {
+		rc = init_mc(cfg, i);
+		if (rc) {
+			dev_err(dev, "%s: init_mc failed rc=%d index=%d\n",
+				__func__, rc, i);
+			goto err1;
+		}
 	}
 
-	/* Map the entire MMIO space of the AFU */
-	afu->afu_map = cxl_psa_map(cfg->mcctx);
+	/* Map the entire MMIO space of the AFU using the first context */
+	hwq = get_hwq(afu, PRIMARY_HWQ);
+	afu->afu_map = cxl_psa_map(hwq->ctx);
 	if (!afu->afu_map) {
 		dev_err(dev, "%s: cxl_psa_map failed\n", __func__);
 		rc = -ENOMEM;
@@ -1711,6 +1972,8 @@ static int init_afu(struct cxlflash_cfg *cfg)
 	dev_dbg(dev, "%s: afu_ver=%s interface_ver=%016llx\n", __func__,
 		afu->version, afu->interface_version);
 
+	get_num_afu_ports(cfg);
+
 	rc = start_afu(cfg);
 	if (rc) {
 		dev_err(dev, "%s: start_afu failed, rc=%d\n", __func__, rc);
@@ -1718,8 +1981,12 @@ static int init_afu(struct cxlflash_cfg *cfg)
 	}
 
 	afu_err_intr_init(cfg->afu);
-	spin_lock_init(&afu->rrin_slock);
-	afu->room = readq_be(&afu->host_map->cmd_room);
+	for (i = 0; i < afu->num_hwqs; i++) {
+		hwq = get_hwq(afu, i);
+
+		spin_lock_init(&hwq->rrin_slock);
+		hwq->room = readq_be(&hwq->host_map->cmd_room);
+	}
 
 	/* Restore the LUN mappings */
 	cxlflash_restore_luntable(cfg);
@@ -1728,8 +1995,10 @@ out:
 	return rc;
 
 err1:
-	term_intr(cfg, UNMAP_THREE);
-	term_mc(cfg);
+	for (i = afu->num_hwqs - 1; i >= 0; i--) {
+		term_intr(cfg, UNMAP_THREE, i);
+		term_mc(cfg, i);
+	}
 	goto out;
 }
 
@@ -1761,6 +2030,7 @@ int cxlflash_afu_sync(struct afu *afu, ctx_hndl_t ctx_hndl_u,
 	struct cxlflash_cfg *cfg = afu->parent;
 	struct device *dev = &cfg->dev->dev;
 	struct afu_cmd *cmd = NULL;
+	struct hwq *hwq = get_hwq(afu, PRIMARY_HWQ);
 	char *buf = NULL;
 	int rc = 0;
 	static DEFINE_MUTEX(sync_active);
@@ -1783,11 +2053,12 @@ int cxlflash_afu_sync(struct afu *afu, ctx_hndl_t ctx_hndl_u,
 	cmd = (struct afu_cmd *)PTR_ALIGN(buf, __alignof__(*cmd));
 	init_completion(&cmd->cevent);
 	cmd->parent = afu;
+	cmd->hwq_index = hwq->index;
 
 	dev_dbg(dev, "%s: afu=%p cmd=%p %d\n", __func__, afu, cmd, ctx_hndl_u);
 
 	cmd->rcb.req_flags = SISL_REQ_FLAGS_AFU_CMD;
-	cmd->rcb.ctx_id = afu->ctx_hndl;
+	cmd->rcb.ctx_id = hwq->ctx_hndl;
 	cmd->rcb.msi = SISL_MSI_RRQ_UPDATED;
 	cmd->rcb.timeout = MC_AFU_SYNC_TIMEOUT;
 
@@ -1971,22 +2242,30 @@ static int cxlflash_change_queue_depth(struct scsi_device *sdev, int qdepth)
 /**
  * cxlflash_show_port_status() - queries and presents the current port status
  * @port:	Desired port for status reporting.
- * @afu:	AFU owning the specified port.
+ * @cfg:	Internal structure associated with the host.
  * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
  *
- * Return: The size of the ASCII string returned in @buf.
+ * Return: The size of the ASCII string returned in @buf or -EINVAL.
  */
-static ssize_t cxlflash_show_port_status(u32 port, struct afu *afu, char *buf)
+static ssize_t cxlflash_show_port_status(u32 port,
+					 struct cxlflash_cfg *cfg,
+					 char *buf)
 {
+	struct device *dev = &cfg->dev->dev;
 	char *disp_status;
 	u64 status;
-	__be64 __iomem *fc_regs;
+	__be64 __iomem *fc_port_regs;
 
-	if (port >= NUM_FC_PORTS)
-		return 0;
+	WARN_ON(port >= MAX_FC_PORTS);
+
+	if (port >= cfg->num_fc_ports) {
+		dev_info(dev, "%s: Port %d not supported on this card.\n",
+			__func__, port);
+		return -EINVAL;
+	}
 
-	fc_regs = &afu->afu_map->global.fc_regs[port][0];
-	status = readq_be(&fc_regs[FC_MTIP_STATUS / 8]);
+	fc_port_regs = get_fc_port_regs(cfg, port);
+	status = readq_be(&fc_port_regs[FC_MTIP_STATUS / 8]);
 	status &= FC_MTIP_STATUS_MASK;
 
 	if (status == FC_MTIP_STATUS_ONLINE)
@@ -2012,9 +2291,8 @@ static ssize_t port0_show(struct device *dev,
 			  char *buf)
 {
 	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
-	struct afu *afu = cfg->afu;
 
-	return cxlflash_show_port_status(0, afu, buf);
+	return cxlflash_show_port_status(0, cfg, buf);
 }
 
 /**
@@ -2030,9 +2308,42 @@ static ssize_t port1_show(struct device *dev,
 			  char *buf)
 {
 	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
-	struct afu *afu = cfg->afu;
 
-	return cxlflash_show_port_status(1, afu, buf);
+	return cxlflash_show_port_status(1, cfg, buf);
+}
+
+/**
+ * port2_show() - queries and presents the current status of port 2
+ * @dev:	Generic device associated with the host owning the port.
+ * @attr:	Device attribute representing the port.
+ * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t port2_show(struct device *dev,
+			  struct device_attribute *attr,
+			  char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+
+	return cxlflash_show_port_status(2, cfg, buf);
+}
+
+/**
+ * port3_show() - queries and presents the current status of port 3
+ * @dev:	Generic device associated with the host owning the port.
+ * @attr:	Device attribute representing the port.
+ * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t port3_show(struct device *dev,
+			  struct device_attribute *attr,
+			  char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+
+	return cxlflash_show_port_status(3, cfg, buf);
 }
 
 /**
@@ -2090,12 +2401,13 @@ static ssize_t lun_mode_store(struct device *dev,
 
 		/*
 		 * When configured for internal LUN, there is only one channel,
-		 * channel number 0, else there will be 2 (default).
+		 * channel number 0, else there will be one less than the number
+		 * of fc ports for this card.
 		 */
 		if (afu->internal_lun)
 			shost->max_channel = 0;
 		else
-			shost->max_channel = NUM_FC_PORTS - 1;
+			shost->max_channel = PORTNUM2CHAN(cfg->num_fc_ports);
 
 		afu_reset(cfg);
 		scsi_scan_host(cfg->host);
@@ -2121,27 +2433,34 @@ static ssize_t ioctl_version_show(struct device *dev,
 /**
  * cxlflash_show_port_lun_table() - queries and presents the port LUN table
  * @port:	Desired port for status reporting.
- * @afu:	AFU owning the specified port.
+ * @cfg:	Internal structure associated with the host.
  * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
  *
- * Return: The size of the ASCII string returned in @buf.
+ * Return: The size of the ASCII string returned in @buf or -EINVAL.
  */
 static ssize_t cxlflash_show_port_lun_table(u32 port,
-					    struct afu *afu,
+					    struct cxlflash_cfg *cfg,
 					    char *buf)
 {
+	struct device *dev = &cfg->dev->dev;
+	__be64 __iomem *fc_port_luns;
 	int i;
 	ssize_t bytes = 0;
-	__be64 __iomem *fc_port;
 
-	if (port >= NUM_FC_PORTS)
-		return 0;
+	WARN_ON(port >= MAX_FC_PORTS);
 
-	fc_port = &afu->afu_map->global.fc_port[port][0];
+	if (port >= cfg->num_fc_ports) {
+		dev_info(dev, "%s: Port %d not supported on this card.\n",
+			__func__, port);
+		return -EINVAL;
+	}
+
+	fc_port_luns = get_fc_port_luns(cfg, port);
 
 	for (i = 0; i < CXLFLASH_NUM_VLUNS; i++)
 		bytes += scnprintf(buf + bytes, PAGE_SIZE - bytes,
-				   "%03d: %016llx\n", i, readq_be(&fc_port[i]));
+				   "%03d: %016llx\n",
+				   i, readq_be(&fc_port_luns[i]));
 	return bytes;
 }
 
@@ -2158,9 +2477,8 @@ static ssize_t port0_lun_table_show(struct device *dev,
 				    char *buf)
 {
 	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
-	struct afu *afu = cfg->afu;
 
-	return cxlflash_show_port_lun_table(0, afu, buf);
+	return cxlflash_show_port_lun_table(0, cfg, buf);
 }
 
 /**
@@ -2176,9 +2494,272 @@ static ssize_t port1_lun_table_show(struct device *dev,
 				    char *buf)
 {
 	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+
+	return cxlflash_show_port_lun_table(1, cfg, buf);
+}
+
+/**
+ * port2_lun_table_show() - presents the current LUN table of port 2
+ * @dev:	Generic device associated with the host owning the port.
+ * @attr:	Device attribute representing the port.
+ * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t port2_lun_table_show(struct device *dev,
+				    struct device_attribute *attr,
+				    char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+
+	return cxlflash_show_port_lun_table(2, cfg, buf);
+}
+
+/**
+ * port3_lun_table_show() - presents the current LUN table of port 3
+ * @dev:	Generic device associated with the host owning the port.
+ * @attr:	Device attribute representing the port.
+ * @buf:	Buffer of length PAGE_SIZE to report back port status in ASCII.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t port3_lun_table_show(struct device *dev,
+				    struct device_attribute *attr,
+				    char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+
+	return cxlflash_show_port_lun_table(3, cfg, buf);
+}
+
+/**
+ * irqpoll_weight_show() - presents the current IRQ poll weight for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the IRQ poll weight.
+ * @buf:	Buffer of length PAGE_SIZE to report back the current IRQ poll
+ *		weight in ASCII.
+ *
+ * An IRQ poll weight of 0 indicates polling is disabled.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t irqpoll_weight_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+	struct afu *afu = cfg->afu;
+
+	return scnprintf(buf, PAGE_SIZE, "%u\n", afu->irqpoll_weight);
+}
+
+/**
+ * irqpoll_weight_store() - sets the current IRQ poll weight for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the IRQ poll weight.
+ * @buf:	Buffer of length PAGE_SIZE containing the desired IRQ poll
+ *		weight in ASCII.
+ * @count:	Length of data resizing in @buf.
+ *
+ * An IRQ poll weight of 0 indicates polling is disabled.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t irqpoll_weight_store(struct device *dev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t count)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+	struct device *cfgdev = &cfg->dev->dev;
+	struct afu *afu = cfg->afu;
+	struct hwq *hwq;
+	u32 weight;
+	int rc, i;
+
+	rc = kstrtouint(buf, 10, &weight);
+	if (rc)
+		return -EINVAL;
+
+	if (weight > 256) {
+		dev_info(cfgdev,
+			 "Invalid IRQ poll weight. It must be 256 or less.\n");
+		return -EINVAL;
+	}
+
+	if (weight == afu->irqpoll_weight) {
+		dev_info(cfgdev,
+			 "Current IRQ poll weight has the same weight.\n");
+		return -EINVAL;
+	}
+
+	if (afu_is_irqpoll_enabled(afu)) {
+		for (i = 0; i < afu->num_hwqs; i++) {
+			hwq = get_hwq(afu, i);
+
+			irq_poll_disable(&hwq->irqpoll);
+		}
+	}
+
+	afu->irqpoll_weight = weight;
+
+	if (weight > 0) {
+		for (i = 0; i < afu->num_hwqs; i++) {
+			hwq = get_hwq(afu, i);
+
+			irq_poll_init(&hwq->irqpoll, weight, cxlflash_irqpoll);
+		}
+	}
+
+	return count;
+}
+
+/**
+ * num_hwqs_show() - presents the number of hardware queues for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the number of hardware queues.
+ * @buf:	Buffer of length PAGE_SIZE to report back the number of hardware
+ *		queues in ASCII.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t num_hwqs_show(struct device *dev,
+			     struct device_attribute *attr, char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
 	struct afu *afu = cfg->afu;
 
-	return cxlflash_show_port_lun_table(1, afu, buf);
+	return scnprintf(buf, PAGE_SIZE, "%u\n", afu->num_hwqs);
+}
+
+/**
+ * num_hwqs_store() - sets the number of hardware queues for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the number of hardware queues.
+ * @buf:	Buffer of length PAGE_SIZE containing the number of hardware
+ *		queues in ASCII.
+ * @count:	Length of data resizing in @buf.
+ *
+ * n > 0: num_hwqs = n
+ * n = 0: num_hwqs = num_online_cpus()
+ * n < 0: num_online_cpus() / abs(n)
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t num_hwqs_store(struct device *dev,
+			      struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+	struct afu *afu = cfg->afu;
+	int rc;
+	int nhwqs, num_hwqs;
+
+	rc = kstrtoint(buf, 10, &nhwqs);
+	if (rc)
+		return -EINVAL;
+
+	if (nhwqs >= 1)
+		num_hwqs = nhwqs;
+	else if (nhwqs == 0)
+		num_hwqs = num_online_cpus();
+	else
+		num_hwqs = num_online_cpus() / abs(nhwqs);
+
+	afu->desired_hwqs = min(num_hwqs, CXLFLASH_MAX_HWQS);
+	WARN_ON_ONCE(afu->desired_hwqs == 0);
+
+retry:
+	switch (cfg->state) {
+	case STATE_NORMAL:
+		cfg->state = STATE_RESET;
+		drain_ioctls(cfg);
+		cxlflash_mark_contexts_error(cfg);
+		rc = afu_reset(cfg);
+		if (rc)
+			cfg->state = STATE_FAILTERM;
+		else
+			cfg->state = STATE_NORMAL;
+		wake_up_all(&cfg->reset_waitq);
+		break;
+	case STATE_RESET:
+		wait_event(cfg->reset_waitq, cfg->state != STATE_RESET);
+		if (cfg->state == STATE_NORMAL)
+			goto retry;
+	default:
+		/* Ideally should not happen */
+		dev_err(dev, "%s: Device is not ready, state=%d\n",
+			__func__, cfg->state);
+		break;
+	}
+
+	return count;
+}
+
+static const char *hwq_mode_name[MAX_HWQ_MODE] = { "rr", "tag", "cpu" };
+
+/**
+ * hwq_mode_show() - presents the HWQ steering mode for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the HWQ steering mode.
+ * @buf:	Buffer of length PAGE_SIZE to report back the HWQ steering mode
+ *		as a character string.
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t hwq_mode_show(struct device *dev,
+			     struct device_attribute *attr, char *buf)
+{
+	struct cxlflash_cfg *cfg = shost_priv(class_to_shost(dev));
+	struct afu *afu = cfg->afu;
+
+	return scnprintf(buf, PAGE_SIZE, "%s\n", hwq_mode_name[afu->hwq_mode]);
+}
+
+/**
+ * hwq_mode_store() - sets the HWQ steering mode for the host
+ * @dev:	Generic device associated with the host.
+ * @attr:	Device attribute representing the HWQ steering mode.
+ * @buf:	Buffer of length PAGE_SIZE containing the HWQ steering mode
+ *		as a character string.
+ * @count:	Length of data resizing in @buf.
+ *
+ * rr = Round-Robin
+ * tag = Block MQ Tagging
+ * cpu = CPU Affinity
+ *
+ * Return: The size of the ASCII string returned in @buf.
+ */
+static ssize_t hwq_mode_store(struct device *dev,
+			      struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	struct Scsi_Host *shost = class_to_shost(dev);
+	struct cxlflash_cfg *cfg = shost_priv(shost);
+	struct device *cfgdev = &cfg->dev->dev;
+	struct afu *afu = cfg->afu;
+	int i;
+	u32 mode = MAX_HWQ_MODE;
+
+	for (i = 0; i < MAX_HWQ_MODE; i++) {
+		if (!strncmp(hwq_mode_name[i], buf, strlen(hwq_mode_name[i]))) {
+			mode = i;
+			break;
+		}
+	}
+
+	if (mode >= MAX_HWQ_MODE) {
+		dev_info(cfgdev, "Invalid HWQ steering mode.\n");
+		return -EINVAL;
+	}
+
+	if ((mode == HWQ_MODE_TAG) && !shost_use_blk_mq(shost)) {
+		dev_info(cfgdev, "SCSI-MQ is not enabled, use a different "
+			 "HWQ steering mode.\n");
+		return -EINVAL;
+	}
+
+	afu->hwq_mode = mode;
+
+	return count;
 }
 
 /**
@@ -2203,18 +2784,32 @@ static ssize_t mode_show(struct device *dev,
  */
 static DEVICE_ATTR_RO(port0);
 static DEVICE_ATTR_RO(port1);
+static DEVICE_ATTR_RO(port2);
+static DEVICE_ATTR_RO(port3);
 static DEVICE_ATTR_RW(lun_mode);
 static DEVICE_ATTR_RO(ioctl_version);
 static DEVICE_ATTR_RO(port0_lun_table);
 static DEVICE_ATTR_RO(port1_lun_table);
+static DEVICE_ATTR_RO(port2_lun_table);
+static DEVICE_ATTR_RO(port3_lun_table);
+static DEVICE_ATTR_RW(irqpoll_weight);
+static DEVICE_ATTR_RW(num_hwqs);
+static DEVICE_ATTR_RW(hwq_mode);
 
 static struct device_attribute *cxlflash_host_attrs[] = {
 	&dev_attr_port0,
 	&dev_attr_port1,
+	&dev_attr_port2,
+	&dev_attr_port3,
 	&dev_attr_lun_mode,
 	&dev_attr_ioctl_version,
 	&dev_attr_port0_lun_table,
 	&dev_attr_port1_lun_table,
+	&dev_attr_port2_lun_table,
+	&dev_attr_port3_lun_table,
+	&dev_attr_irqpoll_weight,
+	&dev_attr_num_hwqs,
+	&dev_attr_hwq_mode,
 	NULL
 };
 
@@ -2292,6 +2887,7 @@ static void cxlflash_worker_thread(struct work_struct *work)
 						work_q);
 	struct afu *afu = cfg->afu;
 	struct device *dev = &cfg->dev->dev;
+	__be64 __iomem *fc_port_regs;
 	int port;
 	ulong lock_flags;
 
@@ -2312,8 +2908,8 @@ static void cxlflash_worker_thread(struct work_struct *work)
 					       lock_flags);
 
 			/* The reset can block... */
-			afu_link_reset(afu, port,
-				       &afu->afu_map->global.fc_regs[port][0]);
+			fc_port_regs = get_fc_port_regs(cfg, port);
+			afu_link_reset(afu, port, fc_port_regs);
 			spin_lock_irqsave(cfg->host->host_lock, lock_flags);
 		}
 
@@ -2331,6 +2927,15 @@ static void cxlflash_worker_thread(struct work_struct *work)
  * @pdev:	PCI device associated with the host.
  * @dev_id:	PCI device id associated with device.
  *
+ * The device will initially start out in a 'probing' state and
+ * transition to the 'normal' state at the end of a successful
+ * probe. Should an EEH event occur during probe, the notification
+ * thread (error_detected()) will wait until the probe handler
+ * is nearly complete. At that time, the device will be moved to
+ * a 'probed' state and the EEH thread woken up to drive the slot
+ * reset and recovery (device moves to 'normal' state). Meanwhile,
+ * the probe will be allowed to exit successfully.
+ *
  * Return: 0 on success, -errno on failure
  */
 static int cxlflash_probe(struct pci_dev *pdev,
@@ -2341,6 +2946,7 @@ static int cxlflash_probe(struct pci_dev *pdev,
 	struct device *dev = &pdev->dev;
 	struct dev_dependent_vals *ddv;
 	int rc = 0;
+	int k;
 
 	dev_dbg(&pdev->dev, "%s: Found CXLFLASH with IRQ: %d\n",
 		__func__, pdev->irq);
@@ -2357,7 +2963,6 @@ static int cxlflash_probe(struct pci_dev *pdev,
 
 	host->max_id = CXLFLASH_MAX_NUM_TARGETS_PER_BUS;
 	host->max_lun = CXLFLASH_MAX_NUM_LUNS_PER_TARGET;
-	host->max_channel = NUM_FC_PORTS - 1;
 	host->unique_id = host->host_no;
 	host->max_cmd_len = CXLFLASH_MAX_CDB_LEN;
 
@@ -2376,14 +2981,16 @@ static int cxlflash_probe(struct pci_dev *pdev,
 	cfg->cxl_fops = cxlflash_cxl_fops;
 
 	/*
-	 * The promoted LUNs move to the top of the LUN table. The rest stay
-	 * on the bottom half. The bottom half grows from the end
-	 * (index = 255), whereas the top half grows from the beginning
-	 * (index = 0).
+	 * Promoted LUNs move to the top of the LUN table. The rest stay on
+	 * the bottom half. The bottom half grows from the end (index = 255),
+	 * whereas the top half grows from the beginning (index = 0).
+	 *
+	 * Initialize the last LUN index for all possible ports.
 	 */
-	cfg->promote_lun_index  = 0;
-	cfg->last_lun_index[0] = CXLFLASH_NUM_VLUNS/2 - 1;
-	cfg->last_lun_index[1] = CXLFLASH_NUM_VLUNS/2 - 1;
+	cfg->promote_lun_index = 0;
+
+	for (k = 0; k < MAX_FC_PORTS; k++)
+		cfg->last_lun_index[k] = CXLFLASH_NUM_VLUNS/2 - 1;
 
 	cfg->dev_id = (struct pci_device_id *)dev_id;
 
@@ -2412,7 +3019,7 @@ static int cxlflash_probe(struct pci_dev *pdev,
 	cfg->init_state = INIT_STATE_PCI;
 
 	rc = init_afu(cfg);
-	if (rc) {
+	if (rc && !wq_has_sleeper(&cfg->reset_waitq)) {
 		dev_err(dev, "%s: init_afu failed rc=%d\n", __func__, rc);
 		goto out_remove;
 	}
@@ -2425,6 +3032,11 @@ static int cxlflash_probe(struct pci_dev *pdev,
 	}
 	cfg->init_state = INIT_STATE_SCSI;
 
+	if (wq_has_sleeper(&cfg->reset_waitq)) {
+		cfg->state = STATE_PROBED;
+		wake_up_all(&cfg->reset_waitq);
+	} else
+		cfg->state = STATE_NORMAL;
 out:
 	dev_dbg(dev, "%s: returning rc=%d\n", __func__, rc);
 	return rc;
@@ -2455,7 +3067,8 @@ static pci_ers_result_t cxlflash_pci_error_detected(struct pci_dev *pdev,
 
 	switch (state) {
 	case pci_channel_io_frozen:
-		wait_event(cfg->reset_waitq, cfg->state != STATE_RESET);
+		wait_event(cfg->reset_waitq, cfg->state != STATE_RESET &&
+					     cfg->state != STATE_PROBING);
 		if (cfg->state == STATE_FAILTERM)
 			return PCI_ERS_RESULT_DISCONNECT;
 
@@ -2546,6 +3159,7 @@ static struct pci_driver cxlflash_driver = {
  */
 static int __init init_cxlflash(void)
 {
+	check_sizes();
 	cxlflash_list_init();
 
 	return pci_register_driver(&cxlflash_driver);
diff --git a/drivers/scsi/cxlflash/main.h b/drivers/scsi/cxlflash/main.h
index 0be2261e6312..49657f1f409e 100644
--- a/drivers/scsi/cxlflash/main.h
+++ b/drivers/scsi/cxlflash/main.h
@@ -37,8 +37,6 @@
 
 #define CXLFLASH_PCI_ERROR_RECOVERY_TIMEOUT	(120 * HZ)
 
-#define NUM_FC_PORTS	CXLFLASH_NUM_FC_PORTS	/* ports per AFU */
-
 /* FC defines */
 #define FC_MTIP_CMDCONFIG 0x010
 #define FC_MTIP_STATUS 0x018
diff --git a/drivers/scsi/cxlflash/sislite.h b/drivers/scsi/cxlflash/sislite.h
index a6e48a893fef..a768360d2fa6 100644
--- a/drivers/scsi/cxlflash/sislite.h
+++ b/drivers/scsi/cxlflash/sislite.h
@@ -90,15 +90,15 @@ struct sisl_rc {
 #define SISL_AFU_RC_RHT_UNALIGNED         0x02U	/* should never happen */
 #define SISL_AFU_RC_RHT_OUT_OF_BOUNDS     0x03u	/* user error */
 #define SISL_AFU_RC_RHT_DMA_ERR           0x04u	/* see afu_extra
-						   may retry if afu_retry is off
-						   possible on master exit
+						 * may retry if afu_retry is off
+						 * possible on master exit
 						 */
 #define SISL_AFU_RC_RHT_RW_PERM           0x05u	/* no RW perms, user error */
 #define SISL_AFU_RC_LXT_UNALIGNED         0x12U	/* should never happen */
 #define SISL_AFU_RC_LXT_OUT_OF_BOUNDS     0x13u	/* user error */
 #define SISL_AFU_RC_LXT_DMA_ERR           0x14u	/* see afu_extra
-						   may retry if afu_retry is off
-						   possible on master exit
+						 * may retry if afu_retry is off
+						 * possible on master exit
 						 */
 #define SISL_AFU_RC_LXT_RW_PERM           0x15u	/* no RW perms, user error */
 
@@ -111,11 +111,11 @@ struct sisl_rc {
 	 */
 #define SISL_AFU_RC_NO_CHANNELS           0x20U	/* see afu_extra, may retry */
 #define SISL_AFU_RC_CAP_VIOLATION         0x21U	/* either user error or
-						   afu reset/master restart
+						 * afu reset/master restart
 						 */
 #define SISL_AFU_RC_OUT_OF_DATA_BUFS      0x30U	/* always retry */
 #define SISL_AFU_RC_DATA_DMA_ERR          0x31U	/* see afu_extra
-						   may retry if afu_retry is off
+						 * may retry if afu_retry is off
 						 */
 
 	u8 scsi_rc;		/* SCSI status byte, retry as appropriate */
@@ -149,8 +149,9 @@ struct sisl_rc {
 #define SISL_FC_RC_ABORTFAIL	0x59	/* pending abort completed w/fail */
 #define SISL_FC_RC_RESID	0x5A	/* ioasa underrun/overrun flags set */
 #define SISL_FC_RC_RESIDERR	0x5B	/* actual data len does not match SCSI
-					   reported len, possibly due to dropped
-					   frames */
+					 * reported len, possibly due to dropped
+					 * frames
+					 */
 #define SISL_FC_RC_TGTABORT	0x5C	/* command aborted by target */
 };
 
@@ -227,10 +228,10 @@ struct sisl_ioasa {
 
 /* per context host transport MMIO  */
 struct sisl_host_map {
-	__be64 endian_ctrl;     /* Per context Endian Control. The AFU will
-			      * operate on whatever the context is of the
-			      * host application.
-			      */
+	__be64 endian_ctrl;	/* Per context Endian Control. The AFU will
+				 * operate on whatever the context is of the
+				 * host application.
+				 */
 
 	__be64 intr_status;	/* this sends LISN# programmed in ctx_ctrl.
 				 * Only recovery in a PERM_ERR is a context
@@ -292,28 +293,54 @@ struct sisl_ctrl_map {
 /* single copy global regs */
 struct sisl_global_regs {
 	__be64 aintr_status;
-	/* In cxlflash, each FC port/link gets a byte of status */
-#define SISL_ASTATUS_FC0_OTHER	 0x8000ULL /* b48, other err,
-					      FC_ERRCAP[31:20] */
-#define SISL_ASTATUS_FC0_LOGO    0x4000ULL /* b49, target sent FLOGI/PLOGI/LOGO
-						   while logged in */
-#define SISL_ASTATUS_FC0_CRC_T   0x2000ULL /* b50, CRC threshold exceeded */
-#define SISL_ASTATUS_FC0_LOGI_R  0x1000ULL /* b51, login state machine timed out
-						   and retrying */
-#define SISL_ASTATUS_FC0_LOGI_F  0x0800ULL /* b52, login failed,
-					      FC_ERROR[19:0] */
-#define SISL_ASTATUS_FC0_LOGI_S  0x0400ULL /* b53, login succeeded */
-#define SISL_ASTATUS_FC0_LINK_DN 0x0200ULL /* b54, link online to offline */
-#define SISL_ASTATUS_FC0_LINK_UP 0x0100ULL /* b55, link offline to online */
-
-#define SISL_ASTATUS_FC1_OTHER   0x0080ULL /* b56 */
-#define SISL_ASTATUS_FC1_LOGO    0x0040ULL /* b57 */
-#define SISL_ASTATUS_FC1_CRC_T   0x0020ULL /* b58 */
-#define SISL_ASTATUS_FC1_LOGI_R  0x0010ULL /* b59 */
-#define SISL_ASTATUS_FC1_LOGI_F  0x0008ULL /* b60 */
-#define SISL_ASTATUS_FC1_LOGI_S  0x0004ULL /* b61 */
-#define SISL_ASTATUS_FC1_LINK_DN 0x0002ULL /* b62 */
-#define SISL_ASTATUS_FC1_LINK_UP 0x0001ULL /* b63 */
+	/*
+	 * In cxlflash, FC port/link are arranged in port pairs, each
+	 * gets a byte of status:
+	 *
+	 *	*_OTHER:	other err, FC_ERRCAP[31:20]
+	 *	*_LOGO:		target sent FLOGI/PLOGI/LOGO while logged in
+	 *	*_CRC_T:	CRC threshold exceeded
+	 *	*_LOGI_R:	login state machine timed out and retrying
+	 *	*_LOGI_F:	login failed, FC_ERROR[19:0]
+	 *	*_LOGI_S:	login succeeded
+	 *	*_LINK_DN:	link online to offline
+	 *	*_LINK_UP:	link offline to online
+	 */
+#define SISL_ASTATUS_FC2_OTHER	 0x80000000ULL /* b32 */
+#define SISL_ASTATUS_FC2_LOGO    0x40000000ULL /* b33 */
+#define SISL_ASTATUS_FC2_CRC_T   0x20000000ULL /* b34 */
+#define SISL_ASTATUS_FC2_LOGI_R  0x10000000ULL /* b35 */
+#define SISL_ASTATUS_FC2_LOGI_F  0x08000000ULL /* b36 */
+#define SISL_ASTATUS_FC2_LOGI_S  0x04000000ULL /* b37 */
+#define SISL_ASTATUS_FC2_LINK_DN 0x02000000ULL /* b38 */
+#define SISL_ASTATUS_FC2_LINK_UP 0x01000000ULL /* b39 */
+
+#define SISL_ASTATUS_FC3_OTHER   0x00800000ULL /* b40 */
+#define SISL_ASTATUS_FC3_LOGO    0x00400000ULL /* b41 */
+#define SISL_ASTATUS_FC3_CRC_T   0x00200000ULL /* b42 */
+#define SISL_ASTATUS_FC3_LOGI_R  0x00100000ULL /* b43 */
+#define SISL_ASTATUS_FC3_LOGI_F  0x00080000ULL /* b44 */
+#define SISL_ASTATUS_FC3_LOGI_S  0x00040000ULL /* b45 */
+#define SISL_ASTATUS_FC3_LINK_DN 0x00020000ULL /* b46 */
+#define SISL_ASTATUS_FC3_LINK_UP 0x00010000ULL /* b47 */
+
+#define SISL_ASTATUS_FC0_OTHER	 0x00008000ULL /* b48 */
+#define SISL_ASTATUS_FC0_LOGO    0x00004000ULL /* b49 */
+#define SISL_ASTATUS_FC0_CRC_T   0x00002000ULL /* b50 */
+#define SISL_ASTATUS_FC0_LOGI_R  0x00001000ULL /* b51 */
+#define SISL_ASTATUS_FC0_LOGI_F  0x00000800ULL /* b52 */
+#define SISL_ASTATUS_FC0_LOGI_S  0x00000400ULL /* b53 */
+#define SISL_ASTATUS_FC0_LINK_DN 0x00000200ULL /* b54 */
+#define SISL_ASTATUS_FC0_LINK_UP 0x00000100ULL /* b55 */
+
+#define SISL_ASTATUS_FC1_OTHER   0x00000080ULL /* b56 */
+#define SISL_ASTATUS_FC1_LOGO    0x00000040ULL /* b57 */
+#define SISL_ASTATUS_FC1_CRC_T   0x00000020ULL /* b58 */
+#define SISL_ASTATUS_FC1_LOGI_R  0x00000010ULL /* b59 */
+#define SISL_ASTATUS_FC1_LOGI_F  0x00000008ULL /* b60 */
+#define SISL_ASTATUS_FC1_LOGI_S  0x00000004ULL /* b61 */
+#define SISL_ASTATUS_FC1_LINK_DN 0x00000002ULL /* b62 */
+#define SISL_ASTATUS_FC1_LINK_UP 0x00000001ULL /* b63 */
 
 #define SISL_FC_INTERNAL_UNMASK	0x0000000300000000ULL	/* 1 means unmasked */
 #define SISL_FC_INTERNAL_MASK	~(SISL_FC_INTERNAL_UNMASK)
@@ -325,7 +352,7 @@ struct sisl_global_regs {
 #define SISL_STATUS_SHUTDOWN_ACTIVE	0x0000000000000010ULL
 #define SISL_STATUS_SHUTDOWN_COMPLETE	0x0000000000000020ULL
 
-#define SISL_ASTATUS_UNMASK	0xFFFFULL		/* 1 means unmasked */
+#define SISL_ASTATUS_UNMASK	0xFFFFFFFFULL		/* 1 means unmasked */
 #define SISL_ASTATUS_MASK	~(SISL_ASTATUS_UNMASK)	/* 1 means masked */
 
 	__be64 aintr_clear;
@@ -367,9 +394,18 @@ struct sisl_global_regs {
 #define SISL_INTVER_CAP_RESERVED_CMD_MODE_B	0x100000000000ULL
 };
 
-#define CXLFLASH_NUM_FC_PORTS   2
-#define CXLFLASH_MAX_CONTEXT  512	/* how many contexts per afu */
-#define CXLFLASH_NUM_VLUNS    512
+#define CXLFLASH_NUM_FC_PORTS_PER_BANK	2	/* fixed # of ports per bank */
+#define CXLFLASH_MAX_FC_BANKS		2	/* max # of banks supported */
+#define CXLFLASH_MAX_FC_PORTS	(CXLFLASH_NUM_FC_PORTS_PER_BANK *	\
+				 CXLFLASH_MAX_FC_BANKS)
+#define CXLFLASH_MAX_CONTEXT	512	/* number of contexts per AFU */
+#define CXLFLASH_NUM_VLUNS	512	/* number of vluns per AFU/port */
+#define CXLFLASH_NUM_REGS	512	/* number of registers per port */
+
+struct fc_port_bank {
+	__be64 fc_port_regs[CXLFLASH_NUM_FC_PORTS_PER_BANK][CXLFLASH_NUM_REGS];
+	__be64 fc_port_luns[CXLFLASH_NUM_FC_PORTS_PER_BANK][CXLFLASH_NUM_VLUNS];
+};
 
 struct sisl_global_map {
 	union {
@@ -379,11 +415,9 @@ struct sisl_global_map {
 
 	char page1[SIZE_4K];	/* page 1 */
 
-	/* pages 2 & 3 */
-	__be64 fc_regs[CXLFLASH_NUM_FC_PORTS][CXLFLASH_NUM_VLUNS];
+	struct fc_port_bank bank[CXLFLASH_MAX_FC_BANKS]; /* pages 2 - 9 */
 
-	/* pages 4 & 5 (lun tbl) */
-	__be64 fc_port[CXLFLASH_NUM_FC_PORTS][CXLFLASH_NUM_VLUNS];
+	/* pages 10 - 15 are reserved */
 
 };
 
@@ -402,7 +436,7 @@ struct sisl_global_map {
  *	|         64 KB Global          |
  *	|   Trusted Process accessible  |
  *	+-------------------------------+
-*/
+ */
 struct cxlflash_afu_map {
 	union {
 		struct sisl_host_map host;
@@ -478,7 +512,9 @@ struct sisl_rht_entry_f1 {
 
 #define PORT0  0x01U
 #define PORT1  0x02U
-#define BOTH_PORTS    (PORT0 | PORT1)
+#define PORT2  0x04U
+#define PORT3  0x08U
+#define PORT_MASK(_n)	((1 << (_n)) - 1)
 
 /* AFU Sync Mode byte */
 #define AFU_LW_SYNC 0x0U
diff --git a/drivers/scsi/cxlflash/superpipe.c b/drivers/scsi/cxlflash/superpipe.c
index b46fd2f45628..fe9f17a6268b 100644
--- a/drivers/scsi/cxlflash/superpipe.c
+++ b/drivers/scsi/cxlflash/superpipe.c
@@ -78,17 +78,18 @@ void cxlflash_free_errpage(void)
  * memory freed. This is accomplished by putting the contexts in error
  * state which will notify the user and let them 'drive' the tear down.
  * Meanwhile, this routine camps until all user contexts have been removed.
+ *
+ * Note that the main loop in this routine will always execute at least once
+ * to flush the reset_waitq.
  */
 void cxlflash_stop_term_user_contexts(struct cxlflash_cfg *cfg)
 {
 	struct device *dev = &cfg->dev->dev;
-	int i, found;
+	int i, found = true;
 
 	cxlflash_mark_contexts_error(cfg);
 
 	while (true) {
-		found = false;
-
 		for (i = 0; i < MAX_CONTEXT; i++)
 			if (cfg->ctx_tbl[i]) {
 				found = true;
@@ -102,6 +103,7 @@ void cxlflash_stop_term_user_contexts(struct cxlflash_cfg *cfg)
 			__func__);
 		wake_up_all(&cfg->reset_waitq);
 		ssleep(1);
+		found = false;
 	}
 }
 
@@ -252,6 +254,7 @@ static int afu_attach(struct cxlflash_cfg *cfg, struct ctx_info *ctxi)
 	struct afu *afu = cfg->afu;
 	struct sisl_ctrl_map __iomem *ctrl_map = ctxi->ctrl_map;
 	int rc = 0;
+	struct hwq *hwq = get_hwq(afu, PRIMARY_HWQ);
 	u64 val;
 
 	/* Unlock cap and restrict user to read/write cmds in translated mode */
@@ -268,7 +271,7 @@ static int afu_attach(struct cxlflash_cfg *cfg, struct ctx_info *ctxi)
 
 	/* Set up MMIO registers pointing to the RHT */
 	writeq_be((u64)ctxi->rht_start, &ctrl_map->rht_start);
-	val = SISL_RHT_CNT_ID((u64)MAX_RHT_PER_CONTEXT, (u64)(afu->ctx_hndl));
+	val = SISL_RHT_CNT_ID((u64)MAX_RHT_PER_CONTEXT, (u64)(hwq->ctx_hndl));
 	writeq_be(val, &ctrl_map->rht_cnt_id);
 out:
 	dev_dbg(dev, "%s: returning rc=%d\n", __func__, rc);
@@ -1624,6 +1627,7 @@ static int cxlflash_afu_recover(struct scsi_device *sdev,
 	struct afu *afu = cfg->afu;
 	struct ctx_info *ctxi = NULL;
 	struct mutex *mutex = &cfg->ctx_recovery_mutex;
+	struct hwq *hwq = get_hwq(afu, PRIMARY_HWQ);
 	u64 flags;
 	u64 ctxid = DECODE_CTXID(recover->context_id),
 	    rctxid = recover->context_id;
@@ -1694,7 +1698,7 @@ retry_recover:
 	}
 
 	/* Test if in error state */
-	reg = readq_be(&afu->ctrl_map->mbox_r);
+	reg = readq_be(&hwq->ctrl_map->mbox_r);
 	if (reg == -1) {
 		dev_dbg(dev, "%s: MMIO fail, wait for recovery.\n", __func__);
 
@@ -1933,7 +1937,7 @@ static int cxlflash_disk_direct_open(struct scsi_device *sdev, void *arg)
 	u64 lun_size = 0;
 	u64 last_lba = 0;
 	u64 rsrc_handle = -1;
-	u32 port = CHAN2PORT(sdev->channel);
+	u32 port = CHAN2PORTMASK(sdev->channel);
 
 	int rc = 0;
 
diff --git a/drivers/scsi/cxlflash/superpipe.h b/drivers/scsi/cxlflash/superpipe.h
index 9e62ff304e4b..0b5976829913 100644
--- a/drivers/scsi/cxlflash/superpipe.h
+++ b/drivers/scsi/cxlflash/superpipe.h
@@ -24,8 +24,8 @@ extern struct cxlflash_global global;
  */
 
 /* Chunk size parms: note sislite minimum chunk size is
-   0x10000 LBAs corresponding to a NMASK or 16.
-*/
+ * 0x10000 LBAs corresponding to a NMASK or 16.
+ */
 #define MC_CHUNK_SIZE     (1 << MC_RHT_NMASK)	/* in LBAs */
 
 #define CMD_TIMEOUT 30  /* 30 secs */
@@ -33,9 +33,6 @@ extern struct cxlflash_global global;
 
 #define MAX_SECTOR_UNIT  512 /* max_sector is in 512 byte multiples */
 
-#define CHAN2PORT(_x)	((_x) + 1)
-#define PORT2CHAN(_x)	((_x) - 1)
-
 enum lun_mode {
 	MODE_NONE = 0,
 	MODE_VIRTUAL,
@@ -59,7 +56,7 @@ struct glun_info {
 
 /* Local (per-adapter) lun_info structure */
 struct llun_info {
-	u64 lun_id[CXLFLASH_NUM_FC_PORTS]; /* from REPORT_LUNS */
+	u64 lun_id[MAX_FC_PORTS]; /* from REPORT_LUNS */
 	u32 lun_index;		/* Index in the LUN table */
 	u32 host_no;		/* host_no from Scsi_host */
 	u32 port_sel;		/* What port to use for this LUN */
@@ -92,7 +89,8 @@ enum ctx_ctrl {
 struct ctx_info {
 	struct sisl_ctrl_map __iomem *ctrl_map; /* initialized at startup */
 	struct sisl_rht_entry *rht_start; /* 1 page (req'd for alignment),
-					     alloc/free on attach/detach */
+					   * alloc/free on attach/detach
+					   */
 	u32 rht_out;		/* Number of checked out RHT entries */
 	u32 rht_perms;		/* User-defined permissions for RHT entries */
 	struct llun_info **rht_lun;       /* Mapping of RHT entries to LUNs */
@@ -120,34 +118,40 @@ struct cxlflash_global {
 	struct page *err_page; /* One page of all 0xF for error notification */
 };
 
-int cxlflash_vlun_resize(struct scsi_device *, struct dk_cxlflash_resize *);
-int _cxlflash_vlun_resize(struct scsi_device *, struct ctx_info *,
-			  struct dk_cxlflash_resize *);
+int cxlflash_vlun_resize(struct scsi_device *sdev,
+			 struct dk_cxlflash_resize *resize);
+int _cxlflash_vlun_resize(struct scsi_device *sdev, struct ctx_info *ctxi,
+			  struct dk_cxlflash_resize *resize);
 
-int cxlflash_disk_release(struct scsi_device *, struct dk_cxlflash_release *);
-int _cxlflash_disk_release(struct scsi_device *, struct ctx_info *,
-			   struct dk_cxlflash_release *);
+int cxlflash_disk_release(struct scsi_device *sdev,
+			  struct dk_cxlflash_release *release);
+int _cxlflash_disk_release(struct scsi_device *sdev, struct ctx_info *ctxi,
+			   struct dk_cxlflash_release *release);
 
-int cxlflash_disk_clone(struct scsi_device *, struct dk_cxlflash_clone *);
+int cxlflash_disk_clone(struct scsi_device *sdev,
+			struct dk_cxlflash_clone *clone);
 
-int cxlflash_disk_virtual_open(struct scsi_device *, void *);
+int cxlflash_disk_virtual_open(struct scsi_device *sdev, void *arg);
 
-int cxlflash_lun_attach(struct glun_info *, enum lun_mode, bool);
-void cxlflash_lun_detach(struct glun_info *);
+int cxlflash_lun_attach(struct glun_info *gli, enum lun_mode mode, bool locked);
+void cxlflash_lun_detach(struct glun_info *gli);
 
-struct ctx_info *get_context(struct cxlflash_cfg *, u64, void *, enum ctx_ctrl);
-void put_context(struct ctx_info *);
+struct ctx_info *get_context(struct cxlflash_cfg *cfg, u64 rctxit, void *arg,
+			     enum ctx_ctrl ctrl);
+void put_context(struct ctx_info *ctxi);
 
-struct sisl_rht_entry *get_rhte(struct ctx_info *, res_hndl_t,
-				struct llun_info *);
+struct sisl_rht_entry *get_rhte(struct ctx_info *ctxi, res_hndl_t rhndl,
+				struct llun_info *lli);
 
-struct sisl_rht_entry *rhte_checkout(struct ctx_info *, struct llun_info *);
-void rhte_checkin(struct ctx_info *, struct sisl_rht_entry *);
+struct sisl_rht_entry *rhte_checkout(struct ctx_info *ctxi,
+				     struct llun_info *lli);
+void rhte_checkin(struct ctx_info *ctxi, struct sisl_rht_entry *rhte);
 
-void cxlflash_ba_terminate(struct ba_lun *);
+void cxlflash_ba_terminate(struct ba_lun *ba_lun);
 
-int cxlflash_manage_lun(struct scsi_device *, struct dk_cxlflash_manage_lun *);
+int cxlflash_manage_lun(struct scsi_device *sdev,
+			struct dk_cxlflash_manage_lun *manage);
 
-int check_state(struct cxlflash_cfg *);
+int check_state(struct cxlflash_cfg *cfg);
 
 #endif /* ifndef _CXLFLASH_SUPERPIPE_H */
diff --git a/drivers/scsi/cxlflash/vlun.c b/drivers/scsi/cxlflash/vlun.c
index 7aa06ef229fd..90b5c19f81f0 100644
--- a/drivers/scsi/cxlflash/vlun.c
+++ b/drivers/scsi/cxlflash/vlun.c
@@ -819,11 +819,10 @@ int cxlflash_vlun_resize(struct scsi_device *sdev,
 void cxlflash_restore_luntable(struct cxlflash_cfg *cfg)
 {
 	struct llun_info *lli, *temp;
-	u32 chan;
 	u32 lind;
-	struct afu *afu = cfg->afu;
+	int k;
 	struct device *dev = &cfg->dev->dev;
-	struct sisl_global_map __iomem *agm = &afu->afu_map->global;
+	__be64 __iomem *fc_port_luns;
 
 	mutex_lock(&global.mutex);
 
@@ -832,33 +831,41 @@ void cxlflash_restore_luntable(struct cxlflash_cfg *cfg)
 			continue;
 
 		lind = lli->lun_index;
+		dev_dbg(dev, "%s: Virtual LUNs on slot %d:\n", __func__, lind);
 
-		if (lli->port_sel == BOTH_PORTS) {
-			writeq_be(lli->lun_id[0], &agm->fc_port[0][lind]);
-			writeq_be(lli->lun_id[1], &agm->fc_port[1][lind]);
-			dev_dbg(dev, "%s: Virtual LUN on slot %d  id0=%llx "
-				"id1=%llx\n", __func__, lind,
-				lli->lun_id[0], lli->lun_id[1]);
-		} else {
-			chan = PORT2CHAN(lli->port_sel);
-			writeq_be(lli->lun_id[chan], &agm->fc_port[chan][lind]);
-			dev_dbg(dev, "%s: Virtual LUN on slot %d chan=%d "
-				"id=%llx\n", __func__, lind, chan,
-				lli->lun_id[chan]);
-		}
+		for (k = 0; k < cfg->num_fc_ports; k++)
+			if (lli->port_sel & (1 << k)) {
+				fc_port_luns = get_fc_port_luns(cfg, k);
+				writeq_be(lli->lun_id[k], &fc_port_luns[lind]);
+				dev_dbg(dev, "\t%d=%llx\n", k, lli->lun_id[k]);
+			}
 	}
 
 	mutex_unlock(&global.mutex);
 }
 
 /**
+ * get_num_ports() - compute number of ports from port selection mask
+ * @psm:	Port selection mask.
+ *
+ * Return: Population count of port selection mask
+ */
+static inline u8 get_num_ports(u32 psm)
+{
+	static const u8 bits[16] = { 0, 1, 1, 2, 1, 2, 2, 3,
+				     1, 2, 2, 3, 2, 3, 3, 4 };
+
+	return bits[psm & 0xf];
+}
+
+/**
  * init_luntable() - write an entry in the LUN table
  * @cfg:	Internal structure associated with the host.
  * @lli:	Per adapter LUN information structure.
  *
- * On successful return, a LUN table entry is created.
- * At the top for LUNs visible on both ports.
- * At the bottom for LUNs visible only on one port.
+ * On successful return, a LUN table entry is created:
+ *	- at the top for LUNs visible on multiple ports.
+ *	- at the bottom for LUNs visible only on one port.
  *
  * Return: 0 on success, -errno on failure
  */
@@ -866,48 +873,68 @@ static int init_luntable(struct cxlflash_cfg *cfg, struct llun_info *lli)
 {
 	u32 chan;
 	u32 lind;
+	u32 nports;
 	int rc = 0;
-	struct afu *afu = cfg->afu;
+	int k;
 	struct device *dev = &cfg->dev->dev;
-	struct sisl_global_map __iomem *agm = &afu->afu_map->global;
+	__be64 __iomem *fc_port_luns;
 
 	mutex_lock(&global.mutex);
 
 	if (lli->in_table)
 		goto out;
 
-	if (lli->port_sel == BOTH_PORTS) {
+	nports = get_num_ports(lli->port_sel);
+	if (nports == 0 || nports > cfg->num_fc_ports) {
+		WARN(1, "Unsupported port configuration nports=%u", nports);
+		rc = -EIO;
+		goto out;
+	}
+
+	if (nports > 1) {
 		/*
-		 * If this LUN is visible from both ports, we will put
+		 * When LUN is visible from multiple ports, we will put
 		 * it in the top half of the LUN table.
 		 */
-		if ((cfg->promote_lun_index == cfg->last_lun_index[0]) ||
-		    (cfg->promote_lun_index == cfg->last_lun_index[1])) {
-			rc = -ENOSPC;
-			goto out;
+		for (k = 0; k < cfg->num_fc_ports; k++) {
+			if (!(lli->port_sel & (1 << k)))
+				continue;
+
+			if (cfg->promote_lun_index == cfg->last_lun_index[k]) {
+				rc = -ENOSPC;
+				goto out;
+			}
 		}
 
 		lind = lli->lun_index = cfg->promote_lun_index;
-		writeq_be(lli->lun_id[0], &agm->fc_port[0][lind]);
-		writeq_be(lli->lun_id[1], &agm->fc_port[1][lind]);
+		dev_dbg(dev, "%s: Virtual LUNs on slot %d:\n", __func__, lind);
+
+		for (k = 0; k < cfg->num_fc_ports; k++) {
+			if (!(lli->port_sel & (1 << k)))
+				continue;
+
+			fc_port_luns = get_fc_port_luns(cfg, k);
+			writeq_be(lli->lun_id[k], &fc_port_luns[lind]);
+			dev_dbg(dev, "\t%d=%llx\n", k, lli->lun_id[k]);
+		}
+
 		cfg->promote_lun_index++;
-		dev_dbg(dev, "%s: Virtual LUN on slot %d  id0=%llx id1=%llx\n",
-			__func__, lind, lli->lun_id[0], lli->lun_id[1]);
 	} else {
 		/*
-		 * If this LUN is visible only from one port, we will put
+		 * When LUN is visible only from one port, we will put
 		 * it in the bottom half of the LUN table.
 		 */
-		chan = PORT2CHAN(lli->port_sel);
+		chan = PORTMASK2CHAN(lli->port_sel);
 		if (cfg->promote_lun_index == cfg->last_lun_index[chan]) {
 			rc = -ENOSPC;
 			goto out;
 		}
 
 		lind = lli->lun_index = cfg->last_lun_index[chan];
-		writeq_be(lli->lun_id[chan], &agm->fc_port[chan][lind]);
+		fc_port_luns = get_fc_port_luns(cfg, chan);
+		writeq_be(lli->lun_id[chan], &fc_port_luns[lind]);
 		cfg->last_lun_index[chan]--;
-		dev_dbg(dev, "%s: Virtual LUN on slot %d  chan=%d id=%llx\n",
+		dev_dbg(dev, "%s: Virtual LUNs on slot %d:\n\t%d=%llx\n",
 			__func__, lind, chan, lli->lun_id[chan]);
 	}
 
@@ -1016,7 +1043,7 @@ int cxlflash_disk_virtual_open(struct scsi_device *sdev, void *arg)
 	virt->last_lba = last_lba;
 	virt->rsrc_handle = rsrc_handle;
 
-	if (lli->port_sel == BOTH_PORTS)
+	if (get_num_ports(lli->port_sel) > 1)
 		virt->hdr.return_flags |= DK_CXLFLASH_ALL_PORTS_ACTIVE;
 out:
 	if (likely(ctxi))
diff --git a/drivers/scsi/cxlflash/vlun.h b/drivers/scsi/cxlflash/vlun.h
index 8b29a74946e4..27a63a0155ce 100644
--- a/drivers/scsi/cxlflash/vlun.h
+++ b/drivers/scsi/cxlflash/vlun.h
@@ -47,7 +47,7 @@
  * not stored anywhere.
  *
  * The LXT table is re-allocated whenever it needs to cross into another group.
-*/
+ */
 #define LXT_GROUP_SIZE          8
 #define LXT_NUM_GROUPS(lxt_cnt) (((lxt_cnt) + 7)/8)	/* alloc'ed groups */
 #define LXT_LUNIDX_SHIFT  8	/* LXT entry, shift for LUN index */
diff --git a/drivers/scsi/esas2r/esas2r_log.c b/drivers/scsi/esas2r/esas2r_log.c
index a82030aa8577..65fdf22b0ba9 100644
--- a/drivers/scsi/esas2r/esas2r_log.c
+++ b/drivers/scsi/esas2r/esas2r_log.c
@@ -130,11 +130,6 @@ static int esas2r_log_master(const long level,
 
 		spin_lock_irqsave(&event_buffer_lock, flags);
 
-		if (buffer == NULL) {
-			spin_unlock_irqrestore(&event_buffer_lock, flags);
-			return -1;
-		}
-
 		memset(buffer, 0, buflen);
 
 		/*
diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c
index ab7bc1505e0b..90939f66bc0d 100644
--- a/drivers/scsi/fcoe/fcoe.c
+++ b/drivers/scsi/fcoe/fcoe.c
@@ -63,11 +63,11 @@ unsigned int fcoe_debug_logging;
 module_param_named(debug_logging, fcoe_debug_logging, int, S_IRUGO|S_IWUSR);
 MODULE_PARM_DESC(debug_logging, "a bit mask of logging levels");
 
-unsigned int fcoe_e_d_tov = 2 * 1000;
+static unsigned int fcoe_e_d_tov = 2 * 1000;
 module_param_named(e_d_tov, fcoe_e_d_tov, int, S_IRUGO|S_IWUSR);
 MODULE_PARM_DESC(e_d_tov, "E_D_TOV in ms, default 2000");
 
-unsigned int fcoe_r_a_tov = 2 * 2 * 1000;
+static unsigned int fcoe_r_a_tov = 2 * 2 * 1000;
 module_param_named(r_a_tov, fcoe_r_a_tov, int, S_IRUGO|S_IWUSR);
 MODULE_PARM_DESC(r_a_tov, "R_A_TOV in ms, default 4000");
 
diff --git a/drivers/scsi/fnic/fnic.h b/drivers/scsi/fnic/fnic.h
index 9e4b7709043e..67aab965c0f4 100644
--- a/drivers/scsi/fnic/fnic.h
+++ b/drivers/scsi/fnic/fnic.h
@@ -39,7 +39,7 @@
 
 #define DRV_NAME		"fnic"
 #define DRV_DESCRIPTION		"Cisco FCoE HBA Driver"
-#define DRV_VERSION		"1.6.0.21"
+#define DRV_VERSION		"1.6.0.34"
 #define PFX			DRV_NAME ": "
 #define DFX                     DRV_NAME "%d: "
 
@@ -217,7 +217,6 @@ struct fnic {
 	struct fcoe_ctlr ctlr;		/* FIP FCoE controller structure */
 	struct vnic_dev_bar bar0;
 
-	struct msix_entry msix_entry[FNIC_MSIX_INTR_MAX];
 	struct fnic_msix_entry msix[FNIC_MSIX_INTR_MAX];
 
 	struct vnic_stats *stats;
diff --git a/drivers/scsi/fnic/fnic_fcs.c b/drivers/scsi/fnic/fnic_fcs.c
index 3b7da66e2771..245dcd95e11f 100644
--- a/drivers/scsi/fnic/fnic_fcs.c
+++ b/drivers/scsi/fnic/fnic_fcs.c
@@ -342,8 +342,11 @@ static void fnic_fcoe_send_vlan_req(struct fnic *fnic)
 
 	fnic_fcoe_reset_vlans(fnic);
 	fnic->set_vlan(fnic, 0);
-	FNIC_FCS_DBG(KERN_INFO, fnic->lport->host,
-		  "Sending VLAN request...\n");
+
+	if (printk_ratelimit())
+		FNIC_FCS_DBG(KERN_INFO, fnic->lport->host,
+			  "Sending VLAN request...\n");
+
 	skb = dev_alloc_skb(sizeof(struct fip_vlan));
 	if (!skb)
 		return;
@@ -359,7 +362,7 @@ static void fnic_fcoe_send_vlan_req(struct fnic *fnic)
 
 	vlan->fip.fip_ver = FIP_VER_ENCAPS(FIP_VER);
 	vlan->fip.fip_op = htons(FIP_OP_VLAN);
-	vlan->fip.fip_subcode = FIP_SC_VL_NOTE;
+	vlan->fip.fip_subcode = FIP_SC_VL_REQ;
 	vlan->fip.fip_dl_len = htons(sizeof(vlan->desc) / FIP_BPW);
 
 	vlan->desc.mac.fd_desc.fip_dtype = FIP_DT_MAC;
@@ -1313,10 +1316,11 @@ void fnic_handle_fip_timer(struct fnic *fnic)
 
 	spin_lock_irqsave(&fnic->vlans_lock, flags);
 	if (list_empty(&fnic->vlans)) {
-		/* no vlans available, try again */
-		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
-			  "Start VLAN Discovery\n");
 		spin_unlock_irqrestore(&fnic->vlans_lock, flags);
+		/* no vlans available, try again */
+		if (printk_ratelimit())
+			FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+				  "Start VLAN Discovery\n");
 		fnic_event_enq(fnic, FNIC_EVT_START_VLAN_DISC);
 		return;
 	}
@@ -1332,10 +1336,11 @@ void fnic_handle_fip_timer(struct fnic *fnic)
 		spin_unlock_irqrestore(&fnic->vlans_lock, flags);
 		break;
 	case FIP_VLAN_FAILED:
-		/* if all vlans are in failed state, restart vlan disc */
-		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
-			  "Start VLAN Discovery\n");
 		spin_unlock_irqrestore(&fnic->vlans_lock, flags);
+		/* if all vlans are in failed state, restart vlan disc */
+		if (printk_ratelimit())
+			FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+				  "Start VLAN Discovery\n");
 		fnic_event_enq(fnic, FNIC_EVT_START_VLAN_DISC);
 		break;
 	case FIP_VLAN_SENT:
diff --git a/drivers/scsi/fnic/fnic_isr.c b/drivers/scsi/fnic/fnic_isr.c
index a0dd1b67a467..4e3a50202e8c 100644
--- a/drivers/scsi/fnic/fnic_isr.c
+++ b/drivers/scsi/fnic/fnic_isr.c
@@ -154,13 +154,13 @@ void fnic_free_intr(struct fnic *fnic)
 	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
 	case VNIC_DEV_INTR_MODE_INTX:
 	case VNIC_DEV_INTR_MODE_MSI:
-		free_irq(fnic->pdev->irq, fnic);
+		free_irq(pci_irq_vector(fnic->pdev, 0), fnic);
 		break;
 
 	case VNIC_DEV_INTR_MODE_MSIX:
 		for (i = 0; i < ARRAY_SIZE(fnic->msix); i++)
 			if (fnic->msix[i].requested)
-				free_irq(fnic->msix_entry[i].vector,
+				free_irq(pci_irq_vector(fnic->pdev, i),
 					 fnic->msix[i].devid);
 		break;
 
@@ -177,12 +177,12 @@ int fnic_request_intr(struct fnic *fnic)
 	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
 
 	case VNIC_DEV_INTR_MODE_INTX:
-		err = request_irq(fnic->pdev->irq, &fnic_isr_legacy,
-				  IRQF_SHARED, DRV_NAME, fnic);
+		err = request_irq(pci_irq_vector(fnic->pdev, 0),
+				&fnic_isr_legacy, IRQF_SHARED, DRV_NAME, fnic);
 		break;
 
 	case VNIC_DEV_INTR_MODE_MSI:
-		err = request_irq(fnic->pdev->irq, &fnic_isr_msi,
+		err = request_irq(pci_irq_vector(fnic->pdev, 0), &fnic_isr_msi,
 				  0, fnic->name, fnic);
 		break;
 
@@ -210,7 +210,7 @@ int fnic_request_intr(struct fnic *fnic)
 		fnic->msix[FNIC_MSIX_ERR_NOTIFY].devid = fnic;
 
 		for (i = 0; i < ARRAY_SIZE(fnic->msix); i++) {
-			err = request_irq(fnic->msix_entry[i].vector,
+			err = request_irq(pci_irq_vector(fnic->pdev, i),
 					  fnic->msix[i].isr, 0,
 					  fnic->msix[i].devname,
 					  fnic->msix[i].devid);
@@ -237,7 +237,6 @@ int fnic_set_intr_mode(struct fnic *fnic)
 	unsigned int n = ARRAY_SIZE(fnic->rq);
 	unsigned int m = ARRAY_SIZE(fnic->wq);
 	unsigned int o = ARRAY_SIZE(fnic->wq_copy);
-	unsigned int i;
 
 	/*
 	 * Set interrupt mode (INTx, MSI, MSI-X) depending
@@ -248,23 +247,20 @@ int fnic_set_intr_mode(struct fnic *fnic)
 	 * We need n RQs, m WQs, o Copy WQs, n+m+o CQs, and n+m+o+1 INTRs
 	 * (last INTR is used for WQ/RQ errors and notification area)
 	 */
-
-	BUG_ON(ARRAY_SIZE(fnic->msix_entry) < n + m + o + 1);
-	for (i = 0; i < n + m + o + 1; i++)
-		fnic->msix_entry[i].entry = i;
-
 	if (fnic->rq_count >= n &&
 	    fnic->raw_wq_count >= m &&
 	    fnic->wq_copy_count >= o &&
 	    fnic->cq_count >= n + m + o) {
-		if (!pci_enable_msix_exact(fnic->pdev, fnic->msix_entry,
-					   n + m + o + 1)) {
+		int vecs = n + m + o + 1;
+
+		if (pci_alloc_irq_vectors(fnic->pdev, vecs, vecs,
+				PCI_IRQ_MSIX) < 0) {
 			fnic->rq_count = n;
 			fnic->raw_wq_count = m;
 			fnic->wq_copy_count = o;
 			fnic->wq_count = m + o;
 			fnic->cq_count = n + m + o;
-			fnic->intr_count = n + m + o + 1;
+			fnic->intr_count = vecs;
 			fnic->err_intr_offset = FNIC_MSIX_ERR_NOTIFY;
 
 			FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host,
@@ -284,8 +280,7 @@ int fnic_set_intr_mode(struct fnic *fnic)
 	    fnic->wq_copy_count >= 1 &&
 	    fnic->cq_count >= 3 &&
 	    fnic->intr_count >= 1 &&
-	    !pci_enable_msi(fnic->pdev)) {
-
+	    pci_alloc_irq_vectors(fnic->pdev, 1, 1, PCI_IRQ_MSI) < 0) {
 		fnic->rq_count = 1;
 		fnic->raw_wq_count = 1;
 		fnic->wq_copy_count = 1;
@@ -334,17 +329,7 @@ int fnic_set_intr_mode(struct fnic *fnic)
 
 void fnic_clear_intr_mode(struct fnic *fnic)
 {
-	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
-	case VNIC_DEV_INTR_MODE_MSIX:
-		pci_disable_msix(fnic->pdev);
-		break;
-	case VNIC_DEV_INTR_MODE_MSI:
-		pci_disable_msi(fnic->pdev);
-		break;
-	default:
-		break;
-	}
-
+	pci_free_irq_vectors(fnic->pdev);
 	vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_INTX);
 }
 
diff --git a/drivers/scsi/fnic/fnic_scsi.c b/drivers/scsi/fnic/fnic_scsi.c
index adb3d5871e74..d048f3b5006f 100644
--- a/drivers/scsi/fnic/fnic_scsi.c
+++ b/drivers/scsi/fnic/fnic_scsi.c
@@ -823,6 +823,7 @@ static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic,
 	spinlock_t *io_lock;
 	u64 cmd_trace;
 	unsigned long start_time;
+	unsigned long io_duration_time;
 
 	/* Decode the cmpl description to get the io_req id */
 	fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag);
@@ -876,32 +877,28 @@ static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic,
 
 	/*
 	 *  if SCSI-ML has already issued abort on this command,
-	 * ignore completion of the IO. The abts path will clean it up
+	 *  set completion of the IO. The abts path will clean it up
 	 */
 	if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
-		spin_unlock_irqrestore(io_lock, flags);
+
+		/*
+		 * set the FNIC_IO_DONE so that this doesn't get
+		 * flagged as 'out of order' if it was not aborted
+		 */
+		CMD_FLAGS(sc) |= FNIC_IO_DONE;
 		CMD_FLAGS(sc) |= FNIC_IO_ABTS_PENDING;
-		switch (hdr_status) {
-		case FCPIO_SUCCESS:
-			CMD_FLAGS(sc) |= FNIC_IO_DONE;
-			FNIC_SCSI_DBG(KERN_INFO, fnic->lport->host,
-				  "icmnd_cmpl ABTS pending hdr status = %s "
-				  "sc  0x%p scsi_status %x  residual %d\n",
-				  fnic_fcpio_status_to_str(hdr_status), sc,
-				  icmnd_cmpl->scsi_status,
-				  icmnd_cmpl->residual);
-			break;
-		case FCPIO_ABORTED:
+		spin_unlock_irqrestore(io_lock, flags);
+		if(FCPIO_ABORTED == hdr_status)
 			CMD_FLAGS(sc) |= FNIC_IO_ABORTED;
-			break;
-		default:
-			FNIC_SCSI_DBG(KERN_INFO, fnic->lport->host,
-					  "icmnd_cmpl abts pending "
-					  "hdr status = %s tag = 0x%x sc = 0x%p\n",
-					  fnic_fcpio_status_to_str(hdr_status),
-					  id, sc);
-			break;
-		}
+
+		FNIC_SCSI_DBG(KERN_INFO, fnic->lport->host,
+			"icmnd_cmpl abts pending "
+			  "hdr status = %s tag = 0x%x sc = 0x%p"
+			  "scsi_status = %x residual = %d\n",
+			  fnic_fcpio_status_to_str(hdr_status),
+			  id, sc,
+			  icmnd_cmpl->scsi_status,
+			  icmnd_cmpl->residual);
 		return;
 	}
 
@@ -919,6 +916,9 @@ static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic,
 		if (icmnd_cmpl->flags & FCPIO_ICMND_CMPL_RESID_UNDER)
 			xfer_len -= icmnd_cmpl->residual;
 
+		if (icmnd_cmpl->scsi_status == SAM_STAT_CHECK_CONDITION)
+			atomic64_inc(&fnic_stats->misc_stats.check_condition);
+
 		if (icmnd_cmpl->scsi_status == SAM_STAT_TASK_SET_FULL)
 			atomic64_inc(&fnic_stats->misc_stats.queue_fulls);
 		break;
@@ -1017,6 +1017,28 @@ static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic,
 	else
 		atomic64_inc(&fnic_stats->io_stats.io_completions);
 
+
+	io_duration_time = jiffies_to_msecs(jiffies) - jiffies_to_msecs(io_req->start_time);
+
+	if(io_duration_time <= 10)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_0_to_10_msec);
+	else if(io_duration_time <= 100)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_10_to_100_msec);
+	else if(io_duration_time <= 500)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_100_to_500_msec);
+	else if(io_duration_time <= 5000)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_500_to_5000_msec);
+	else if(io_duration_time <= 10000)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_5000_to_10000_msec);
+	else if(io_duration_time <= 30000)
+		atomic64_inc(&fnic_stats->io_stats.io_btw_10000_to_30000_msec);
+	else {
+		atomic64_inc(&fnic_stats->io_stats.io_greater_than_30000_msec);
+
+		if(io_duration_time > atomic64_read(&fnic_stats->io_stats.current_max_io_time))
+			atomic64_set(&fnic_stats->io_stats.current_max_io_time, io_duration_time);
+	}
+
 	/* Call SCSI completion function to complete the IO */
 	if (sc->scsi_done)
 		sc->scsi_done(sc);
@@ -1128,18 +1150,11 @@ static void fnic_fcpio_itmf_cmpl_handler(struct fnic *fnic,
 		}
 
 		CMD_FLAGS(sc) |= FNIC_IO_ABT_TERM_DONE;
+		CMD_ABTS_STATUS(sc) = hdr_status;
 
 		/* If the status is IO not found consider it as success */
 		if (hdr_status == FCPIO_IO_NOT_FOUND)
 			CMD_ABTS_STATUS(sc) = FCPIO_SUCCESS;
-		else
-			CMD_ABTS_STATUS(sc) = hdr_status;
-
-		atomic64_dec(&fnic_stats->io_stats.active_ios);
-		if (atomic64_read(&fnic->io_cmpl_skip))
-			atomic64_dec(&fnic->io_cmpl_skip);
-		else
-			atomic64_inc(&fnic_stats->io_stats.io_completions);
 
 		if (!(CMD_FLAGS(sc) & (FNIC_IO_ABORTED | FNIC_IO_DONE)))
 			atomic64_inc(&misc_stats->no_icmnd_itmf_cmpls);
@@ -1181,6 +1196,11 @@ static void fnic_fcpio_itmf_cmpl_handler(struct fnic *fnic,
 					(((u64)CMD_FLAGS(sc) << 32) |
 					CMD_STATE(sc)));
 				sc->scsi_done(sc);
+				atomic64_dec(&fnic_stats->io_stats.active_ios);
+				if (atomic64_read(&fnic->io_cmpl_skip))
+					atomic64_dec(&fnic->io_cmpl_skip);
+				else
+					atomic64_inc(&fnic_stats->io_stats.io_completions);
 			}
 		}
 
@@ -1793,6 +1813,7 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	struct terminate_stats *term_stats;
 	enum fnic_ioreq_state old_ioreq_state;
 	int tag;
+	unsigned long abt_issued_time;
 	DECLARE_COMPLETION_ONSTACK(tm_done);
 
 	/* Wait for rport to unblock */
@@ -1846,6 +1867,25 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 		spin_unlock_irqrestore(io_lock, flags);
 		goto wait_pending;
 	}
+
+	abt_issued_time = jiffies_to_msecs(jiffies) - jiffies_to_msecs(io_req->start_time);
+	if (abt_issued_time <= 6000)
+		atomic64_inc(&abts_stats->abort_issued_btw_0_to_6_sec);
+	else if (abt_issued_time > 6000 && abt_issued_time <= 20000)
+		atomic64_inc(&abts_stats->abort_issued_btw_6_to_20_sec);
+	else if (abt_issued_time > 20000 && abt_issued_time <= 30000)
+		atomic64_inc(&abts_stats->abort_issued_btw_20_to_30_sec);
+	else if (abt_issued_time > 30000 && abt_issued_time <= 40000)
+		atomic64_inc(&abts_stats->abort_issued_btw_30_to_40_sec);
+	else if (abt_issued_time > 40000 && abt_issued_time <= 50000)
+		atomic64_inc(&abts_stats->abort_issued_btw_40_to_50_sec);
+	else if (abt_issued_time > 50000 && abt_issued_time <= 60000)
+		atomic64_inc(&abts_stats->abort_issued_btw_50_to_60_sec);
+	else
+		atomic64_inc(&abts_stats->abort_issued_greater_than_60_sec);
+
+	FNIC_SCSI_DBG(KERN_INFO, fnic->lport->host,
+		"CBD Opcode: %02x Abort issued time: %lu msec\n", sc->cmnd[0], abt_issued_time);
 	/*
 	 * Command is still pending, need to abort it
 	 * If the firmware completes the command after this point,
@@ -1970,6 +2010,11 @@ int fnic_abort_cmd(struct scsi_cmnd *sc)
 	/* Call SCSI completion function to complete the IO */
 		sc->result = (DID_ABORT << 16);
 		sc->scsi_done(sc);
+		atomic64_dec(&fnic_stats->io_stats.active_ios);
+		if (atomic64_read(&fnic->io_cmpl_skip))
+			atomic64_dec(&fnic->io_cmpl_skip);
+		else
+			atomic64_inc(&fnic_stats->io_stats.io_completions);
 	}
 
 fnic_abort_cmd_end:
diff --git a/drivers/scsi/fnic/fnic_stats.h b/drivers/scsi/fnic/fnic_stats.h
index 540cceb843cd..88c73cccb015 100644
--- a/drivers/scsi/fnic/fnic_stats.h
+++ b/drivers/scsi/fnic/fnic_stats.h
@@ -26,6 +26,14 @@ struct io_path_stats {
 	atomic64_t sc_null;
 	atomic64_t io_not_found;
 	atomic64_t num_ios;
+	atomic64_t io_btw_0_to_10_msec;
+	atomic64_t io_btw_10_to_100_msec;
+	atomic64_t io_btw_100_to_500_msec;
+	atomic64_t io_btw_500_to_5000_msec;
+	atomic64_t io_btw_5000_to_10000_msec;
+	atomic64_t io_btw_10000_to_30000_msec;
+	atomic64_t io_greater_than_30000_msec;
+	atomic64_t current_max_io_time;
 };
 
 struct abort_stats {
@@ -34,6 +42,13 @@ struct abort_stats {
 	atomic64_t abort_drv_timeouts;
 	atomic64_t abort_fw_timeouts;
 	atomic64_t abort_io_not_found;
+	atomic64_t abort_issued_btw_0_to_6_sec;
+	atomic64_t abort_issued_btw_6_to_20_sec;
+	atomic64_t abort_issued_btw_20_to_30_sec;
+	atomic64_t abort_issued_btw_30_to_40_sec;
+	atomic64_t abort_issued_btw_40_to_50_sec;
+	atomic64_t abort_issued_btw_50_to_60_sec;
+	atomic64_t abort_issued_greater_than_60_sec;
 };
 
 struct terminate_stats {
@@ -88,6 +103,7 @@ struct misc_stats {
 	atomic64_t devrst_cpwq_alloc_failures;
 	atomic64_t io_cpwq_alloc_failures;
 	atomic64_t no_icmnd_itmf_cmpls;
+	atomic64_t check_condition;
 	atomic64_t queue_fulls;
 	atomic64_t rport_not_ready;
 	atomic64_t frame_errors;
diff --git a/drivers/scsi/fnic/fnic_trace.c b/drivers/scsi/fnic/fnic_trace.c
index 5a5fa01576b7..b5ac5381a0d7 100644
--- a/drivers/scsi/fnic/fnic_trace.c
+++ b/drivers/scsi/fnic/fnic_trace.c
@@ -229,7 +229,16 @@ int fnic_get_stats_data(struct stats_debug_info *debug,
 		  "Number of IO Failures: %lld\nNumber of IO NOT Found: %lld\n"
 		  "Number of Memory alloc Failures: %lld\n"
 		  "Number of IOREQ Null: %lld\n"
-		  "Number of SCSI cmd pointer Null: %lld\n",
+		  "Number of SCSI cmd pointer Null: %lld\n"
+
+		  "\nIO completion times: \n"
+		  "            < 10 ms : %lld\n"
+		  "     10 ms - 100 ms : %lld\n"
+		  "    100 ms - 500 ms : %lld\n"
+		  "    500 ms -   5 sec: %lld\n"
+		  "     5 sec -  10 sec: %lld\n"
+		  "    10 sec -  30 sec: %lld\n"
+		  "            > 30 sec: %lld\n",
 		  (u64)atomic64_read(&stats->io_stats.active_ios),
 		  (u64)atomic64_read(&stats->io_stats.max_active_ios),
 		  (u64)atomic64_read(&stats->io_stats.num_ios),
@@ -238,28 +247,58 @@ int fnic_get_stats_data(struct stats_debug_info *debug,
 		  (u64)atomic64_read(&stats->io_stats.io_not_found),
 		  (u64)atomic64_read(&stats->io_stats.alloc_failures),
 		  (u64)atomic64_read(&stats->io_stats.ioreq_null),
-		  (u64)atomic64_read(&stats->io_stats.sc_null));
+		  (u64)atomic64_read(&stats->io_stats.sc_null),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_0_to_10_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_10_to_100_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_100_to_500_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_500_to_5000_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_5000_to_10000_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_btw_10000_to_30000_msec),
+		  (u64)atomic64_read(&stats->io_stats.io_greater_than_30000_msec));
+
+	len += snprintf(debug->debug_buffer + len, buf_size - len,
+		  "\nCurrent Max IO time : %lld\n",
+		  (u64)atomic64_read(&stats->io_stats.current_max_io_time));
 
 	len += snprintf(debug->debug_buffer + len, buf_size - len,
 		  "\n------------------------------------------\n"
 		  "\t\tAbort Statistics\n"
 		  "------------------------------------------\n");
+
 	len += snprintf(debug->debug_buffer + len, buf_size - len,
 		  "Number of Aborts: %lld\n"
 		  "Number of Abort Failures: %lld\n"
 		  "Number of Abort Driver Timeouts: %lld\n"
 		  "Number of Abort FW Timeouts: %lld\n"
-		  "Number of Abort IO NOT Found: %lld\n",
+		  "Number of Abort IO NOT Found: %lld\n"
+
+		  "Abord issued times: \n"
+		  "            < 6 sec : %lld\n"
+		  "     6 sec - 20 sec : %lld\n"
+		  "    20 sec - 30 sec : %lld\n"
+		  "    30 sec - 40 sec : %lld\n"
+		  "    40 sec - 50 sec : %lld\n"
+		  "    50 sec - 60 sec : %lld\n"
+		  "            > 60 sec: %lld\n",
+
 		  (u64)atomic64_read(&stats->abts_stats.aborts),
 		  (u64)atomic64_read(&stats->abts_stats.abort_failures),
 		  (u64)atomic64_read(&stats->abts_stats.abort_drv_timeouts),
 		  (u64)atomic64_read(&stats->abts_stats.abort_fw_timeouts),
-		  (u64)atomic64_read(&stats->abts_stats.abort_io_not_found));
+		  (u64)atomic64_read(&stats->abts_stats.abort_io_not_found),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_0_to_6_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_6_to_20_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_20_to_30_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_30_to_40_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_40_to_50_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_btw_50_to_60_sec),
+		  (u64)atomic64_read(&stats->abts_stats.abort_issued_greater_than_60_sec));
 
 	len += snprintf(debug->debug_buffer + len, buf_size - len,
 		  "\n------------------------------------------\n"
 		  "\t\tTerminate Statistics\n"
 		  "------------------------------------------\n");
+
 	len += snprintf(debug->debug_buffer + len, buf_size - len,
 		  "Number of Terminates: %lld\n"
 		  "Maximum Terminates: %lld\n"
@@ -357,6 +396,7 @@ int fnic_get_stats_data(struct stats_debug_info *debug,
 		  "Number of Copy WQ Alloc Failures for Device Reset: %lld\n"
 		  "Number of Copy WQ Alloc Failures for IOs: %lld\n"
 		  "Number of no icmnd itmf Completions: %lld\n"
+		  "Number of Check Conditions encountered: %lld\n"
 		  "Number of QUEUE Fulls: %lld\n"
 		  "Number of rport not ready: %lld\n"
 		  "Number of receive frame errors: %lld\n",
@@ -377,6 +417,7 @@ int fnic_get_stats_data(struct stats_debug_info *debug,
 			  &stats->misc_stats.devrst_cpwq_alloc_failures),
 		  (u64)atomic64_read(&stats->misc_stats.io_cpwq_alloc_failures),
 		  (u64)atomic64_read(&stats->misc_stats.no_icmnd_itmf_cmpls),
+		  (u64)atomic64_read(&stats->misc_stats.check_condition),
 		  (u64)atomic64_read(&stats->misc_stats.queue_fulls),
 		  (u64)atomic64_read(&stats->misc_stats.rport_not_ready),
 		  (u64)atomic64_read(&stats->misc_stats.frame_errors));
diff --git a/drivers/scsi/hisi_sas/Kconfig b/drivers/scsi/hisi_sas/Kconfig
index d1dd1616f983..374a329b91fc 100644
--- a/drivers/scsi/hisi_sas/Kconfig
+++ b/drivers/scsi/hisi_sas/Kconfig
@@ -4,5 +4,6 @@ config SCSI_HISI_SAS
 	depends on ARM64 || COMPILE_TEST
 	select SCSI_SAS_LIBSAS
 	select BLK_DEV_INTEGRITY
+	depends on ATA
 	help
 		This driver supports HiSilicon's SAS HBA
diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index 9216deaa3ff5..4e28f32e90b0 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -31,6 +31,7 @@
 #define HISI_SAS_QUEUE_SLOTS 512
 #define HISI_SAS_MAX_ITCT_ENTRIES 2048
 #define HISI_SAS_MAX_DEVICES HISI_SAS_MAX_ITCT_ENTRIES
+#define HISI_SAS_RESET_BIT	0
 
 #define HISI_SAS_STATUS_BUF_SZ \
 		(sizeof(struct hisi_sas_err_record) + 1024)
@@ -90,7 +91,6 @@ struct hisi_sas_port {
 	struct asd_sas_port	sas_port;
 	u8	port_attached;
 	u8	id; /* from hw */
-	struct list_head	list;
 };
 
 struct hisi_sas_cq {
@@ -113,7 +113,9 @@ struct hisi_sas_device {
 	u64 attached_phy;
 	u64 device_id;
 	atomic64_t running_req;
+	struct list_head	list;
 	u8 dev_status;
+	int sata_idx;
 };
 
 struct hisi_sas_slot {
@@ -136,6 +138,7 @@ struct hisi_sas_slot {
 	struct hisi_sas_sge_page *sge_page;
 	dma_addr_t sge_page_dma;
 	struct work_struct abort_slot;
+	struct timer_list internal_abort_timer;
 };
 
 struct hisi_sas_tmf_task {
@@ -165,7 +168,8 @@ struct hisi_sas_hw {
 			  struct hisi_sas_slot *slot,
 			  int device_id, int abort_flag, int tag_to_abort);
 	int (*slot_complete)(struct hisi_hba *hisi_hba,
-			     struct hisi_sas_slot *slot, int abort);
+			     struct hisi_sas_slot *slot);
+	void (*phys_init)(struct hisi_hba *hisi_hba);
 	void (*phy_enable)(struct hisi_hba *hisi_hba, int phy_no);
 	void (*phy_disable)(struct hisi_hba *hisi_hba, int phy_no);
 	void (*phy_hard_reset)(struct hisi_hba *hisi_hba, int phy_no);
@@ -175,6 +179,7 @@ struct hisi_sas_hw {
 	void (*free_device)(struct hisi_hba *hisi_hba,
 			    struct hisi_sas_device *dev);
 	int (*get_wideport_bitmap)(struct hisi_hba *hisi_hba, int port_id);
+	int (*soft_reset)(struct hisi_hba *hisi_hba);
 	int max_command_entries;
 	int complete_hdr_size;
 };
@@ -193,7 +198,6 @@ struct hisi_hba {
 	u8 sas_addr[SAS_ADDR_SIZE];
 
 	int n_phy;
-	int scan_finished;
 	spinlock_t lock;
 
 	struct timer_list timer;
@@ -201,6 +205,7 @@ struct hisi_hba {
 
 	int slot_index_count;
 	unsigned long *slot_index_tags;
+	unsigned long reject_stp_links_msk;
 
 	/* SCSI/SAS glue */
 	struct sas_ha_struct sha;
@@ -233,7 +238,10 @@ struct hisi_hba {
 	struct hisi_sas_breakpoint *sata_breakpoint;
 	dma_addr_t sata_breakpoint_dma;
 	struct hisi_sas_slot	*slot_info;
+	unsigned long flags;
 	const struct hisi_sas_hw *hw;	/* Low level hw interface */
+	unsigned long sata_dev_bitmap[BITS_TO_LONGS(HISI_SAS_MAX_DEVICES)];
+	struct work_struct rst_work;
 };
 
 /* Generic HW DMA host memory structures */
@@ -346,6 +354,8 @@ union hisi_sas_command_table {
 	struct hisi_sas_command_table_smp smp;
 	struct hisi_sas_command_table_stp stp;
 };
+
+extern struct hisi_sas_port *to_hisi_sas_port(struct asd_sas_port *sas_port);
 extern int hisi_sas_probe(struct platform_device *pdev,
 			  const struct hisi_sas_hw *ops);
 extern int hisi_sas_remove(struct platform_device *pdev);
@@ -354,4 +364,7 @@ extern void hisi_sas_phy_down(struct hisi_hba *hisi_hba, int phy_no, int rdy);
 extern void hisi_sas_slot_task_free(struct hisi_hba *hisi_hba,
 				    struct sas_task *task,
 				    struct hisi_sas_slot *slot);
+extern void hisi_sas_init_mem(struct hisi_hba *hisi_hba);
+extern void hisi_sas_rescan_topology(struct hisi_hba *hisi_hba, u32 old_state,
+				     u32 state);
 #endif
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 53637a941b94..d622db502ec9 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -21,12 +21,19 @@ static int
 hisi_sas_internal_task_abort(struct hisi_hba *hisi_hba,
 			     struct domain_device *device,
 			     int abort_flag, int tag);
+static int hisi_sas_softreset_ata_disk(struct domain_device *device);
 
 static struct hisi_hba *dev_to_hisi_hba(struct domain_device *device)
 {
 	return device->port->ha->lldd_ha;
 }
 
+struct hisi_sas_port *to_hisi_sas_port(struct asd_sas_port *sas_port)
+{
+	return container_of(sas_port, struct hisi_sas_port, sas_port);
+}
+EXPORT_SYMBOL_GPL(to_hisi_sas_port);
+
 static void hisi_sas_slot_index_clear(struct hisi_hba *hisi_hba, int slot_idx)
 {
 	void *bitmap = hisi_hba->slot_index_tags;
@@ -70,17 +77,22 @@ static void hisi_sas_slot_index_init(struct hisi_hba *hisi_hba)
 void hisi_sas_slot_task_free(struct hisi_hba *hisi_hba, struct sas_task *task,
 			     struct hisi_sas_slot *slot)
 {
-	struct device *dev = &hisi_hba->pdev->dev;
-	struct domain_device *device = task->dev;
-	struct hisi_sas_device *sas_dev = device->lldd_dev;
 
-	if (!slot->task)
-		return;
+	if (task) {
+		struct device *dev = &hisi_hba->pdev->dev;
+		struct domain_device *device = task->dev;
+		struct hisi_sas_device *sas_dev = device->lldd_dev;
 
-	if (!sas_protocol_ata(task->task_proto))
-		if (slot->n_elem)
-			dma_unmap_sg(dev, task->scatter, slot->n_elem,
-				     task->data_dir);
+		if (!sas_protocol_ata(task->task_proto))
+			if (slot->n_elem)
+				dma_unmap_sg(dev, task->scatter, slot->n_elem,
+					     task->data_dir);
+
+		task->lldd_task = NULL;
+
+		if (sas_dev)
+			atomic64_dec(&sas_dev->running_req);
+	}
 
 	if (slot->command_table)
 		dma_pool_free(hisi_hba->command_table_pool,
@@ -95,12 +107,10 @@ void hisi_sas_slot_task_free(struct hisi_hba *hisi_hba, struct sas_task *task,
 			      slot->sge_page_dma);
 
 	list_del_init(&slot->entry);
-	task->lldd_task = NULL;
 	slot->task = NULL;
 	slot->port = NULL;
 	hisi_sas_slot_index_free(hisi_hba, slot->idx);
-	if (sas_dev)
-		atomic64_dec(&sas_dev->running_req);
+
 	/* slot memory is fully zeroed when it is reused */
 }
 EXPORT_SYMBOL_GPL(hisi_sas_slot_task_free);
@@ -178,10 +188,12 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 	struct hisi_sas_port *port;
 	struct hisi_sas_slot *slot;
 	struct hisi_sas_cmd_hdr	*cmd_hdr_base;
+	struct asd_sas_port *sas_port = device->port;
 	struct device *dev = &hisi_hba->pdev->dev;
 	int dlvry_queue_slot, dlvry_queue, n_elem = 0, rc, slot_idx;
+	unsigned long flags;
 
-	if (!device->port) {
+	if (!sas_port) {
 		struct task_status_struct *ts = &task->task_status;
 
 		ts->resp = SAS_TASK_UNDELIVERED;
@@ -192,7 +204,7 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 		 */
 		if (device->dev_type != SAS_SATA_DEV)
 			task->task_done(task);
-		return 0;
+		return SAS_PHY_DOWN;
 	}
 
 	if (DEV_IS_GONE(sas_dev)) {
@@ -203,13 +215,13 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 			dev_info(dev, "task prep: device %016llx not ready\n",
 				 SAS_ADDR(device->sas_addr));
 
-		rc = SAS_PHY_DOWN;
-		return rc;
+		return SAS_PHY_DOWN;
 	}
-	port = device->port->lldd_port;
+
+	port = to_hisi_sas_port(sas_port);
 	if (port && !port->port_attached) {
 		dev_info(dev, "task prep: %s port%d not attach device\n",
-			 (sas_protocol_ata(task->task_proto)) ?
+			 (dev_is_sata(device)) ?
 			 "SATA/STP" : "SAS",
 			 device->port->id);
 
@@ -299,10 +311,10 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_hba *hisi_hba,
 		goto err_out_command_table;
 	}
 
-	list_add_tail(&slot->entry, &port->list);
-	spin_lock(&task->task_state_lock);
+	list_add_tail(&slot->entry, &sas_dev->list);
+	spin_lock_irqsave(&task->task_state_lock, flags);
 	task->task_state_flags |= SAS_TASK_AT_INITIATOR;
-	spin_unlock(&task->task_state_lock);
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
 
 	hisi_hba->slot_prep = slot;
 
@@ -343,6 +355,9 @@ static int hisi_sas_task_exec(struct sas_task *task, gfp_t gfp_flags,
 	struct hisi_hba *hisi_hba = dev_to_hisi_hba(task->dev);
 	struct device *dev = &hisi_hba->pdev->dev;
 
+	if (unlikely(test_bit(HISI_SAS_RESET_BIT, &hisi_hba->flags)))
+		return -EINVAL;
+
 	/* protect task_prep and start_delivery sequence */
 	spin_lock_irqsave(&hisi_hba->lock, flags);
 	rc = hisi_sas_task_prep(task, hisi_hba, is_tmf, tmf, &pass);
@@ -412,6 +427,7 @@ static struct hisi_sas_device *hisi_sas_alloc_dev(struct domain_device *device)
 			sas_dev->dev_type = device->dev_type;
 			sas_dev->hisi_hba = hisi_hba;
 			sas_dev->sas_device = device;
+			INIT_LIST_HEAD(&hisi_hba->devices[i].list);
 			break;
 		}
 	}
@@ -482,12 +498,8 @@ static int hisi_sas_slave_configure(struct scsi_device *sdev)
 static void hisi_sas_scan_start(struct Scsi_Host *shost)
 {
 	struct hisi_hba *hisi_hba = shost_priv(shost);
-	int i;
-
-	for (i = 0; i < hisi_hba->n_phy; ++i)
-		hisi_sas_bytes_dmaed(hisi_hba, i);
 
-	hisi_hba->scan_finished = 1;
+	hisi_hba->hw->phys_init(hisi_hba);
 }
 
 static int hisi_sas_scan_finished(struct Scsi_Host *shost, unsigned long time)
@@ -495,7 +507,8 @@ static int hisi_sas_scan_finished(struct Scsi_Host *shost, unsigned long time)
 	struct hisi_hba *hisi_hba = shost_priv(shost);
 	struct sas_ha_struct *sha = &hisi_hba->sha;
 
-	if (hisi_hba->scan_finished == 0)
+	/* Wait for PHY up interrupt to occur */
+	if (time < HZ)
 		return 0;
 
 	sas_drain_work(sha);
@@ -545,7 +558,7 @@ static void hisi_sas_port_notify_formed(struct asd_sas_phy *sas_phy)
 	struct hisi_hba *hisi_hba = sas_ha->lldd_ha;
 	struct hisi_sas_phy *phy = sas_phy->lldd_phy;
 	struct asd_sas_port *sas_port = sas_phy->port;
-	struct hisi_sas_port *port = &hisi_hba->port[phy->port_id];
+	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
 	unsigned long flags;
 
 	if (!sas_port)
@@ -559,50 +572,54 @@ static void hisi_sas_port_notify_formed(struct asd_sas_phy *sas_phy)
 	spin_unlock_irqrestore(&hisi_hba->lock, flags);
 }
 
-static void hisi_sas_do_release_task(struct hisi_hba *hisi_hba, int phy_no,
-				     struct domain_device *device)
+static void hisi_sas_do_release_task(struct hisi_hba *hisi_hba, struct sas_task *task,
+				     struct hisi_sas_slot *slot)
 {
-	struct hisi_sas_phy *phy;
-	struct hisi_sas_port *port;
-	struct hisi_sas_slot *slot, *slot2;
-	struct device *dev = &hisi_hba->pdev->dev;
-
-	phy = &hisi_hba->phy[phy_no];
-	port = phy->port;
-	if (!port)
-		return;
+	if (task) {
+		unsigned long flags;
+		struct task_status_struct *ts;
+
+		ts = &task->task_status;
+
+		ts->resp = SAS_TASK_COMPLETE;
+		ts->stat = SAS_ABORTED_TASK;
+		spin_lock_irqsave(&task->task_state_lock, flags);
+		task->task_state_flags &=
+			~(SAS_TASK_STATE_PENDING | SAS_TASK_AT_INITIATOR);
+		task->task_state_flags |= SAS_TASK_STATE_DONE;
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
+	}
 
-	list_for_each_entry_safe(slot, slot2, &port->list, entry) {
-		struct sas_task *task;
+	hisi_sas_slot_task_free(hisi_hba, task, slot);
+}
 
-		task = slot->task;
-		if (device && task->dev != device)
-			continue;
+/* hisi_hba.lock should be locked */
+static void hisi_sas_release_task(struct hisi_hba *hisi_hba,
+			struct domain_device *device)
+{
+	struct hisi_sas_slot *slot, *slot2;
+	struct hisi_sas_device *sas_dev = device->lldd_dev;
 
-		dev_info(dev, "Release slot [%d:%d], task [%p]:\n",
-			 slot->dlvry_queue, slot->dlvry_queue_slot, task);
-		hisi_hba->hw->slot_complete(hisi_hba, slot, 1);
-	}
+	list_for_each_entry_safe(slot, slot2, &sas_dev->list, entry)
+		hisi_sas_do_release_task(hisi_hba, slot->task, slot);
 }
 
-static void hisi_sas_port_notify_deformed(struct asd_sas_phy *sas_phy)
+static void hisi_sas_release_tasks(struct hisi_hba *hisi_hba)
 {
+	struct hisi_sas_device *sas_dev;
 	struct domain_device *device;
-	struct hisi_sas_phy *phy = sas_phy->lldd_phy;
-	struct asd_sas_port *sas_port = sas_phy->port;
+	int i;
 
-	list_for_each_entry(device, &sas_port->dev_list, dev_list_node)
-		hisi_sas_do_release_task(phy->hisi_hba, sas_phy->id, device);
-}
+	for (i = 0; i < HISI_SAS_MAX_DEVICES; i++) {
+		sas_dev = &hisi_hba->devices[i];
+		device = sas_dev->sas_device;
 
-static void hisi_sas_release_task(struct hisi_hba *hisi_hba,
-			struct domain_device *device)
-{
-	struct asd_sas_port *port = device->port;
-	struct asd_sas_phy *sas_phy;
+		if ((sas_dev->dev_type == SAS_PHY_UNUSED) ||
+		    !device)
+			continue;
 
-	list_for_each_entry(sas_phy, &port->phy_list, port_phy_el)
-		hisi_sas_do_release_task(hisi_hba, sas_phy->id, device);
+		hisi_sas_release_task(hisi_hba, device);
+	}
 }
 
 static void hisi_sas_dev_gone(struct domain_device *device)
@@ -644,8 +661,9 @@ static int hisi_sas_control_phy(struct asd_sas_phy *sas_phy, enum phy_func func,
 		break;
 
 	case PHY_FUNC_LINK_RESET:
+		hisi_hba->hw->phy_disable(hisi_hba, phy_no);
+		msleep(100);
 		hisi_hba->hw->phy_enable(hisi_hba, phy_no);
-		hisi_hba->hw->phy_hard_reset(hisi_hba, phy_no);
 		break;
 
 	case PHY_FUNC_DISABLE:
@@ -698,7 +716,12 @@ static int hisi_sas_exec_internal_tmf_task(struct domain_device *device,
 		task->dev = device;
 		task->task_proto = device->tproto;
 
-		memcpy(&task->ssp_task, parameter, para_len);
+		if (dev_is_sata(device)) {
+			task->ata_task.device_control_reg_update = 1;
+			memcpy(&task->ata_task.fis, parameter, para_len);
+		} else {
+			memcpy(&task->ssp_task, parameter, para_len);
+		}
 		task->task_done = hisi_sas_task_done;
 
 		task->slow_task->timer.data = (unsigned long) task;
@@ -720,15 +743,11 @@ static int hisi_sas_exec_internal_tmf_task(struct domain_device *device,
 		/* Even TMF timed out, return direct. */
 		if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) {
 			if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) {
-				dev_err(dev, "abort tmf: TMF task[%d] timeout\n",
-					tmf->tag_of_task_to_be_managed);
-				if (task->lldd_task) {
-					struct hisi_sas_slot *slot =
-						task->lldd_task;
+				struct hisi_sas_slot *slot = task->lldd_task;
 
-					hisi_sas_slot_task_free(hisi_hba,
-								task, slot);
-				}
+				dev_err(dev, "abort tmf: TMF task timeout\n");
+				if (slot)
+					slot->task = NULL;
 
 				goto ex_err;
 			}
@@ -781,6 +800,63 @@ ex_err:
 	return res;
 }
 
+static void hisi_sas_fill_ata_reset_cmd(struct ata_device *dev,
+		bool reset, int pmp, u8 *fis)
+{
+	struct ata_taskfile tf;
+
+	ata_tf_init(dev, &tf);
+	if (reset)
+		tf.ctl |= ATA_SRST;
+	else
+		tf.ctl &= ~ATA_SRST;
+	tf.command = ATA_CMD_DEV_RESET;
+	ata_tf_to_fis(&tf, pmp, 0, fis);
+}
+
+static int hisi_sas_softreset_ata_disk(struct domain_device *device)
+{
+	u8 fis[20] = {0};
+	struct ata_port *ap = device->sata_dev.ap;
+	struct ata_link *link;
+	int rc = TMF_RESP_FUNC_FAILED;
+	struct hisi_hba *hisi_hba = dev_to_hisi_hba(device);
+	struct device *dev = &hisi_hba->pdev->dev;
+	int s = sizeof(struct host_to_dev_fis);
+	unsigned long flags;
+
+	ata_for_each_link(link, ap, EDGE) {
+		int pmp = sata_srst_pmp(link);
+
+		hisi_sas_fill_ata_reset_cmd(link->device, 1, pmp, fis);
+		rc = hisi_sas_exec_internal_tmf_task(device, fis, s, NULL);
+		if (rc != TMF_RESP_FUNC_COMPLETE)
+			break;
+	}
+
+	if (rc == TMF_RESP_FUNC_COMPLETE) {
+		ata_for_each_link(link, ap, EDGE) {
+			int pmp = sata_srst_pmp(link);
+
+			hisi_sas_fill_ata_reset_cmd(link->device, 0, pmp, fis);
+			rc = hisi_sas_exec_internal_tmf_task(device, fis,
+							     s, NULL);
+			if (rc != TMF_RESP_FUNC_COMPLETE)
+				dev_err(dev, "ata disk de-reset failed\n");
+		}
+	} else {
+		dev_err(dev, "ata disk reset failed\n");
+	}
+
+	if (rc == TMF_RESP_FUNC_COMPLETE) {
+		spin_lock_irqsave(&hisi_hba->lock, flags);
+		hisi_sas_release_task(hisi_hba, device);
+		spin_unlock_irqrestore(&hisi_hba->lock, flags);
+	}
+
+	return rc;
+}
+
 static int hisi_sas_debug_issue_ssp_tmf(struct domain_device *device,
 				u8 *lun, struct hisi_sas_tmf_task *tmf)
 {
@@ -795,6 +871,40 @@ static int hisi_sas_debug_issue_ssp_tmf(struct domain_device *device,
 				sizeof(ssp_task), tmf);
 }
 
+static int hisi_sas_controller_reset(struct hisi_hba *hisi_hba)
+{
+	int rc;
+
+	if (!hisi_hba->hw->soft_reset)
+		return -1;
+
+	if (!test_and_set_bit(HISI_SAS_RESET_BIT, &hisi_hba->flags)) {
+		struct device *dev = &hisi_hba->pdev->dev;
+		struct sas_ha_struct *sas_ha = &hisi_hba->sha;
+		unsigned long flags;
+
+		dev_dbg(dev, "controller reset begins!\n");
+		scsi_block_requests(hisi_hba->shost);
+		rc = hisi_hba->hw->soft_reset(hisi_hba);
+		if (rc) {
+			dev_warn(dev, "controller reset failed (%d)\n", rc);
+			goto out;
+		}
+		spin_lock_irqsave(&hisi_hba->lock, flags);
+		hisi_sas_release_tasks(hisi_hba);
+		spin_unlock_irqrestore(&hisi_hba->lock, flags);
+
+		sas_ha->notify_ha_event(sas_ha, HAE_RESET);
+		dev_dbg(dev, "controller reset successful!\n");
+	} else
+		return -1;
+
+out:
+	scsi_unblock_requests(hisi_hba->shost);
+	clear_bit(HISI_SAS_RESET_BIT, &hisi_hba->flags);
+	return rc;
+}
+
 static int hisi_sas_abort_task(struct sas_task *task)
 {
 	struct scsi_lun lun;
@@ -811,19 +921,17 @@ static int hisi_sas_abort_task(struct sas_task *task)
 		return TMF_RESP_FUNC_FAILED;
 	}
 
-	spin_lock_irqsave(&task->task_state_lock, flags);
 	if (task->task_state_flags & SAS_TASK_STATE_DONE) {
-		spin_unlock_irqrestore(&task->task_state_lock, flags);
 		rc = TMF_RESP_FUNC_COMPLETE;
 		goto out;
 	}
 
-	spin_unlock_irqrestore(&task->task_state_lock, flags);
 	sas_dev->dev_status = HISI_SAS_DEV_EH;
 	if (task->lldd_task && task->task_proto & SAS_PROTOCOL_SSP) {
 		struct scsi_cmnd *cmnd = task->uldd_task;
 		struct hisi_sas_slot *slot = task->lldd_task;
 		u32 tag = slot->idx;
+		int rc2;
 
 		int_to_scsilun(cmnd->device->lun, &lun);
 		tmf_task.tmf = TMF_ABORT_TASK;
@@ -832,35 +940,41 @@ static int hisi_sas_abort_task(struct sas_task *task)
 		rc = hisi_sas_debug_issue_ssp_tmf(task->dev, lun.scsi_lun,
 						  &tmf_task);
 
-		/* if successful, clear the task and callback forwards.*/
-		if (rc == TMF_RESP_FUNC_COMPLETE) {
+		rc2 = hisi_sas_internal_task_abort(hisi_hba, device,
+						   HISI_SAS_INT_ABT_CMD, tag);
+		/*
+		 * If the TMF finds that the IO is not in the device and also
+		 * the internal abort does not succeed, then it is safe to
+		 * free the slot.
+		 * Note: if the internal abort succeeds then the slot
+		 * will have already been completed
+		 */
+		if (rc == TMF_RESP_FUNC_COMPLETE && rc2 != TMF_RESP_FUNC_SUCC) {
 			if (task->lldd_task) {
-				struct hisi_sas_slot *slot;
-
-				slot = &hisi_hba->slot_info
-					[tmf_task.tag_of_task_to_be_managed];
 				spin_lock_irqsave(&hisi_hba->lock, flags);
-				hisi_hba->hw->slot_complete(hisi_hba, slot, 1);
+				hisi_sas_do_release_task(hisi_hba, task, slot);
 				spin_unlock_irqrestore(&hisi_hba->lock, flags);
 			}
 		}
-
-		hisi_sas_internal_task_abort(hisi_hba, device,
-					     HISI_SAS_INT_ABT_CMD, tag);
 	} else if (task->task_proto & SAS_PROTOCOL_SATA ||
 		task->task_proto & SAS_PROTOCOL_STP) {
 		if (task->dev->dev_type == SAS_SATA_DEV) {
 			hisi_sas_internal_task_abort(hisi_hba, device,
 						     HISI_SAS_INT_ABT_DEV, 0);
-			rc = TMF_RESP_FUNC_COMPLETE;
+			rc = hisi_sas_softreset_ata_disk(device);
 		}
 	} else if (task->task_proto & SAS_PROTOCOL_SMP) {
 		/* SMP */
 		struct hisi_sas_slot *slot = task->lldd_task;
 		u32 tag = slot->idx;
 
-		hisi_sas_internal_task_abort(hisi_hba, device,
-					     HISI_SAS_INT_ABT_CMD, tag);
+		rc = hisi_sas_internal_task_abort(hisi_hba, device,
+			     HISI_SAS_INT_ABT_CMD, tag);
+		if (rc == TMF_RESP_FUNC_FAILED) {
+			spin_lock_irqsave(&hisi_hba->lock, flags);
+			hisi_sas_do_release_task(hisi_hba, task, slot);
+			spin_unlock_irqrestore(&hisi_hba->lock, flags);
+		}
 	}
 
 out:
@@ -915,37 +1029,66 @@ static int hisi_sas_I_T_nexus_reset(struct domain_device *device)
 
 	rc = hisi_sas_debug_I_T_nexus_reset(device);
 
-	spin_lock_irqsave(&hisi_hba->lock, flags);
-	hisi_sas_release_task(hisi_hba, device);
-	spin_unlock_irqrestore(&hisi_hba->lock, flags);
-
-	return 0;
+	if (rc == TMF_RESP_FUNC_COMPLETE) {
+		spin_lock_irqsave(&hisi_hba->lock, flags);
+		hisi_sas_release_task(hisi_hba, device);
+		spin_unlock_irqrestore(&hisi_hba->lock, flags);
+	}
+	return rc;
 }
 
 static int hisi_sas_lu_reset(struct domain_device *device, u8 *lun)
 {
-	struct hisi_sas_tmf_task tmf_task;
 	struct hisi_sas_device *sas_dev = device->lldd_dev;
 	struct hisi_hba *hisi_hba = dev_to_hisi_hba(device);
 	struct device *dev = &hisi_hba->pdev->dev;
 	unsigned long flags;
 	int rc = TMF_RESP_FUNC_FAILED;
 
-	tmf_task.tmf = TMF_LU_RESET;
 	sas_dev->dev_status = HISI_SAS_DEV_EH;
-	rc = hisi_sas_debug_issue_ssp_tmf(device, lun, &tmf_task);
-	if (rc == TMF_RESP_FUNC_COMPLETE) {
-		spin_lock_irqsave(&hisi_hba->lock, flags);
-		hisi_sas_release_task(hisi_hba, device);
-		spin_unlock_irqrestore(&hisi_hba->lock, flags);
-	}
+	if (dev_is_sata(device)) {
+		struct sas_phy *phy;
+
+		/* Clear internal IO and then hardreset */
+		rc = hisi_sas_internal_task_abort(hisi_hba, device,
+						  HISI_SAS_INT_ABT_DEV, 0);
+		if (rc == TMF_RESP_FUNC_FAILED)
+			goto out;
+
+		phy = sas_get_local_phy(device);
 
-	/* If failed, fall-through I_T_Nexus reset */
-	dev_err(dev, "lu_reset: for device[%llx]:rc= %d\n",
-		sas_dev->device_id, rc);
+		rc = sas_phy_reset(phy, 1);
+
+		if (rc == 0) {
+			spin_lock_irqsave(&hisi_hba->lock, flags);
+			hisi_sas_release_task(hisi_hba, device);
+			spin_unlock_irqrestore(&hisi_hba->lock, flags);
+		}
+		sas_put_local_phy(phy);
+	} else {
+		struct hisi_sas_tmf_task tmf_task = { .tmf =  TMF_LU_RESET };
+
+		rc = hisi_sas_debug_issue_ssp_tmf(device, lun, &tmf_task);
+		if (rc == TMF_RESP_FUNC_COMPLETE) {
+			spin_lock_irqsave(&hisi_hba->lock, flags);
+			hisi_sas_release_task(hisi_hba, device);
+			spin_unlock_irqrestore(&hisi_hba->lock, flags);
+		}
+	}
+out:
+	if (rc != TMF_RESP_FUNC_COMPLETE)
+		dev_err(dev, "lu_reset: for device[%llx]:rc= %d\n",
+			     sas_dev->device_id, rc);
 	return rc;
 }
 
+static int hisi_sas_clear_nexus_ha(struct sas_ha_struct *sas_ha)
+{
+	struct hisi_hba *hisi_hba = sas_ha->lldd_ha;
+
+	return hisi_sas_controller_reset(hisi_hba);
+}
+
 static int hisi_sas_query_task(struct sas_task *task)
 {
 	struct scsi_lun lun;
@@ -990,13 +1133,18 @@ hisi_sas_internal_abort_task_exec(struct hisi_hba *hisi_hba, u64 device_id,
 	struct device *dev = &hisi_hba->pdev->dev;
 	struct hisi_sas_port *port;
 	struct hisi_sas_slot *slot;
+	struct asd_sas_port *sas_port = device->port;
 	struct hisi_sas_cmd_hdr *cmd_hdr_base;
 	int dlvry_queue_slot, dlvry_queue, n_elem = 0, rc, slot_idx;
+	unsigned long flags;
+
+	if (unlikely(test_bit(HISI_SAS_RESET_BIT, &hisi_hba->flags)))
+		return -EINVAL;
 
 	if (!device->port)
 		return -1;
 
-	port = device->port->lldd_port;
+	port = to_hisi_sas_port(sas_port);
 
 	/* simply get a slot and send abort command */
 	rc = hisi_sas_slot_index_alloc(hisi_hba, &slot_idx);
@@ -1027,14 +1175,11 @@ hisi_sas_internal_abort_task_exec(struct hisi_hba *hisi_hba, u64 device_id,
 	if (rc)
 		goto err_out_tag;
 
-	/* Port structure is static for the HBA, so
-	*  even if the port is deformed it is ok
-	*  to reference.
-	*/
-	list_add_tail(&slot->entry, &port->list);
-	spin_lock(&task->task_state_lock);
+
+	list_add_tail(&slot->entry, &sas_dev->list);
+	spin_lock_irqsave(&task->task_state_lock, flags);
 	task->task_state_flags |= SAS_TASK_AT_INITIATOR;
-	spin_unlock(&task->task_state_lock);
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
 
 	hisi_hba->slot_prep = slot;
 
@@ -1085,7 +1230,7 @@ hisi_sas_internal_task_abort(struct hisi_hba *hisi_hba,
 	task->task_done = hisi_sas_task_done;
 	task->slow_task->timer.data = (unsigned long)task;
 	task->slow_task->timer.function = hisi_sas_tmf_timedout;
-	task->slow_task->timer.expires = jiffies + 20*HZ;
+	task->slow_task->timer.expires = jiffies + msecs_to_jiffies(110);
 	add_timer(&task->slow_task->timer);
 
 	/* Lock as we are alloc'ing a slot, which cannot be interrupted */
@@ -1108,15 +1253,16 @@ hisi_sas_internal_task_abort(struct hisi_hba *hisi_hba,
 		goto exit;
 	}
 
-	/* TMF timed out, return direct. */
+	if (task->task_status.resp == SAS_TASK_COMPLETE &&
+		task->task_status.stat == TMF_RESP_FUNC_SUCC) {
+		res = TMF_RESP_FUNC_SUCC;
+		goto exit;
+	}
+
+	/* Internal abort timed out */
 	if ((task->task_state_flags & SAS_TASK_STATE_ABORTED)) {
 		if (!(task->task_state_flags & SAS_TASK_STATE_DONE)) {
 			dev_err(dev, "internal task abort: timeout.\n");
-			if (task->lldd_task) {
-				struct hisi_sas_slot *slot = task->lldd_task;
-
-				hisi_sas_slot_task_free(hisi_hba, task, slot);
-			}
 		}
 	}
 
@@ -1137,11 +1283,6 @@ static void hisi_sas_port_formed(struct asd_sas_phy *sas_phy)
 	hisi_sas_port_notify_formed(sas_phy);
 }
 
-static void hisi_sas_port_deformed(struct asd_sas_phy *sas_phy)
-{
-	hisi_sas_port_notify_deformed(sas_phy);
-}
-
 static void hisi_sas_phy_disconnected(struct hisi_sas_phy *phy)
 {
 	phy->phy_attached = 0;
@@ -1181,6 +1322,37 @@ void hisi_sas_phy_down(struct hisi_hba *hisi_hba, int phy_no, int rdy)
 }
 EXPORT_SYMBOL_GPL(hisi_sas_phy_down);
 
+void hisi_sas_rescan_topology(struct hisi_hba *hisi_hba, u32 old_state,
+			      u32 state)
+{
+	struct sas_ha_struct *sas_ha = &hisi_hba->sha;
+	int phy_no;
+
+	for (phy_no = 0; phy_no < hisi_hba->n_phy; phy_no++) {
+		struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
+		struct asd_sas_phy *sas_phy = &phy->sas_phy;
+		struct asd_sas_port *sas_port = sas_phy->port;
+		struct domain_device *dev;
+
+		if (sas_phy->enabled) {
+			/* Report PHY state change to libsas */
+			if (state & (1 << phy_no))
+				continue;
+
+			if (old_state & (1 << phy_no))
+				/* PHY down but was up before */
+				hisi_sas_phy_down(hisi_hba, phy_no, 0);
+		}
+		if (!sas_port)
+			continue;
+		dev = sas_port->port_dev;
+
+		if (DEV_IS_EXPANDER(dev->dev_type))
+			sas_ha->notify_phy_event(sas_phy, PORTE_BROADCAST_RCVD);
+	}
+}
+EXPORT_SYMBOL_GPL(hisi_sas_rescan_topology);
+
 static struct scsi_transport_template *hisi_sas_stt;
 
 static struct scsi_host_template hisi_sas_sht = {
@@ -1215,10 +1387,41 @@ static struct sas_domain_function_template hisi_sas_transport_ops = {
 	.lldd_I_T_nexus_reset	= hisi_sas_I_T_nexus_reset,
 	.lldd_lu_reset		= hisi_sas_lu_reset,
 	.lldd_query_task	= hisi_sas_query_task,
+	.lldd_clear_nexus_ha = hisi_sas_clear_nexus_ha,
 	.lldd_port_formed	= hisi_sas_port_formed,
-	.lldd_port_deformed	= hisi_sas_port_deformed,
 };
 
+void hisi_sas_init_mem(struct hisi_hba *hisi_hba)
+{
+	int i, s, max_command_entries = hisi_hba->hw->max_command_entries;
+
+	for (i = 0; i < hisi_hba->queue_count; i++) {
+		struct hisi_sas_cq *cq = &hisi_hba->cq[i];
+		struct hisi_sas_dq *dq = &hisi_hba->dq[i];
+
+		s = sizeof(struct hisi_sas_cmd_hdr) * HISI_SAS_QUEUE_SLOTS;
+		memset(hisi_hba->cmd_hdr[i], 0, s);
+		dq->wr_point = 0;
+
+		s = hisi_hba->hw->complete_hdr_size * HISI_SAS_QUEUE_SLOTS;
+		memset(hisi_hba->complete_hdr[i], 0, s);
+		cq->rd_point = 0;
+	}
+
+	s = sizeof(struct hisi_sas_initial_fis) * hisi_hba->n_phy;
+	memset(hisi_hba->initial_fis, 0, s);
+
+	s = max_command_entries * sizeof(struct hisi_sas_iost);
+	memset(hisi_hba->iost, 0, s);
+
+	s = max_command_entries * sizeof(struct hisi_sas_breakpoint);
+	memset(hisi_hba->breakpoint, 0, s);
+
+	s = max_command_entries * sizeof(struct hisi_sas_breakpoint) * 2;
+	memset(hisi_hba->sata_breakpoint, 0, s);
+}
+EXPORT_SYMBOL_GPL(hisi_sas_init_mem);
+
 static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 {
 	struct platform_device *pdev = hisi_hba->pdev;
@@ -1230,7 +1433,6 @@ static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 		hisi_sas_phy_init(hisi_hba, i);
 		hisi_hba->port[i].port_attached = 0;
 		hisi_hba->port[i].id = -1;
-		INIT_LIST_HEAD(&hisi_hba->port[i].list);
 	}
 
 	for (i = 0; i < HISI_SAS_MAX_DEVICES; i++) {
@@ -1257,7 +1459,6 @@ static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 					&hisi_hba->cmd_hdr_dma[i], GFP_KERNEL);
 		if (!hisi_hba->cmd_hdr[i])
 			goto err_out;
-		memset(hisi_hba->cmd_hdr[i], 0, s);
 
 		/* Completion queue */
 		s = hisi_hba->hw->complete_hdr_size * HISI_SAS_QUEUE_SLOTS;
@@ -1265,7 +1466,6 @@ static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 				&hisi_hba->complete_hdr_dma[i], GFP_KERNEL);
 		if (!hisi_hba->complete_hdr[i])
 			goto err_out;
-		memset(hisi_hba->complete_hdr[i], 0, s);
 	}
 
 	s = HISI_SAS_STATUS_BUF_SZ;
@@ -1300,16 +1500,12 @@ static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 	if (!hisi_hba->iost)
 		goto err_out;
 
-	memset(hisi_hba->iost, 0, s);
-
 	s = max_command_entries * sizeof(struct hisi_sas_breakpoint);
 	hisi_hba->breakpoint = dma_alloc_coherent(dev, s,
 				&hisi_hba->breakpoint_dma, GFP_KERNEL);
 	if (!hisi_hba->breakpoint)
 		goto err_out;
 
-	memset(hisi_hba->breakpoint, 0, s);
-
 	hisi_hba->slot_index_count = max_command_entries;
 	s = hisi_hba->slot_index_count / BITS_PER_BYTE;
 	hisi_hba->slot_index_tags = devm_kzalloc(dev, s, GFP_KERNEL);
@@ -1326,14 +1522,13 @@ static int hisi_sas_alloc(struct hisi_hba *hisi_hba, struct Scsi_Host *shost)
 				&hisi_hba->initial_fis_dma, GFP_KERNEL);
 	if (!hisi_hba->initial_fis)
 		goto err_out;
-	memset(hisi_hba->initial_fis, 0, s);
 
 	s = max_command_entries * sizeof(struct hisi_sas_breakpoint) * 2;
 	hisi_hba->sata_breakpoint = dma_alloc_coherent(dev, s,
 				&hisi_hba->sata_breakpoint_dma, GFP_KERNEL);
 	if (!hisi_hba->sata_breakpoint)
 		goto err_out;
-	memset(hisi_hba->sata_breakpoint, 0, s);
+	hisi_sas_init_mem(hisi_hba);
 
 	hisi_sas_slot_index_init(hisi_hba);
 
@@ -1404,6 +1599,14 @@ static void hisi_sas_free(struct hisi_hba *hisi_hba)
 		destroy_workqueue(hisi_hba->wq);
 }
 
+static void hisi_sas_rst_work_handler(struct work_struct *work)
+{
+	struct hisi_hba *hisi_hba =
+		container_of(work, struct hisi_hba, rst_work);
+
+	hisi_sas_controller_reset(hisi_hba);
+}
+
 static struct Scsi_Host *hisi_sas_shost_alloc(struct platform_device *pdev,
 					      const struct hisi_sas_hw *hw)
 {
@@ -1421,6 +1624,7 @@ static struct Scsi_Host *hisi_sas_shost_alloc(struct platform_device *pdev,
 	}
 	hisi_hba = shost_priv(shost);
 
+	INIT_WORK(&hisi_hba->rst_work, hisi_sas_rst_work_handler);
 	hisi_hba->hw = hw;
 	hisi_hba->pdev = pdev;
 	hisi_hba->shost = shost;
@@ -1583,7 +1787,6 @@ int hisi_sas_remove(struct platform_device *pdev)
 	struct hisi_hba *hisi_hba = sha->lldd_ha;
 	struct Scsi_Host *shost = sha->core.shost;
 
-	scsi_remove_host(sha->core.shost);
 	sas_unregister_ha(sha);
 	sas_remove_host(sha->core.shost);
 
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index 854fbeaade3e..fc1c1b2c1a19 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -508,6 +508,8 @@ static void setup_itct_v1_hw(struct hisi_hba *hisi_hba,
 	struct device *dev = &hisi_hba->pdev->dev;
 	u64 qw0, device_id = sas_dev->device_id;
 	struct hisi_sas_itct *itct = &hisi_hba->itct[device_id];
+	struct asd_sas_port *sas_port = device->port;
+	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
 
 	memset(itct, 0, sizeof(*itct));
 
@@ -528,7 +530,7 @@ static void setup_itct_v1_hw(struct hisi_hba *hisi_hba,
 		(1 << ITCT_HDR_AWT_CONTROL_OFF) |
 		(device->max_linkrate << ITCT_HDR_MAX_CONN_RATE_OFF) |
 		(1 << ITCT_HDR_VALID_LINK_NUM_OFF) |
-		(device->port->id << ITCT_HDR_PORT_ID_OFF));
+		(port->id << ITCT_HDR_PORT_ID_OFF));
 	itct->qw0 = cpu_to_le64(qw0);
 
 	/* qw1 */
@@ -1275,7 +1277,7 @@ static void slot_err_v1_hw(struct hisi_hba *hisi_hba,
 }
 
 static int slot_complete_v1_hw(struct hisi_hba *hisi_hba,
-			       struct hisi_sas_slot *slot, int abort)
+			       struct hisi_sas_slot *slot)
 {
 	struct sas_task *task = slot->task;
 	struct hisi_sas_device *sas_dev;
@@ -1286,6 +1288,7 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba,
 	struct hisi_sas_complete_v1_hdr *complete_queue =
 			hisi_hba->complete_hdr[slot->cmplt_queue];
 	struct hisi_sas_complete_v1_hdr *complete_hdr;
+	unsigned long flags;
 	u32 cmplt_hdr_data;
 
 	complete_hdr = &complete_queue[slot->cmplt_queue_slot];
@@ -1298,16 +1301,17 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba,
 	device = task->dev;
 	sas_dev = device->lldd_dev;
 
+	spin_lock_irqsave(&task->task_state_lock, flags);
 	task->task_state_flags &=
 		~(SAS_TASK_STATE_PENDING | SAS_TASK_AT_INITIATOR);
 	task->task_state_flags |= SAS_TASK_STATE_DONE;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
 
 	memset(ts, 0, sizeof(*ts));
 	ts->resp = SAS_TASK_COMPLETE;
 
-	if (unlikely(!sas_dev || abort)) {
-		if (!sas_dev)
-			dev_dbg(dev, "slot complete: port has not device\n");
+	if (unlikely(!sas_dev)) {
+		dev_dbg(dev, "slot complete: port has no device\n");
 		ts->stat = SAS_PHY_DOWN;
 		goto out;
 	}
@@ -1620,7 +1624,7 @@ static irqreturn_t cq_interrupt_v1_hw(int irq, void *p)
 		 */
 		slot->cmplt_queue_slot = rd_point;
 		slot->cmplt_queue = queue;
-		slot_complete_v1_hw(hisi_hba, slot, 0);
+		slot_complete_v1_hw(hisi_hba, slot);
 
 		if (++rd_point >= HISI_SAS_QUEUE_SLOTS)
 			rd_point = 0;
@@ -1845,8 +1849,6 @@ static int hisi_sas_v1_init(struct hisi_hba *hisi_hba)
 	if (rc)
 		return rc;
 
-	phys_init_v1_hw(hisi_hba);
-
 	return 0;
 }
 
@@ -1860,6 +1862,7 @@ static const struct hisi_sas_hw hisi_sas_v1_hw = {
 	.get_free_slot = get_free_slot_v1_hw,
 	.start_delivery = start_delivery_v1_hw,
 	.slot_complete = slot_complete_v1_hw,
+	.phys_init = phys_init_v1_hw,
 	.phy_enable = enable_phy_v1_hw,
 	.phy_disable = disable_phy_v1_hw,
 	.phy_hard_reset = phy_hard_reset_v1_hw,
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 1b214450dcb5..e241921bee10 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -194,9 +194,9 @@
 #define SL_CONTROL_NOTIFY_EN_MSK	(0x1 << SL_CONTROL_NOTIFY_EN_OFF)
 #define SL_CONTROL_CTA_OFF		17
 #define SL_CONTROL_CTA_MSK		(0x1 << SL_CONTROL_CTA_OFF)
-#define RX_PRIMS_STATUS         (PORT_BASE + 0x98)
-#define RX_BCAST_CHG_OFF        1
-#define RX_BCAST_CHG_MSK        (0x1 << RX_BCAST_CHG_OFF)
+#define RX_PRIMS_STATUS			(PORT_BASE + 0x98)
+#define RX_BCAST_CHG_OFF		1
+#define RX_BCAST_CHG_MSK		(0x1 << RX_BCAST_CHG_OFF)
 #define TX_ID_DWORD0			(PORT_BASE + 0x9c)
 #define TX_ID_DWORD1			(PORT_BASE + 0xa0)
 #define TX_ID_DWORD2			(PORT_BASE + 0xa4)
@@ -207,8 +207,10 @@
 #define TXID_AUTO			(PORT_BASE + 0xb8)
 #define TXID_AUTO_CT3_OFF		1
 #define TXID_AUTO_CT3_MSK		(0x1 << TXID_AUTO_CT3_OFF)
-#define TX_HARDRST_OFF          2
-#define TX_HARDRST_MSK          (0x1 << TX_HARDRST_OFF)
+#define TXID_AUTO_CTB_OFF		11
+#define TXID_AUTO_CTB_MSK		(0x1 << TXID_AUTO_CTB_OFF)
+#define TX_HARDRST_OFF			2
+#define TX_HARDRST_MSK			(0x1 << TX_HARDRST_OFF)
 #define RX_IDAF_DWORD0			(PORT_BASE + 0xc4)
 #define RX_IDAF_DWORD1			(PORT_BASE + 0xc8)
 #define RX_IDAF_DWORD2			(PORT_BASE + 0xcc)
@@ -218,6 +220,9 @@
 #define RX_IDAF_DWORD6			(PORT_BASE + 0xdc)
 #define RXOP_CHECK_CFG_H		(PORT_BASE + 0xfc)
 #define CON_CONTROL			(PORT_BASE + 0x118)
+#define CON_CONTROL_CFG_OPEN_ACC_STP_OFF	0
+#define CON_CONTROL_CFG_OPEN_ACC_STP_MSK	\
+		(0x01 << CON_CONTROL_CFG_OPEN_ACC_STP_OFF)
 #define DONE_RECEIVED_TIME		(PORT_BASE + 0x11c)
 #define CHL_INT0			(PORT_BASE + 0x1b4)
 #define CHL_INT0_HOTPLUG_TOUT_OFF	0
@@ -240,6 +245,17 @@
 #define CHL_INT1_MSK			(PORT_BASE + 0x1c4)
 #define CHL_INT2_MSK			(PORT_BASE + 0x1c8)
 #define CHL_INT_COAL_EN			(PORT_BASE + 0x1d0)
+#define DMA_TX_DFX0				(PORT_BASE + 0x200)
+#define DMA_TX_DFX1				(PORT_BASE + 0x204)
+#define DMA_TX_DFX1_IPTT_OFF		0
+#define DMA_TX_DFX1_IPTT_MSK		(0xffff << DMA_TX_DFX1_IPTT_OFF)
+#define DMA_TX_FIFO_DFX0		(PORT_BASE + 0x240)
+#define PORT_DFX0				(PORT_BASE + 0x258)
+#define LINK_DFX2					(PORT_BASE + 0X264)
+#define LINK_DFX2_RCVR_HOLD_STS_OFF	9
+#define LINK_DFX2_RCVR_HOLD_STS_MSK	(0x1 << LINK_DFX2_RCVR_HOLD_STS_OFF)
+#define LINK_DFX2_SEND_HOLD_STS_OFF	10
+#define LINK_DFX2_SEND_HOLD_STS_MSK	(0x1 << LINK_DFX2_SEND_HOLD_STS_OFF)
 #define PHY_CTRL_RDY_MSK		(PORT_BASE + 0x2b0)
 #define PHYCTRL_NOT_RDY_MSK		(PORT_BASE + 0x2b4)
 #define PHYCTRL_DWS_RESET_MSK		(PORT_BASE + 0x2b8)
@@ -257,6 +273,10 @@
 #define AM_CFG_MAX_TRANS		(0x5010)
 #define AM_CFG_SINGLE_PORT_MAX_TRANS	(0x5014)
 
+#define AXI_MASTER_CFG_BASE		(0x5000)
+#define AM_CTRL_GLOBAL			(0x0)
+#define AM_CURR_TRANS_RETURN	(0x150)
+
 /* HW dma structures */
 /* Delivery queue header */
 /* dw0 */
@@ -309,6 +329,8 @@
 
 /* Completion header */
 /* dw0 */
+#define CMPLT_HDR_ERR_PHASE_OFF	2
+#define CMPLT_HDR_ERR_PHASE_MSK	(0xff << CMPLT_HDR_ERR_PHASE_OFF)
 #define CMPLT_HDR_RSPNS_XFRD_OFF	10
 #define CMPLT_HDR_RSPNS_XFRD_MSK	(0x1 << CMPLT_HDR_RSPNS_XFRD_OFF)
 #define CMPLT_HDR_ERX_OFF		12
@@ -385,10 +407,10 @@ enum {
 
 enum {
 	TRANS_TX_FAIL_BASE = 0x0, /* dw0 */
-	TRANS_RX_FAIL_BASE = 0x100, /* dw1 */
-	DMA_TX_ERR_BASE = 0x200, /* dw2 bit 15-0 */
-	SIPC_RX_ERR_BASE = 0x300, /* dw2 bit 31-16*/
-	DMA_RX_ERR_BASE = 0x400, /* dw3 */
+	TRANS_RX_FAIL_BASE = 0x20, /* dw1 */
+	DMA_TX_ERR_BASE = 0x40, /* dw2 bit 15-0 */
+	SIPC_RX_ERR_BASE = 0x50, /* dw2 bit 31-16*/
+	DMA_RX_ERR_BASE = 0x60, /* dw3 */
 
 	/* trans tx*/
 	TRANS_TX_OPEN_FAIL_WITH_IT_NEXUS_LOSS = TRANS_TX_FAIL_BASE, /* 0x0 */
@@ -428,100 +450,104 @@ enum {
 	TRANS_TX_ERR_WITH_WAIT_RECV_TIMEOUT, /* 0x1f for sata/stp */
 
 	/* trans rx */
-	TRANS_RX_ERR_WITH_RXFRAME_CRC_ERR = TRANS_RX_FAIL_BASE, /* 0x100 */
-	TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR, /* 0x101 for sata/stp */
-	TRANS_RX_ERR_WITH_RXFRAME_HAVE_ERRPRM, /* 0x102 for ssp/smp */
-	/*IO_ERR_WITH_RXFIS_8B10B_CODE_ERR, [> 0x102 <] for sata/stp */
-	TRANS_RX_ERR_WITH_RXFIS_DECODE_ERROR, /* 0x103 for sata/stp */
-	TRANS_RX_ERR_WITH_RXFIS_CRC_ERR, /* 0x104 for sata/stp */
-	TRANS_RX_ERR_WITH_RXFRAME_LENGTH_OVERRUN, /* 0x105 for smp */
-	/*IO_ERR_WITH_RXFIS_TX SYNCP, [> 0x105 <] for sata/stp */
-	TRANS_RX_ERR_WITH_RXFIS_RX_SYNCP, /* 0x106 for sata/stp*/
-	TRANS_RX_ERR_WITH_LINK_BUF_OVERRUN, /* 0x107 */
-	TRANS_RX_ERR_WITH_BREAK_TIMEOUT, /* 0x108 */
-	TRANS_RX_ERR_WITH_BREAK_REQUEST, /* 0x109 */
-	TRANS_RX_ERR_WITH_BREAK_RECEVIED, /* 0x10a */
-	RESERVED1, /* 0x10b */
-	TRANS_RX_ERR_WITH_CLOSE_NORMAL, /* 0x10c */
-	TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE, /* 0x10d */
-	TRANS_RX_ERR_WITH_CLOSE_DWS_TIMEOUT, /* 0x10e */
-	TRANS_RX_ERR_WITH_CLOSE_COMINIT, /* 0x10f */
-	TRANS_RX_ERR_WITH_DATA_LEN0, /* 0x110 for ssp/smp */
-	TRANS_RX_ERR_WITH_BAD_HASH, /* 0x111 for ssp */
-	/*IO_RX_ERR_WITH_FIS_TOO_SHORT, [> 0x111 <] for sata/stp */
-	TRANS_RX_XRDY_WLEN_ZERO_ERR, /* 0x112 for ssp*/
-	/*IO_RX_ERR_WITH_FIS_TOO_LONG, [> 0x112 <] for sata/stp */
-	TRANS_RX_SSP_FRM_LEN_ERR, /* 0x113 for ssp */
-	/*IO_RX_ERR_WITH_SATA_DEVICE_LOST, [> 0x113 <] for sata */
-	RESERVED2, /* 0x114 */
-	RESERVED3, /* 0x115 */
-	RESERVED4, /* 0x116 */
-	RESERVED5, /* 0x117 */
-	TRANS_RX_ERR_WITH_BAD_FRM_TYPE, /* 0x118 */
-	TRANS_RX_SMP_FRM_LEN_ERR, /* 0x119 */
-	TRANS_RX_SMP_RESP_TIMEOUT_ERR, /* 0x11a */
-	RESERVED6, /* 0x11b */
-	RESERVED7, /* 0x11c */
-	RESERVED8, /* 0x11d */
-	RESERVED9, /* 0x11e */
-	TRANS_RX_R_ERR, /* 0x11f */
+	TRANS_RX_ERR_WITH_RXFRAME_CRC_ERR = TRANS_RX_FAIL_BASE, /* 0x20 */
+	TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR, /* 0x21 for sata/stp */
+	TRANS_RX_ERR_WITH_RXFRAME_HAVE_ERRPRM, /* 0x22 for ssp/smp */
+	/*IO_ERR_WITH_RXFIS_8B10B_CODE_ERR, [> 0x22 <] for sata/stp */
+	TRANS_RX_ERR_WITH_RXFIS_DECODE_ERROR, /* 0x23 for sata/stp */
+	TRANS_RX_ERR_WITH_RXFIS_CRC_ERR, /* 0x24 for sata/stp */
+	TRANS_RX_ERR_WITH_RXFRAME_LENGTH_OVERRUN, /* 0x25 for smp */
+	/*IO_ERR_WITH_RXFIS_TX SYNCP, [> 0x25 <] for sata/stp */
+	TRANS_RX_ERR_WITH_RXFIS_RX_SYNCP, /* 0x26 for sata/stp*/
+	TRANS_RX_ERR_WITH_LINK_BUF_OVERRUN, /* 0x27 */
+	TRANS_RX_ERR_WITH_BREAK_TIMEOUT, /* 0x28 */
+	TRANS_RX_ERR_WITH_BREAK_REQUEST, /* 0x29 */
+	TRANS_RX_ERR_WITH_BREAK_RECEVIED, /* 0x2a */
+	RESERVED1, /* 0x2b */
+	TRANS_RX_ERR_WITH_CLOSE_NORMAL, /* 0x2c */
+	TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE, /* 0x2d */
+	TRANS_RX_ERR_WITH_CLOSE_DWS_TIMEOUT, /* 0x2e */
+	TRANS_RX_ERR_WITH_CLOSE_COMINIT, /* 0x2f */
+	TRANS_RX_ERR_WITH_DATA_LEN0, /* 0x30 for ssp/smp */
+	TRANS_RX_ERR_WITH_BAD_HASH, /* 0x31 for ssp */
+	/*IO_RX_ERR_WITH_FIS_TOO_SHORT, [> 0x31 <] for sata/stp */
+	TRANS_RX_XRDY_WLEN_ZERO_ERR, /* 0x32 for ssp*/
+	/*IO_RX_ERR_WITH_FIS_TOO_LONG, [> 0x32 <] for sata/stp */
+	TRANS_RX_SSP_FRM_LEN_ERR, /* 0x33 for ssp */
+	/*IO_RX_ERR_WITH_SATA_DEVICE_LOST, [> 0x33 <] for sata */
+	RESERVED2, /* 0x34 */
+	RESERVED3, /* 0x35 */
+	RESERVED4, /* 0x36 */
+	RESERVED5, /* 0x37 */
+	TRANS_RX_ERR_WITH_BAD_FRM_TYPE, /* 0x38 */
+	TRANS_RX_SMP_FRM_LEN_ERR, /* 0x39 */
+	TRANS_RX_SMP_RESP_TIMEOUT_ERR, /* 0x3a */
+	RESERVED6, /* 0x3b */
+	RESERVED7, /* 0x3c */
+	RESERVED8, /* 0x3d */
+	RESERVED9, /* 0x3e */
+	TRANS_RX_R_ERR, /* 0x3f */
 
 	/* dma tx */
-	DMA_TX_DIF_CRC_ERR = DMA_TX_ERR_BASE, /* 0x200 */
-	DMA_TX_DIF_APP_ERR, /* 0x201 */
-	DMA_TX_DIF_RPP_ERR, /* 0x202 */
-	DMA_TX_DATA_SGL_OVERFLOW, /* 0x203 */
-	DMA_TX_DIF_SGL_OVERFLOW, /* 0x204 */
-	DMA_TX_UNEXP_XFER_ERR, /* 0x205 */
-	DMA_TX_UNEXP_RETRANS_ERR, /* 0x206 */
-	DMA_TX_XFER_LEN_OVERFLOW, /* 0x207 */
-	DMA_TX_XFER_OFFSET_ERR, /* 0x208 */
-	DMA_TX_RAM_ECC_ERR, /* 0x209 */
-	DMA_TX_DIF_LEN_ALIGN_ERR, /* 0x20a */
+	DMA_TX_DIF_CRC_ERR = DMA_TX_ERR_BASE, /* 0x40 */
+	DMA_TX_DIF_APP_ERR, /* 0x41 */
+	DMA_TX_DIF_RPP_ERR, /* 0x42 */
+	DMA_TX_DATA_SGL_OVERFLOW, /* 0x43 */
+	DMA_TX_DIF_SGL_OVERFLOW, /* 0x44 */
+	DMA_TX_UNEXP_XFER_ERR, /* 0x45 */
+	DMA_TX_UNEXP_RETRANS_ERR, /* 0x46 */
+	DMA_TX_XFER_LEN_OVERFLOW, /* 0x47 */
+	DMA_TX_XFER_OFFSET_ERR, /* 0x48 */
+	DMA_TX_RAM_ECC_ERR, /* 0x49 */
+	DMA_TX_DIF_LEN_ALIGN_ERR, /* 0x4a */
+	DMA_TX_MAX_ERR_CODE,
 
 	/* sipc rx */
-	SIPC_RX_FIS_STATUS_ERR_BIT_VLD = SIPC_RX_ERR_BASE, /* 0x300 */
-	SIPC_RX_PIO_WRSETUP_STATUS_DRQ_ERR, /* 0x301 */
-	SIPC_RX_FIS_STATUS_BSY_BIT_ERR, /* 0x302 */
-	SIPC_RX_WRSETUP_LEN_ODD_ERR, /* 0x303 */
-	SIPC_RX_WRSETUP_LEN_ZERO_ERR, /* 0x304 */
-	SIPC_RX_WRDATA_LEN_NOT_MATCH_ERR, /* 0x305 */
-	SIPC_RX_NCQ_WRSETUP_OFFSET_ERR, /* 0x306 */
-	SIPC_RX_NCQ_WRSETUP_AUTO_ACTIVE_ERR, /* 0x307 */
-	SIPC_RX_SATA_UNEXP_FIS_ERR, /* 0x308 */
-	SIPC_RX_WRSETUP_ESTATUS_ERR, /* 0x309 */
-	SIPC_RX_DATA_UNDERFLOW_ERR, /* 0x30a */
+	SIPC_RX_FIS_STATUS_ERR_BIT_VLD = SIPC_RX_ERR_BASE, /* 0x50 */
+	SIPC_RX_PIO_WRSETUP_STATUS_DRQ_ERR, /* 0x51 */
+	SIPC_RX_FIS_STATUS_BSY_BIT_ERR, /* 0x52 */
+	SIPC_RX_WRSETUP_LEN_ODD_ERR, /* 0x53 */
+	SIPC_RX_WRSETUP_LEN_ZERO_ERR, /* 0x54 */
+	SIPC_RX_WRDATA_LEN_NOT_MATCH_ERR, /* 0x55 */
+	SIPC_RX_NCQ_WRSETUP_OFFSET_ERR, /* 0x56 */
+	SIPC_RX_NCQ_WRSETUP_AUTO_ACTIVE_ERR, /* 0x57 */
+	SIPC_RX_SATA_UNEXP_FIS_ERR, /* 0x58 */
+	SIPC_RX_WRSETUP_ESTATUS_ERR, /* 0x59 */
+	SIPC_RX_DATA_UNDERFLOW_ERR, /* 0x5a */
+	SIPC_RX_MAX_ERR_CODE,
 
 	/* dma rx */
-	DMA_RX_DIF_CRC_ERR = DMA_RX_ERR_BASE, /* 0x400 */
-	DMA_RX_DIF_APP_ERR, /* 0x401 */
-	DMA_RX_DIF_RPP_ERR, /* 0x402 */
-	DMA_RX_DATA_SGL_OVERFLOW, /* 0x403 */
-	DMA_RX_DIF_SGL_OVERFLOW, /* 0x404 */
-	DMA_RX_DATA_LEN_OVERFLOW, /* 0x405 */
-	DMA_RX_DATA_LEN_UNDERFLOW, /* 0x406 */
-	DMA_RX_DATA_OFFSET_ERR, /* 0x407 */
-	RESERVED10, /* 0x408 */
-	DMA_RX_SATA_FRAME_TYPE_ERR, /* 0x409 */
-	DMA_RX_RESP_BUF_OVERFLOW, /* 0x40a */
-	DMA_RX_UNEXP_RETRANS_RESP_ERR, /* 0x40b */
-	DMA_RX_UNEXP_NORM_RESP_ERR, /* 0x40c */
-	DMA_RX_UNEXP_RDFRAME_ERR, /* 0x40d */
-	DMA_RX_PIO_DATA_LEN_ERR, /* 0x40e */
-	DMA_RX_RDSETUP_STATUS_ERR, /* 0x40f */
-	DMA_RX_RDSETUP_STATUS_DRQ_ERR, /* 0x410 */
-	DMA_RX_RDSETUP_STATUS_BSY_ERR, /* 0x411 */
-	DMA_RX_RDSETUP_LEN_ODD_ERR, /* 0x412 */
-	DMA_RX_RDSETUP_LEN_ZERO_ERR, /* 0x413 */
-	DMA_RX_RDSETUP_LEN_OVER_ERR, /* 0x414 */
-	DMA_RX_RDSETUP_OFFSET_ERR, /* 0x415 */
-	DMA_RX_RDSETUP_ACTIVE_ERR, /* 0x416 */
-	DMA_RX_RDSETUP_ESTATUS_ERR, /* 0x417 */
-	DMA_RX_RAM_ECC_ERR, /* 0x418 */
-	DMA_RX_UNKNOWN_FRM_ERR, /* 0x419 */
+	DMA_RX_DIF_CRC_ERR = DMA_RX_ERR_BASE, /* 0x60 */
+	DMA_RX_DIF_APP_ERR, /* 0x61 */
+	DMA_RX_DIF_RPP_ERR, /* 0x62 */
+	DMA_RX_DATA_SGL_OVERFLOW, /* 0x63 */
+	DMA_RX_DIF_SGL_OVERFLOW, /* 0x64 */
+	DMA_RX_DATA_LEN_OVERFLOW, /* 0x65 */
+	DMA_RX_DATA_LEN_UNDERFLOW, /* 0x66 */
+	DMA_RX_DATA_OFFSET_ERR, /* 0x67 */
+	RESERVED10, /* 0x68 */
+	DMA_RX_SATA_FRAME_TYPE_ERR, /* 0x69 */
+	DMA_RX_RESP_BUF_OVERFLOW, /* 0x6a */
+	DMA_RX_UNEXP_RETRANS_RESP_ERR, /* 0x6b */
+	DMA_RX_UNEXP_NORM_RESP_ERR, /* 0x6c */
+	DMA_RX_UNEXP_RDFRAME_ERR, /* 0x6d */
+	DMA_RX_PIO_DATA_LEN_ERR, /* 0x6e */
+	DMA_RX_RDSETUP_STATUS_ERR, /* 0x6f */
+	DMA_RX_RDSETUP_STATUS_DRQ_ERR, /* 0x70 */
+	DMA_RX_RDSETUP_STATUS_BSY_ERR, /* 0x71 */
+	DMA_RX_RDSETUP_LEN_ODD_ERR, /* 0x72 */
+	DMA_RX_RDSETUP_LEN_ZERO_ERR, /* 0x73 */
+	DMA_RX_RDSETUP_LEN_OVER_ERR, /* 0x74 */
+	DMA_RX_RDSETUP_OFFSET_ERR, /* 0x75 */
+	DMA_RX_RDSETUP_ACTIVE_ERR, /* 0x76 */
+	DMA_RX_RDSETUP_ESTATUS_ERR, /* 0x77 */
+	DMA_RX_RAM_ECC_ERR, /* 0x78 */
+	DMA_RX_UNKNOWN_FRM_ERR, /* 0x79 */
+	DMA_RX_MAX_ERR_CODE,
 };
 
 #define HISI_SAS_COMMAND_ENTRIES_V2_HW 4096
+#define HISI_MAX_SATA_SUPPORT_V2_HW	(HISI_SAS_COMMAND_ENTRIES_V2_HW/64 - 1)
 
 #define DIR_NO_DATA 0
 #define DIR_TO_INI 1
@@ -534,7 +560,13 @@ enum {
 #define SATA_PROTOCOL_FPDMA		0x8
 #define SATA_PROTOCOL_ATAPI		0x10
 
-static void hisi_sas_link_timeout_disable_link(unsigned long data);
+#define ERR_ON_TX_PHASE(err_phase) (err_phase == 0x2 || \
+		err_phase == 0x4 || err_phase == 0x8 ||\
+		err_phase == 0x6 || err_phase == 0xa)
+#define ERR_ON_RX_PHASE(err_phase) (err_phase == 0x10 || \
+		err_phase == 0x20 || err_phase == 0x40)
+
+static void link_timeout_disable_link(unsigned long data);
 
 static u32 hisi_sas_read32(struct hisi_hba *hisi_hba, u32 off)
 {
@@ -576,38 +608,86 @@ static u32 hisi_sas_phy_read32(struct hisi_hba *hisi_hba,
 /* This function needs to be protected from pre-emption. */
 static int
 slot_index_alloc_quirk_v2_hw(struct hisi_hba *hisi_hba, int *slot_idx,
-		       struct domain_device *device)
+			     struct domain_device *device)
 {
-	unsigned int index = 0;
-	void *bitmap = hisi_hba->slot_index_tags;
 	int sata_dev = dev_is_sata(device);
+	void *bitmap = hisi_hba->slot_index_tags;
+	struct hisi_sas_device *sas_dev = device->lldd_dev;
+	int sata_idx = sas_dev->sata_idx;
+	int start, end;
+
+	if (!sata_dev) {
+		/*
+		 * STP link SoC bug workaround: index starts from 1.
+		 * additionally, we can only allocate odd IPTT(1~4095)
+		 * for SAS/SMP device.
+		 */
+		start = 1;
+		end = hisi_hba->slot_index_count;
+	} else {
+		if (sata_idx >= HISI_MAX_SATA_SUPPORT_V2_HW)
+			return -EINVAL;
+
+		/*
+		 * For SATA device: allocate even IPTT in this interval
+		 * [64*(sata_idx+1), 64*(sata_idx+2)], then each SATA device
+		 * own 32 IPTTs. IPTT 0 shall not be used duing to STP link
+		 * SoC bug workaround. So we ignore the first 32 even IPTTs.
+		 */
+		start = 64 * (sata_idx + 1);
+		end = 64 * (sata_idx + 2);
+	}
 
 	while (1) {
-		index = find_next_zero_bit(bitmap, hisi_hba->slot_index_count,
-					   index);
-		if (index >= hisi_hba->slot_index_count)
+		start = find_next_zero_bit(bitmap,
+					hisi_hba->slot_index_count, start);
+		if (start >= end)
 			return -SAS_QUEUE_FULL;
 		/*
-		 * SAS IPTT bit0 should be 1
-		 */
-		if (sata_dev || (index & 1))
+		  * SAS IPTT bit0 should be 1, and SATA IPTT bit0 should be 0.
+		  */
+		if (sata_dev ^ (start & 1))
 			break;
-		index++;
+		start++;
 	}
 
-	set_bit(index, bitmap);
-	*slot_idx = index;
+	set_bit(start, bitmap);
+	*slot_idx = start;
 	return 0;
 }
 
+static bool sata_index_alloc_v2_hw(struct hisi_hba *hisi_hba, int *idx)
+{
+	unsigned int index;
+	struct device *dev = &hisi_hba->pdev->dev;
+	void *bitmap = hisi_hba->sata_dev_bitmap;
+
+	index = find_first_zero_bit(bitmap, HISI_MAX_SATA_SUPPORT_V2_HW);
+	if (index >= HISI_MAX_SATA_SUPPORT_V2_HW) {
+		dev_warn(dev, "alloc sata index failed, index=%d\n", index);
+		return false;
+	}
+
+	set_bit(index, bitmap);
+	*idx = index;
+	return true;
+}
+
+
 static struct
 hisi_sas_device *alloc_dev_quirk_v2_hw(struct domain_device *device)
 {
 	struct hisi_hba *hisi_hba = device->port->ha->lldd_ha;
 	struct hisi_sas_device *sas_dev = NULL;
 	int i, sata_dev = dev_is_sata(device);
+	int sata_idx = -1;
 
 	spin_lock(&hisi_hba->lock);
+
+	if (sata_dev)
+		if (!sata_index_alloc_v2_hw(hisi_hba, &sata_idx))
+			goto out;
+
 	for (i = 0; i < HISI_SAS_MAX_DEVICES; i++) {
 		/*
 		 * SATA device id bit0 should be 0
@@ -621,9 +701,13 @@ hisi_sas_device *alloc_dev_quirk_v2_hw(struct domain_device *device)
 			sas_dev->dev_type = device->dev_type;
 			sas_dev->hisi_hba = hisi_hba;
 			sas_dev->sas_device = device;
+			sas_dev->sata_idx = sata_idx;
+			INIT_LIST_HEAD(&hisi_hba->devices[i].list);
 			break;
 		}
 	}
+
+out:
 	spin_unlock(&hisi_hba->lock);
 
 	return sas_dev;
@@ -676,7 +760,8 @@ static void setup_itct_v2_hw(struct hisi_hba *hisi_hba,
 	u64 qw0, device_id = sas_dev->device_id;
 	struct hisi_sas_itct *itct = &hisi_hba->itct[device_id];
 	struct domain_device *parent_dev = device->parent;
-	struct hisi_sas_port *port = device->port->lldd_port;
+	struct asd_sas_port *sas_port = device->port;
+	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
 
 	memset(itct, 0, sizeof(*itct));
 
@@ -729,6 +814,10 @@ static void free_device_v2_hw(struct hisi_hba *hisi_hba,
 	u32 reg_val = hisi_sas_read32(hisi_hba, ENT_INT_SRC3);
 	int i;
 
+	/* SoC bug workaround */
+	if (dev_is_sata(sas_dev->sas_device))
+		clear_bit(sas_dev->sata_idx, hisi_hba->sata_dev_bitmap);
+
 	/* clear the itct interrupt state */
 	if (ENT_INT_SRC3_ITC_INT_MSK & reg_val)
 		hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
@@ -858,6 +947,46 @@ static int reset_hw_v2_hw(struct hisi_hba *hisi_hba)
 	return 0;
 }
 
+/* This function needs to be called after resetting SAS controller. */
+static void phys_reject_stp_links_v2_hw(struct hisi_hba *hisi_hba)
+{
+	u32 cfg;
+	int phy_no;
+
+	hisi_hba->reject_stp_links_msk = (1 << hisi_hba->n_phy) - 1;
+	for (phy_no = 0; phy_no < hisi_hba->n_phy; phy_no++) {
+		cfg = hisi_sas_phy_read32(hisi_hba, phy_no, CON_CONTROL);
+		if (!(cfg & CON_CONTROL_CFG_OPEN_ACC_STP_MSK))
+			continue;
+
+		cfg &= ~CON_CONTROL_CFG_OPEN_ACC_STP_MSK;
+		hisi_sas_phy_write32(hisi_hba, phy_no, CON_CONTROL, cfg);
+	}
+}
+
+static void phys_try_accept_stp_links_v2_hw(struct hisi_hba *hisi_hba)
+{
+	int phy_no;
+	u32 dma_tx_dfx1;
+
+	for (phy_no = 0; phy_no < hisi_hba->n_phy; phy_no++) {
+		if (!(hisi_hba->reject_stp_links_msk & BIT(phy_no)))
+			continue;
+
+		dma_tx_dfx1 = hisi_sas_phy_read32(hisi_hba, phy_no,
+						DMA_TX_DFX1);
+		if (dma_tx_dfx1 & DMA_TX_DFX1_IPTT_MSK) {
+			u32 cfg = hisi_sas_phy_read32(hisi_hba,
+				phy_no, CON_CONTROL);
+
+			cfg |= CON_CONTROL_CFG_OPEN_ACC_STP_MSK;
+			hisi_sas_phy_write32(hisi_hba, phy_no,
+				CON_CONTROL, cfg);
+			clear_bit(phy_no, &hisi_hba->reject_stp_links_msk);
+		}
+	}
+}
+
 static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 {
 	struct device *dev = &hisi_hba->pdev->dev;
@@ -876,7 +1005,7 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 			 (u32)((1ULL << hisi_hba->queue_count) - 1));
 	hisi_sas_write32(hisi_hba, AXI_USER1, 0xc0000000);
 	hisi_sas_write32(hisi_hba, AXI_USER2, 0x10000);
-	hisi_sas_write32(hisi_hba, HGC_SAS_TXFAIL_RETRY_CTRL, 0x108);
+	hisi_sas_write32(hisi_hba, HGC_SAS_TXFAIL_RETRY_CTRL, 0x0);
 	hisi_sas_write32(hisi_hba, HGC_SAS_TX_OPEN_FAIL_RETRY_CTRL, 0x7FF);
 	hisi_sas_write32(hisi_hba, OPENA_WT_CONTI_TIME, 0x1);
 	hisi_sas_write32(hisi_hba, I_T_NEXUS_LOSS_TIME, 0x1F4);
@@ -885,9 +1014,9 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 	hisi_sas_write32(hisi_hba, CFG_AGING_TIME, 0x1);
 	hisi_sas_write32(hisi_hba, HGC_ERR_STAT_EN, 0x1);
 	hisi_sas_write32(hisi_hba, HGC_GET_ITV_TIME, 0x1);
-	hisi_sas_write32(hisi_hba, INT_COAL_EN, 0x1);
-	hisi_sas_write32(hisi_hba, OQ_INT_COAL_TIME, 0x1);
-	hisi_sas_write32(hisi_hba, OQ_INT_COAL_CNT, 0x1);
+	hisi_sas_write32(hisi_hba, INT_COAL_EN, 0xc);
+	hisi_sas_write32(hisi_hba, OQ_INT_COAL_TIME, 0x60);
+	hisi_sas_write32(hisi_hba, OQ_INT_COAL_CNT, 0x3);
 	hisi_sas_write32(hisi_hba, ENT_INT_COAL_TIME, 0x1);
 	hisi_sas_write32(hisi_hba, ENT_INT_COAL_CNT, 0x1);
 	hisi_sas_write32(hisi_hba, OQ_INT_SRC, 0x0);
@@ -910,14 +1039,14 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 		hisi_sas_phy_write32(hisi_hba, i, SL_TOUT_CFG, 0x7d7d7d7d);
 		hisi_sas_phy_write32(hisi_hba, i, SL_CONTROL, 0x0);
 		hisi_sas_phy_write32(hisi_hba, i, TXID_AUTO, 0x2);
-		hisi_sas_phy_write32(hisi_hba, i, DONE_RECEIVED_TIME, 0x10);
+		hisi_sas_phy_write32(hisi_hba, i, DONE_RECEIVED_TIME, 0x8);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT0, 0xffffffff);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT1, 0xffffffff);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2, 0xfff87fff);
 		hisi_sas_phy_write32(hisi_hba, i, RXOP_CHECK_CFG_H, 0x1000);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT1_MSK, 0xffffffff);
 		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2_MSK, 0x8ffffbff);
-		hisi_sas_phy_write32(hisi_hba, i, SL_CFG, 0x23f801fc);
+		hisi_sas_phy_write32(hisi_hba, i, SL_CFG, 0x13f801fc);
 		hisi_sas_phy_write32(hisi_hba, i, PHY_CTRL_RDY_MSK, 0x0);
 		hisi_sas_phy_write32(hisi_hba, i, PHYCTRL_NOT_RDY_MSK, 0x0);
 		hisi_sas_phy_write32(hisi_hba, i, PHYCTRL_DWS_RESET_MSK, 0x0);
@@ -989,12 +1118,15 @@ static void init_reg_v2_hw(struct hisi_hba *hisi_hba)
 			 upper_32_bits(hisi_hba->initial_fis_dma));
 }
 
-static void hisi_sas_link_timeout_enable_link(unsigned long data)
+static void link_timeout_enable_link(unsigned long data)
 {
 	struct hisi_hba *hisi_hba = (struct hisi_hba *)data;
 	int i, reg_val;
 
 	for (i = 0; i < hisi_hba->n_phy; i++) {
+		if (hisi_hba->reject_stp_links_msk & BIT(i))
+			continue;
+
 		reg_val = hisi_sas_phy_read32(hisi_hba, i, CON_CONTROL);
 		if (!(reg_val & BIT(0))) {
 			hisi_sas_phy_write32(hisi_hba, i,
@@ -1003,17 +1135,20 @@ static void hisi_sas_link_timeout_enable_link(unsigned long data)
 		}
 	}
 
-	hisi_hba->timer.function = hisi_sas_link_timeout_disable_link;
+	hisi_hba->timer.function = link_timeout_disable_link;
 	mod_timer(&hisi_hba->timer, jiffies + msecs_to_jiffies(900));
 }
 
-static void hisi_sas_link_timeout_disable_link(unsigned long data)
+static void link_timeout_disable_link(unsigned long data)
 {
 	struct hisi_hba *hisi_hba = (struct hisi_hba *)data;
 	int i, reg_val;
 
 	reg_val = hisi_sas_read32(hisi_hba, PHY_STATE);
 	for (i = 0; i < hisi_hba->n_phy && reg_val; i++) {
+		if (hisi_hba->reject_stp_links_msk & BIT(i))
+			continue;
+
 		if (reg_val & BIT(i)) {
 			hisi_sas_phy_write32(hisi_hba, i,
 					CON_CONTROL, 0x6);
@@ -1021,14 +1156,14 @@ static void hisi_sas_link_timeout_disable_link(unsigned long data)
 		}
 	}
 
-	hisi_hba->timer.function = hisi_sas_link_timeout_enable_link;
+	hisi_hba->timer.function = link_timeout_enable_link;
 	mod_timer(&hisi_hba->timer, jiffies + msecs_to_jiffies(100));
 }
 
 static void set_link_timer_quirk(struct hisi_hba *hisi_hba)
 {
 	hisi_hba->timer.data = (unsigned long)hisi_hba;
-	hisi_hba->timer.function = hisi_sas_link_timeout_disable_link;
+	hisi_hba->timer.function = link_timeout_disable_link;
 	hisi_hba->timer.expires = jiffies + msecs_to_jiffies(1000);
 	add_timer(&hisi_hba->timer);
 }
@@ -1058,12 +1193,138 @@ static void enable_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
 	hisi_sas_phy_write32(hisi_hba, phy_no, PHY_CFG, cfg);
 }
 
+static bool is_sata_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
+{
+	u32 context;
+
+	context = hisi_sas_read32(hisi_hba, PHY_CONTEXT);
+	if (context & (1 << phy_no))
+		return true;
+
+	return false;
+}
+
+static bool tx_fifo_is_empty_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
+{
+	u32 dfx_val;
+
+	dfx_val = hisi_sas_phy_read32(hisi_hba, phy_no, DMA_TX_DFX1);
+
+	if (dfx_val & BIT(16))
+		return false;
+
+	return true;
+}
+
+static bool axi_bus_is_idle_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
+{
+	int i, max_loop = 1000;
+	struct device *dev = &hisi_hba->pdev->dev;
+	u32 status, axi_status, dfx_val, dfx_tx_val;
+
+	for (i = 0; i < max_loop; i++) {
+		status = hisi_sas_read32_relaxed(hisi_hba,
+			AXI_MASTER_CFG_BASE + AM_CURR_TRANS_RETURN);
+
+		axi_status = hisi_sas_read32(hisi_hba, AXI_CFG);
+		dfx_val = hisi_sas_phy_read32(hisi_hba, phy_no, DMA_TX_DFX1);
+		dfx_tx_val = hisi_sas_phy_read32(hisi_hba,
+			phy_no, DMA_TX_FIFO_DFX0);
+
+		if ((status == 0x3) && (axi_status == 0x0) &&
+		    (dfx_val & BIT(20)) && (dfx_tx_val & BIT(10)))
+			return true;
+		udelay(10);
+	}
+	dev_err(dev, "bus is not idle phy%d, axi150:0x%x axi100:0x%x port204:0x%x port240:0x%x\n",
+			phy_no, status, axi_status,
+			dfx_val, dfx_tx_val);
+	return false;
+}
+
+static bool wait_io_done_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
+{
+	int i, max_loop = 1000;
+	struct device *dev = &hisi_hba->pdev->dev;
+	u32 status, tx_dfx0;
+
+	for (i = 0; i < max_loop; i++) {
+		status = hisi_sas_phy_read32(hisi_hba, phy_no, LINK_DFX2);
+		status = (status & 0x3fc0) >> 6;
+
+		if (status != 0x1)
+			return true;
+
+		tx_dfx0 = hisi_sas_phy_read32(hisi_hba, phy_no, DMA_TX_DFX0);
+		if ((tx_dfx0 & 0x1ff) == 0x2)
+			return true;
+		udelay(10);
+	}
+	dev_err(dev, "IO not done phy%d, port264:0x%x port200:0x%x\n",
+			phy_no, status, tx_dfx0);
+	return false;
+}
+
+static bool allowed_disable_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
+{
+	if (tx_fifo_is_empty_v2_hw(hisi_hba, phy_no))
+		return true;
+
+	if (!axi_bus_is_idle_v2_hw(hisi_hba, phy_no))
+		return false;
+
+	if (!wait_io_done_v2_hw(hisi_hba, phy_no))
+		return false;
+
+	return true;
+}
+
+
 static void disable_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
 {
-	u32 cfg = hisi_sas_phy_read32(hisi_hba, phy_no, PHY_CFG);
+	u32 cfg, axi_val, dfx0_val, txid_auto;
+	struct device *dev = &hisi_hba->pdev->dev;
+
+	/* Close axi bus. */
+	axi_val = hisi_sas_read32(hisi_hba, AXI_MASTER_CFG_BASE +
+				AM_CTRL_GLOBAL);
+	axi_val |= 0x1;
+	hisi_sas_write32(hisi_hba, AXI_MASTER_CFG_BASE +
+		AM_CTRL_GLOBAL, axi_val);
+
+	if (is_sata_phy_v2_hw(hisi_hba, phy_no)) {
+		if (allowed_disable_phy_v2_hw(hisi_hba, phy_no))
+			goto do_disable;
+
+		/* Reset host controller. */
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
+		return;
+	}
+
+	dfx0_val = hisi_sas_phy_read32(hisi_hba, phy_no, PORT_DFX0);
+	dfx0_val = (dfx0_val & 0x1fc0) >> 6;
+	if (dfx0_val != 0x4)
+		goto do_disable;
 
+	if (!tx_fifo_is_empty_v2_hw(hisi_hba, phy_no)) {
+		dev_warn(dev, "phy%d, wait tx fifo need send break\n",
+			phy_no);
+		txid_auto = hisi_sas_phy_read32(hisi_hba, phy_no,
+					TXID_AUTO);
+		txid_auto |= TXID_AUTO_CTB_MSK;
+		hisi_sas_phy_write32(hisi_hba, phy_no, TXID_AUTO,
+					txid_auto);
+	}
+
+do_disable:
+	cfg = hisi_sas_phy_read32(hisi_hba, phy_no, PHY_CFG);
 	cfg &= ~PHY_CFG_ENA_MSK;
 	hisi_sas_phy_write32(hisi_hba, phy_no, PHY_CFG, cfg);
+
+	/* Open axi bus. */
+	axi_val &= ~0x1;
+	hisi_sas_write32(hisi_hba, AXI_MASTER_CFG_BASE +
+		AM_CTRL_GLOBAL, axi_val);
 }
 
 static void start_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
@@ -1078,6 +1339,14 @@ static void stop_phy_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
 	disable_phy_v2_hw(hisi_hba, phy_no);
 }
 
+static void stop_phys_v2_hw(struct hisi_hba *hisi_hba)
+{
+	int i;
+
+	for (i = 0; i < hisi_hba->n_phy; i++)
+		stop_phy_v2_hw(hisi_hba, i);
+}
+
 static void phy_hard_reset_v2_hw(struct hisi_hba *hisi_hba, int phy_no)
 {
 	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
@@ -1437,10 +1706,205 @@ static void sata_done_v2_hw(struct hisi_hba *hisi_hba, struct sas_task *task,
 	ts->buf_valid_size = sizeof(*resp);
 }
 
+#define TRANS_TX_ERR	0
+#define TRANS_RX_ERR	1
+#define DMA_TX_ERR		2
+#define SIPC_RX_ERR		3
+#define DMA_RX_ERR		4
+
+#define DMA_TX_ERR_OFF	0
+#define DMA_TX_ERR_MSK	(0xffff << DMA_TX_ERR_OFF)
+#define SIPC_RX_ERR_OFF	16
+#define SIPC_RX_ERR_MSK (0xffff << SIPC_RX_ERR_OFF)
+
+static int parse_trans_tx_err_code_v2_hw(u32 err_msk)
+{
+	const u8 trans_tx_err_code_prio[] = {
+		TRANS_TX_OPEN_FAIL_WITH_IT_NEXUS_LOSS,
+		TRANS_TX_ERR_PHY_NOT_ENABLE,
+		TRANS_TX_OPEN_CNX_ERR_WRONG_DESTINATION,
+		TRANS_TX_OPEN_CNX_ERR_ZONE_VIOLATION,
+		TRANS_TX_OPEN_CNX_ERR_BY_OTHER,
+		RESERVED0,
+		TRANS_TX_OPEN_CNX_ERR_AIP_TIMEOUT,
+		TRANS_TX_OPEN_CNX_ERR_STP_RESOURCES_BUSY,
+		TRANS_TX_OPEN_CNX_ERR_PROTOCOL_NOT_SUPPORTED,
+		TRANS_TX_OPEN_CNX_ERR_CONNECTION_RATE_NOT_SUPPORTED,
+		TRANS_TX_OPEN_CNX_ERR_BAD_DESTINATION,
+		TRANS_TX_OPEN_CNX_ERR_BREAK_RCVD,
+		TRANS_TX_OPEN_CNX_ERR_LOW_PHY_POWER,
+		TRANS_TX_OPEN_CNX_ERR_PATHWAY_BLOCKED,
+		TRANS_TX_OPEN_CNX_ERR_OPEN_TIMEOUT,
+		TRANS_TX_OPEN_CNX_ERR_NO_DESTINATION,
+		TRANS_TX_OPEN_RETRY_ERR_THRESHOLD_REACHED,
+		TRANS_TX_ERR_WITH_CLOSE_PHYDISALE,
+		TRANS_TX_ERR_WITH_CLOSE_DWS_TIMEOUT,
+		TRANS_TX_ERR_WITH_CLOSE_COMINIT,
+		TRANS_TX_ERR_WITH_BREAK_TIMEOUT,
+		TRANS_TX_ERR_WITH_BREAK_REQUEST,
+		TRANS_TX_ERR_WITH_BREAK_RECEVIED,
+		TRANS_TX_ERR_WITH_CLOSE_TIMEOUT,
+		TRANS_TX_ERR_WITH_CLOSE_NORMAL,
+		TRANS_TX_ERR_WITH_NAK_RECEVIED,
+		TRANS_TX_ERR_WITH_ACK_NAK_TIMEOUT,
+		TRANS_TX_ERR_WITH_CREDIT_TIMEOUT,
+		TRANS_TX_ERR_WITH_IPTT_CONFLICT,
+		TRANS_TX_ERR_WITH_OPEN_BY_DES_OR_OTHERS,
+		TRANS_TX_ERR_WITH_WAIT_RECV_TIMEOUT,
+	};
+	int index, i;
+
+	for (i = 0; i < ARRAY_SIZE(trans_tx_err_code_prio); i++) {
+		index = trans_tx_err_code_prio[i] - TRANS_TX_FAIL_BASE;
+		if (err_msk & (1 << index))
+			return trans_tx_err_code_prio[i];
+	}
+	return -1;
+}
+
+static int parse_trans_rx_err_code_v2_hw(u32 err_msk)
+{
+	const u8 trans_rx_err_code_prio[] = {
+		TRANS_RX_ERR_WITH_RXFRAME_CRC_ERR,
+		TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR,
+		TRANS_RX_ERR_WITH_RXFRAME_HAVE_ERRPRM,
+		TRANS_RX_ERR_WITH_RXFIS_DECODE_ERROR,
+		TRANS_RX_ERR_WITH_RXFIS_CRC_ERR,
+		TRANS_RX_ERR_WITH_RXFRAME_LENGTH_OVERRUN,
+		TRANS_RX_ERR_WITH_RXFIS_RX_SYNCP,
+		TRANS_RX_ERR_WITH_LINK_BUF_OVERRUN,
+		TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE,
+		TRANS_RX_ERR_WITH_CLOSE_DWS_TIMEOUT,
+		TRANS_RX_ERR_WITH_CLOSE_COMINIT,
+		TRANS_RX_ERR_WITH_BREAK_TIMEOUT,
+		TRANS_RX_ERR_WITH_BREAK_REQUEST,
+		TRANS_RX_ERR_WITH_BREAK_RECEVIED,
+		RESERVED1,
+		TRANS_RX_ERR_WITH_CLOSE_NORMAL,
+		TRANS_RX_ERR_WITH_DATA_LEN0,
+		TRANS_RX_ERR_WITH_BAD_HASH,
+		TRANS_RX_XRDY_WLEN_ZERO_ERR,
+		TRANS_RX_SSP_FRM_LEN_ERR,
+		RESERVED2,
+		RESERVED3,
+		RESERVED4,
+		RESERVED5,
+		TRANS_RX_ERR_WITH_BAD_FRM_TYPE,
+		TRANS_RX_SMP_FRM_LEN_ERR,
+		TRANS_RX_SMP_RESP_TIMEOUT_ERR,
+		RESERVED6,
+		RESERVED7,
+		RESERVED8,
+		RESERVED9,
+		TRANS_RX_R_ERR,
+	};
+	int index, i;
+
+	for (i = 0; i < ARRAY_SIZE(trans_rx_err_code_prio); i++) {
+		index = trans_rx_err_code_prio[i] - TRANS_RX_FAIL_BASE;
+		if (err_msk & (1 << index))
+			return trans_rx_err_code_prio[i];
+	}
+	return -1;
+}
+
+static int parse_dma_tx_err_code_v2_hw(u32 err_msk)
+{
+	const u8 dma_tx_err_code_prio[] = {
+		DMA_TX_UNEXP_XFER_ERR,
+		DMA_TX_UNEXP_RETRANS_ERR,
+		DMA_TX_XFER_LEN_OVERFLOW,
+		DMA_TX_XFER_OFFSET_ERR,
+		DMA_TX_RAM_ECC_ERR,
+		DMA_TX_DIF_LEN_ALIGN_ERR,
+		DMA_TX_DIF_CRC_ERR,
+		DMA_TX_DIF_APP_ERR,
+		DMA_TX_DIF_RPP_ERR,
+		DMA_TX_DATA_SGL_OVERFLOW,
+		DMA_TX_DIF_SGL_OVERFLOW,
+	};
+	int index, i;
+
+	for (i = 0; i < ARRAY_SIZE(dma_tx_err_code_prio); i++) {
+		index = dma_tx_err_code_prio[i] - DMA_TX_ERR_BASE;
+		err_msk = err_msk & DMA_TX_ERR_MSK;
+		if (err_msk & (1 << index))
+			return dma_tx_err_code_prio[i];
+	}
+	return -1;
+}
+
+static int parse_sipc_rx_err_code_v2_hw(u32 err_msk)
+{
+	const u8 sipc_rx_err_code_prio[] = {
+		SIPC_RX_FIS_STATUS_ERR_BIT_VLD,
+		SIPC_RX_PIO_WRSETUP_STATUS_DRQ_ERR,
+		SIPC_RX_FIS_STATUS_BSY_BIT_ERR,
+		SIPC_RX_WRSETUP_LEN_ODD_ERR,
+		SIPC_RX_WRSETUP_LEN_ZERO_ERR,
+		SIPC_RX_WRDATA_LEN_NOT_MATCH_ERR,
+		SIPC_RX_NCQ_WRSETUP_OFFSET_ERR,
+		SIPC_RX_NCQ_WRSETUP_AUTO_ACTIVE_ERR,
+		SIPC_RX_SATA_UNEXP_FIS_ERR,
+		SIPC_RX_WRSETUP_ESTATUS_ERR,
+		SIPC_RX_DATA_UNDERFLOW_ERR,
+	};
+	int index, i;
+
+	for (i = 0; i < ARRAY_SIZE(sipc_rx_err_code_prio); i++) {
+		index = sipc_rx_err_code_prio[i] - SIPC_RX_ERR_BASE;
+		err_msk = err_msk & SIPC_RX_ERR_MSK;
+		if (err_msk & (1 << (index + 0x10)))
+			return sipc_rx_err_code_prio[i];
+	}
+	return -1;
+}
+
+static int parse_dma_rx_err_code_v2_hw(u32 err_msk)
+{
+	const u8 dma_rx_err_code_prio[] = {
+		DMA_RX_UNKNOWN_FRM_ERR,
+		DMA_RX_DATA_LEN_OVERFLOW,
+		DMA_RX_DATA_LEN_UNDERFLOW,
+		DMA_RX_DATA_OFFSET_ERR,
+		RESERVED10,
+		DMA_RX_SATA_FRAME_TYPE_ERR,
+		DMA_RX_RESP_BUF_OVERFLOW,
+		DMA_RX_UNEXP_RETRANS_RESP_ERR,
+		DMA_RX_UNEXP_NORM_RESP_ERR,
+		DMA_RX_UNEXP_RDFRAME_ERR,
+		DMA_RX_PIO_DATA_LEN_ERR,
+		DMA_RX_RDSETUP_STATUS_ERR,
+		DMA_RX_RDSETUP_STATUS_DRQ_ERR,
+		DMA_RX_RDSETUP_STATUS_BSY_ERR,
+		DMA_RX_RDSETUP_LEN_ODD_ERR,
+		DMA_RX_RDSETUP_LEN_ZERO_ERR,
+		DMA_RX_RDSETUP_LEN_OVER_ERR,
+		DMA_RX_RDSETUP_OFFSET_ERR,
+		DMA_RX_RDSETUP_ACTIVE_ERR,
+		DMA_RX_RDSETUP_ESTATUS_ERR,
+		DMA_RX_RAM_ECC_ERR,
+		DMA_RX_DIF_CRC_ERR,
+		DMA_RX_DIF_APP_ERR,
+		DMA_RX_DIF_RPP_ERR,
+		DMA_RX_DATA_SGL_OVERFLOW,
+		DMA_RX_DIF_SGL_OVERFLOW,
+	};
+	int index, i;
+
+	for (i = 0; i < ARRAY_SIZE(dma_rx_err_code_prio); i++) {
+		index = dma_rx_err_code_prio[i] - DMA_RX_ERR_BASE;
+		if (err_msk & (1 << index))
+			return dma_rx_err_code_prio[i];
+	}
+	return -1;
+}
+
 /* by default, task resp is complete */
 static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 			   struct sas_task *task,
-			   struct hisi_sas_slot *slot)
+			   struct hisi_sas_slot *slot,
+			   int err_phase)
 {
 	struct task_status_struct *ts = &task->task_status;
 	struct hisi_sas_err_record_v2 *err_record = slot->status_buffer;
@@ -1451,21 +1915,23 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 	u32 dma_rx_err_type = cpu_to_le32(err_record->dma_rx_err_type);
 	int error = -1;
 
-	if (dma_rx_err_type) {
-		error = ffs(dma_rx_err_type)
-			- 1 + DMA_RX_ERR_BASE;
-	} else if (sipc_rx_err_type) {
-		error = ffs(sipc_rx_err_type)
-			- 1 + SIPC_RX_ERR_BASE;
-	}  else if (dma_tx_err_type) {
-		error = ffs(dma_tx_err_type)
-			- 1 + DMA_TX_ERR_BASE;
-	} else if (trans_rx_fail_type) {
-		error = ffs(trans_rx_fail_type)
-			- 1 + TRANS_RX_FAIL_BASE;
-	} else if (trans_tx_fail_type) {
-		error = ffs(trans_tx_fail_type)
-			- 1 + TRANS_TX_FAIL_BASE;
+	if (err_phase == 1) {
+		/* error in TX phase, the priority of error is: DW2 > DW0 */
+		error = parse_dma_tx_err_code_v2_hw(dma_tx_err_type);
+		if (error == -1)
+			error = parse_trans_tx_err_code_v2_hw(
+					trans_tx_fail_type);
+	} else if (err_phase == 2) {
+		/* error in RX phase, the priority is: DW1 > DW3 > DW2 */
+		error = parse_trans_rx_err_code_v2_hw(
+					trans_rx_fail_type);
+		if (error == -1) {
+			error = parse_dma_rx_err_code_v2_hw(
+					dma_rx_err_type);
+			if (error == -1)
+				error = parse_sipc_rx_err_code_v2_hw(
+						sipc_rx_err_type);
+		}
 	}
 
 	switch (task->task_proto) {
@@ -1478,12 +1944,6 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 			ts->open_rej_reason = SAS_OREJ_NO_DEST;
 			break;
 		}
-		case TRANS_TX_OPEN_CNX_ERR_PATHWAY_BLOCKED:
-		{
-			ts->stat = SAS_OPEN_REJECT;
-			ts->open_rej_reason = SAS_OREJ_PATH_BLOCKED;
-			break;
-		}
 		case TRANS_TX_OPEN_CNX_ERR_PROTOCOL_NOT_SUPPORTED:
 		{
 			ts->stat = SAS_OPEN_REJECT;
@@ -1502,19 +1962,15 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 			ts->open_rej_reason = SAS_OREJ_BAD_DEST;
 			break;
 		}
-		case TRANS_TX_OPEN_CNX_ERR_BREAK_RCVD:
-		{
-			ts->stat = SAS_OPEN_REJECT;
-			ts->open_rej_reason = SAS_OREJ_RSVD_RETRY;
-			break;
-		}
 		case TRANS_TX_OPEN_CNX_ERR_WRONG_DESTINATION:
 		{
 			ts->stat = SAS_OPEN_REJECT;
 			ts->open_rej_reason = SAS_OREJ_WRONG_DEST;
 			break;
 		}
+		case DMA_RX_UNEXP_NORM_RESP_ERR:
 		case TRANS_TX_OPEN_CNX_ERR_ZONE_VIOLATION:
+		case DMA_RX_RESP_BUF_OVERFLOW:
 		{
 			ts->stat = SAS_OPEN_REJECT;
 			ts->open_rej_reason = SAS_OREJ_UNKNOWN;
@@ -1526,16 +1982,6 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 			ts->stat = SAS_DEV_NO_RESPONSE;
 			break;
 		}
-		case TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE:
-		{
-			ts->stat = SAS_PHY_DOWN;
-			break;
-		}
-		case TRANS_TX_OPEN_CNX_ERR_OPEN_TIMEOUT:
-		{
-			ts->stat = SAS_OPEN_TO;
-			break;
-		}
 		case DMA_RX_DATA_LEN_OVERFLOW:
 		{
 			ts->stat = SAS_DATA_OVERRUN;
@@ -1543,60 +1989,65 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 			break;
 		}
 		case DMA_RX_DATA_LEN_UNDERFLOW:
-		case SIPC_RX_DATA_UNDERFLOW_ERR:
 		{
-			ts->residual = trans_tx_fail_type;
+			ts->residual = dma_rx_err_type;
 			ts->stat = SAS_DATA_UNDERRUN;
 			break;
 		}
-		case TRANS_TX_ERR_FRAME_TXED:
-		{
-			/* This will request a retry */
-			ts->stat = SAS_QUEUE_FULL;
-			slot->abort = 1;
-			break;
-		}
 		case TRANS_TX_OPEN_FAIL_WITH_IT_NEXUS_LOSS:
 		case TRANS_TX_ERR_PHY_NOT_ENABLE:
 		case TRANS_TX_OPEN_CNX_ERR_BY_OTHER:
 		case TRANS_TX_OPEN_CNX_ERR_AIP_TIMEOUT:
+		case TRANS_TX_OPEN_CNX_ERR_BREAK_RCVD:
+		case TRANS_TX_OPEN_CNX_ERR_PATHWAY_BLOCKED:
+		case TRANS_TX_OPEN_CNX_ERR_OPEN_TIMEOUT:
 		case TRANS_TX_OPEN_RETRY_ERR_THRESHOLD_REACHED:
 		case TRANS_TX_ERR_WITH_BREAK_TIMEOUT:
 		case TRANS_TX_ERR_WITH_BREAK_REQUEST:
 		case TRANS_TX_ERR_WITH_BREAK_RECEVIED:
 		case TRANS_TX_ERR_WITH_CLOSE_TIMEOUT:
 		case TRANS_TX_ERR_WITH_CLOSE_NORMAL:
+		case TRANS_TX_ERR_WITH_CLOSE_PHYDISALE:
 		case TRANS_TX_ERR_WITH_CLOSE_DWS_TIMEOUT:
 		case TRANS_TX_ERR_WITH_CLOSE_COMINIT:
 		case TRANS_TX_ERR_WITH_NAK_RECEVIED:
 		case TRANS_TX_ERR_WITH_ACK_NAK_TIMEOUT:
-		case TRANS_TX_ERR_WITH_IPTT_CONFLICT:
 		case TRANS_TX_ERR_WITH_CREDIT_TIMEOUT:
+		case TRANS_TX_ERR_WITH_IPTT_CONFLICT:
 		case TRANS_RX_ERR_WITH_RXFRAME_CRC_ERR:
 		case TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR:
 		case TRANS_RX_ERR_WITH_RXFRAME_HAVE_ERRPRM:
+		case TRANS_RX_ERR_WITH_LINK_BUF_OVERRUN:
 		case TRANS_RX_ERR_WITH_BREAK_TIMEOUT:
 		case TRANS_RX_ERR_WITH_BREAK_REQUEST:
 		case TRANS_RX_ERR_WITH_BREAK_RECEVIED:
 		case TRANS_RX_ERR_WITH_CLOSE_NORMAL:
 		case TRANS_RX_ERR_WITH_CLOSE_DWS_TIMEOUT:
 		case TRANS_RX_ERR_WITH_CLOSE_COMINIT:
+		case TRANS_TX_ERR_FRAME_TXED:
+		case TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE:
 		case TRANS_RX_ERR_WITH_DATA_LEN0:
 		case TRANS_RX_ERR_WITH_BAD_HASH:
 		case TRANS_RX_XRDY_WLEN_ZERO_ERR:
 		case TRANS_RX_SSP_FRM_LEN_ERR:
 		case TRANS_RX_ERR_WITH_BAD_FRM_TYPE:
+		case DMA_TX_DATA_SGL_OVERFLOW:
 		case DMA_TX_UNEXP_XFER_ERR:
 		case DMA_TX_UNEXP_RETRANS_ERR:
 		case DMA_TX_XFER_LEN_OVERFLOW:
 		case DMA_TX_XFER_OFFSET_ERR:
+		case SIPC_RX_DATA_UNDERFLOW_ERR:
+		case DMA_RX_DATA_SGL_OVERFLOW:
 		case DMA_RX_DATA_OFFSET_ERR:
-		case DMA_RX_UNEXP_NORM_RESP_ERR:
-		case DMA_RX_UNEXP_RDFRAME_ERR:
+		case DMA_RX_RDSETUP_LEN_ODD_ERR:
+		case DMA_RX_RDSETUP_LEN_ZERO_ERR:
+		case DMA_RX_RDSETUP_LEN_OVER_ERR:
+		case DMA_RX_SATA_FRAME_TYPE_ERR:
 		case DMA_RX_UNKNOWN_FRM_ERR:
 		{
-			ts->stat = SAS_OPEN_REJECT;
-			ts->open_rej_reason = SAS_OREJ_UNKNOWN;
+			/* This will request a retry */
+			ts->stat = SAS_QUEUE_FULL;
+			slot->abort = 1;
 			break;
 		}
 		default:
@@ -1613,57 +2064,92 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 	case SAS_PROTOCOL_SATA | SAS_PROTOCOL_STP:
 	{
 		switch (error) {
-		case TRANS_TX_OPEN_CNX_ERR_LOW_PHY_POWER:
-		case TRANS_TX_OPEN_CNX_ERR_PATHWAY_BLOCKED:
 		case TRANS_TX_OPEN_CNX_ERR_NO_DESTINATION:
 		{
+			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_NO_DEST;
+			break;
+		}
+		case TRANS_TX_OPEN_CNX_ERR_LOW_PHY_POWER:
+		{
 			ts->resp = SAS_TASK_UNDELIVERED;
 			ts->stat = SAS_DEV_NO_RESPONSE;
 			break;
 		}
 		case TRANS_TX_OPEN_CNX_ERR_PROTOCOL_NOT_SUPPORTED:
+		{
+			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_EPROTO;
+			break;
+		}
 		case TRANS_TX_OPEN_CNX_ERR_CONNECTION_RATE_NOT_SUPPORTED:
+		{
+			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_CONN_RATE;
+			break;
+		}
 		case TRANS_TX_OPEN_CNX_ERR_BAD_DESTINATION:
-		case TRANS_TX_OPEN_CNX_ERR_BREAK_RCVD:
+		{
+			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_CONN_RATE;
+			break;
+		}
 		case TRANS_TX_OPEN_CNX_ERR_WRONG_DESTINATION:
-		case TRANS_TX_OPEN_CNX_ERR_ZONE_VIOLATION:
-		case TRANS_TX_OPEN_CNX_ERR_STP_RESOURCES_BUSY:
 		{
 			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_WRONG_DEST;
 			break;
 		}
-		case TRANS_TX_OPEN_CNX_ERR_OPEN_TIMEOUT:
+		case DMA_RX_RESP_BUF_OVERFLOW:
+		case DMA_RX_UNEXP_NORM_RESP_ERR:
+		case TRANS_TX_OPEN_CNX_ERR_ZONE_VIOLATION:
 		{
-			ts->stat = SAS_OPEN_TO;
+			ts->stat = SAS_OPEN_REJECT;
+			ts->open_rej_reason = SAS_OREJ_UNKNOWN;
 			break;
 		}
 		case DMA_RX_DATA_LEN_OVERFLOW:
 		{
 			ts->stat = SAS_DATA_OVERRUN;
+			ts->residual = 0;
+			break;
+		}
+		case DMA_RX_DATA_LEN_UNDERFLOW:
+		{
+			ts->residual = dma_rx_err_type;
+			ts->stat = SAS_DATA_UNDERRUN;
 			break;
 		}
 		case TRANS_TX_OPEN_FAIL_WITH_IT_NEXUS_LOSS:
 		case TRANS_TX_ERR_PHY_NOT_ENABLE:
 		case TRANS_TX_OPEN_CNX_ERR_BY_OTHER:
 		case TRANS_TX_OPEN_CNX_ERR_AIP_TIMEOUT:
+		case TRANS_TX_OPEN_CNX_ERR_BREAK_RCVD:
+		case TRANS_TX_OPEN_CNX_ERR_PATHWAY_BLOCKED:
+		case TRANS_TX_OPEN_CNX_ERR_OPEN_TIMEOUT:
 		case TRANS_TX_OPEN_RETRY_ERR_THRESHOLD_REACHED:
 		case TRANS_TX_ERR_WITH_BREAK_TIMEOUT:
 		case TRANS_TX_ERR_WITH_BREAK_REQUEST:
 		case TRANS_TX_ERR_WITH_BREAK_RECEVIED:
 		case TRANS_TX_ERR_WITH_CLOSE_TIMEOUT:
 		case TRANS_TX_ERR_WITH_CLOSE_NORMAL:
+		case TRANS_TX_ERR_WITH_CLOSE_PHYDISALE:
 		case TRANS_TX_ERR_WITH_CLOSE_DWS_TIMEOUT:
 		case TRANS_TX_ERR_WITH_CLOSE_COMINIT:
-		case TRANS_TX_ERR_WITH_NAK_RECEVIED:
 		case TRANS_TX_ERR_WITH_ACK_NAK_TIMEOUT:
 		case TRANS_TX_ERR_WITH_CREDIT_TIMEOUT:
+		case TRANS_TX_ERR_WITH_OPEN_BY_DES_OR_OTHERS:
 		case TRANS_TX_ERR_WITH_WAIT_RECV_TIMEOUT:
-		case TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR:
 		case TRANS_RX_ERR_WITH_RXFRAME_HAVE_ERRPRM:
+		case TRANS_RX_ERR_WITH_RXFIS_8B10B_DISP_ERR:
 		case TRANS_RX_ERR_WITH_RXFIS_DECODE_ERROR:
 		case TRANS_RX_ERR_WITH_RXFIS_CRC_ERR:
 		case TRANS_RX_ERR_WITH_RXFRAME_LENGTH_OVERRUN:
 		case TRANS_RX_ERR_WITH_RXFIS_RX_SYNCP:
+		case TRANS_RX_ERR_WITH_LINK_BUF_OVERRUN:
+		case TRANS_RX_ERR_WITH_BREAK_TIMEOUT:
+		case TRANS_RX_ERR_WITH_BREAK_REQUEST:
+		case TRANS_RX_ERR_WITH_BREAK_RECEVIED:
 		case TRANS_RX_ERR_WITH_CLOSE_NORMAL:
 		case TRANS_RX_ERR_WITH_CLOSE_PHY_DISABLE:
 		case TRANS_RX_ERR_WITH_CLOSE_DWS_TIMEOUT:
@@ -1671,7 +2157,12 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 		case TRANS_RX_ERR_WITH_DATA_LEN0:
 		case TRANS_RX_ERR_WITH_BAD_HASH:
 		case TRANS_RX_XRDY_WLEN_ZERO_ERR:
-		case TRANS_RX_SSP_FRM_LEN_ERR:
+		case TRANS_RX_ERR_WITH_BAD_FRM_TYPE:
+		case DMA_TX_DATA_SGL_OVERFLOW:
+		case DMA_TX_UNEXP_XFER_ERR:
+		case DMA_TX_UNEXP_RETRANS_ERR:
+		case DMA_TX_XFER_LEN_OVERFLOW:
+		case DMA_TX_XFER_OFFSET_ERR:
 		case SIPC_RX_FIS_STATUS_ERR_BIT_VLD:
 		case SIPC_RX_PIO_WRSETUP_STATUS_DRQ_ERR:
 		case SIPC_RX_FIS_STATUS_BSY_BIT_ERR:
@@ -1679,6 +2170,8 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 		case SIPC_RX_WRSETUP_LEN_ZERO_ERR:
 		case SIPC_RX_WRDATA_LEN_NOT_MATCH_ERR:
 		case SIPC_RX_SATA_UNEXP_FIS_ERR:
+		case DMA_RX_DATA_SGL_OVERFLOW:
+		case DMA_RX_DATA_OFFSET_ERR:
 		case DMA_RX_SATA_FRAME_TYPE_ERR:
 		case DMA_RX_UNEXP_RDFRAME_ERR:
 		case DMA_RX_PIO_DATA_LEN_ERR:
@@ -1692,8 +2185,11 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 		case DMA_RX_RDSETUP_ACTIVE_ERR:
 		case DMA_RX_RDSETUP_ESTATUS_ERR:
 		case DMA_RX_UNKNOWN_FRM_ERR:
+		case TRANS_RX_SSP_FRM_LEN_ERR:
+		case TRANS_TX_OPEN_CNX_ERR_STP_RESOURCES_BUSY:
 		{
-			ts->stat = SAS_OPEN_REJECT;
+			slot->abort = 1;
+			ts->stat = SAS_PHY_DOWN;
 			break;
 		}
 		default:
@@ -1711,8 +2207,7 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
 }
 
 static int
-slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
-		    int abort)
+slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot)
 {
 	struct sas_task *task = slot->task;
 	struct hisi_sas_device *sas_dev;
@@ -1724,6 +2219,8 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
 			hisi_hba->complete_hdr[slot->cmplt_queue];
 	struct hisi_sas_complete_v2_hdr *complete_hdr =
 			&complete_queue[slot->cmplt_queue_slot];
+	unsigned long flags;
+	int aborted;
 
 	if (unlikely(!task || !task->lldd_task || !task->dev))
 		return -EINVAL;
@@ -1732,16 +2229,23 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
 	device = task->dev;
 	sas_dev = device->lldd_dev;
 
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	aborted = task->task_state_flags & SAS_TASK_STATE_ABORTED;
 	task->task_state_flags &=
 		~(SAS_TASK_STATE_PENDING | SAS_TASK_AT_INITIATOR);
-	task->task_state_flags |= SAS_TASK_STATE_DONE;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
 
 	memset(ts, 0, sizeof(*ts));
 	ts->resp = SAS_TASK_COMPLETE;
 
-	if (unlikely(!sas_dev || abort)) {
-		if (!sas_dev)
-			dev_dbg(dev, "slot complete: port has not device\n");
+	if (unlikely(aborted)) {
+		ts->stat = SAS_ABORTED_TASK;
+		hisi_sas_slot_task_free(hisi_hba, task, slot);
+		return -1;
+	}
+
+	if (unlikely(!sas_dev)) {
+		dev_dbg(dev, "slot complete: port has no device\n");
 		ts->stat = SAS_PHY_DOWN;
 		goto out;
 	}
@@ -1755,16 +2259,19 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
 		goto out;
 	case STAT_IO_COMPLETE:
 		/* internal abort command complete */
-		ts->stat = TMF_RESP_FUNC_COMPLETE;
+		ts->stat = TMF_RESP_FUNC_SUCC;
+		del_timer(&slot->internal_abort_timer);
 		goto out;
 	case STAT_IO_NO_DEVICE:
 		ts->stat = TMF_RESP_FUNC_COMPLETE;
+		del_timer(&slot->internal_abort_timer);
 		goto out;
 	case STAT_IO_NOT_VALID:
 		/* abort single io, controller don't find
 		 * the io need to abort
 		 */
 		ts->stat = TMF_RESP_FUNC_FAILED;
+		del_timer(&slot->internal_abort_timer);
 		goto out;
 	default:
 		break;
@@ -1772,13 +2279,17 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
 
 	if ((complete_hdr->dw0 & CMPLT_HDR_ERX_MSK) &&
 		(!(complete_hdr->dw0 & CMPLT_HDR_RSPNS_XFRD_MSK))) {
+		u32 err_phase = (complete_hdr->dw0 & CMPLT_HDR_ERR_PHASE_MSK)
+				>> CMPLT_HDR_ERR_PHASE_OFF;
+
+		/* Analyse error happens on which phase TX or RX */
+		if (ERR_ON_TX_PHASE(err_phase))
+			slot_err_v2_hw(hisi_hba, task, slot, 1);
+		else if (ERR_ON_RX_PHASE(err_phase))
+			slot_err_v2_hw(hisi_hba, task, slot, 2);
 
-		slot_err_v2_hw(hisi_hba, task, slot);
-		if (unlikely(slot->abort)) {
-			queue_work(hisi_hba->wq, &slot->abort_slot);
-			/* immediately return and do not complete */
+		if (unlikely(slot->abort))
 			return ts->stat;
-		}
 		goto out;
 	}
 
@@ -1830,7 +2341,9 @@ slot_complete_v2_hw(struct hisi_hba *hisi_hba, struct hisi_sas_slot *slot,
 	}
 
 out:
-
+	spin_lock_irqsave(&task->task_state_lock, flags);
+	task->task_state_flags |= SAS_TASK_STATE_DONE;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
 	hisi_sas_slot_task_free(hisi_hba, task, slot);
 	sts = ts->stat;
 
@@ -1920,7 +2433,8 @@ static int prep_ata_v2_hw(struct hisi_hba *hisi_hba,
 	struct domain_device *parent_dev = device->parent;
 	struct hisi_sas_device *sas_dev = device->lldd_dev;
 	struct hisi_sas_cmd_hdr *hdr = slot->cmd_hdr;
-	struct hisi_sas_port *port = device->port->lldd_port;
+	struct asd_sas_port *sas_port = device->port;
+	struct hisi_sas_port *port = to_hisi_sas_port(sas_port);
 	u8 *buf_cmd;
 	int has_data = 0, rc = 0, hdr_tag = 0;
 	u32 dw1 = 0, dw2 = 0;
@@ -1947,7 +2461,8 @@ static int prep_ata_v2_hw(struct hisi_hba *hisi_hba,
 		dw1 &= ~CMD_HDR_DIR_MSK;
 	}
 
-	if (0 == task->ata_task.fis.command)
+	if ((task->ata_task.fis.command == ATA_CMD_DEV_RESET) &&
+			(task->ata_task.fis.control & ATA_SRST))
 		dw1 |= 1 << CMD_HDR_RESET_OFF;
 
 	dw1 |= (get_ata_protocol(task->ata_task.fis.command, task->data_dir))
@@ -1990,6 +2505,40 @@ static int prep_ata_v2_hw(struct hisi_hba *hisi_hba,
 	return 0;
 }
 
+static void hisi_sas_internal_abort_quirk_timeout(unsigned long data)
+{
+	struct hisi_sas_slot *slot = (struct hisi_sas_slot *)data;
+	struct hisi_sas_port *port = slot->port;
+	struct asd_sas_port *asd_sas_port;
+	struct asd_sas_phy *sas_phy;
+
+	if (!port)
+		return;
+
+	asd_sas_port = &port->sas_port;
+
+	/* Kick the hardware - send break command */
+	list_for_each_entry(sas_phy, &asd_sas_port->phy_list, port_phy_el) {
+		struct hisi_sas_phy *phy = sas_phy->lldd_phy;
+		struct hisi_hba *hisi_hba = phy->hisi_hba;
+		int phy_no = sas_phy->id;
+		u32 link_dfx2;
+
+		link_dfx2 = hisi_sas_phy_read32(hisi_hba, phy_no, LINK_DFX2);
+		if ((link_dfx2 == LINK_DFX2_RCVR_HOLD_STS_MSK) ||
+		    (link_dfx2 & LINK_DFX2_SEND_HOLD_STS_MSK)) {
+			u32 txid_auto;
+
+			txid_auto = hisi_sas_phy_read32(hisi_hba, phy_no,
+							TXID_AUTO);
+			txid_auto |= TXID_AUTO_CTB_MSK;
+			hisi_sas_phy_write32(hisi_hba, phy_no, TXID_AUTO,
+					     txid_auto);
+			return;
+		}
+	}
+}
+
 static int prep_abort_v2_hw(struct hisi_hba *hisi_hba,
 		struct hisi_sas_slot *slot,
 		int device_id, int abort_flag, int tag_to_abort)
@@ -1998,6 +2547,13 @@ static int prep_abort_v2_hw(struct hisi_hba *hisi_hba,
 	struct domain_device *dev = task->dev;
 	struct hisi_sas_cmd_hdr *hdr = slot->cmd_hdr;
 	struct hisi_sas_port *port = slot->port;
+	struct timer_list *timer = &slot->internal_abort_timer;
+
+	/* setup the quirk timer */
+	setup_timer(timer, hisi_sas_internal_abort_quirk_timeout,
+		    (unsigned long)slot);
+	/* Set the timeout to 10ms less than internal abort timeout */
+	mod_timer(timer, jiffies + msecs_to_jiffies(100));
 
 	/* dw0 */
 	hdr->dw0 = cpu_to_le32((5 << CMD_HDR_CMD_OFF) | /*abort*/
@@ -2018,8 +2574,8 @@ static int prep_abort_v2_hw(struct hisi_hba *hisi_hba,
 
 static int phy_up_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
 {
-	int i, res = 0;
-	u32 context, port_id, link_rate, hard_phy_linkrate;
+	int i, res = IRQ_HANDLED;
+	u32 port_id, link_rate, hard_phy_linkrate;
 	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
 	struct asd_sas_phy *sas_phy = &phy->sas_phy;
 	struct device *dev = &hisi_hba->pdev->dev;
@@ -2028,9 +2584,7 @@ static int phy_up_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
 
 	hisi_sas_phy_write32(hisi_hba, phy_no, PHYCTRL_PHY_ENA_MSK, 1);
 
-	/* Check for SATA dev */
-	context = hisi_sas_read32(hisi_hba, PHY_CONTEXT);
-	if (context & (1 << phy_no))
+	if (is_sata_phy_v2_hw(hisi_hba, phy_no))
 		goto end;
 
 	if (phy_no == 8) {
@@ -2106,7 +2660,6 @@ static bool check_any_wideports_v2_hw(struct hisi_hba *hisi_hba)
 
 static int phy_down_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
 {
-	int res = 0;
 	u32 phy_state, sl_ctrl, txid_auto;
 	struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
 	struct hisi_sas_port *port = phy->port;
@@ -2131,7 +2684,7 @@ static int phy_down_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
 	hisi_sas_phy_write32(hisi_hba, phy_no, CHL_INT0, CHL_INT0_NOT_RDY_MSK);
 	hisi_sas_phy_write32(hisi_hba, phy_no, PHYCTRL_NOT_RDY_MSK, 0);
 
-	return res;
+	return IRQ_HANDLED;
 }
 
 static irqreturn_t int_phy_updown_v2_hw(int irq_no, void *p)
@@ -2139,35 +2692,58 @@ static irqreturn_t int_phy_updown_v2_hw(int irq_no, void *p)
 	struct hisi_hba *hisi_hba = p;
 	u32 irq_msk;
 	int phy_no = 0;
-	irqreturn_t res = IRQ_HANDLED;
 
 	irq_msk = (hisi_sas_read32(hisi_hba, HGC_INVLD_DQE_INFO)
 		   >> HGC_INVLD_DQE_INFO_FB_CH0_OFF) & 0x1ff;
 	while (irq_msk) {
 		if (irq_msk  & 1) {
-			u32 irq_value = hisi_sas_phy_read32(hisi_hba, phy_no,
-							    CHL_INT0);
+			u32 reg_value = hisi_sas_phy_read32(hisi_hba, phy_no,
+					    CHL_INT0);
 
-			if (irq_value & CHL_INT0_SL_PHY_ENABLE_MSK)
+			switch (reg_value & (CHL_INT0_NOT_RDY_MSK |
+					CHL_INT0_SL_PHY_ENABLE_MSK)) {
+
+			case CHL_INT0_SL_PHY_ENABLE_MSK:
 				/* phy up */
-				if (phy_up_v2_hw(phy_no, hisi_hba)) {
-					res = IRQ_NONE;
-					goto end;
-				}
+				if (phy_up_v2_hw(phy_no, hisi_hba) ==
+				    IRQ_NONE)
+					return IRQ_NONE;
+				break;
 
-			if (irq_value & CHL_INT0_NOT_RDY_MSK)
+			case CHL_INT0_NOT_RDY_MSK:
 				/* phy down */
-				if (phy_down_v2_hw(phy_no, hisi_hba)) {
-					res = IRQ_NONE;
-					goto end;
+				if (phy_down_v2_hw(phy_no, hisi_hba) ==
+				    IRQ_NONE)
+					return IRQ_NONE;
+				break;
+
+			case (CHL_INT0_NOT_RDY_MSK |
+					CHL_INT0_SL_PHY_ENABLE_MSK):
+				reg_value = hisi_sas_read32(hisi_hba,
+						PHY_STATE);
+				if (reg_value & BIT(phy_no)) {
+					/* phy up */
+					if (phy_up_v2_hw(phy_no, hisi_hba) ==
+					    IRQ_NONE)
+						return IRQ_NONE;
+				} else {
+					/* phy down */
+					if (phy_down_v2_hw(phy_no, hisi_hba) ==
+					    IRQ_NONE)
+						return IRQ_NONE;
 				}
+				break;
+
+			default:
+				break;
+			}
+
 		}
 		irq_msk >>= 1;
 		phy_no++;
 	}
 
-end:
-	return res;
+	return IRQ_HANDLED;
 }
 
 static void phy_bcast_v2_hw(int phy_no, struct hisi_hba *hisi_hba)
@@ -2342,94 +2918,105 @@ static void multi_bit_ecc_error_process_v2_hw(struct hisi_hba *hisi_hba,
 
 	if (irq_value & BIT(SAS_ECC_INTR_DQE_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_DQE_ECC_ADDR);
-		panic("%s: hgc_dqe_accbad_intr (0x%x) found: \
+		dev_warn(dev, "hgc_dqe_accbad_intr (0x%x) found: \
 		       Ram address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_DQE_ECC_MB_ADDR_MSK) >>
 		      HGC_DQE_ECC_MB_ADDR_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_IOST_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_IOST_ECC_ADDR);
-		panic("%s: hgc_iost_accbad_intr (0x%x) found: \
+		dev_warn(dev, "hgc_iost_accbad_intr (0x%x) found: \
 		       Ram address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_IOST_ECC_MB_ADDR_MSK) >>
 		      HGC_IOST_ECC_MB_ADDR_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_ITCT_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_ITCT_ECC_ADDR);
-		panic("%s: hgc_itct_accbad_intr (0x%x) found: \
+		dev_warn(dev,"hgc_itct_accbad_intr (0x%x) found: \
 		       Ram address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_ITCT_ECC_MB_ADDR_MSK) >>
 		      HGC_ITCT_ECC_MB_ADDR_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_IOSTLIST_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
-		panic("%s: hgc_iostl_accbad_intr (0x%x) found: \
+		dev_warn(dev, "hgc_iostl_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_LM_DFX_STATUS2_IOSTLIST_MSK) >>
 		      HGC_LM_DFX_STATUS2_IOSTLIST_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_ITCTLIST_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_LM_DFX_STATUS2);
-		panic("%s: hgc_itctl_accbad_intr (0x%x) found: \
+		dev_warn(dev, "hgc_itctl_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_LM_DFX_STATUS2_ITCTLIST_MSK) >>
 		      HGC_LM_DFX_STATUS2_ITCTLIST_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_CQE_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_CQE_ECC_ADDR);
-		panic("%s: hgc_cqe_accbad_intr (0x%x) found: \
+		dev_warn(dev, "hgc_cqe_accbad_intr (0x%x) found: \
 		       Ram address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_CQE_ECC_MB_ADDR_MSK) >>
 		      HGC_CQE_ECC_MB_ADDR_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM0_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
-		panic("%s: rxm_mem0_accbad_intr (0x%x) found: \
+		dev_warn(dev, "rxm_mem0_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_RXM_DFX_STATUS14_MEM0_MSK) >>
 		      HGC_RXM_DFX_STATUS14_MEM0_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM1_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
-		panic("%s: rxm_mem1_accbad_intr (0x%x) found: \
+		dev_warn(dev, "rxm_mem1_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_RXM_DFX_STATUS14_MEM1_MSK) >>
 		      HGC_RXM_DFX_STATUS14_MEM1_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM2_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS14);
-		panic("%s: rxm_mem2_accbad_intr (0x%x) found: \
+		dev_warn(dev, "rxm_mem2_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_RXM_DFX_STATUS14_MEM2_MSK) >>
 		      HGC_RXM_DFX_STATUS14_MEM2_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
 	if (irq_value & BIT(SAS_ECC_INTR_NCQ_MEM3_ECC_MB_OFF)) {
 		reg_val = hisi_sas_read32(hisi_hba, HGC_RXM_DFX_STATUS15);
-		panic("%s: rxm_mem3_accbad_intr (0x%x) found: \
+		dev_warn(dev, "rxm_mem3_accbad_intr (0x%x) found: \
 		       memory address is 0x%08X\n",
-		      dev_name(dev), irq_value,
+		      irq_value,
 		      (reg_val & HGC_RXM_DFX_STATUS15_MEM3_MSK) >>
 		      HGC_RXM_DFX_STATUS15_MEM3_OFF);
+		queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 	}
 
+	return;
 }
 
 static irqreturn_t fatal_ecc_int_v2_hw(int irq_no, void *p)
@@ -2487,23 +3074,27 @@ static irqreturn_t fatal_axi_int_v2_hw(int irq_no, void *p)
 		if (irq_value & BIT(ENT_INT_SRC3_WP_DEPTH_OFF)) {
 			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
 					1 << ENT_INT_SRC3_WP_DEPTH_OFF);
-			panic("%s: write pointer and depth error (0x%x) \
+			dev_warn(dev, "write pointer and depth error (0x%x) \
 			       found!\n",
-			      dev_name(dev), irq_value);
+			       irq_value);
+			queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 		}
 
 		if (irq_value & BIT(ENT_INT_SRC3_IPTT_SLOT_NOMATCH_OFF)) {
 			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
 					 1 <<
 					 ENT_INT_SRC3_IPTT_SLOT_NOMATCH_OFF);
-			panic("%s: iptt no match slot error (0x%x) found!\n",
-			      dev_name(dev), irq_value);
+			dev_warn(dev, "iptt no match slot error (0x%x) found!\n",
+			      irq_value);
+			queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 		}
 
-		if (irq_value & BIT(ENT_INT_SRC3_RP_DEPTH_OFF))
-			panic("%s: read pointer and depth error (0x%x) \
+		if (irq_value & BIT(ENT_INT_SRC3_RP_DEPTH_OFF)) {
+			dev_warn(dev, "read pointer and depth error (0x%x) \
 			       found!\n",
-			      dev_name(dev), irq_value);
+			       irq_value);
+			queue_work(hisi_hba->wq, &hisi_hba->rst_work);
+		}
 
 		if (irq_value & BIT(ENT_INT_SRC3_AXI_OFF)) {
 			int i;
@@ -2514,10 +3105,11 @@ static irqreturn_t fatal_axi_int_v2_hw(int irq_no, void *p)
 						    HGC_AXI_FIFO_ERR_INFO);
 
 			for (i = 0; i < AXI_ERR_NR; i++) {
-				if (err_value & BIT(i))
-					panic("%s: %s (0x%x) found!\n",
-					       dev_name(dev),
+				if (err_value & BIT(i)) {
+					dev_warn(dev, "%s (0x%x) found!\n",
 					      axi_err_info[i], irq_value);
+					queue_work(hisi_hba->wq, &hisi_hba->rst_work);
+				}
 			}
 		}
 
@@ -2530,10 +3122,11 @@ static irqreturn_t fatal_axi_int_v2_hw(int irq_no, void *p)
 						    HGC_AXI_FIFO_ERR_INFO);
 
 			for (i = 0; i < FIFO_ERR_NR; i++) {
-				if (err_value & BIT(AXI_ERR_NR + i))
-					panic("%s: %s (0x%x) found!\n",
-					      dev_name(dev),
+				if (err_value & BIT(AXI_ERR_NR + i)) {
+					dev_warn(dev, "%s (0x%x) found!\n",
 					      fifo_err_info[i], irq_value);
+					queue_work(hisi_hba->wq, &hisi_hba->rst_work);
+				}
 			}
 
 		}
@@ -2541,15 +3134,17 @@ static irqreturn_t fatal_axi_int_v2_hw(int irq_no, void *p)
 		if (irq_value & BIT(ENT_INT_SRC3_LM_OFF)) {
 			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
 					1 << ENT_INT_SRC3_LM_OFF);
-			panic("%s: LM add/fetch list error (0x%x) found!\n",
-			      dev_name(dev), irq_value);
+			dev_warn(dev, "LM add/fetch list error (0x%x) found!\n",
+			      irq_value);
+			queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 		}
 
 		if (irq_value & BIT(ENT_INT_SRC3_ABT_OFF)) {
 			hisi_sas_write32(hisi_hba, ENT_INT_SRC3,
 					1 << ENT_INT_SRC3_ABT_OFF);
-			panic("%s: SAS_HGC_ABT fetch LM list error (0x%x) found!\n",
-			      dev_name(dev), irq_value);
+			dev_warn(dev, "SAS_HGC_ABT fetch LM list error (0x%x) found!\n",
+			      irq_value);
+			queue_work(hisi_hba->wq, &hisi_hba->rst_work);
 		}
 	}
 
@@ -2568,6 +3163,9 @@ static void cq_tasklet_v2_hw(unsigned long val)
 	u32 rd_point = cq->rd_point, wr_point, dev_id;
 	int queue = cq->id;
 
+	if (unlikely(hisi_hba->reject_stp_links_msk))
+		phys_try_accept_stp_links_v2_hw(hisi_hba);
+
 	complete_queue = hisi_hba->complete_hdr[queue];
 
 	spin_lock(&hisi_hba->lock);
@@ -2600,7 +3198,7 @@ static void cq_tasklet_v2_hw(unsigned long val)
 				slot = &hisi_hba->slot_info[iptt];
 				slot->cmplt_queue_slot = rd_point;
 				slot->cmplt_queue = queue;
-				slot_complete_v2_hw(hisi_hba, slot, 0);
+				slot_complete_v2_hw(hisi_hba, slot);
 
 				act_tmp &= ~(1 << ncq_tag_count);
 				ncq_tag_count = ffs(act_tmp);
@@ -2610,7 +3208,7 @@ static void cq_tasklet_v2_hw(unsigned long val)
 			slot = &hisi_hba->slot_info[iptt];
 			slot->cmplt_queue_slot = rd_point;
 			slot->cmplt_queue = queue;
-			slot_complete_v2_hw(hisi_hba, slot, 0);
+			slot_complete_v2_hw(hisi_hba, slot);
 		}
 
 		if (++rd_point >= HISI_SAS_QUEUE_SLOTS)
@@ -2842,6 +3440,8 @@ static int hisi_sas_v2_init(struct hisi_hba *hisi_hba)
 {
 	int rc;
 
+	memset(hisi_hba->sata_dev_bitmap, 0, sizeof(hisi_hba->sata_dev_bitmap));
+
 	rc = hw_init_v2_hw(hisi_hba);
 	if (rc)
 		return rc;
@@ -2850,7 +3450,88 @@ static int hisi_sas_v2_init(struct hisi_hba *hisi_hba)
 	if (rc)
 		return rc;
 
-	phys_init_v2_hw(hisi_hba);
+	return 0;
+}
+
+static void interrupt_disable_v2_hw(struct hisi_hba *hisi_hba)
+{
+	struct platform_device *pdev = hisi_hba->pdev;
+	int i;
+
+	for (i = 0; i < hisi_hba->queue_count; i++)
+		hisi_sas_write32(hisi_hba, OQ0_INT_SRC_MSK + 0x4 * i, 0x1);
+
+	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK1, 0xffffffff);
+	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK2, 0xffffffff);
+	hisi_sas_write32(hisi_hba, ENT_INT_SRC_MSK3, 0xffffffff);
+	hisi_sas_write32(hisi_hba, SAS_ECC_INTR_MSK, 0xffffffff);
+
+	for (i = 0; i < hisi_hba->n_phy; i++) {
+		hisi_sas_phy_write32(hisi_hba, i, CHL_INT1_MSK, 0xffffffff);
+		hisi_sas_phy_write32(hisi_hba, i, CHL_INT2_MSK, 0xffffffff);
+	}
+
+	for (i = 0; i < 128; i++)
+		synchronize_irq(platform_get_irq(pdev, i));
+}
+
+static int soft_reset_v2_hw(struct hisi_hba *hisi_hba)
+{
+	struct device *dev = &hisi_hba->pdev->dev;
+	u32 old_state, state;
+	int rc, cnt;
+	int phy_no;
+
+	old_state = hisi_sas_read32(hisi_hba, PHY_STATE);
+
+	interrupt_disable_v2_hw(hisi_hba);
+	hisi_sas_write32(hisi_hba, DLVRY_QUEUE_ENABLE, 0x0);
+
+	stop_phys_v2_hw(hisi_hba);
+
+	mdelay(10);
+
+	hisi_sas_write32(hisi_hba, AXI_MASTER_CFG_BASE + AM_CTRL_GLOBAL, 0x1);
+
+	/* wait until bus idle */
+	cnt = 0;
+	while (1) {
+		u32 status = hisi_sas_read32_relaxed(hisi_hba,
+				AXI_MASTER_CFG_BASE + AM_CURR_TRANS_RETURN);
+
+		if (status == 0x3)
+			break;
+
+		udelay(10);
+		if (cnt++ > 10) {
+			dev_info(dev, "wait axi bus state to idle timeout!\n");
+			return -1;
+		}
+	}
+
+	hisi_sas_init_mem(hisi_hba);
+
+	rc = hw_init_v2_hw(hisi_hba);
+	if (rc)
+		return rc;
+
+	phys_reject_stp_links_v2_hw(hisi_hba);
+
+	/* Re-enable the PHYs */
+	for (phy_no = 0; phy_no < hisi_hba->n_phy; phy_no++) {
+		struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no];
+		struct asd_sas_phy *sas_phy = &phy->sas_phy;
+
+		if (sas_phy->enabled)
+			start_phy_v2_hw(hisi_hba, phy_no);
+	}
+
+	/* Wait for the PHYs to come up and read the PHY state */
+	msleep(1000);
+
+	state = hisi_sas_read32(hisi_hba, PHY_STATE);
+
+	hisi_sas_rescan_topology(hisi_hba, old_state, state);
 
 	return 0;
 }
@@ -2870,6 +3551,7 @@ static const struct hisi_sas_hw hisi_sas_v2_hw = {
 	.get_free_slot = get_free_slot_v2_hw,
 	.start_delivery = start_delivery_v2_hw,
 	.slot_complete = slot_complete_v2_hw,
+	.phys_init = phys_init_v2_hw,
 	.phy_enable = enable_phy_v2_hw,
 	.phy_disable = disable_phy_v2_hw,
 	.phy_hard_reset = phy_hard_reset_v2_hw,
@@ -2877,6 +3559,7 @@ static const struct hisi_sas_hw hisi_sas_v2_hw = {
 	.phy_get_max_linkrate = phy_get_max_linkrate_v2_hw,
 	.max_command_entries = HISI_SAS_COMMAND_ENTRIES_V2_HW,
 	.complete_hdr_size = sizeof(struct hisi_sas_complete_v2_hdr),
+	.soft_reset = soft_reset_v2_hw,
 };
 
 static int hisi_sas_v2_probe(struct platform_device *pdev)
diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 9d659aaace15..73daace478cb 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -60,7 +60,7 @@
  * HPSA_DRIVER_VERSION must be 3 byte values (0-255) separated by '.'
  * with an optional trailing '-' followed by a byte value (0-255).
  */
-#define HPSA_DRIVER_VERSION "3.4.16-0"
+#define HPSA_DRIVER_VERSION "3.4.18-0"
 #define DRIVER_NAME "HP HPSA Driver (v " HPSA_DRIVER_VERSION ")"
 #define HPSA "hpsa"
 
@@ -108,10 +108,12 @@ static const struct pci_device_id hpsa_pci_device_id[] = {
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3354},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3355},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSF,     0x103C, 0x3356},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103c, 0x1920},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1921},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1922},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1923},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1924},
+	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103c, 0x1925},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1926},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1928},
 	{PCI_VENDOR_ID_HP,     PCI_DEVICE_ID_HP_CISSH,     0x103C, 0x1929},
@@ -171,10 +173,12 @@ static struct board_type products[] = {
 	{0x3354103C, "Smart Array P420i", &SA5_access},
 	{0x3355103C, "Smart Array P220i", &SA5_access},
 	{0x3356103C, "Smart Array P721m", &SA5_access},
+	{0x1920103C, "Smart Array P430i", &SA5_access},
 	{0x1921103C, "Smart Array P830i", &SA5_access},
 	{0x1922103C, "Smart Array P430", &SA5_access},
 	{0x1923103C, "Smart Array P431", &SA5_access},
 	{0x1924103C, "Smart Array P830", &SA5_access},
+	{0x1925103C, "Smart Array P831", &SA5_access},
 	{0x1926103C, "Smart Array P731m", &SA5_access},
 	{0x1928103C, "Smart Array P230i", &SA5_access},
 	{0x1929103C, "Smart Array P530", &SA5_access},
diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index 2c92dabb55f6..26cd3c28186a 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -3910,12 +3910,6 @@ static int ibmvfc_alloc_target(struct ibmvfc_host *vhost, u64 scsi_id)
 	spin_unlock_irqrestore(vhost->host->host_lock, flags);
 
 	tgt = mempool_alloc(vhost->tgt_pool, GFP_NOIO);
-	if (!tgt) {
-		dev_err(vhost->dev, "Target allocation failure for scsi id %08llx\n",
-			scsi_id);
-		return -ENOMEM;
-	}
-
 	memset(tgt, 0, sizeof(*tgt));
 	tgt->scsi_id = scsi_id;
 	tgt->new_scsi_id = scsi_id;
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 5d5e272fd815..b0c68d24db01 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -820,7 +820,7 @@ static int ipr_set_pcix_cmd_reg(struct ipr_ioa_cfg *ioa_cfg)
 }
 
 /**
- * ipr_sata_eh_done - done function for aborted SATA commands
+ * __ipr_sata_eh_done - done function for aborted SATA commands
  * @ipr_cmd:	ipr command struct
  *
  * This function is invoked for ops generated to SATA
@@ -829,19 +829,41 @@ static int ipr_set_pcix_cmd_reg(struct ipr_ioa_cfg *ioa_cfg)
  * Return value:
  * 	none
  **/
-static void ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
+static void __ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
 {
 	struct ata_queued_cmd *qc = ipr_cmd->qc;
 	struct ipr_sata_port *sata_port = qc->ap->private_data;
 
 	qc->err_mask |= AC_ERR_OTHER;
 	sata_port->ioasa.status |= ATA_BUSY;
-	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 	ata_qc_complete(qc);
+	if (ipr_cmd->eh_comp)
+		complete(ipr_cmd->eh_comp);
+	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 }
 
 /**
- * ipr_scsi_eh_done - mid-layer done function for aborted ops
+ * ipr_sata_eh_done - done function for aborted SATA commands
+ * @ipr_cmd:	ipr command struct
+ *
+ * This function is invoked for ops generated to SATA
+ * devices which are being aborted.
+ *
+ * Return value:
+ * 	none
+ **/
+static void ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
+{
+	struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
+	unsigned long hrrq_flags;
+
+	spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
+	__ipr_sata_eh_done(ipr_cmd);
+	spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
+}
+
+/**
+ * __ipr_scsi_eh_done - mid-layer done function for aborted ops
  * @ipr_cmd:	ipr command struct
  *
  * This function is invoked by the interrupt handler for
@@ -850,7 +872,7 @@ static void ipr_sata_eh_done(struct ipr_cmnd *ipr_cmd)
  * Return value:
  * 	none
  **/
-static void ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
+static void __ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
 {
 	struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
 
@@ -864,6 +886,26 @@ static void ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
 }
 
 /**
+ * ipr_scsi_eh_done - mid-layer done function for aborted ops
+ * @ipr_cmd:	ipr command struct
+ *
+ * This function is invoked by the interrupt handler for
+ * ops generated by the SCSI mid-layer which are being aborted.
+ *
+ * Return value:
+ * 	none
+ **/
+static void ipr_scsi_eh_done(struct ipr_cmnd *ipr_cmd)
+{
+	unsigned long hrrq_flags;
+	struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
+
+	spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
+	__ipr_scsi_eh_done(ipr_cmd);
+	spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
+}
+
+/**
  * ipr_fail_all_ops - Fails all outstanding ops.
  * @ioa_cfg:	ioa config struct
  *
@@ -890,9 +932,9 @@ static void ipr_fail_all_ops(struct ipr_ioa_cfg *ioa_cfg)
 				cpu_to_be32(IPR_DRIVER_ILID);
 
 			if (ipr_cmd->scsi_cmd)
-				ipr_cmd->done = ipr_scsi_eh_done;
+				ipr_cmd->done = __ipr_scsi_eh_done;
 			else if (ipr_cmd->qc)
-				ipr_cmd->done = ipr_sata_eh_done;
+				ipr_cmd->done = __ipr_sata_eh_done;
 
 			ipr_trc_hook(ipr_cmd, IPR_TRACE_FINISH,
 				     IPR_IOASC_IOA_WAS_RESET);
@@ -5006,6 +5048,42 @@ static int ipr_match_lun(struct ipr_cmnd *ipr_cmd, void *device)
 }
 
 /**
+ * ipr_cmnd_is_free - Check if a command is free or not
+ * @ipr_cmd	ipr command struct
+ *
+ * Returns:
+ *	true / false
+ **/
+static bool ipr_cmnd_is_free(struct ipr_cmnd *ipr_cmd)
+{
+	struct ipr_cmnd *loop_cmd;
+
+	list_for_each_entry(loop_cmd, &ipr_cmd->hrrq->hrrq_free_q, queue) {
+		if (loop_cmd == ipr_cmd)
+			return true;
+	}
+
+	return false;
+}
+
+/**
+ * ipr_match_res - Match function for specified resource entry
+ * @ipr_cmd:	ipr command struct
+ * @resource:	resource entry to match
+ *
+ * Returns:
+ *	1 if command matches sdev / 0 if command does not match sdev
+ **/
+static int ipr_match_res(struct ipr_cmnd *ipr_cmd, void *resource)
+{
+	struct ipr_resource_entry *res = resource;
+
+	if (res && ipr_cmd->ioarcb.res_handle == res->res_handle)
+		return 1;
+	return 0;
+}
+
+/**
  * ipr_wait_for_ops - Wait for matching commands to complete
  * @ipr_cmd:	ipr command struct
  * @device:		device to match (sdev)
@@ -5018,7 +5096,7 @@ static int ipr_wait_for_ops(struct ipr_ioa_cfg *ioa_cfg, void *device,
 			    int (*match)(struct ipr_cmnd *, void *))
 {
 	struct ipr_cmnd *ipr_cmd;
-	int wait;
+	int wait, i;
 	unsigned long flags;
 	struct ipr_hrr_queue *hrrq;
 	signed long timeout = IPR_ABORT_TASK_TIMEOUT;
@@ -5030,10 +5108,13 @@ static int ipr_wait_for_ops(struct ipr_ioa_cfg *ioa_cfg, void *device,
 
 		for_each_hrrq(hrrq, ioa_cfg) {
 			spin_lock_irqsave(hrrq->lock, flags);
-			list_for_each_entry(ipr_cmd, &hrrq->hrrq_pending_q, queue) {
-				if (match(ipr_cmd, device)) {
-					ipr_cmd->eh_comp = &comp;
-					wait++;
+			for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
+				ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
+				if (!ipr_cmnd_is_free(ipr_cmd)) {
+					if (match(ipr_cmd, device)) {
+						ipr_cmd->eh_comp = &comp;
+						wait++;
+					}
 				}
 			}
 			spin_unlock_irqrestore(hrrq->lock, flags);
@@ -5047,10 +5128,13 @@ static int ipr_wait_for_ops(struct ipr_ioa_cfg *ioa_cfg, void *device,
 
 				for_each_hrrq(hrrq, ioa_cfg) {
 					spin_lock_irqsave(hrrq->lock, flags);
-					list_for_each_entry(ipr_cmd, &hrrq->hrrq_pending_q, queue) {
-						if (match(ipr_cmd, device)) {
-							ipr_cmd->eh_comp = NULL;
-							wait++;
+					for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
+						ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
+						if (!ipr_cmnd_is_free(ipr_cmd)) {
+							if (match(ipr_cmd, device)) {
+								ipr_cmd->eh_comp = NULL;
+								wait++;
+							}
 						}
 					}
 					spin_unlock_irqrestore(hrrq->lock, flags);
@@ -5179,7 +5263,7 @@ static int ipr_sata_reset(struct ata_link *link, unsigned int *classes,
 	struct ipr_ioa_cfg *ioa_cfg = sata_port->ioa_cfg;
 	struct ipr_resource_entry *res;
 	unsigned long lock_flags = 0;
-	int rc = -ENXIO;
+	int rc = -ENXIO, ret;
 
 	ENTER;
 	spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
@@ -5193,9 +5277,19 @@ static int ipr_sata_reset(struct ata_link *link, unsigned int *classes,
 	if (res) {
 		rc = ipr_device_reset(ioa_cfg, res);
 		*classes = res->ata_class;
-	}
+		spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
+
+		ret = ipr_wait_for_ops(ioa_cfg, res, ipr_match_res);
+		if (ret != SUCCESS) {
+			spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
+			ipr_initiate_ioa_reset(ioa_cfg, IPR_SHUTDOWN_ABBREV);
+			spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
+
+			wait_event(ioa_cfg->reset_wait_q, !ioa_cfg->in_reset_reload);
+		}
+	} else
+		spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
 
-	spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
 	LEAVE;
 	return rc;
 }
@@ -5217,16 +5311,13 @@ static int __ipr_eh_dev_reset(struct scsi_cmnd *scsi_cmd)
 	struct ipr_ioa_cfg *ioa_cfg;
 	struct ipr_resource_entry *res;
 	struct ata_port *ap;
-	int rc = 0;
+	int rc = 0, i;
 	struct ipr_hrr_queue *hrrq;
 
 	ENTER;
 	ioa_cfg = (struct ipr_ioa_cfg *) scsi_cmd->device->host->hostdata;
 	res = scsi_cmd->device->hostdata;
 
-	if (!res)
-		return FAILED;
-
 	/*
 	 * If we are currently going through reset/reload, return failed. This will force the
 	 * mid-layer to call ipr_eh_host_reset, which will then go to sleep and wait for the
@@ -5239,14 +5330,17 @@ static int __ipr_eh_dev_reset(struct scsi_cmnd *scsi_cmd)
 
 	for_each_hrrq(hrrq, ioa_cfg) {
 		spin_lock(&hrrq->_lock);
-		list_for_each_entry(ipr_cmd, &hrrq->hrrq_pending_q, queue) {
+		for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
+			ipr_cmd = ioa_cfg->ipr_cmnd_list[i];
+
 			if (ipr_cmd->ioarcb.res_handle == res->res_handle) {
-				if (ipr_cmd->scsi_cmd)
-					ipr_cmd->done = ipr_scsi_eh_done;
-				if (ipr_cmd->qc)
-					ipr_cmd->done = ipr_sata_eh_done;
-				if (ipr_cmd->qc &&
-				    !(ipr_cmd->qc->flags & ATA_QCFLAG_FAILED)) {
+				if (!ipr_cmd->qc)
+					continue;
+				if (ipr_cmnd_is_free(ipr_cmd))
+					continue;
+
+				ipr_cmd->done = ipr_sata_eh_done;
+				if (!(ipr_cmd->qc->flags & ATA_QCFLAG_FAILED)) {
 					ipr_cmd->qc->err_mask |= AC_ERR_TIMEOUT;
 					ipr_cmd->qc->flags |= ATA_QCFLAG_FAILED;
 				}
@@ -5262,19 +5356,6 @@ static int __ipr_eh_dev_reset(struct scsi_cmnd *scsi_cmd)
 		spin_unlock_irq(scsi_cmd->device->host->host_lock);
 		ata_std_error_handler(ap);
 		spin_lock_irq(scsi_cmd->device->host->host_lock);
-
-		for_each_hrrq(hrrq, ioa_cfg) {
-			spin_lock(&hrrq->_lock);
-			list_for_each_entry(ipr_cmd,
-					    &hrrq->hrrq_pending_q, queue) {
-				if (ipr_cmd->ioarcb.res_handle ==
-				    res->res_handle) {
-					rc = -EIO;
-					break;
-				}
-			}
-			spin_unlock(&hrrq->_lock);
-		}
 	} else
 		rc = ipr_device_reset(ioa_cfg, res);
 	res->resetting_device = 0;
@@ -5288,15 +5369,24 @@ static int ipr_eh_dev_reset(struct scsi_cmnd *cmd)
 {
 	int rc;
 	struct ipr_ioa_cfg *ioa_cfg;
+	struct ipr_resource_entry *res;
 
 	ioa_cfg = (struct ipr_ioa_cfg *) cmd->device->host->hostdata;
+	res = cmd->device->hostdata;
+
+	if (!res)
+		return FAILED;
 
 	spin_lock_irq(cmd->device->host->host_lock);
 	rc = __ipr_eh_dev_reset(cmd);
 	spin_unlock_irq(cmd->device->host->host_lock);
 
-	if (rc == SUCCESS)
-		rc = ipr_wait_for_ops(ioa_cfg, cmd->device, ipr_match_lun);
+	if (rc == SUCCESS) {
+		if (ipr_is_gata(res) && res->sata_port)
+			rc = ipr_wait_for_ops(ioa_cfg, res, ipr_match_res);
+		else
+			rc = ipr_wait_for_ops(ioa_cfg, cmd->device, ipr_match_lun);
+	}
 
 	return rc;
 }
@@ -5393,7 +5483,7 @@ static int ipr_cancel_op(struct scsi_cmnd *scsi_cmd)
 	struct ipr_resource_entry *res;
 	struct ipr_cmd_pkt *cmd_pkt;
 	u32 ioasc, int_reg;
-	int op_found = 0;
+	int i, op_found = 0;
 	struct ipr_hrr_queue *hrrq;
 
 	ENTER;
@@ -5422,11 +5512,12 @@ static int ipr_cancel_op(struct scsi_cmnd *scsi_cmd)
 
 	for_each_hrrq(hrrq, ioa_cfg) {
 		spin_lock(&hrrq->_lock);
-		list_for_each_entry(ipr_cmd, &hrrq->hrrq_pending_q, queue) {
-			if (ipr_cmd->scsi_cmd == scsi_cmd) {
-				ipr_cmd->done = ipr_scsi_eh_done;
-				op_found = 1;
-				break;
+		for (i = hrrq->min_cmd_id; i <= hrrq->max_cmd_id; i++) {
+			if (ioa_cfg->ipr_cmnd_list[i]->scsi_cmd == scsi_cmd) {
+				if (!ipr_cmnd_is_free(ioa_cfg->ipr_cmnd_list[i])) {
+					op_found = 1;
+					break;
+				}
 			}
 		}
 		spin_unlock(&hrrq->_lock);
@@ -5917,7 +6008,7 @@ static int ipr_build_ioadl(struct ipr_ioa_cfg *ioa_cfg,
 }
 
 /**
- * ipr_erp_done - Process completion of ERP for a device
+ * __ipr_erp_done - Process completion of ERP for a device
  * @ipr_cmd:		ipr command struct
  *
  * This function copies the sense buffer into the scsi_cmd
@@ -5926,7 +6017,7 @@ static int ipr_build_ioadl(struct ipr_ioa_cfg *ioa_cfg,
  * Return value:
  * 	nothing
  **/
-static void ipr_erp_done(struct ipr_cmnd *ipr_cmd)
+static void __ipr_erp_done(struct ipr_cmnd *ipr_cmd)
 {
 	struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd;
 	struct ipr_resource_entry *res = scsi_cmd->device->hostdata;
@@ -5947,8 +6038,30 @@ static void ipr_erp_done(struct ipr_cmnd *ipr_cmd)
 		res->in_erp = 0;
 	}
 	scsi_dma_unmap(ipr_cmd->scsi_cmd);
-	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 	scsi_cmd->scsi_done(scsi_cmd);
+	if (ipr_cmd->eh_comp)
+		complete(ipr_cmd->eh_comp);
+	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
+}
+
+/**
+ * ipr_erp_done - Process completion of ERP for a device
+ * @ipr_cmd:		ipr command struct
+ *
+ * This function copies the sense buffer into the scsi_cmd
+ * struct and pushes the scsi_done function.
+ *
+ * Return value:
+ * 	nothing
+ **/
+static void ipr_erp_done(struct ipr_cmnd *ipr_cmd)
+{
+	struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
+	unsigned long hrrq_flags;
+
+	spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
+	__ipr_erp_done(ipr_cmd);
+	spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
 }
 
 /**
@@ -5983,7 +6096,7 @@ static void ipr_reinit_ipr_cmnd_for_erp(struct ipr_cmnd *ipr_cmd)
 }
 
 /**
- * ipr_erp_request_sense - Send request sense to a device
+ * __ipr_erp_request_sense - Send request sense to a device
  * @ipr_cmd:	ipr command struct
  *
  * This function sends a request sense to a device as a result
@@ -5992,13 +6105,13 @@ static void ipr_reinit_ipr_cmnd_for_erp(struct ipr_cmnd *ipr_cmd)
  * Return value:
  * 	nothing
  **/
-static void ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
+static void __ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
 {
 	struct ipr_cmd_pkt *cmd_pkt = &ipr_cmd->ioarcb.cmd_pkt;
 	u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc);
 
 	if (IPR_IOASC_SENSE_KEY(ioasc) > 0) {
-		ipr_erp_done(ipr_cmd);
+		__ipr_erp_done(ipr_cmd);
 		return;
 	}
 
@@ -6019,6 +6132,26 @@ static void ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
 }
 
 /**
+ * ipr_erp_request_sense - Send request sense to a device
+ * @ipr_cmd:	ipr command struct
+ *
+ * This function sends a request sense to a device as a result
+ * of a check condition.
+ *
+ * Return value:
+ * 	nothing
+ **/
+static void ipr_erp_request_sense(struct ipr_cmnd *ipr_cmd)
+{
+	struct ipr_hrr_queue *hrrq = ipr_cmd->hrrq;
+	unsigned long hrrq_flags;
+
+	spin_lock_irqsave(&hrrq->_lock, hrrq_flags);
+	__ipr_erp_request_sense(ipr_cmd);
+	spin_unlock_irqrestore(&hrrq->_lock, hrrq_flags);
+}
+
+/**
  * ipr_erp_cancel_all - Send cancel all to a device
  * @ipr_cmd:	ipr command struct
  *
@@ -6041,7 +6174,7 @@ static void ipr_erp_cancel_all(struct ipr_cmnd *ipr_cmd)
 	ipr_reinit_ipr_cmnd_for_erp(ipr_cmd);
 
 	if (!scsi_cmd->device->simple_tags) {
-		ipr_erp_request_sense(ipr_cmd);
+		__ipr_erp_request_sense(ipr_cmd);
 		return;
 	}
 
@@ -6261,7 +6394,7 @@ static void ipr_erp_start(struct ipr_ioa_cfg *ioa_cfg,
 	u32 masked_ioasc = ioasc & IPR_IOASC_IOASC_MASK;
 
 	if (!res) {
-		ipr_scsi_eh_done(ipr_cmd);
+		__ipr_scsi_eh_done(ipr_cmd);
 		return;
 	}
 
@@ -6343,8 +6476,10 @@ static void ipr_erp_start(struct ipr_ioa_cfg *ioa_cfg,
 	}
 
 	scsi_dma_unmap(ipr_cmd->scsi_cmd);
-	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 	scsi_cmd->scsi_done(scsi_cmd);
+	if (ipr_cmd->eh_comp)
+		complete(ipr_cmd->eh_comp);
+	list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 }
 
 /**
@@ -6370,8 +6505,10 @@ static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd)
 		scsi_dma_unmap(scsi_cmd);
 
 		spin_lock_irqsave(ipr_cmd->hrrq->lock, lock_flags);
-		list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 		scsi_cmd->scsi_done(scsi_cmd);
+		if (ipr_cmd->eh_comp)
+			complete(ipr_cmd->eh_comp);
+		list_add_tail(&ipr_cmd->queue, &ipr_cmd->hrrq->hrrq_free_q);
 		spin_unlock_irqrestore(ipr_cmd->hrrq->lock, lock_flags);
 	} else {
 		spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h
index b7d2e98eb45b..e98a87a65335 100644
--- a/drivers/scsi/ipr.h
+++ b/drivers/scsi/ipr.h
@@ -39,8 +39,8 @@
 /*
  * Literals
  */
-#define IPR_DRIVER_VERSION "2.6.3"
-#define IPR_DRIVER_DATE "(October 17, 2015)"
+#define IPR_DRIVER_VERSION "2.6.4"
+#define IPR_DRIVER_DATE "(March 14, 2017)"
 
 /*
  * IPR_MAX_CMD_PER_LUN: This defines the maximum number of outstanding
diff --git a/drivers/scsi/isci/init.c b/drivers/scsi/isci/init.c
index 0b5b5db0d0f8..45371179ab87 100644
--- a/drivers/scsi/isci/init.c
+++ b/drivers/scsi/isci/init.c
@@ -272,7 +272,6 @@ static void isci_unregister(struct isci_host *isci_host)
 		return;
 
 	shost = to_shost(isci_host);
-	scsi_remove_host(shost);
 	sas_unregister_ha(&isci_host->sas_ha);
 
 	sas_remove_host(shost);
diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c
index 0e67621477a8..a808e8ef1d08 100644
--- a/drivers/scsi/libfc/fc_fcp.c
+++ b/drivers/scsi/libfc/fc_fcp.c
@@ -154,7 +154,7 @@ static struct fc_fcp_pkt *fc_fcp_pkt_alloc(struct fc_lport *lport, gfp_t gfp)
 		memset(fsp, 0, sizeof(*fsp));
 		fsp->lp = lport;
 		fsp->xfer_ddp = FC_XID_UNKNOWN;
-		atomic_set(&fsp->ref_cnt, 1);
+		refcount_set(&fsp->ref_cnt, 1);
 		init_timer(&fsp->timer);
 		fsp->timer.data = (unsigned long)fsp;
 		INIT_LIST_HEAD(&fsp->list);
@@ -175,7 +175,7 @@ static struct fc_fcp_pkt *fc_fcp_pkt_alloc(struct fc_lport *lport, gfp_t gfp)
  */
 static void fc_fcp_pkt_release(struct fc_fcp_pkt *fsp)
 {
-	if (atomic_dec_and_test(&fsp->ref_cnt)) {
+	if (refcount_dec_and_test(&fsp->ref_cnt)) {
 		struct fc_fcp_internal *si = fc_get_scsi_internal(fsp->lp);
 
 		mempool_free(fsp, si->scsi_pkt_pool);
@@ -188,7 +188,7 @@ static void fc_fcp_pkt_release(struct fc_fcp_pkt *fsp)
  */
 static void fc_fcp_pkt_hold(struct fc_fcp_pkt *fsp)
 {
-	atomic_inc(&fsp->ref_cnt);
+	refcount_inc(&fsp->ref_cnt);
 }
 
 /**
diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c
index aa76f36abe03..2fd0ec651170 100644
--- a/drivers/scsi/libfc/fc_lport.c
+++ b/drivers/scsi/libfc/fc_lport.c
@@ -887,8 +887,6 @@ out:
 static void fc_lport_recv_els_req(struct fc_lport *lport,
 				  struct fc_frame *fp)
 {
-	void (*recv)(struct fc_lport *, struct fc_frame *);
-
 	mutex_lock(&lport->lp_mutex);
 
 	/*
@@ -902,31 +900,31 @@ static void fc_lport_recv_els_req(struct fc_lport *lport,
 		/*
 		 * Check opcode.
 		 */
-		recv = fc_rport_recv_req;
 		switch (fc_frame_payload_op(fp)) {
 		case ELS_FLOGI:
 			if (!lport->point_to_multipoint)
-				recv = fc_lport_recv_flogi_req;
+				fc_lport_recv_flogi_req(lport, fp);
 			break;
 		case ELS_LOGO:
 			if (fc_frame_sid(fp) == FC_FID_FLOGI)
-				recv = fc_lport_recv_logo_req;
+				fc_lport_recv_logo_req(lport, fp);
 			break;
 		case ELS_RSCN:
-			recv = lport->tt.disc_recv_req;
+			lport->tt.disc_recv_req(lport, fp);
 			break;
 		case ELS_ECHO:
-			recv = fc_lport_recv_echo_req;
+			fc_lport_recv_echo_req(lport, fp);
 			break;
 		case ELS_RLIR:
-			recv = fc_lport_recv_rlir_req;
+			fc_lport_recv_rlir_req(lport, fp);
 			break;
 		case ELS_RNID:
-			recv = fc_lport_recv_rnid_req;
+			fc_lport_recv_rnid_req(lport, fp);
+			break;
+		default:
+			fc_rport_recv_req(lport, fp);
 			break;
 		}
-
-		recv(lport, fp);
 	}
 	mutex_unlock(&lport->lp_mutex);
 }
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 894b1e3ebd56..dd6828f7f772 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -517,13 +517,13 @@ static void iscsi_free_task(struct iscsi_task *task)
 
 void __iscsi_get_task(struct iscsi_task *task)
 {
-	atomic_inc(&task->refcount);
+	refcount_inc(&task->refcount);
 }
 EXPORT_SYMBOL_GPL(__iscsi_get_task);
 
 void __iscsi_put_task(struct iscsi_task *task)
 {
-	if (atomic_dec_and_test(&task->refcount))
+	if (refcount_dec_and_test(&task->refcount))
 		iscsi_free_task(task);
 }
 EXPORT_SYMBOL_GPL(__iscsi_put_task);
@@ -749,7 +749,7 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 	 * released by the lld when it has transmitted the task for
 	 * pdus we do not expect a response for.
 	 */
-	atomic_set(&task->refcount, 1);
+	refcount_set(&task->refcount, 1);
 	task->conn = conn;
 	task->sc = NULL;
 	INIT_LIST_HEAD(&task->running);
@@ -1638,7 +1638,7 @@ static inline struct iscsi_task *iscsi_alloc_task(struct iscsi_conn *conn,
 	sc->SCp.phase = conn->session->age;
 	sc->SCp.ptr = (char *) task;
 
-	atomic_set(&task->refcount, 1);
+	refcount_set(&task->refcount, 1);
 	task->state = ISCSI_TASK_PENDING;
 	task->conn = conn;
 	task->sc = sc;
diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c
index 15ef8e2e685c..64e9cdda1c3c 100644
--- a/drivers/scsi/libsas/sas_init.c
+++ b/drivers/scsi/libsas/sas_init.c
@@ -566,13 +566,6 @@ sas_domain_attach_transport(struct sas_domain_function_template *dft)
 }
 EXPORT_SYMBOL_GPL(sas_domain_attach_transport);
 
-
-void sas_domain_release_transport(struct scsi_transport_template *stt)
-{
-	sas_release_transport(stt);
-}
-EXPORT_SYMBOL_GPL(sas_domain_release_transport);
-
 /* ---------- SAS Class register/unregister ---------- */
 
 static int __init sas_class_init(void)
diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
index 9bd55bce83af..87e5079d816b 100644
--- a/drivers/scsi/libsas/sas_scsi_host.c
+++ b/drivers/scsi/libsas/sas_scsi_host.c
@@ -491,9 +491,6 @@ int sas_eh_abort_handler(struct scsi_cmnd *cmd)
 	struct Scsi_Host *host = cmd->device->host;
 	struct sas_internal *i = to_sas_internal(host->transportt);
 
-	if (current != host->ehandler)
-		return FAILED;
-
 	if (!i->dft->lldd_abort_task)
 		return FAILED;
 
@@ -616,8 +613,6 @@ static void sas_eh_handle_sas_errors(struct Scsi_Host *shost, struct list_head *
 		SAS_DPRINTK("trying to find task 0x%p\n", task);
 		res = sas_scsi_find_task(task);
 
-		cmd->eh_eflags = 0;
-
 		switch (res) {
 		case TASK_IS_DONE:
 			SAS_DPRINTK("%s: task 0x%p is done\n", __func__,
diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c
index 513fd07715cd..4830370bfab1 100644
--- a/drivers/scsi/lpfc/lpfc_attr.c
+++ b/drivers/scsi/lpfc/lpfc_attr.c
@@ -181,7 +181,7 @@ lpfc_nvme_info_show(struct device *dev, struct device_attribute *attr,
 				wwn_to_u64(vport->fc_nodename.u.wwn),
 				phba->targetport->port_id);
 
-		len += snprintf(buf + len, PAGE_SIZE,
+		len += snprintf(buf + len, PAGE_SIZE - len,
 				"\nNVME Target: Statistics\n");
 		tgtp = (struct lpfc_nvmet_tgtport *)phba->targetport->private;
 		len += snprintf(buf+len, PAGE_SIZE-len,
@@ -326,7 +326,7 @@ lpfc_nvme_info_show(struct device *dev, struct device_attribute *attr,
 	}
 	spin_unlock_irq(shost->host_lock);
 
-	len += snprintf(buf + len, PAGE_SIZE, "\nNVME Statistics\n");
+	len += snprintf(buf + len, PAGE_SIZE - len, "\nNVME Statistics\n");
 	len += snprintf(buf+len, PAGE_SIZE-len,
 			"LS: Xmt %016llx Cmpl %016llx\n",
 			phba->fc4NvmeLsRequests,
diff --git a/drivers/scsi/mac_esp.c b/drivers/scsi/mac_esp.c
index bb567d3b0693..cdb61eaa2d1f 100644
--- a/drivers/scsi/mac_esp.c
+++ b/drivers/scsi/mac_esp.c
@@ -55,6 +55,7 @@ struct mac_esp_priv {
 	int error;
 };
 static struct esp *esp_chips[2];
+static DEFINE_SPINLOCK(esp_chips_lock);
 
 #define MAC_ESP_GET_PRIV(esp) ((struct mac_esp_priv *) \
 			       platform_get_drvdata((struct platform_device *) \
@@ -562,15 +563,18 @@ static int esp_mac_probe(struct platform_device *dev)
 	}
 
 	host->irq = IRQ_MAC_SCSI;
-	esp_chips[dev->id] = esp;
-	mb();
-	if (esp_chips[!dev->id] == NULL) {
-		err = request_irq(host->irq, mac_scsi_esp_intr, 0, "ESP", NULL);
-		if (err < 0) {
-			esp_chips[dev->id] = NULL;
-			goto fail_free_priv;
-		}
+
+	/* The request_irq() call is intended to succeed for the first device
+	 * and fail for the second device.
+	 */
+	err = request_irq(host->irq, mac_scsi_esp_intr, 0, "ESP", NULL);
+	spin_lock(&esp_chips_lock);
+	if (err < 0 && esp_chips[!dev->id] == NULL) {
+		spin_unlock(&esp_chips_lock);
+		goto fail_free_priv;
 	}
+	esp_chips[dev->id] = esp;
+	spin_unlock(&esp_chips_lock);
 
 	err = scsi_esp_register(esp, &dev->dev);
 	if (err)
@@ -579,8 +583,13 @@ static int esp_mac_probe(struct platform_device *dev)
 	return 0;
 
 fail_free_irq:
-	if (esp_chips[!dev->id] == NULL)
-		free_irq(host->irq, esp);
+	spin_lock(&esp_chips_lock);
+	esp_chips[dev->id] = NULL;
+	if (esp_chips[!dev->id] == NULL) {
+		spin_unlock(&esp_chips_lock);
+		free_irq(host->irq, NULL);
+	} else
+		spin_unlock(&esp_chips_lock);
 fail_free_priv:
 	kfree(mep);
 fail_free_command_block:
@@ -599,9 +608,13 @@ static int esp_mac_remove(struct platform_device *dev)
 
 	scsi_esp_unregister(esp);
 
+	spin_lock(&esp_chips_lock);
 	esp_chips[dev->id] = NULL;
-	if (!(esp_chips[0] || esp_chips[1]))
+	if (esp_chips[!dev->id] == NULL) {
+		spin_unlock(&esp_chips_lock);
 		free_irq(irq, NULL);
+	} else
+		spin_unlock(&esp_chips_lock);
 
 	kfree(mep);
 
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c
index 5b7aec5d575a..18039bba26c4 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_base.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_base.c
@@ -1025,7 +1025,6 @@ _base_interrupt(int irq, void *bus_id)
 				    0 : ioc->reply_free_host_index + 1;
 				ioc->reply_free[ioc->reply_free_host_index] =
 				    cpu_to_le32(reply);
-				wmb();
 				writel(ioc->reply_free_host_index,
 				    &ioc->chip->ReplyFreeHostIndex);
 			}
@@ -1074,7 +1073,6 @@ _base_interrupt(int irq, void *bus_id)
 		return IRQ_NONE;
 	}
 
-	wmb();
 	if (ioc->is_warpdrive) {
 		writel(reply_q->reply_post_host_index,
 		ioc->reply_post_host_index[msix_index]);
diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
index 919ba2bb15f1..a5d872664257 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
@@ -8283,7 +8283,6 @@ static void scsih_remove(struct pci_dev *pdev)
 	}
 
 	sas_remove_host(shost);
-	scsi_remove_host(shost);
 	mpt3sas_base_detach(ioc);
 	spin_lock(&gioc_lock);
 	list_del(&ioc->list);
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 8280046fd1f0..4e047b5001a6 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -642,7 +642,6 @@ static void mvs_pci_remove(struct pci_dev *pdev)
 	tasklet_kill(&((struct mvs_prv_info *)sha->lldd_ha)->mv_tasklet);
 #endif
 
-	scsi_remove_host(mvi->shost);
 	sas_unregister_ha(sha);
 	sas_remove_host(mvi->shost);
 
diff --git a/drivers/scsi/mvumi.c b/drivers/scsi/mvumi.c
index 247df5e79b71..fe97401ad192 100644
--- a/drivers/scsi/mvumi.c
+++ b/drivers/scsi/mvumi.c
@@ -210,39 +210,27 @@ static int mvumi_make_sgl(struct mvumi_hba *mhba, struct scsi_cmnd *scmd,
 	unsigned int sgnum = scsi_sg_count(scmd);
 	dma_addr_t busaddr;
 
-	if (sgnum) {
-		sg = scsi_sglist(scmd);
-		*sg_count = pci_map_sg(mhba->pdev, sg, sgnum,
-				(int) scmd->sc_data_direction);
-		if (*sg_count > mhba->max_sge) {
-			dev_err(&mhba->pdev->dev, "sg count[0x%x] is bigger "
-						"than max sg[0x%x].\n",
-						*sg_count, mhba->max_sge);
-			return -1;
-		}
-		for (i = 0; i < *sg_count; i++) {
-			busaddr = sg_dma_address(&sg[i]);
-			m_sg->baseaddr_l = cpu_to_le32(lower_32_bits(busaddr));
-			m_sg->baseaddr_h = cpu_to_le32(upper_32_bits(busaddr));
-			m_sg->flags = 0;
-			sgd_setsz(mhba, m_sg, cpu_to_le32(sg_dma_len(&sg[i])));
-			if ((i + 1) == *sg_count)
-				m_sg->flags |= 1U << mhba->eot_flag;
-
-			sgd_inc(mhba, m_sg);
-		}
-	} else {
-		scmd->SCp.dma_handle = scsi_bufflen(scmd) ?
-			pci_map_single(mhba->pdev, scsi_sglist(scmd),
-				scsi_bufflen(scmd),
-				(int) scmd->sc_data_direction)
-			: 0;
-		busaddr = scmd->SCp.dma_handle;
+	sg = scsi_sglist(scmd);
+	*sg_count = pci_map_sg(mhba->pdev, sg, sgnum,
+			       (int) scmd->sc_data_direction);
+	if (*sg_count > mhba->max_sge) {
+		dev_err(&mhba->pdev->dev,
+			"sg count[0x%x] is bigger than max sg[0x%x].\n",
+			*sg_count, mhba->max_sge);
+		pci_unmap_sg(mhba->pdev, sg, sgnum,
+			     (int) scmd->sc_data_direction);
+		return -1;
+	}
+	for (i = 0; i < *sg_count; i++) {
+		busaddr = sg_dma_address(&sg[i]);
 		m_sg->baseaddr_l = cpu_to_le32(lower_32_bits(busaddr));
 		m_sg->baseaddr_h = cpu_to_le32(upper_32_bits(busaddr));
-		m_sg->flags = 1U << mhba->eot_flag;
-		sgd_setsz(mhba, m_sg, cpu_to_le32(scsi_bufflen(scmd)));
-		*sg_count = 1;
+		m_sg->flags = 0;
+		sgd_setsz(mhba, m_sg, cpu_to_le32(sg_dma_len(&sg[i])));
+		if ((i + 1) == *sg_count)
+			m_sg->flags |= 1U << mhba->eot_flag;
+
+		sgd_inc(mhba, m_sg);
 	}
 
 	return 0;
@@ -1350,21 +1338,10 @@ static void mvumi_complete_cmd(struct mvumi_hba *mhba, struct mvumi_cmd *cmd,
 		break;
 	}
 
-	if (scsi_bufflen(scmd)) {
-		if (scsi_sg_count(scmd)) {
-			pci_unmap_sg(mhba->pdev,
-				scsi_sglist(scmd),
-				scsi_sg_count(scmd),
-				(int) scmd->sc_data_direction);
-		} else {
-			pci_unmap_single(mhba->pdev,
-				scmd->SCp.dma_handle,
-				scsi_bufflen(scmd),
-				(int) scmd->sc_data_direction);
-
-			scmd->SCp.dma_handle = 0;
-		}
-	}
+	if (scsi_bufflen(scmd))
+		pci_unmap_sg(mhba->pdev, scsi_sglist(scmd),
+			     scsi_sg_count(scmd),
+			     (int) scmd->sc_data_direction);
 	cmd->scmd->scsi_done(scmd);
 	mvumi_return_cmd(mhba, cmd);
 }
@@ -2171,19 +2148,9 @@ static enum blk_eh_timer_return mvumi_timed_out(struct scsi_cmnd *scmd)
 	scmd->result = (DRIVER_INVALID << 24) | (DID_ABORT << 16);
 	scmd->SCp.ptr = NULL;
 	if (scsi_bufflen(scmd)) {
-		if (scsi_sg_count(scmd)) {
-			pci_unmap_sg(mhba->pdev,
-				scsi_sglist(scmd),
-				scsi_sg_count(scmd),
-				(int)scmd->sc_data_direction);
-		} else {
-			pci_unmap_single(mhba->pdev,
-				scmd->SCp.dma_handle,
-				scsi_bufflen(scmd),
-				(int)scmd->sc_data_direction);
-
-			scmd->SCp.dma_handle = 0;
-		}
+		pci_unmap_sg(mhba->pdev, scsi_sglist(scmd),
+			     scsi_sg_count(scmd),
+			     (int)scmd->sc_data_direction);
 	}
 	mvumi_return_cmd(mhba, cmd);
 	spin_unlock_irqrestore(mhba->shost->host_lock, flags);
diff --git a/drivers/scsi/osd/osd_uld.c b/drivers/scsi/osd/osd_uld.c
index e0ce5d2fd14d..ed948025112c 100644
--- a/drivers/scsi/osd/osd_uld.c
+++ b/drivers/scsi/osd/osd_uld.c
@@ -464,14 +464,15 @@ static int osd_probe(struct device *dev)
 	/* hold one more reference to the scsi_device that will get released
 	 * in __release, in case a logout is happening while fs is mounted
 	 */
-	scsi_device_get(scsi_device);
+	if (scsi_device_get(scsi_device))
+		goto err_put_disk;
 	osd_dev_init(&oud->od, scsi_device);
 
 	/* Detect the OSD Version */
 	error = __detect_osd(oud);
 	if (error) {
 		OSD_ERR("osd detection failed, non-compatible OSD device\n");
-		goto err_put_disk;
+		goto err_put_sdev;
 	}
 
 	/* init the char-device for communication with user-mode */
@@ -508,8 +509,9 @@ static int osd_probe(struct device *dev)
 
 err_put_cdev:
 	cdev_del(&oud->cdev);
-err_put_disk:
+err_put_sdev:
 	scsi_device_put(scsi_device);
+err_put_disk:
 	put_disk(disk);
 err_free_osd:
 	dev_set_drvdata(dev, NULL);
@@ -524,10 +526,9 @@ static int osd_remove(struct device *dev)
 	struct scsi_device *scsi_device = to_scsi_device(dev);
 	struct osd_uld_device *oud = dev_get_drvdata(dev);
 
-	if (!oud || (oud->od.scsi_device != scsi_device)) {
-		OSD_ERR("Half cooked osd-device %p,%p || %p!=%p",
-			dev, oud, oud ? oud->od.scsi_device : NULL,
-			scsi_device);
+	if (oud->od.scsi_device != scsi_device) {
+		OSD_ERR("Half cooked osd-device %p, || %p!=%p",
+			dev, oud->od.scsi_device, scsi_device);
 	}
 
 	device_unregister(&oud->class_dev);
diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c
index 417368ccb686..034b2f7d1135 100644
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -1088,7 +1088,6 @@ static void pm8001_pci_remove(struct pci_dev *pdev)
 	struct pm8001_hba_info *pm8001_ha;
 	int i, j;
 	pm8001_ha = sha->lldd_ha;
-	scsi_remove_host(pm8001_ha->shost);
 	sas_unregister_ha(sha);
 	sas_remove_host(pm8001_ha->shost);
 	list_del(&pm8001_ha->list);
diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c
index 49e70a383afa..a4aadf5f4dc6 100644
--- a/drivers/scsi/pmcraid.c
+++ b/drivers/scsi/pmcraid.c
@@ -77,7 +77,7 @@ static atomic_t pmcraid_adapter_count = ATOMIC_INIT(0);
  */
 static unsigned int pmcraid_major;
 static struct class *pmcraid_class;
-DECLARE_BITMAP(pmcraid_minor, PMCRAID_MAX_ADAPTERS);
+static DECLARE_BITMAP(pmcraid_minor, PMCRAID_MAX_ADAPTERS);
 
 /*
  * Module parameters
@@ -175,7 +175,7 @@ static int pmcraid_slave_alloc(struct scsi_device *scsi_dev)
 			if (fw_version <= PMCRAID_FW_VERSION_1)
 				target = temp->cfg_entry.unique_flags1;
 			else
-				target = temp->cfg_entry.array_id & 0xFF;
+				target = le16_to_cpu(temp->cfg_entry.array_id) & 0xFF;
 
 			if (target > PMCRAID_MAX_VSET_TARGETS)
 				continue;
@@ -330,7 +330,7 @@ static void pmcraid_init_cmdblk(struct pmcraid_cmd *cmd, int index)
 		ioarcb->request_flags0 = 0;
 		ioarcb->request_flags1 = 0;
 		ioarcb->cmd_timeout = 0;
-		ioarcb->ioarcb_bus_addr &= (~0x1FULL);
+		ioarcb->ioarcb_bus_addr &= cpu_to_le64(~0x1FULL);
 		ioarcb->ioadl_bus_addr = 0;
 		ioarcb->ioadl_length = 0;
 		ioarcb->data_transfer_length = 0;
@@ -345,7 +345,7 @@ static void pmcraid_init_cmdblk(struct pmcraid_cmd *cmd, int index)
 	cmd->scsi_cmd = NULL;
 	cmd->release = 0;
 	cmd->completion_req = 0;
-	cmd->sense_buffer = 0;
+	cmd->sense_buffer = NULL;
 	cmd->sense_buffer_dma = 0;
 	cmd->dma_handle = 0;
 	init_timer(&cmd->timer);
@@ -898,8 +898,7 @@ static void _pmcraid_fire_command(struct pmcraid_cmd *cmd)
 
 	/* driver writes lower 32-bit value of IOARCB address only */
 	mb();
-	iowrite32(le32_to_cpu(cmd->ioa_cb->ioarcb.ioarcb_bus_addr),
-		  pinstance->ioarrin);
+	iowrite32(le64_to_cpu(cmd->ioa_cb->ioarcb.ioarcb_bus_addr), pinstance->ioarrin);
 }
 
 /**
@@ -1051,7 +1050,7 @@ static void pmcraid_get_fwversion(struct pmcraid_cmd *cmd)
 					offsetof(struct pmcraid_ioarcb,
 						add_data.u.ioadl[0]));
 	ioarcb->ioadl_length = cpu_to_le32(sizeof(struct pmcraid_ioadl_desc));
-	ioarcb->ioarcb_bus_addr &= ~(0x1FULL);
+	ioarcb->ioarcb_bus_addr &= cpu_to_le64(~(0x1FULL));
 
 	ioarcb->request_flags0 |= NO_LINK_DESCS;
 	ioarcb->data_transfer_length = cpu_to_le32(data_size);
@@ -1077,7 +1076,7 @@ static void pmcraid_identify_hrrq(struct pmcraid_cmd *cmd)
 	struct pmcraid_ioarcb *ioarcb = &cmd->ioa_cb->ioarcb;
 	int index = cmd->hrrq_index;
 	__be64 hrrq_addr = cpu_to_be64(pinstance->hrrq_start_bus_addr[index]);
-	u32 hrrq_size = cpu_to_be32(sizeof(u32) * PMCRAID_MAX_CMD);
+	__be32 hrrq_size = cpu_to_be32(sizeof(u32) * PMCRAID_MAX_CMD);
 	void (*done_function)(struct pmcraid_cmd *);
 
 	pmcraid_reinit_cmdblk(cmd);
@@ -1202,7 +1201,7 @@ static struct pmcraid_cmd *pmcraid_init_hcam
 
 	ioadl[0].flags |= IOADL_FLAGS_READ_LAST;
 	ioadl[0].data_len = cpu_to_le32(rcb_size);
-	ioadl[0].address = cpu_to_le32(dma);
+	ioadl[0].address = cpu_to_le64(dma);
 
 	cmd->cmd_done = cmd_done;
 	return cmd;
@@ -1237,7 +1236,13 @@ static void pmcraid_prepare_cancel_cmd(
 )
 {
 	struct pmcraid_ioarcb *ioarcb = &cmd->ioa_cb->ioarcb;
-	__be64 ioarcb_addr = cmd_to_cancel->ioa_cb->ioarcb.ioarcb_bus_addr;
+	__be64 ioarcb_addr;
+
+	/* IOARCB address of the command to be cancelled is given in
+	 * cdb[2]..cdb[9] is Big-Endian format. Note that length bits in
+	 * IOARCB address are not masked.
+	 */
+	ioarcb_addr = cpu_to_be64(le64_to_cpu(cmd_to_cancel->ioa_cb->ioarcb.ioarcb_bus_addr));
 
 	/* Get the resource handle to where the command to be aborted has been
 	 * sent.
@@ -1247,11 +1252,6 @@ static void pmcraid_prepare_cancel_cmd(
 	memset(ioarcb->cdb, 0, PMCRAID_MAX_CDB_LEN);
 	ioarcb->cdb[0] = PMCRAID_ABORT_CMD;
 
-	/* IOARCB address of the command to be cancelled is given in
-	 * cdb[2]..cdb[9] is Big-Endian format. Note that length bits in
-	 * IOARCB address are not masked.
-	 */
-	ioarcb_addr = cpu_to_be64(ioarcb_addr);
 	memcpy(&(ioarcb->cdb[2]), &ioarcb_addr, sizeof(ioarcb_addr));
 }
 
@@ -1493,7 +1493,7 @@ static int pmcraid_notify_ccn(struct pmcraid_instance *pinstance)
 {
 	return pmcraid_notify_aen(pinstance,
 				pinstance->ccn.msg,
-				pinstance->ccn.hcam->data_len +
+				le32_to_cpu(pinstance->ccn.hcam->data_len) +
 				sizeof(struct pmcraid_hcam_hdr));
 }
 
@@ -1508,7 +1508,7 @@ static int pmcraid_notify_ldn(struct pmcraid_instance *pinstance)
 {
 	return pmcraid_notify_aen(pinstance,
 				pinstance->ldn.msg,
-				pinstance->ldn.hcam->data_len +
+				le32_to_cpu(pinstance->ldn.hcam->data_len) +
 				sizeof(struct pmcraid_hcam_hdr));
 }
 
@@ -1556,10 +1556,10 @@ static void pmcraid_handle_config_change(struct pmcraid_instance *pinstance)
 
 	pmcraid_info("CCN(%x): %x timestamp: %llx type: %x lost: %x flags: %x \
 		 res: %x:%x:%x:%x\n",
-		 pinstance->ccn.hcam->ilid,
+		 le32_to_cpu(pinstance->ccn.hcam->ilid),
 		 pinstance->ccn.hcam->op_code,
-		((pinstance->ccn.hcam->timestamp1) |
-		((pinstance->ccn.hcam->timestamp2 & 0xffffffffLL) << 32)),
+		(le32_to_cpu(pinstance->ccn.hcam->timestamp1) |
+		((le32_to_cpu(pinstance->ccn.hcam->timestamp2) & 0xffffffffLL) << 32)),
 		 pinstance->ccn.hcam->notification_type,
 		 pinstance->ccn.hcam->notification_lost,
 		 pinstance->ccn.hcam->flags,
@@ -1570,7 +1570,7 @@ static void pmcraid_handle_config_change(struct pmcraid_instance *pinstance)
 		 RES_IS_VSET(*cfg_entry) ?
 			(fw_version <= PMCRAID_FW_VERSION_1 ?
 				cfg_entry->unique_flags1 :
-					cfg_entry->array_id & 0xFF) :
+				le16_to_cpu(cfg_entry->array_id) & 0xFF) :
 			RES_TARGET(cfg_entry->resource_address),
 		 RES_LUN(cfg_entry->resource_address));
 
@@ -1658,7 +1658,7 @@ static void pmcraid_handle_config_change(struct pmcraid_instance *pinstance)
 			if (fw_version <= PMCRAID_FW_VERSION_1)
 				res->cfg_entry.unique_flags1 &= 0x7F;
 			else
-				res->cfg_entry.array_id &= 0xFF;
+				res->cfg_entry.array_id &= cpu_to_le16(0xFF);
 			res->change_detected = RES_CHANGE_DEL;
 			res->cfg_entry.resource_handle =
 				PMCRAID_INVALID_RES_HANDLE;
@@ -1716,8 +1716,8 @@ static void pmcraid_ioasc_logger(u32 ioasc, struct pmcraid_cmd *cmd)
 	/* log the error string */
 	pmcraid_err("cmd [%x] for resource %x failed with %x(%s)\n",
 		cmd->ioa_cb->ioarcb.cdb[0],
-		cmd->ioa_cb->ioarcb.resource_handle,
-		le32_to_cpu(ioasc), error_info->error_string);
+		le32_to_cpu(cmd->ioa_cb->ioarcb.resource_handle),
+		ioasc, error_info->error_string);
 }
 
 /**
@@ -2034,7 +2034,7 @@ static void pmcraid_fail_outstanding_cmds(struct pmcraid_instance *pinstance)
 		cmd->ioa_cb->ioasa.ioasc =
 			cpu_to_le32(PMCRAID_IOASC_IOA_WAS_RESET);
 		cmd->ioa_cb->ioasa.ilid =
-			cpu_to_be32(PMCRAID_DRIVER_ILID);
+			cpu_to_le32(PMCRAID_DRIVER_ILID);
 
 		/* In case the command timer is still running */
 		del_timer(&cmd->timer);
@@ -2373,46 +2373,43 @@ static int pmcraid_reset_reload(
 		spin_lock_irqsave(pinstance->host->host_lock, lock_flags);
 
 		if (pinstance->ioa_state == IOA_STATE_DEAD) {
-			spin_unlock_irqrestore(pinstance->host->host_lock,
-					       lock_flags);
 			pmcraid_info("reset_reload: IOA is dead\n");
-			return reset;
-		} else if (pinstance->ioa_state == target_state) {
+			goto out_unlock;
+		}
+
+		if (pinstance->ioa_state == target_state) {
 			reset = 0;
+			goto out_unlock;
 		}
 	}
 
-	if (reset) {
-		pmcraid_info("reset_reload: proceeding with reset\n");
-		scsi_block_requests(pinstance->host);
-		reset_cmd = pmcraid_get_free_cmd(pinstance);
-
-		if (reset_cmd == NULL) {
-			pmcraid_err("no free cmnd for reset_reload\n");
-			spin_unlock_irqrestore(pinstance->host->host_lock,
-					       lock_flags);
-			return reset;
-		}
+	pmcraid_info("reset_reload: proceeding with reset\n");
+	scsi_block_requests(pinstance->host);
+	reset_cmd = pmcraid_get_free_cmd(pinstance);
+	if (reset_cmd == NULL) {
+		pmcraid_err("no free cmnd for reset_reload\n");
+		goto out_unlock;
+	}
 
-		if (shutdown_type == SHUTDOWN_NORMAL)
-			pinstance->ioa_bringdown = 1;
+	if (shutdown_type == SHUTDOWN_NORMAL)
+		pinstance->ioa_bringdown = 1;
 
-		pinstance->ioa_shutdown_type = shutdown_type;
-		pinstance->reset_cmd = reset_cmd;
-		pinstance->force_ioa_reset = reset;
-		pmcraid_info("reset_reload: initiating reset\n");
-		pmcraid_ioa_reset(reset_cmd);
-		spin_unlock_irqrestore(pinstance->host->host_lock, lock_flags);
-		pmcraid_info("reset_reload: waiting for reset to complete\n");
-		wait_event(pinstance->reset_wait_q,
-			   !pinstance->ioa_reset_in_progress);
+	pinstance->ioa_shutdown_type = shutdown_type;
+	pinstance->reset_cmd = reset_cmd;
+	pinstance->force_ioa_reset = reset;
+	pmcraid_info("reset_reload: initiating reset\n");
+	pmcraid_ioa_reset(reset_cmd);
+	spin_unlock_irqrestore(pinstance->host->host_lock, lock_flags);
+	pmcraid_info("reset_reload: waiting for reset to complete\n");
+	wait_event(pinstance->reset_wait_q,
+		   !pinstance->ioa_reset_in_progress);
 
-		pmcraid_info("reset_reload: reset is complete !!\n");
-		scsi_unblock_requests(pinstance->host);
-		if (pinstance->ioa_state == target_state)
-			reset = 0;
-	}
+	pmcraid_info("reset_reload: reset is complete !!\n");
+	scsi_unblock_requests(pinstance->host);
+	return pinstance->ioa_state != target_state;
 
+out_unlock:
+	spin_unlock_irqrestore(pinstance->host->host_lock, lock_flags);
 	return reset;
 }
 
@@ -2529,7 +2526,7 @@ static void pmcraid_cancel_all(struct pmcraid_cmd *cmd, u32 sense)
 	ioarcb->ioadl_bus_addr = 0;
 	ioarcb->ioadl_length = 0;
 	ioarcb->data_transfer_length = 0;
-	ioarcb->ioarcb_bus_addr &= (~0x1FULL);
+	ioarcb->ioarcb_bus_addr &= cpu_to_le64((~0x1FULL));
 
 	/* writing to IOARRIN must be protected by host_lock, as mid-layer
 	 * schedule queuecommand while we are doing this
@@ -2692,8 +2689,8 @@ static int pmcraid_error_handler(struct pmcraid_cmd *cmd)
 		 * mid-layer
 		 */
 		if (ioasa->auto_sense_length != 0) {
-			short sense_len = ioasa->auto_sense_length;
-			int data_size = min_t(u16, le16_to_cpu(sense_len),
+			short sense_len = le16_to_cpu(ioasa->auto_sense_length);
+			int data_size = min_t(u16, sense_len,
 					      SCSI_SENSE_BUFFERSIZE);
 
 			memcpy(scsi_cmd->sense_buffer,
@@ -2915,7 +2912,7 @@ static struct pmcraid_cmd *pmcraid_abort_cmd(struct pmcraid_cmd *cmd)
 
 	pmcraid_info("aborting command CDB[0]= %x with index = %d\n",
 		cmd->ioa_cb->ioarcb.cdb[0],
-		cmd->ioa_cb->ioarcb.response_handle >> 2);
+		le32_to_cpu(cmd->ioa_cb->ioarcb.response_handle) >> 2);
 
 	init_completion(&cancel_cmd->wait_for_completion);
 	cancel_cmd->completion_req = 1;
@@ -3140,9 +3137,8 @@ pmcraid_init_ioadls(struct pmcraid_cmd *cmd, int sgcount)
 	int ioadl_count = 0;
 
 	if (ioarcb->add_cmd_param_length)
-		ioadl_count = DIV_ROUND_UP(ioarcb->add_cmd_param_length, 16);
-	ioarcb->ioadl_length =
-		sizeof(struct pmcraid_ioadl_desc) * sgcount;
+		ioadl_count = DIV_ROUND_UP(le16_to_cpu(ioarcb->add_cmd_param_length), 16);
+	ioarcb->ioadl_length = cpu_to_le32(sizeof(struct pmcraid_ioadl_desc) * sgcount);
 
 	if ((sgcount + ioadl_count) > (ARRAY_SIZE(ioarcb->add_data.u.ioadl))) {
 		/* external ioadls start at offset 0x80 from control_block
@@ -3150,7 +3146,7 @@ pmcraid_init_ioadls(struct pmcraid_cmd *cmd, int sgcount)
 		 * It is necessary to indicate to firmware that driver is
 		 * using ioadls to be treated as external to IOARCB.
 		 */
-		ioarcb->ioarcb_bus_addr &= ~(0x1FULL);
+		ioarcb->ioarcb_bus_addr &= cpu_to_le64(~(0x1FULL));
 		ioarcb->ioadl_bus_addr =
 			cpu_to_le64((cmd->ioa_cb_bus_addr) +
 				offsetof(struct pmcraid_ioarcb,
@@ -3164,7 +3160,7 @@ pmcraid_init_ioadls(struct pmcraid_cmd *cmd, int sgcount)
 
 		ioadl = &ioarcb->add_data.u.ioadl[ioadl_count];
 		ioarcb->ioarcb_bus_addr |=
-				DIV_ROUND_CLOSEST(sgcount + ioadl_count, 8);
+			cpu_to_le64(DIV_ROUND_CLOSEST(sgcount + ioadl_count, 8));
 	}
 
 	return ioadl;
@@ -3325,7 +3321,7 @@ static struct pmcraid_sglist *pmcraid_alloc_sglist(int buflen)
  */
 static int pmcraid_copy_sglist(
 	struct pmcraid_sglist *sglist,
-	unsigned long buffer,
+	void __user *buffer,
 	u32 len,
 	int direction
 )
@@ -3346,11 +3342,9 @@ static int pmcraid_copy_sglist(
 
 		kaddr = kmap(page);
 		if (direction == DMA_TO_DEVICE)
-			rc = __copy_from_user(kaddr,
-					      (void *)buffer,
-					      bsize_elem);
+			rc = copy_from_user(kaddr, buffer, bsize_elem);
 		else
-			rc = __copy_to_user((void *)buffer, kaddr, bsize_elem);
+			rc = copy_to_user(buffer, kaddr, bsize_elem);
 
 		kunmap(page);
 
@@ -3368,13 +3362,9 @@ static int pmcraid_copy_sglist(
 		kaddr = kmap(page);
 
 		if (direction == DMA_TO_DEVICE)
-			rc = __copy_from_user(kaddr,
-					      (void *)buffer,
-					      len % bsize_elem);
+			rc = copy_from_user(kaddr, buffer, len % bsize_elem);
 		else
-			rc = __copy_to_user((void *)buffer,
-					    kaddr,
-					    len % bsize_elem);
+			rc = copy_to_user(buffer, kaddr, len % bsize_elem);
 
 		kunmap(page);
 
@@ -3496,7 +3486,7 @@ static int pmcraid_queuecommand_lck(
 		     RES_IS_VSET(res->cfg_entry) ?
 			(fw_version <= PMCRAID_FW_VERSION_1 ?
 				res->cfg_entry.unique_flags1 :
-					res->cfg_entry.array_id & 0xFF) :
+				le16_to_cpu(res->cfg_entry.array_id) & 0xFF) :
 			RES_TARGET(res->cfg_entry.resource_address),
 		     RES_LUN(res->cfg_entry.resource_address));
 
@@ -3652,17 +3642,17 @@ static long pmcraid_ioctl_passthrough(
 	struct pmcraid_instance *pinstance,
 	unsigned int ioctl_cmd,
 	unsigned int buflen,
-	unsigned long arg
+	void __user *arg
 )
 {
 	struct pmcraid_passthrough_ioctl_buffer *buffer;
 	struct pmcraid_ioarcb *ioarcb;
 	struct pmcraid_cmd *cmd;
 	struct pmcraid_cmd *cancel_cmd;
-	unsigned long request_buffer;
+	void __user *request_buffer;
 	unsigned long request_offset;
 	unsigned long lock_flags;
-	void *ioasa;
+	void __user *ioasa;
 	u32 ioasc;
 	int request_size;
 	int buffer_size;
@@ -3701,13 +3691,10 @@ static long pmcraid_ioctl_passthrough(
 
 	request_buffer = arg + request_offset;
 
-	rc = __copy_from_user(buffer,
-			     (struct pmcraid_passthrough_ioctl_buffer *) arg,
+	rc = copy_from_user(buffer, arg,
 			     sizeof(struct pmcraid_passthrough_ioctl_buffer));
 
-	ioasa =
-	(void *)(arg +
-		offsetof(struct pmcraid_passthrough_ioctl_buffer, ioasa));
+	ioasa = arg + offsetof(struct pmcraid_passthrough_ioctl_buffer, ioasa);
 
 	if (rc) {
 		pmcraid_err("ioctl: can't copy passthrough buffer\n");
@@ -3715,7 +3702,7 @@ static long pmcraid_ioctl_passthrough(
 		goto out_free_buffer;
 	}
 
-	request_size = buffer->ioarcb.data_transfer_length;
+	request_size = le32_to_cpu(buffer->ioarcb.data_transfer_length);
 
 	if (buffer->ioarcb.request_flags0 & TRANSFER_DIR_WRITE) {
 		access = VERIFY_READ;
@@ -3725,20 +3712,14 @@ static long pmcraid_ioctl_passthrough(
 		direction = DMA_FROM_DEVICE;
 	}
 
-	if (request_size > 0) {
-		rc = access_ok(access, arg, request_offset + request_size);
-
-		if (!rc) {
-			rc = -EFAULT;
-			goto out_free_buffer;
-		}
-	} else if (request_size < 0) {
+	if (request_size < 0) {
 		rc = -EINVAL;
 		goto out_free_buffer;
 	}
 
 	/* check if we have any additional command parameters */
-	if (buffer->ioarcb.add_cmd_param_length > PMCRAID_ADD_CMD_PARAM_LEN) {
+	if (le16_to_cpu(buffer->ioarcb.add_cmd_param_length)
+	     > PMCRAID_ADD_CMD_PARAM_LEN) {
 		rc = -EINVAL;
 		goto out_free_buffer;
 	}
@@ -3770,7 +3751,7 @@ static long pmcraid_ioctl_passthrough(
 			buffer->ioarcb.add_cmd_param_offset;
 		memcpy(ioarcb->add_data.u.add_cmd_params,
 			buffer->ioarcb.add_data.u.add_cmd_params,
-			buffer->ioarcb.add_cmd_param_length);
+			le16_to_cpu(buffer->ioarcb.add_cmd_param_length));
 	}
 
 	/* set hrrq number where the IOA should respond to. Note that all cmds
@@ -3840,10 +3821,10 @@ static long pmcraid_ioctl_passthrough(
 		wait_for_completion(&cmd->wait_for_completion);
 	} else if (!wait_for_completion_timeout(
 			&cmd->wait_for_completion,
-			msecs_to_jiffies(buffer->ioarcb.cmd_timeout * 1000))) {
+			msecs_to_jiffies(le16_to_cpu(buffer->ioarcb.cmd_timeout) * 1000))) {
 
 		pmcraid_info("aborting cmd %d (CDB[0] = %x) due to timeout\n",
-			le32_to_cpu(cmd->ioa_cb->ioarcb.response_handle >> 2),
+			le32_to_cpu(cmd->ioa_cb->ioarcb.response_handle) >> 2,
 			cmd->ioa_cb->ioarcb.cdb[0]);
 
 		spin_lock_irqsave(pinstance->host->host_lock, lock_flags);
@@ -3852,7 +3833,7 @@ static long pmcraid_ioctl_passthrough(
 
 		if (cancel_cmd) {
 			wait_for_completion(&cancel_cmd->wait_for_completion);
-			ioasc = cancel_cmd->ioa_cb->ioasa.ioasc;
+			ioasc = le32_to_cpu(cancel_cmd->ioa_cb->ioasa.ioasc);
 			pmcraid_return_cmd(cancel_cmd);
 
 			/* if abort task couldn't find the command i.e it got
@@ -3941,11 +3922,6 @@ static long pmcraid_ioctl_driver(
 {
 	int rc = -ENOSYS;
 
-	if (!access_ok(VERIFY_READ, user_buffer, _IOC_SIZE(cmd))) {
-		pmcraid_err("ioctl_driver: access fault in request buffer\n");
-		return -EFAULT;
-	}
-
 	switch (cmd) {
 	case PMCRAID_IOCTL_RESET_ADAPTER:
 		pmcraid_reset_bringup(pinstance);
@@ -3977,8 +3953,7 @@ static int pmcraid_check_ioctl_buffer(
 	struct pmcraid_ioctl_header *hdr
 )
 {
-	int rc = 0;
-	int access = VERIFY_READ;
+	int rc;
 
 	if (copy_from_user(hdr, arg, sizeof(struct pmcraid_ioctl_header))) {
 		pmcraid_err("couldn't copy ioctl header from user buffer\n");
@@ -3994,19 +3969,6 @@ static int pmcraid_check_ioctl_buffer(
 		return -EINVAL;
 	}
 
-	/* check for appropriate buffer access */
-	if ((_IOC_DIR(cmd) & _IOC_READ) == _IOC_READ)
-		access = VERIFY_WRITE;
-
-	rc = access_ok(access,
-		       (arg + sizeof(struct pmcraid_ioctl_header)),
-		       hdr->buffer_length);
-	if (!rc) {
-		pmcraid_err("access failed for user buffer of size %d\n",
-			     hdr->buffer_length);
-		return -EFAULT;
-	}
-
 	return 0;
 }
 
@@ -4021,6 +3983,7 @@ static long pmcraid_chr_ioctl(
 {
 	struct pmcraid_instance *pinstance = NULL;
 	struct pmcraid_ioctl_header *hdr = NULL;
+	void __user *argp = (void __user *)arg;
 	int retval = -ENOTTY;
 
 	hdr = kmalloc(sizeof(struct pmcraid_ioctl_header), GFP_KERNEL);
@@ -4030,7 +3993,7 @@ static long pmcraid_chr_ioctl(
 		return -ENOMEM;
 	}
 
-	retval = pmcraid_check_ioctl_buffer(cmd, (void *)arg, hdr);
+	retval = pmcraid_check_ioctl_buffer(cmd, argp, hdr);
 
 	if (retval) {
 		pmcraid_info("chr_ioctl: header check failed\n");
@@ -4055,10 +4018,8 @@ static long pmcraid_chr_ioctl(
 		if (cmd == PMCRAID_IOCTL_DOWNLOAD_MICROCODE)
 			scsi_block_requests(pinstance->host);
 
-		retval = pmcraid_ioctl_passthrough(pinstance,
-						   cmd,
-						   hdr->buffer_length,
-						   arg);
+		retval = pmcraid_ioctl_passthrough(pinstance, cmd,
+						   hdr->buffer_length, argp);
 
 		if (cmd == PMCRAID_IOCTL_DOWNLOAD_MICROCODE)
 			scsi_unblock_requests(pinstance->host);
@@ -4066,10 +4027,8 @@ static long pmcraid_chr_ioctl(
 
 	case PMCRAID_DRIVER_IOCTL:
 		arg += sizeof(struct pmcraid_ioctl_header);
-		retval = pmcraid_ioctl_driver(pinstance,
-					      cmd,
-					      hdr->buffer_length,
-					      (void __user *)arg);
+		retval = pmcraid_ioctl_driver(pinstance, cmd,
+					      hdr->buffer_length, argp);
 		break;
 
 	default:
@@ -4470,7 +4429,7 @@ static void pmcraid_worker_function(struct work_struct *workp)
 				if (fw_version <= PMCRAID_FW_VERSION_1)
 					target = res->cfg_entry.unique_flags1;
 				else
-					target = res->cfg_entry.array_id & 0xFF;
+					target = le16_to_cpu(res->cfg_entry.array_id) & 0xFF;
 				lun = PMCRAID_VSET_LUN_ID;
 			} else {
 				bus = PMCRAID_PHYS_BUS_ID;
@@ -4509,7 +4468,7 @@ static void pmcraid_tasklet_function(unsigned long instance)
 	unsigned long host_lock_flags;
 	spinlock_t *lockp; /* hrrq buffer lock */
 	int id;
-	__le32 resp;
+	u32 resp;
 
 	hrrq_vector = (struct pmcraid_isr_param *)instance;
 	pinstance = hrrq_vector->drv_inst;
@@ -4833,7 +4792,7 @@ static int pmcraid_allocate_host_rrqs(struct pmcraid_instance *pinstance)
 					buffer_size,
 					&(pinstance->hrrq_start_bus_addr[i]));
 
-		if (pinstance->hrrq_start[i] == 0) {
+		if (!pinstance->hrrq_start[i]) {
 			pmcraid_err("pci_alloc failed for hrrq vector : %d\n",
 				    i);
 			pmcraid_release_host_rrqs(pinstance, i);
@@ -5549,8 +5508,7 @@ static void pmcraid_set_timestamp(struct pmcraid_cmd *cmd)
 	struct pmcraid_ioarcb *ioarcb = &cmd->ioa_cb->ioarcb;
 	__be32 time_stamp_len = cpu_to_be32(PMCRAID_TIMESTAMP_LEN);
 	struct pmcraid_ioadl_desc *ioadl = ioarcb->add_data.u.ioadl;
-
-	__le64 timestamp;
+	u64 timestamp;
 
 	timestamp = ktime_get_real_seconds() * 1000;
 
@@ -5572,7 +5530,7 @@ static void pmcraid_set_timestamp(struct pmcraid_cmd *cmd)
 					offsetof(struct pmcraid_ioarcb,
 						add_data.u.ioadl[0]));
 	ioarcb->ioadl_length = cpu_to_le32(sizeof(struct pmcraid_ioadl_desc));
-	ioarcb->ioarcb_bus_addr &= ~(0x1FULL);
+	ioarcb->ioarcb_bus_addr &= cpu_to_le64(~(0x1FULL));
 
 	ioarcb->request_flags0 |= NO_LINK_DESCS;
 	ioarcb->request_flags0 |= TRANSFER_DIR_WRITE;
@@ -5631,7 +5589,7 @@ static void pmcraid_init_res_table(struct pmcraid_cmd *cmd)
 	list_for_each_entry_safe(res, temp, &pinstance->used_res_q, queue)
 		list_move_tail(&res->queue, &old_res);
 
-	for (i = 0; i < pinstance->cfg_table->num_entries; i++) {
+	for (i = 0; i < le16_to_cpu(pinstance->cfg_table->num_entries); i++) {
 		if (be16_to_cpu(pinstance->inq_data->fw_version) <=
 						PMCRAID_FW_VERSION_1)
 			cfgte = &pinstance->cfg_table->entries[i];
@@ -5686,7 +5644,7 @@ static void pmcraid_init_res_table(struct pmcraid_cmd *cmd)
 				 res->cfg_entry.resource_type,
 				 (fw_version <= PMCRAID_FW_VERSION_1 ?
 					res->cfg_entry.unique_flags1 :
-						res->cfg_entry.array_id & 0xFF),
+					le16_to_cpu(res->cfg_entry.array_id) & 0xFF),
 				 le32_to_cpu(res->cfg_entry.resource_address));
 		}
 	}
@@ -5724,7 +5682,7 @@ static void pmcraid_querycfg(struct pmcraid_cmd *cmd)
 	struct pmcraid_ioarcb *ioarcb = &cmd->ioa_cb->ioarcb;
 	struct pmcraid_ioadl_desc *ioadl = ioarcb->add_data.u.ioadl;
 	struct pmcraid_instance *pinstance = cmd->drv_inst;
-	int cfg_table_size = cpu_to_be32(sizeof(struct pmcraid_config_table));
+	__be32 cfg_table_size = cpu_to_be32(sizeof(struct pmcraid_config_table));
 
 	if (be16_to_cpu(pinstance->inq_data->fw_version) <=
 					PMCRAID_FW_VERSION_1)
@@ -5749,7 +5707,7 @@ static void pmcraid_querycfg(struct pmcraid_cmd *cmd)
 					offsetof(struct pmcraid_ioarcb,
 						add_data.u.ioadl[0]));
 	ioarcb->ioadl_length = cpu_to_le32(sizeof(struct pmcraid_ioadl_desc));
-	ioarcb->ioarcb_bus_addr &= ~(0x1FULL);
+	ioarcb->ioarcb_bus_addr &= cpu_to_le64(~0x1FULL);
 
 	ioarcb->request_flags0 |= NO_LINK_DESCS;
 	ioarcb->data_transfer_length =
diff --git a/drivers/scsi/pmcraid.h b/drivers/scsi/pmcraid.h
index 568b18a2f47d..01eb2bc16dc1 100644
--- a/drivers/scsi/pmcraid.h
+++ b/drivers/scsi/pmcraid.h
@@ -554,7 +554,7 @@ struct pmcraid_inquiry_data {
 	__u8	add_page_len;
 	__u8	length;
 	__u8	reserved2;
-	__le16	fw_version;
+	__be16	fw_version;
 	__u8	reserved3[16];
 };
 
@@ -697,13 +697,13 @@ struct pmcraid_instance {
 	dma_addr_t hrrq_start_bus_addr[PMCRAID_NUM_MSIX_VECTORS];
 
 	/* Pointer to 1st entry of HRRQ */
-	__be32 *hrrq_start[PMCRAID_NUM_MSIX_VECTORS];
+	__le32 *hrrq_start[PMCRAID_NUM_MSIX_VECTORS];
 
 	/* Pointer to last entry of HRRQ */
-	__be32 *hrrq_end[PMCRAID_NUM_MSIX_VECTORS];
+	__le32 *hrrq_end[PMCRAID_NUM_MSIX_VECTORS];
 
 	/* Pointer to current pointer of hrrq */
-	__be32 *hrrq_curr[PMCRAID_NUM_MSIX_VECTORS];
+	__le32 *hrrq_curr[PMCRAID_NUM_MSIX_VECTORS];
 
 	/* Lock for HRRQ access */
 	spinlock_t hrrq_lock[PMCRAID_NUM_MSIX_VECTORS];
diff --git a/drivers/scsi/qedf/qedf_debugfs.c b/drivers/scsi/qedf/qedf_debugfs.c
index cb08b625c594..00a1d6405ebe 100644
--- a/drivers/scsi/qedf/qedf_debugfs.c
+++ b/drivers/scsi/qedf/qedf_debugfs.c
@@ -449,7 +449,7 @@ const struct file_operations qedf_dbg_fops[] = {
 	qedf_dbg_fileops(qedf, clear_stats),
 	qedf_dbg_fileops_seq(qedf, offload_stats),
 	/* This must be last */
-	{ NULL, NULL },
+	{ },
 };
 
 #else /* CONFIG_DEBUG_FS */
diff --git a/drivers/scsi/qedi/qedi_debugfs.c b/drivers/scsi/qedi/qedi_debugfs.c
index 59417199bf36..39d77818a677 100644
--- a/drivers/scsi/qedi/qedi_debugfs.c
+++ b/drivers/scsi/qedi/qedi_debugfs.c
@@ -240,5 +240,5 @@ const struct file_operations qedi_dbg_fops[] = {
 	qedi_dbg_fileops_seq(qedi, gbl_ctx),
 	qedi_dbg_fileops(qedi, do_not_recover),
 	qedi_dbg_fileops_seq(qedi, io_trace),
-	{ NULL, NULL },
+	{ },
 };
diff --git a/drivers/scsi/qedi/qedi_iscsi.c b/drivers/scsi/qedi/qedi_iscsi.c
index d1de172bebac..3548d46f9b27 100644
--- a/drivers/scsi/qedi/qedi_iscsi.c
+++ b/drivers/scsi/qedi/qedi_iscsi.c
@@ -1370,7 +1370,7 @@ static void qedi_cleanup_task(struct iscsi_task *task)
 {
 	if (!task->sc || task->state == ISCSI_TASK_PENDING) {
 		QEDI_INFO(NULL, QEDI_LOG_IO, "Returning ref_cnt=%d\n",
-			  atomic_read(&task->refcount));
+			  refcount_read(&task->refcount));
 		return;
 	}
 
diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 435ff7fd6384..7c8d6c54ab70 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -695,7 +695,7 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,
 	case 0x2025e:
 		if (!IS_P3P_TYPE(ha) || vha != base_vha) {
 			ql_log(ql_log_info, vha, 0x7071,
-			    "FCoE ctx reset no supported.\n");
+			    "FCoE ctx reset not supported.\n");
 			return -EPERM;
 		}
 
diff --git a/drivers/scsi/qla2xxx/qla_bsg.c b/drivers/scsi/qla2xxx/qla_bsg.c
index b6e40fd4c3c1..16d1cd50feed 100644
--- a/drivers/scsi/qla2xxx/qla_bsg.c
+++ b/drivers/scsi/qla2xxx/qla_bsg.c
@@ -1822,7 +1822,7 @@ qla24xx_process_bidir_cmd(struct bsg_job *bsg_job)
 	/* Check if operating mode is P2P */
 	if (ha->operating_mode != P2P) {
 		ql_log(ql_log_warn, vha, 0x70a4,
-		    "Host is operating mode is not P2p\n");
+		    "Host operating mode is not P2p\n");
 		rval = EXT_STATUS_INVALID_CFG;
 		goto done;
 	}
diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
index ab0f873fd6a1..9bc9aa9e164a 100644
--- a/drivers/scsi/qla2xxx/qla_gs.c
+++ b/drivers/scsi/qla2xxx/qla_gs.c
@@ -144,7 +144,7 @@ qla2x00_chk_ms_status(scsi_qla_host_t *vha, ms_iocb_entry_t *ms_pkt,
 			if (ct_rsp->header.response !=
 			    cpu_to_be16(CT_ACCEPT_RESPONSE)) {
 				ql_dbg(ql_dbg_disc + ql_dbg_buffer, vha, 0x2077,
-				    "%s failed rejected request on port_id: %02x%02x%02x Compeltion status 0x%x, response 0x%x\n",
+				    "%s failed rejected request on port_id: %02x%02x%02x Completion status 0x%x, response 0x%x\n",
 				    routine, vha->d_id.b.domain,
 				    vha->d_id.b.area, vha->d_id.b.al_pa,
 				    comp_status, ct_rsp->header.response);
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index f9d2fe7b1ade..034743309ada 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -2289,7 +2289,7 @@ qla2x00_chip_diag(scsi_qla_host_t *vha)
 		goto chip_diag_failed;
 
 	/* Check product ID of chip */
-	ql_dbg(ql_dbg_init, vha, 0x007d, "Checking product Id of chip.\n");
+	ql_dbg(ql_dbg_init, vha, 0x007d, "Checking product ID of chip.\n");
 
 	mb[1] = RD_MAILBOX_REG(ha, reg, 1);
 	mb[2] = RD_MAILBOX_REG(ha, reg, 2);
diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index 3203367a4f42..aac03504d9a3 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -2100,14 +2100,14 @@ qla25xx_process_bidir_status_iocb(scsi_qla_host_t *vha, void *pkt,
 
 	case CS_DATA_OVERRUN:
 		ql_dbg(ql_dbg_user, vha, 0x70b1,
-		    "Command completed with date overrun thread_id=%d\n",
+		    "Command completed with data overrun thread_id=%d\n",
 		    thread_id);
 		rval = EXT_STATUS_DATA_OVERRUN;
 		break;
 
 	case CS_DATA_UNDERRUN:
 		ql_dbg(ql_dbg_user, vha, 0x70b2,
-		    "Command completed with date underrun thread_id=%d\n",
+		    "Command completed with data underrun thread_id=%d\n",
 		    thread_id);
 		rval = EXT_STATUS_DATA_UNDERRUN;
 		break;
@@ -2134,7 +2134,7 @@ qla25xx_process_bidir_status_iocb(scsi_qla_host_t *vha, void *pkt,
 
 	case CS_BIDIR_RD_UNDERRUN:
 		ql_dbg(ql_dbg_user, vha, 0x70b6,
-		    "Command completed with read data data underrun "
+		    "Command completed with read data underrun "
 		    "thread_id=%d\n", thread_id);
 		rval = EXT_STATUS_DATA_UNDERRUN;
 		break;
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 83d61d2142e9..1c7957903283 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -423,7 +423,6 @@ static void qla2x00_free_req_que(struct qla_hw_data *ha, struct req_que *req)
 		kfree(req->outstanding_cmds);
 
 	kfree(req);
-	req = NULL;
 }
 
 static void qla2x00_free_rsp_que(struct qla_hw_data *ha, struct rsp_que *rsp)
@@ -439,7 +438,6 @@ static void qla2x00_free_rsp_que(struct qla_hw_data *ha, struct rsp_que *rsp)
 		rsp->ring, rsp->dma);
 	}
 	kfree(rsp);
-	rsp = NULL;
 }
 
 static void qla2x00_free_queues(struct qla_hw_data *ha)
@@ -653,7 +651,6 @@ qla2x00_sp_free_dma(void *ptr)
 		ha->gbl_dsd_inuse -= ctx1->dsd_use_cnt;
 		ha->gbl_dsd_avail += ctx1->dsd_use_cnt;
 		mempool_free(ctx1, ha->ctx_mempool);
-		ctx1 = NULL;
 	}
 
 	CMD_SP(cmd) = NULL;
@@ -3256,7 +3253,6 @@ iospace_config_failed:
 	}
 	pci_release_selected_regions(ha->pdev, ha->bars);
 	kfree(ha);
-	ha = NULL;
 
 probe_out:
 	pci_disable_device(pdev);
@@ -3504,7 +3500,6 @@ qla2x00_remove_one(struct pci_dev *pdev)
 
 	pci_release_selected_regions(ha->pdev, ha->bars);
 	kfree(ha);
-	ha = NULL;
 
 	pci_disable_pcie_error_reporting(pdev);
 
@@ -3568,7 +3563,6 @@ void qla2x00_free_fcports(struct scsi_qla_host *vha)
 		list_del(&fcport->list);
 		qla2x00_clear_loop_id(fcport);
 		kfree(fcport);
-		fcport = NULL;
 	}
 }
 
diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c
index 4180d6d9fe78..5d6d158bbfd6 100644
--- a/drivers/scsi/qla4xxx/ql4_init.c
+++ b/drivers/scsi/qla4xxx/ql4_init.c
@@ -389,7 +389,7 @@ void qla4xxx_alloc_fw_dump(struct scsi_qla_host *ha)
 		goto alloc_cleanup;
 
 	DEBUG2(ql4_printk(KERN_INFO, ha,
-			  "Minidump Tempalate Size = 0x%x KB\n",
+			  "Minidump Template Size = 0x%x KB\n",
 			  ha->fw_dump_tmplt_size));
 	DEBUG2(ql4_printk(KERN_INFO, ha,
 			  "Total Minidump size = 0x%x KB\n", ha->fw_dump_size));
diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c
index ac52150d1569..64c6fa563fdb 100644
--- a/drivers/scsi/qla4xxx/ql4_os.c
+++ b/drivers/scsi/qla4xxx/ql4_os.c
@@ -8664,7 +8664,6 @@ static int qla4xxx_probe_adapter(struct pci_dev *pdev,
 	init_completion(&ha->disable_acb_comp);
 	init_completion(&ha->idc_comp);
 	init_completion(&ha->link_up_comp);
-	init_completion(&ha->disable_acb_comp);
 
 	spin_lock_init(&ha->hardware_lock);
 	spin_lock_init(&ha->work_lock);
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index 2db412dd4b44..ecc07dab893d 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -46,6 +46,8 @@
 
 #include <trace/events/scsi.h>
 
+#include <asm/unaligned.h>
+
 static void scsi_eh_done(struct scsi_cmnd *scmd);
 
 /*
@@ -162,13 +164,7 @@ scmd_eh_abort_handler(struct work_struct *work)
 		}
 	}
 
-	if (!scsi_eh_scmd_add(scmd, 0)) {
-		SCSI_LOG_ERROR_RECOVERY(3,
-			scmd_printk(KERN_WARNING, scmd,
-				    "terminate aborted command\n"));
-		set_host_byte(scmd, DID_TIME_OUT);
-		scsi_finish_command(scmd);
-	}
+	scsi_eh_scmd_add(scmd);
 }
 
 /**
@@ -188,7 +184,6 @@ scsi_abort_command(struct scsi_cmnd *scmd)
 		/*
 		 * Retry after abort failed, escalate to next level.
 		 */
-		scmd->eh_eflags &= ~SCSI_EH_ABORT_SCHEDULED;
 		SCSI_LOG_ERROR_RECOVERY(3,
 			scmd_printk(KERN_INFO, scmd,
 				    "previous abort failed\n"));
@@ -196,19 +191,7 @@ scsi_abort_command(struct scsi_cmnd *scmd)
 		return FAILED;
 	}
 
-	/*
-	 * Do not try a command abort if
-	 * SCSI EH has already started.
-	 */
 	spin_lock_irqsave(shost->host_lock, flags);
-	if (scsi_host_in_recovery(shost)) {
-		spin_unlock_irqrestore(shost->host_lock, flags);
-		SCSI_LOG_ERROR_RECOVERY(3,
-			scmd_printk(KERN_INFO, scmd,
-				    "not aborting, host in recovery\n"));
-		return FAILED;
-	}
-
 	if (shost->eh_deadline != -1 && !shost->last_reset)
 		shost->last_reset = jiffies;
 	spin_unlock_irqrestore(shost->host_lock, flags);
@@ -221,40 +204,47 @@ scsi_abort_command(struct scsi_cmnd *scmd)
 }
 
 /**
- * scsi_eh_scmd_add - add scsi cmd to error handling.
+ * scsi_eh_reset - call into ->eh_action to reset internal counters
  * @scmd:	scmd to run eh on.
- * @eh_flag:	optional SCSI_EH flag.
  *
- * Return value:
- *	0 on failure.
+ * The scsi driver might be carrying internal state about the
+ * devices, so we need to call into the driver to reset the
+ * internal state once the error handler is started.
  */
-int scsi_eh_scmd_add(struct scsi_cmnd *scmd, int eh_flag)
+static void scsi_eh_reset(struct scsi_cmnd *scmd)
+{
+	if (!blk_rq_is_passthrough(scmd->request)) {
+		struct scsi_driver *sdrv = scsi_cmd_to_driver(scmd);
+		if (sdrv->eh_reset)
+			sdrv->eh_reset(scmd);
+	}
+}
+
+/**
+ * scsi_eh_scmd_add - add scsi cmd to error handling.
+ * @scmd:	scmd to run eh on.
+ */
+void scsi_eh_scmd_add(struct scsi_cmnd *scmd)
 {
 	struct Scsi_Host *shost = scmd->device->host;
 	unsigned long flags;
-	int ret = 0;
+	int ret;
 
-	if (!shost->ehandler)
-		return 0;
+	WARN_ON_ONCE(!shost->ehandler);
 
 	spin_lock_irqsave(shost->host_lock, flags);
-	if (scsi_host_set_state(shost, SHOST_RECOVERY))
-		if (scsi_host_set_state(shost, SHOST_CANCEL_RECOVERY))
-			goto out_unlock;
-
+	if (scsi_host_set_state(shost, SHOST_RECOVERY)) {
+		ret = scsi_host_set_state(shost, SHOST_CANCEL_RECOVERY);
+		WARN_ON_ONCE(ret);
+	}
 	if (shost->eh_deadline != -1 && !shost->last_reset)
 		shost->last_reset = jiffies;
 
-	ret = 1;
-	if (scmd->eh_eflags & SCSI_EH_ABORT_SCHEDULED)
-		eh_flag &= ~SCSI_EH_CANCEL_CMD;
-	scmd->eh_eflags |= eh_flag;
+	scsi_eh_reset(scmd);
 	list_add_tail(&scmd->eh_entry, &shost->eh_cmd_q);
 	shost->host_failed++;
 	scsi_eh_wakeup(shost);
- out_unlock:
 	spin_unlock_irqrestore(shost->host_lock, flags);
-	return ret;
 }
 
 /**
@@ -283,13 +273,10 @@ enum blk_eh_timer_return scsi_times_out(struct request *req)
 		rtn = host->hostt->eh_timed_out(scmd);
 
 	if (rtn == BLK_EH_NOT_HANDLED) {
-		if (!host->hostt->no_async_abort &&
-		    scsi_abort_command(scmd) == SUCCESS)
-			return BLK_EH_NOT_HANDLED;
-
-		set_host_byte(scmd, DID_TIME_OUT);
-		if (!scsi_eh_scmd_add(scmd, SCSI_EH_CANCEL_CMD))
-			rtn = BLK_EH_HANDLED;
+		if (scsi_abort_command(scmd) != SUCCESS) {
+			set_host_byte(scmd, DID_TIME_OUT);
+			scsi_eh_scmd_add(scmd);
+		}
 	}
 
 	return rtn;
@@ -341,7 +328,7 @@ static inline void scsi_eh_prt_fail_stats(struct Scsi_Host *shost,
 		list_for_each_entry(scmd, work_q, eh_entry) {
 			if (scmd->device == sdev) {
 				++total_failures;
-				if (scmd->eh_eflags & SCSI_EH_CANCEL_CMD)
+				if (scmd->eh_eflags & SCSI_EH_ABORT_SCHEDULED)
 					++cmd_cancel;
 				else
 					++cmd_failed;
@@ -931,6 +918,7 @@ void scsi_eh_prep_cmnd(struct scsi_cmnd *scmd, struct scsi_eh_save *ses,
 	ses->result = scmd->result;
 	ses->underflow = scmd->underflow;
 	ses->prot_op = scmd->prot_op;
+	ses->eh_eflags = scmd->eh_eflags;
 
 	scmd->prot_op = SCSI_PROT_NORMAL;
 	scmd->eh_eflags = 0;
@@ -994,6 +982,7 @@ void scsi_eh_restore_cmnd(struct scsi_cmnd* scmd, struct scsi_eh_save *ses)
 	scmd->result = ses->result;
 	scmd->underflow = ses->underflow;
 	scmd->prot_op = ses->prot_op;
+	scmd->eh_eflags = ses->eh_eflags;
 }
 EXPORT_SYMBOL(scsi_eh_restore_cmnd);
 
@@ -1126,7 +1115,6 @@ static int scsi_eh_action(struct scsi_cmnd *scmd, int rtn)
  */
 void scsi_eh_finish_cmd(struct scsi_cmnd *scmd, struct list_head *done_q)
 {
-	scmd->eh_eflags = 0;
 	list_move_tail(&scmd->eh_entry, done_q);
 }
 EXPORT_SYMBOL(scsi_eh_finish_cmd);
@@ -1163,8 +1151,7 @@ int scsi_eh_get_sense(struct list_head *work_q,
 	 * should not get sense.
 	 */
 	list_for_each_entry_safe(scmd, next, work_q, eh_entry) {
-		if ((scmd->eh_eflags & SCSI_EH_CANCEL_CMD) ||
-		    (scmd->eh_eflags & SCSI_EH_ABORT_SCHEDULED) ||
+		if ((scmd->eh_eflags & SCSI_EH_ABORT_SCHEDULED) ||
 		    SCSI_SENSE_VALID(scmd))
 			continue;
 
@@ -1304,61 +1291,6 @@ static int scsi_eh_test_devices(struct list_head *cmd_list,
 	return list_empty(work_q);
 }
 
-
-/**
- * scsi_eh_abort_cmds - abort pending commands.
- * @work_q:	&list_head for pending commands.
- * @done_q:	&list_head for processed commands.
- *
- * Decription:
- *    Try and see whether or not it makes sense to try and abort the
- *    running command.  This only works out to be the case if we have one
- *    command that has timed out.  If the command simply failed, it makes
- *    no sense to try and abort the command, since as far as the shost
- *    adapter is concerned, it isn't running.
- */
-static int scsi_eh_abort_cmds(struct list_head *work_q,
-			      struct list_head *done_q)
-{
-	struct scsi_cmnd *scmd, *next;
-	LIST_HEAD(check_list);
-	int rtn;
-	struct Scsi_Host *shost;
-
-	list_for_each_entry_safe(scmd, next, work_q, eh_entry) {
-		if (!(scmd->eh_eflags & SCSI_EH_CANCEL_CMD))
-			continue;
-		shost = scmd->device->host;
-		if (scsi_host_eh_past_deadline(shost)) {
-			list_splice_init(&check_list, work_q);
-			SCSI_LOG_ERROR_RECOVERY(3,
-				scmd_printk(KERN_INFO, scmd,
-					    "%s: skip aborting cmd, past eh deadline\n",
-					    current->comm));
-			return list_empty(work_q);
-		}
-		SCSI_LOG_ERROR_RECOVERY(3,
-			scmd_printk(KERN_INFO, scmd,
-				     "%s: aborting cmd\n", current->comm));
-		rtn = scsi_try_to_abort_cmd(shost->hostt, scmd);
-		if (rtn == FAILED) {
-			SCSI_LOG_ERROR_RECOVERY(3,
-				scmd_printk(KERN_INFO, scmd,
-					    "%s: aborting cmd failed\n",
-					     current->comm));
-			list_splice_init(&check_list, work_q);
-			return list_empty(work_q);
-		}
-		scmd->eh_eflags &= ~SCSI_EH_CANCEL_CMD;
-		if (rtn == FAST_IO_FAIL)
-			scsi_eh_finish_cmd(scmd, done_q);
-		else
-			list_move_tail(&scmd->eh_entry, &check_list);
-	}
-
-	return scsi_eh_test_devices(&check_list, work_q, done_q, 0);
-}
-
 /**
  * scsi_eh_try_stu - Send START_UNIT to device.
  * @scmd:	&scsi_cmnd to send START_UNIT
@@ -1701,11 +1633,6 @@ static void scsi_eh_offline_sdevs(struct list_head *work_q,
 		sdev_printk(KERN_INFO, scmd->device, "Device offlined - "
 			    "not ready after error recovery\n");
 		scsi_device_set_state(scmd->device, SDEV_OFFLINE);
-		if (scmd->eh_eflags & SCSI_EH_CANCEL_CMD) {
-			/*
-			 * FIXME: Handle lost cmds.
-			 */
-		}
 		scsi_eh_finish_cmd(scmd, done_q);
 	}
 	return;
@@ -2149,8 +2076,7 @@ static void scsi_unjam_host(struct Scsi_Host *shost)
 	SCSI_LOG_ERROR_RECOVERY(1, scsi_eh_prt_fail_stats(shost, &eh_work_q));
 
 	if (!scsi_eh_get_sense(&eh_work_q, &eh_done_q))
-		if (!scsi_eh_abort_cmds(&eh_work_q, &eh_done_q))
-			scsi_eh_ready_devs(shost, &eh_work_q, &eh_done_q);
+		scsi_eh_ready_devs(shost, &eh_work_q, &eh_done_q);
 
 	spin_lock_irqsave(shost->host_lock, flags);
 	if (shost->eh_deadline != -1)
@@ -2437,44 +2363,34 @@ EXPORT_SYMBOL(scsi_command_normalize_sense);
  *			field will be placed if found.
  *
  * Return value:
- *	1 if information field found, 0 if not found.
+ *	true if information field found, false if not found.
  */
-int scsi_get_sense_info_fld(const u8 * sense_buffer, int sb_len,
-			    u64 * info_out)
+bool scsi_get_sense_info_fld(const u8 *sense_buffer, int sb_len,
+			     u64 *info_out)
 {
-	int j;
 	const u8 * ucp;
-	u64 ull;
 
 	if (sb_len < 7)
-		return 0;
+		return false;
 	switch (sense_buffer[0] & 0x7f) {
 	case 0x70:
 	case 0x71:
 		if (sense_buffer[0] & 0x80) {
-			*info_out = (sense_buffer[3] << 24) +
-				    (sense_buffer[4] << 16) +
-				    (sense_buffer[5] << 8) + sense_buffer[6];
-			return 1;
-		} else
-			return 0;
+			*info_out = get_unaligned_be32(&sense_buffer[3]);
+			return true;
+		}
+		return false;
 	case 0x72:
 	case 0x73:
 		ucp = scsi_sense_desc_find(sense_buffer, sb_len,
 					   0 /* info desc */);
 		if (ucp && (0xa == ucp[1])) {
-			ull = 0;
-			for (j = 0; j < 8; ++j) {
-				if (j > 0)
-					ull <<= 8;
-				ull |= ucp[4 + j];
-			}
-			*info_out = ull;
-			return 1;
-		} else
-			return 0;
+			*info_out = get_unaligned_be64(&ucp[4]);
+			return true;
+		}
+		return false;
 	default:
-		return 0;
+		return false;
 	}
 }
 EXPORT_SYMBOL(scsi_get_sense_info_fld);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 1c3e87d6c48f..a1effcce67b3 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1593,8 +1593,8 @@ static void scsi_softirq_done(struct request *rq)
 			scsi_queue_insert(cmd, SCSI_MLQUEUE_DEVICE_BUSY);
 			break;
 		default:
-			if (!scsi_eh_scmd_add(cmd, 0))
-				scsi_finish_command(cmd);
+			scsi_eh_scmd_add(cmd);
+			break;
 	}
 }
 
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index f11bd102d6d5..59ebc1795bb3 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -18,7 +18,6 @@ struct scsi_nl_hdr;
 /*
  * Scsi Error Handler Flags
  */
-#define SCSI_EH_CANCEL_CMD	0x0001	/* Cancel this cmd */
 #define SCSI_EH_ABORT_SCHEDULED	0x0002	/* Abort has been scheduled */
 
 #define SCSI_SENSE_VALID(scmd) \
@@ -72,7 +71,7 @@ extern enum blk_eh_timer_return scsi_times_out(struct request *req);
 extern int scsi_error_handler(void *host);
 extern int scsi_decide_disposition(struct scsi_cmnd *cmd);
 extern void scsi_eh_wakeup(struct Scsi_Host *shost);
-extern int scsi_eh_scmd_add(struct scsi_cmnd *, int);
+extern void scsi_eh_scmd_add(struct scsi_cmnd *);
 void scsi_eh_ready_devs(struct Scsi_Host *shost,
 			struct list_head *work_q,
 			struct list_head *done_q);
diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 2d753c93e07a..d4cf32d55546 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -289,9 +289,10 @@ static const struct {
 	u32 			value;
 	char			*name;
 } fc_port_role_names[] = {
-	{ FC_PORT_ROLE_FCP_TARGET,	"FCP Target" },
-	{ FC_PORT_ROLE_FCP_INITIATOR,	"FCP Initiator" },
-	{ FC_PORT_ROLE_IP_PORT,		"IP Port" },
+	{ FC_PORT_ROLE_FCP_TARGET,		"FCP Target" },
+	{ FC_PORT_ROLE_FCP_INITIATOR,		"FCP Initiator" },
+	{ FC_PORT_ROLE_IP_PORT,			"IP Port" },
+	{ FC_PORT_ROLE_FCP_DUMMY_INITIATOR,	"FCP Dummy Initiator" },
 };
 fc_bitfield_name_search(port_roles, fc_port_role_names)
 
@@ -850,7 +851,7 @@ static int fc_str_to_dev_loss(const char *buf, unsigned long *val)
 	char *cp;
 
 	*val = simple_strtoul(buf, &cp, 0);
-	if ((*cp && (*cp != '\n')) || (*val < 0))
+	if (*cp && (*cp != '\n'))
 		return -EINVAL;
 	/*
 	 * Check for overflow; dev_loss_tmo is u32
@@ -2628,7 +2629,8 @@ fc_remote_port_create(struct Scsi_Host *shost, int channel,
 	spin_lock_irqsave(shost->host_lock, flags);
 
 	rport->number = fc_host->next_rport_number++;
-	if (rport->roles & FC_PORT_ROLE_FCP_TARGET)
+	if ((rport->roles & FC_PORT_ROLE_FCP_TARGET) ||
+	    (rport->roles & FC_PORT_ROLE_FCP_DUMMY_INITIATOR))
 		rport->scsi_target_id = fc_host->next_target_id++;
 	else
 		rport->scsi_target_id = -1;
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 568c9f26a561..a424eaeafeb0 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -2158,7 +2158,6 @@ static int iscsi_iter_destroy_conn_fn(struct device *dev, void *data)
 
 void iscsi_remove_session(struct iscsi_cls_session *session)
 {
-	struct Scsi_Host *shost = iscsi_session_to_shost(session);
 	unsigned long flags;
 	int err;
 
@@ -2185,7 +2184,7 @@ void iscsi_remove_session(struct iscsi_cls_session *session)
 
 	scsi_target_unblock(&session->dev, SDEV_TRANSPORT_OFFLINE);
 	/* flush running scans then delete devices */
-	scsi_flush_work(shost);
+	flush_work(&session->scan_work);
 	__iscsi_unbind_session(&session->unbind_work);
 
 	/* hw iscsi may not have removed all connections from session */
diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index 9fdbd50c31b4..0ebe2f1bb908 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -370,12 +370,16 @@ EXPORT_SYMBOL(sas_remove_children);
  * sas_remove_host  -  tear down a Scsi_Host's SAS data structures
  * @shost:	Scsi Host that is torn down
  *
- * Removes all SAS PHYs and remote PHYs for a given Scsi_Host.
- * Must be called just before scsi_remove_host for SAS HBAs.
+ * Removes all SAS PHYs and remote PHYs for a given Scsi_Host and remove the
+ * Scsi_Host as well.
+ *
+ * Note: Do not call scsi_remove_host() on the Scsi_Host any more, as it is
+ * already removed.
  */
 void sas_remove_host(struct Scsi_Host *shost)
 {
 	sas_remove_children(&shost->shost_gendev);
+	scsi_remove_host(shost);
 }
 EXPORT_SYMBOL(sas_remove_host);
 
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 0dc95e102e69..f9d1432d7cc5 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -115,6 +115,7 @@ static void sd_rescan(struct device *);
 static int sd_init_command(struct scsi_cmnd *SCpnt);
 static void sd_uninit_command(struct scsi_cmnd *SCpnt);
 static int sd_done(struct scsi_cmnd *);
+static void sd_eh_reset(struct scsi_cmnd *);
 static int sd_eh_action(struct scsi_cmnd *, int);
 static void sd_read_capacity(struct scsi_disk *sdkp, unsigned char *buffer);
 static void scsi_disk_release(struct device *cdev);
@@ -573,6 +574,7 @@ static struct scsi_driver sd_template = {
 	.uninit_command		= sd_uninit_command,
 	.done			= sd_done,
 	.eh_action		= sd_eh_action,
+	.eh_reset		= sd_eh_reset,
 };
 
 /*
@@ -888,8 +890,8 @@ out:
  * sd_setup_write_same_cmnd - write the same data to multiple blocks
  * @cmd: command to prepare
  *
- * Will issue either WRITE SAME(10) or WRITE SAME(16) depending on
- * preference indicated by target device.
+ * Will set up either WRITE SAME(10) or WRITE SAME(16) depending on
+ * the preference indicated by the target device.
  **/
 static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd)
 {
@@ -908,7 +910,7 @@ static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd)
 	BUG_ON(bio_offset(bio) || bio_iovec(bio).bv_len != sdp->sector_size);
 
 	if (sd_is_zoned(sdkp)) {
-		ret = sd_zbc_setup_write_cmnd(cmd);
+		ret = sd_zbc_write_lock_zone(cmd);
 		if (ret != BLKPREP_OK)
 			return ret;
 	}
@@ -980,7 +982,7 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 	unsigned char protect;
 
 	if (zoned_write) {
-		ret = sd_zbc_setup_write_cmnd(SCpnt);
+		ret = sd_zbc_write_lock_zone(SCpnt);
 		if (ret != BLKPREP_OK)
 			return ret;
 	}
@@ -1207,7 +1209,7 @@ static int sd_setup_read_write_cmnd(struct scsi_cmnd *SCpnt)
 	ret = BLKPREP_OK;
  out:
 	if (zoned_write && ret != BLKPREP_OK)
-		sd_zbc_cancel_write_cmnd(SCpnt);
+		sd_zbc_write_unlock_zone(SCpnt);
 
 	return ret;
 }
@@ -1264,8 +1266,8 @@ static void sd_uninit_command(struct scsi_cmnd *SCpnt)
 
 /**
  *	sd_open - open a scsi disk device
- *	@inode: only i_rdev member may be used
- *	@filp: only f_mode and f_flags may be used
+ *	@bdev: Block device of the scsi disk to open
+ *	@mode: FMODE_* mask
  *
  *	Returns 0 if successful. Returns a negated errno value in case 
  *	of error.
@@ -1341,8 +1343,8 @@ error_out:
 /**
  *	sd_release - invoked when the (last) close(2) is called on this
  *	scsi disk.
- *	@inode: only i_rdev member may be used
- *	@filp: only f_mode and f_flags may be used
+ *	@disk: disk to release
+ *	@mode: FMODE_* mask
  *
  *	Returns 0. 
  *
@@ -1398,8 +1400,8 @@ static int sd_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 
 /**
  *	sd_ioctl - process an ioctl
- *	@inode: only i_rdev/i_bdev members may be used
- *	@filp: only f_mode and f_flags may be used
+ *	@bdev: target block device
+ *	@mode: FMODE_* mask
  *	@cmd: ioctl command number
  *	@arg: this is third argument given to ioctl(2) system call.
  *	Often contains a pointer.
@@ -1762,6 +1764,26 @@ static const struct block_device_operations sd_fops = {
 };
 
 /**
+ *	sd_eh_reset - reset error handling callback
+ *	@scmd:		sd-issued command that has failed
+ *
+ *	This function is called by the SCSI midlayer before starting
+ *	SCSI EH. When counting medium access failures we have to be
+ *	careful to register it only only once per device and SCSI EH run;
+ *	there might be several timed out commands which will cause the
+ *	'max_medium_access_timeouts' counter to trigger after the first
+ *	SCSI EH run already and set the device to offline.
+ *	So this function resets the internal counter before starting SCSI EH.
+ **/
+static void sd_eh_reset(struct scsi_cmnd *scmd)
+{
+	struct scsi_disk *sdkp = scsi_disk(scmd->request->rq_disk);
+
+	/* New SCSI EH run, reset gate variable */
+	sdkp->ignore_medium_access_errors = false;
+}
+
+/**
  *	sd_eh_action - error handling callback
  *	@scmd:		sd-issued command that has failed
  *	@eh_disp:	The recovery disposition suggested by the midlayer
@@ -1790,7 +1812,10 @@ static int sd_eh_action(struct scsi_cmnd *scmd, int eh_disp)
 	 * process of recovering or has it suffered an internal failure
 	 * that prevents access to the storage medium.
 	 */
-	sdkp->medium_access_timed_out++;
+	if (!sdkp->ignore_medium_access_errors) {
+		sdkp->medium_access_timed_out++;
+		sdkp->ignore_medium_access_errors = true;
+	}
 
 	/*
 	 * If the device keeps failing read/write commands but TEST UNIT
@@ -1802,7 +1827,7 @@ static int sd_eh_action(struct scsi_cmnd *scmd, int eh_disp)
 			    "Medium access timeout failure. Offlining disk!\n");
 		scsi_device_set_state(scmd->device, SDEV_OFFLINE);
 
-		return FAILED;
+		return SUCCESS;
 	}
 
 	return eh_disp;
@@ -1810,41 +1835,44 @@ static int sd_eh_action(struct scsi_cmnd *scmd, int eh_disp)
 
 static unsigned int sd_completed_bytes(struct scsi_cmnd *scmd)
 {
-	u64 start_lba = blk_rq_pos(scmd->request);
-	u64 end_lba = blk_rq_pos(scmd->request) + (scsi_bufflen(scmd) / 512);
-	u64 factor = scmd->device->sector_size / 512;
-	u64 bad_lba;
-	int info_valid;
+	struct request *req = scmd->request;
+	struct scsi_device *sdev = scmd->device;
+	unsigned int transferred, good_bytes;
+	u64 start_lba, end_lba, bad_lba;
+
 	/*
-	 * resid is optional but mostly filled in.  When it's unused,
-	 * its value is zero, so we assume the whole buffer transferred
+	 * Some commands have a payload smaller than the device logical
+	 * block size (e.g. INQUIRY on a 4K disk).
 	 */
-	unsigned int transferred = scsi_bufflen(scmd) - scsi_get_resid(scmd);
-	unsigned int good_bytes;
-
-	info_valid = scsi_get_sense_info_fld(scmd->sense_buffer,
-					     SCSI_SENSE_BUFFERSIZE,
-					     &bad_lba);
-	if (!info_valid)
+	if (scsi_bufflen(scmd) <= sdev->sector_size)
 		return 0;
 
-	if (scsi_bufflen(scmd) <= scmd->device->sector_size)
+	/* Check if we have a 'bad_lba' information */
+	if (!scsi_get_sense_info_fld(scmd->sense_buffer,
+				     SCSI_SENSE_BUFFERSIZE,
+				     &bad_lba))
 		return 0;
 
-	/* be careful ... don't want any overflows */
-	do_div(start_lba, factor);
-	do_div(end_lba, factor);
-
-	/* The bad lba was reported incorrectly, we have no idea where
+	/*
+	 * If the bad lba was reported incorrectly, we have no idea where
 	 * the error is.
 	 */
-	if (bad_lba < start_lba  || bad_lba >= end_lba)
+	start_lba = sectors_to_logical(sdev, blk_rq_pos(req));
+	end_lba = start_lba + bytes_to_logical(sdev, scsi_bufflen(scmd));
+	if (bad_lba < start_lba || bad_lba >= end_lba)
 		return 0;
 
-	/* This computation should always be done in terms of
-	 * the resolution of the device's medium.
+	/*
+	 * resid is optional but mostly filled in.  When it's unused,
+	 * its value is zero, so we assume the whole buffer transferred
+	 */
+	transferred = scsi_bufflen(scmd) - scsi_get_resid(scmd);
+
+	/* This computation should always be done in terms of the
+	 * resolution of the device's medium.
 	 */
-	good_bytes = (bad_lba - start_lba) * scmd->device->sector_size;
+	good_bytes = logical_to_bytes(sdev, bad_lba - start_lba);
+
 	return min(good_bytes, transferred);
 }
 
@@ -1866,8 +1894,6 @@ static int sd_done(struct scsi_cmnd *SCpnt)
 	struct request *req = SCpnt->request;
 	int sense_valid = 0;
 	int sense_deferred = 0;
-	unsigned char op = SCpnt->cmnd[0];
-	unsigned char unmap = SCpnt->cmnd[1] & 8;
 
 	switch (req_op(req)) {
 	case REQ_OP_DISCARD:
@@ -1941,26 +1967,27 @@ static int sd_done(struct scsi_cmnd *SCpnt)
 			good_bytes = sd_completed_bytes(SCpnt);
 		break;
 	case ILLEGAL_REQUEST:
-		if (sshdr.asc == 0x10)  /* DIX: Host detected corruption */
+		switch (sshdr.asc) {
+		case 0x10:	/* DIX: Host detected corruption */
 			good_bytes = sd_completed_bytes(SCpnt);
-		/* INVALID COMMAND OPCODE or INVALID FIELD IN CDB */
-		if (sshdr.asc == 0x20 || sshdr.asc == 0x24) {
-			switch (op) {
+			break;
+		case 0x20:	/* INVALID COMMAND OPCODE */
+		case 0x24:	/* INVALID FIELD IN CDB */
+			switch (SCpnt->cmnd[0]) {
 			case UNMAP:
 				sd_config_discard(sdkp, SD_LBP_DISABLE);
 				break;
 			case WRITE_SAME_16:
 			case WRITE_SAME:
-				if (unmap)
+				if (SCpnt->cmnd[1] & 8) { /* UNMAP */
 					sd_config_discard(sdkp, SD_LBP_DISABLE);
-				else {
+				} else {
 					sdkp->device->no_write_same = 1;
 					sd_config_write_same(sdkp);
-
-					good_bytes = 0;
 					req->__data_len = blk_rq_bytes(req);
 					req->rq_flags |= RQF_QUIET;
 				}
+				break;
 			}
 		}
 		break;
@@ -2798,7 +2825,7 @@ static void sd_read_app_tag_own(struct scsi_disk *sdkp, unsigned char *buffer)
 
 /**
  * sd_read_block_limits - Query disk device for preferred I/O sizes.
- * @disk: disk to query
+ * @sdkp: disk to query
  */
 static void sd_read_block_limits(struct scsi_disk *sdkp)
 {
@@ -2864,7 +2891,7 @@ static void sd_read_block_limits(struct scsi_disk *sdkp)
 
 /**
  * sd_read_block_characteristics - Query block dev. characteristics
- * @disk: disk to query
+ * @sdkp: disk to query
  */
 static void sd_read_block_characteristics(struct scsi_disk *sdkp)
 {
@@ -2912,7 +2939,7 @@ static void sd_read_block_characteristics(struct scsi_disk *sdkp)
 
 /**
  * sd_read_block_provisioning - Query provisioning VPD page
- * @disk: disk to query
+ * @sdkp: disk to query
  */
 static void sd_read_block_provisioning(struct scsi_disk *sdkp)
 {
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index a2c4b5c35379..61d02efd366c 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -114,6 +114,7 @@ struct scsi_disk {
 	unsigned	rc_basis: 2;
 	unsigned	zoned: 2;
 	unsigned	urswrz : 1;
+	unsigned	ignore_medium_access_errors : 1;
 };
 #define to_scsi_disk(obj) container_of(obj,struct scsi_disk,dev)
 
@@ -176,6 +177,11 @@ static inline unsigned int logical_to_bytes(struct scsi_device *sdev, sector_t b
 	return blocks * sdev->sector_size;
 }
 
+static inline sector_t bytes_to_logical(struct scsi_device *sdev, unsigned int bytes)
+{
+	return bytes >> ilog2(sdev->sector_size);
+}
+
 static inline sector_t sectors_to_logical(struct scsi_device *sdev, sector_t sector)
 {
 	return sector >> (ilog2(sdev->sector_size) - 9);
@@ -274,8 +280,8 @@ static inline int sd_is_zoned(struct scsi_disk *sdkp)
 extern int sd_zbc_read_zones(struct scsi_disk *sdkp, unsigned char *buffer);
 extern void sd_zbc_remove(struct scsi_disk *sdkp);
 extern void sd_zbc_print_zones(struct scsi_disk *sdkp);
-extern int sd_zbc_setup_write_cmnd(struct scsi_cmnd *cmd);
-extern void sd_zbc_cancel_write_cmnd(struct scsi_cmnd *cmd);
+extern int sd_zbc_write_lock_zone(struct scsi_cmnd *cmd);
+extern void sd_zbc_write_unlock_zone(struct scsi_cmnd *cmd);
 extern int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd);
 extern int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd);
 extern void sd_zbc_complete(struct scsi_cmnd *cmd, unsigned int good_bytes,
@@ -293,13 +299,13 @@ static inline void sd_zbc_remove(struct scsi_disk *sdkp) {}
 
 static inline void sd_zbc_print_zones(struct scsi_disk *sdkp) {}
 
-static inline int sd_zbc_setup_write_cmnd(struct scsi_cmnd *cmd)
+static inline int sd_zbc_write_lock_zone(struct scsi_cmnd *cmd)
 {
 	/* Let the drive fail requests */
 	return BLKPREP_OK;
 }
 
-static inline void sd_zbc_cancel_write_cmnd(struct scsi_cmnd *cmd) {}
+static inline void sd_zbc_write_unlock_zone(struct scsi_cmnd *cmd) {}
 
 static inline int sd_zbc_setup_report_cmnd(struct scsi_cmnd *cmd)
 {
diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c
index 1994f7799fce..96855df9f49d 100644
--- a/drivers/scsi/sd_zbc.c
+++ b/drivers/scsi/sd_zbc.c
@@ -237,7 +237,6 @@ int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd)
 	struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
 	sector_t sector = blk_rq_pos(rq);
 	sector_t block = sectors_to_logical(sdkp->device, sector);
-	unsigned int zno = block >> sdkp->zone_shift;
 
 	if (!sd_is_zoned(sdkp))
 		/* Not a zoned device */
@@ -250,11 +249,6 @@ int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd)
 		/* Unaligned request */
 		return BLKPREP_KILL;
 
-	/* Do not allow concurrent reset and writes */
-	if (sdkp->zones_wlock &&
-	    test_and_set_bit(zno, sdkp->zones_wlock))
-		return BLKPREP_DEFER;
-
 	cmd->cmd_len = 16;
 	memset(cmd->cmnd, 0, cmd->cmd_len);
 	cmd->cmnd[0] = ZBC_OUT;
@@ -269,7 +263,7 @@ int sd_zbc_setup_reset_cmnd(struct scsi_cmnd *cmd)
 	return BLKPREP_OK;
 }
 
-int sd_zbc_setup_write_cmnd(struct scsi_cmnd *cmd)
+int sd_zbc_write_lock_zone(struct scsi_cmnd *cmd)
 {
 	struct request *rq = cmd->request;
 	struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
@@ -303,8 +297,9 @@ int sd_zbc_setup_write_cmnd(struct scsi_cmnd *cmd)
 	return BLKPREP_OK;
 }
 
-static void sd_zbc_unlock_zone(struct request *rq)
+void sd_zbc_write_unlock_zone(struct scsi_cmnd *cmd)
 {
+	struct request *rq = cmd->request;
 	struct scsi_disk *sdkp = scsi_disk(rq->rq_disk);
 
 	if (sdkp->zones_wlock) {
@@ -315,11 +310,6 @@ static void sd_zbc_unlock_zone(struct request *rq)
 	}
 }
 
-void sd_zbc_cancel_write_cmnd(struct scsi_cmnd *cmd)
-{
-	sd_zbc_unlock_zone(cmd->request);
-}
-
 void sd_zbc_complete(struct scsi_cmnd *cmd,
 		     unsigned int good_bytes,
 		     struct scsi_sense_hdr *sshdr)
@@ -328,39 +318,35 @@ void sd_zbc_complete(struct scsi_cmnd *cmd,
 	struct request *rq = cmd->request;
 
 	switch (req_op(rq)) {
+	case REQ_OP_ZONE_RESET:
+
+		if (result &&
+		    sshdr->sense_key == ILLEGAL_REQUEST &&
+		    sshdr->asc == 0x24)
+			/*
+			 * INVALID FIELD IN CDB error: reset of a conventional
+			 * zone was attempted. Nothing to worry about, so be
+			 * quiet about the error.
+			 */
+			rq->rq_flags |= RQF_QUIET;
+		break;
+
 	case REQ_OP_WRITE:
 	case REQ_OP_WRITE_ZEROES:
 	case REQ_OP_WRITE_SAME:
-	case REQ_OP_ZONE_RESET:
 
 		/* Unlock the zone */
-		sd_zbc_unlock_zone(rq);
+		sd_zbc_write_unlock_zone(cmd);
 
-		if (!result ||
-		    sshdr->sense_key != ILLEGAL_REQUEST)
-			break;
-
-		switch (sshdr->asc) {
-		case 0x24:
-			/*
-			 * INVALID FIELD IN CDB error: For a zone reset,
-			 * this means that a reset of a conventional
-			 * zone was attempted. Nothing to worry about in
-			 * this case, so be quiet about the error.
-			 */
-			if (req_op(rq) == REQ_OP_ZONE_RESET)
-				rq->rq_flags |= RQF_QUIET;
-			break;
-		case 0x21:
+		if (result &&
+		    sshdr->sense_key == ILLEGAL_REQUEST &&
+		    sshdr->asc == 0x21)
 			/*
 			 * INVALID ADDRESS FOR WRITE error: It is unlikely that
 			 * retrying write requests failed with any kind of
 			 * alignement error will result in success. So don't.
 			 */
 			cmd->allowed = 0;
-			break;
-		}
-
 		break;
 
 	case REQ_OP_ZONE_REPORT:
@@ -565,8 +551,7 @@ static int sd_zbc_setup(struct scsi_disk *sdkp)
 int sd_zbc_read_zones(struct scsi_disk *sdkp,
 		      unsigned char *buf)
 {
-	sector_t capacity;
-	int ret = 0;
+	int ret;
 
 	if (!sd_is_zoned(sdkp))
 		/*
@@ -598,7 +583,6 @@ int sd_zbc_read_zones(struct scsi_disk *sdkp,
 	ret = sd_zbc_check_capacity(sdkp, buf);
 	if (ret)
 		goto err;
-	capacity = logical_to_sectors(sdkp->device, sdkp->capacity);
 
 	/*
 	 * Check zone size: only devices with a constant zone size (except
diff --git a/drivers/scsi/ses.c b/drivers/scsi/ses.c
index 50adabbb5808..f1cdf32d7514 100644
--- a/drivers/scsi/ses.c
+++ b/drivers/scsi/ses.c
@@ -548,7 +548,6 @@ static void ses_enclosure_data_process(struct enclosure_device *edev,
 					ecomp = &edev->component[components++];
 
 				if (!IS_ERR(ecomp)) {
-					ses_get_power_status(edev, ecomp);
 					if (addl_desc_ptr)
 						ses_process_descriptor(
 							ecomp,
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 0b60245bd740..0a38ba01b7b4 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -122,7 +122,7 @@ struct sg_device;		/* forward declarations */
 struct sg_fd;
 
 typedef struct sg_request {	/* SG_MAX_QUEUE requests outstanding per file */
-	struct sg_request *nextrp;	/* NULL -> tail request (slist) */
+	struct list_head entry;	/* list entry */
 	struct sg_fd *parentfp;	/* NULL -> not in use */
 	Sg_scatter_hold data;	/* hold buffer, perhaps scatter list */
 	sg_io_hdr_t header;	/* scsi command+info, see <scsi/sg.h> */
@@ -142,19 +142,19 @@ typedef struct sg_fd {		/* holds the state of a file descriptor */
 	struct sg_device *parentdp;	/* owning device */
 	wait_queue_head_t read_wait;	/* queue read until command done */
 	rwlock_t rq_list_lock;	/* protect access to list in req_arr */
+	struct mutex f_mutex;	/* protect against changes in this fd */
 	int timeout;		/* defaults to SG_DEFAULT_TIMEOUT      */
 	int timeout_user;	/* defaults to SG_DEFAULT_TIMEOUT_USER */
 	Sg_scatter_hold reserve;	/* buffer held for this file descriptor */
-	unsigned save_scat_len;	/* original length of trunc. scat. element */
-	Sg_request *headrp;	/* head of request slist, NULL->empty */
+	struct list_head rq_list; /* head of request list */
 	struct fasync_struct *async_qp;	/* used by asynchronous notification */
 	Sg_request req_arr[SG_MAX_QUEUE];	/* used as singly-linked list */
-	char low_dma;		/* as in parent but possibly overridden to 1 */
 	char force_packid;	/* 1 -> pack_id input to read(), 0 -> ignored */
 	char cmd_q;		/* 1 -> allow command queuing, 0 -> don't */
 	unsigned char next_cmd_len; /* 0: automatic, >0: use on next write() */
 	char keep_orphan;	/* 0 -> drop orphan (def), 1 -> keep for read() */
 	char mmap_called;	/* 0 -> mmap() never called on this fd */
+	char res_in_use;	/* 1 -> 'reserve' array in use */
 	struct kref f_ref;
 	struct execute_work ew;
 } Sg_fd;
@@ -198,7 +198,6 @@ static void sg_remove_sfp(struct kref *);
 static Sg_request *sg_get_rq_mark(Sg_fd * sfp, int pack_id);
 static Sg_request *sg_add_request(Sg_fd * sfp);
 static int sg_remove_request(Sg_fd * sfp, Sg_request * srp);
-static int sg_res_in_use(Sg_fd * sfp);
 static Sg_device *sg_get_dev(int dev);
 static void sg_device_destroy(struct kref *kref);
 
@@ -525,6 +524,7 @@ sg_read(struct file *filp, char __user *buf, size_t count, loff_t * ppos)
 	} else
 		count = (old_hdr->result == 0) ? 0 : -EIO;
 	sg_finish_rem_req(srp);
+	sg_remove_request(sfp, srp);
 	retval = count;
 free_old_hdr:
 	kfree(old_hdr);
@@ -565,6 +565,7 @@ sg_new_read(Sg_fd * sfp, char __user *buf, size_t count, Sg_request * srp)
 	}
 err_out:
 	err2 = sg_finish_rem_req(srp);
+	sg_remove_request(sfp, srp);
 	return err ? : err2 ? : count;
 }
 
@@ -614,6 +615,7 @@ sg_write(struct file *filp, const char __user *buf, size_t count, loff_t * ppos)
 	}
 	buf += SZ_SG_HEADER;
 	__get_user(opcode, buf);
+	mutex_lock(&sfp->f_mutex);
 	if (sfp->next_cmd_len > 0) {
 		cmd_size = sfp->next_cmd_len;
 		sfp->next_cmd_len = 0;	/* reset so only this write() effected */
@@ -622,6 +624,7 @@ sg_write(struct file *filp, const char __user *buf, size_t count, loff_t * ppos)
 		if ((opcode >= 0xc0) && old_hdr.twelve_byte)
 			cmd_size = 12;
 	}
+	mutex_unlock(&sfp->f_mutex);
 	SCSI_LOG_TIMEOUT(4, sg_printk(KERN_INFO, sdp,
 		"sg_write:   scsi opcode=0x%02x, cmd_size=%d\n", (int) opcode, cmd_size));
 /* Determine buffer size.  */
@@ -662,18 +665,14 @@ sg_write(struct file *filp, const char __user *buf, size_t count, loff_t * ppos)
 	 * is a non-zero input_size, so emit a warning.
 	 */
 	if (hp->dxfer_direction == SG_DXFER_TO_FROM_DEV) {
-		static char cmd[TASK_COMM_LEN];
-		if (strcmp(current->comm, cmd)) {
-			printk_ratelimited(KERN_WARNING
-					   "sg_write: data in/out %d/%d bytes "
-					   "for SCSI command 0x%x-- guessing "
-					   "data in;\n   program %s not setting "
-					   "count and/or reply_len properly\n",
-					   old_hdr.reply_len - (int)SZ_SG_HEADER,
-					   input_size, (unsigned int) cmnd[0],
-					   current->comm);
-			strcpy(cmd, current->comm);
-		}
+		printk_ratelimited(KERN_WARNING
+				   "sg_write: data in/out %d/%d bytes "
+				   "for SCSI command 0x%x-- guessing "
+				   "data in;\n   program %s not setting "
+				   "count and/or reply_len properly\n",
+				   old_hdr.reply_len - (int)SZ_SG_HEADER,
+				   input_size, (unsigned int) cmnd[0],
+				   current->comm);
 	}
 	k = sg_common_write(sfp, srp, cmnd, sfp->timeout, blocking);
 	return (k < 0) ? k : count;
@@ -721,7 +720,7 @@ sg_new_write(Sg_fd *sfp, struct file *file, const char __user *buf,
 			sg_remove_request(sfp, srp);
 			return -EINVAL;	/* either MMAP_IO or DIRECT_IO (not both) */
 		}
-		if (sg_res_in_use(sfp)) {
+		if (sfp->res_in_use) {
 			sg_remove_request(sfp, srp);
 			return -EBUSY;	/* reserve buffer already being used */
 		}
@@ -752,6 +751,29 @@ sg_new_write(Sg_fd *sfp, struct file *file, const char __user *buf,
 	return count;
 }
 
+static bool sg_is_valid_dxfer(sg_io_hdr_t *hp)
+{
+	switch (hp->dxfer_direction) {
+	case SG_DXFER_NONE:
+		if (hp->dxferp || hp->dxfer_len > 0)
+			return false;
+		return true;
+	case SG_DXFER_TO_DEV:
+	case SG_DXFER_FROM_DEV:
+	case SG_DXFER_TO_FROM_DEV:
+		if (!hp->dxferp || hp->dxfer_len == 0)
+			return false;
+		return true;
+	case SG_DXFER_UNKNOWN:
+		if ((!hp->dxferp && hp->dxfer_len) ||
+		    (hp->dxferp && hp->dxfer_len == 0))
+			return false;
+		return true;
+	default:
+		return false;
+	}
+}
+
 static int
 sg_common_write(Sg_fd * sfp, Sg_request * srp,
 		unsigned char *cmnd, int timeout, int blocking)
@@ -772,11 +794,15 @@ sg_common_write(Sg_fd * sfp, Sg_request * srp,
 			"sg_common_write:  scsi opcode=0x%02x, cmd_size=%d\n",
 			(int) cmnd[0], (int) hp->cmd_len));
 
+	if (!sg_is_valid_dxfer(hp))
+		return -EINVAL;
+
 	k = sg_start_req(srp, cmnd);
 	if (k) {
 		SCSI_LOG_TIMEOUT(1, sg_printk(KERN_INFO, sfp->parentdp,
 			"sg_common_write: start_req err=%d\n", k));
 		sg_finish_rem_req(srp);
+		sg_remove_request(sfp, srp);
 		return k;	/* probably out of space --> ENOMEM */
 	}
 	if (atomic_read(&sdp->detaching)) {
@@ -787,6 +813,7 @@ sg_common_write(Sg_fd * sfp, Sg_request * srp,
 		}
 
 		sg_finish_rem_req(srp);
+		sg_remove_request(sfp, srp);
 		return -ENODEV;
 	}
 
@@ -885,24 +912,14 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg)
 				/* strange ..., for backward compatibility */
 		return sfp->timeout_user;
 	case SG_SET_FORCE_LOW_DMA:
-		result = get_user(val, ip);
-		if (result)
-			return result;
-		if (val) {
-			sfp->low_dma = 1;
-			if ((0 == sfp->low_dma) && (0 == sg_res_in_use(sfp))) {
-				val = (int) sfp->reserve.bufflen;
-				sg_remove_scat(sfp, &sfp->reserve);
-				sg_build_reserve(sfp, val);
-			}
-		} else {
-			if (atomic_read(&sdp->detaching))
-				return -ENODEV;
-			sfp->low_dma = sdp->device->host->unchecked_isa_dma;
-		}
+		/*
+		 * N.B. This ioctl never worked properly, but failed to
+		 * return an error value. So returning '0' to keep compability
+		 * with legacy applications.
+		 */
 		return 0;
 	case SG_GET_LOW_DMA:
-		return put_user((int) sfp->low_dma, ip);
+		return put_user((int) sdp->device->host->unchecked_isa_dma, ip);
 	case SG_GET_SCSI_ID:
 		if (!access_ok(VERIFY_WRITE, p, sizeof (sg_scsi_id_t)))
 			return -EFAULT;
@@ -936,7 +953,7 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg)
 		if (!access_ok(VERIFY_WRITE, ip, sizeof (int)))
 			return -EFAULT;
 		read_lock_irqsave(&sfp->rq_list_lock, iflags);
-		for (srp = sfp->headrp; srp; srp = srp->nextrp) {
+		list_for_each_entry(srp, &sfp->rq_list, entry) {
 			if ((1 == srp->done) && (!srp->sg_io_owned)) {
 				read_unlock_irqrestore(&sfp->rq_list_lock,
 						       iflags);
@@ -949,7 +966,8 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg)
 		return 0;
 	case SG_GET_NUM_WAITING:
 		read_lock_irqsave(&sfp->rq_list_lock, iflags);
-		for (val = 0, srp = sfp->headrp; srp; srp = srp->nextrp) {
+		val = 0;
+		list_for_each_entry(srp, &sfp->rq_list, entry) {
 			if ((1 == srp->done) && (!srp->sg_io_owned))
 				++val;
 		}
@@ -965,12 +983,18 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg)
                         return -EINVAL;
 		val = min_t(int, val,
 			    max_sectors_bytes(sdp->device->request_queue));
+		mutex_lock(&sfp->f_mutex);
 		if (val != sfp->reserve.bufflen) {
-			if (sg_res_in_use(sfp) || sfp->mmap_called)
+			if (sfp->mmap_called ||
+			    sfp->res_in_use) {
+				mutex_unlock(&sfp->f_mutex);
 				return -EBUSY;
+			}
+
 			sg_remove_scat(sfp, &sfp->reserve);
 			sg_build_reserve(sfp, val);
 		}
+		mutex_unlock(&sfp->f_mutex);
 		return 0;
 	case SG_GET_RESERVED_SIZE:
 		val = min_t(int, sfp->reserve.bufflen,
@@ -1018,35 +1042,33 @@ sg_ioctl(struct file *filp, unsigned int cmd_in, unsigned long arg)
 			if (!rinfo)
 				return -ENOMEM;
 			read_lock_irqsave(&sfp->rq_list_lock, iflags);
-			for (srp = sfp->headrp, val = 0; val < SG_MAX_QUEUE;
-			     ++val, srp = srp ? srp->nextrp : srp) {
+			val = 0;
+			list_for_each_entry(srp, &sfp->rq_list, entry) {
+				if (val > SG_MAX_QUEUE)
+					break;
 				memset(&rinfo[val], 0, SZ_SG_REQ_INFO);
-				if (srp) {
-					rinfo[val].req_state = srp->done + 1;
-					rinfo[val].problem =
-					    srp->header.masked_status & 
-					    srp->header.host_status & 
-					    srp->header.driver_status;
-					if (srp->done)
-						rinfo[val].duration =
-							srp->header.duration;
-					else {
-						ms = jiffies_to_msecs(jiffies);
-						rinfo[val].duration =
-						    (ms > srp->header.duration) ?
-						    (ms - srp->header.duration) : 0;
-					}
-					rinfo[val].orphan = srp->orphan;
-					rinfo[val].sg_io_owned =
-							srp->sg_io_owned;
-					rinfo[val].pack_id =
-							srp->header.pack_id;
-					rinfo[val].usr_ptr =
-							srp->header.usr_ptr;
+				rinfo[val].req_state = srp->done + 1;
+				rinfo[val].problem =
+					srp->header.masked_status &
+					srp->header.host_status &
+					srp->header.driver_status;
+				if (srp->done)
+					rinfo[val].duration =
+						srp->header.duration;
+				else {
+					ms = jiffies_to_msecs(jiffies);
+					rinfo[val].duration =
+						(ms > srp->header.duration) ?
+						(ms - srp->header.duration) : 0;
 				}
+				rinfo[val].orphan = srp->orphan;
+				rinfo[val].sg_io_owned = srp->sg_io_owned;
+				rinfo[val].pack_id = srp->header.pack_id;
+				rinfo[val].usr_ptr = srp->header.usr_ptr;
+				val++;
 			}
 			read_unlock_irqrestore(&sfp->rq_list_lock, iflags);
-			result = __copy_to_user(p, rinfo, 
+			result = __copy_to_user(p, rinfo,
 						SZ_SG_REQ_INFO * SG_MAX_QUEUE);
 			result = result ? -EFAULT : 0;
 			kfree(rinfo);
@@ -1152,7 +1174,7 @@ sg_poll(struct file *filp, poll_table * wait)
 		return POLLERR;
 	poll_wait(filp, &sfp->read_wait, wait);
 	read_lock_irqsave(&sfp->rq_list_lock, iflags);
-	for (srp = sfp->headrp; srp; srp = srp->nextrp) {
+	list_for_each_entry(srp, &sfp->rq_list, entry) {
 		/* if any read waiting, flag it */
 		if ((0 == res) && (1 == srp->done) && (!srp->sg_io_owned))
 			res = POLLIN | POLLRDNORM;
@@ -1269,6 +1291,7 @@ sg_rq_end_io_usercontext(struct work_struct *work)
 	struct sg_fd *sfp = srp->parentfp;
 
 	sg_finish_rem_req(srp);
+	sg_remove_request(sfp, srp);
 	kref_put(&sfp->f_ref, sg_remove_sfp);
 }
 
@@ -1732,13 +1755,22 @@ sg_start_req(Sg_request *srp, unsigned char *cmd)
 		md = &map_data;
 
 	if (md) {
-		if (!sg_res_in_use(sfp) && dxfer_len <= rsv_schp->bufflen)
+		mutex_lock(&sfp->f_mutex);
+		if (dxfer_len <= rsv_schp->bufflen &&
+		    !sfp->res_in_use) {
+			sfp->res_in_use = 1;
 			sg_link_reserve(sfp, srp, dxfer_len);
-		else {
+		} else if ((hp->flags & SG_FLAG_MMAP_IO) && sfp->res_in_use) {
+			mutex_unlock(&sfp->f_mutex);
+			return -EBUSY;
+		} else {
 			res = sg_build_indirect(req_schp, sfp, dxfer_len);
-			if (res)
+			if (res) {
+				mutex_unlock(&sfp->f_mutex);
 				return res;
+			}
 		}
+		mutex_unlock(&sfp->f_mutex);
 
 		md->pages = req_schp->pages;
 		md->page_order = req_schp->page_order;
@@ -1806,8 +1838,6 @@ sg_finish_rem_req(Sg_request *srp)
 	else
 		sg_remove_scat(sfp, req_schp);
 
-	sg_remove_request(sfp, srp);
-
 	return ret;
 }
 
@@ -1831,6 +1861,7 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size)
 	int sg_tablesize = sfp->parentdp->sg_tablesize;
 	int blk_size = buff_size, order;
 	gfp_t gfp_mask = GFP_ATOMIC | __GFP_COMP | __GFP_NOWARN;
+	struct sg_device *sdp = sfp->parentdp;
 
 	if (blk_size < 0)
 		return -EFAULT;
@@ -1856,7 +1887,7 @@ sg_build_indirect(Sg_scatter_hold * schp, Sg_fd * sfp, int buff_size)
 			scatter_elem_sz_prev = num;
 	}
 
-	if (sfp->low_dma)
+	if (sdp->device->host->unchecked_isa_dma)
 		gfp_mask |= GFP_DMA;
 
 	if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
@@ -2026,8 +2057,9 @@ sg_unlink_reserve(Sg_fd * sfp, Sg_request * srp)
 	req_schp->pages = NULL;
 	req_schp->page_order = 0;
 	req_schp->sglist_len = 0;
-	sfp->save_scat_len = 0;
 	srp->res_used = 0;
+	/* Called without mutex lock to avoid deadlock */
+	sfp->res_in_use = 0;
 }
 
 static Sg_request *
@@ -2037,7 +2069,7 @@ sg_get_rq_mark(Sg_fd * sfp, int pack_id)
 	unsigned long iflags;
 
 	write_lock_irqsave(&sfp->rq_list_lock, iflags);
-	for (resp = sfp->headrp; resp; resp = resp->nextrp) {
+	list_for_each_entry(resp, &sfp->rq_list, entry) {
 		/* look for requests that are ready + not SG_IO owned */
 		if ((1 == resp->done) && (!resp->sg_io_owned) &&
 		    ((-1 == pack_id) || (resp->header.pack_id == pack_id))) {
@@ -2055,70 +2087,45 @@ sg_add_request(Sg_fd * sfp)
 {
 	int k;
 	unsigned long iflags;
-	Sg_request *resp;
 	Sg_request *rp = sfp->req_arr;
 
 	write_lock_irqsave(&sfp->rq_list_lock, iflags);
-	resp = sfp->headrp;
-	if (!resp) {
-		memset(rp, 0, sizeof (Sg_request));
-		rp->parentfp = sfp;
-		resp = rp;
-		sfp->headrp = resp;
-	} else {
-		if (0 == sfp->cmd_q)
-			resp = NULL;	/* command queuing disallowed */
-		else {
-			for (k = 0; k < SG_MAX_QUEUE; ++k, ++rp) {
-				if (!rp->parentfp)
-					break;
-			}
-			if (k < SG_MAX_QUEUE) {
-				memset(rp, 0, sizeof (Sg_request));
-				rp->parentfp = sfp;
-				while (resp->nextrp)
-					resp = resp->nextrp;
-				resp->nextrp = rp;
-				resp = rp;
-			} else
-				resp = NULL;
+	if (!list_empty(&sfp->rq_list)) {
+		if (!sfp->cmd_q)
+			goto out_unlock;
+
+		for (k = 0; k < SG_MAX_QUEUE; ++k, ++rp) {
+			if (!rp->parentfp)
+				break;
 		}
+		if (k >= SG_MAX_QUEUE)
+			goto out_unlock;
 	}
-	if (resp) {
-		resp->nextrp = NULL;
-		resp->header.duration = jiffies_to_msecs(jiffies);
-	}
+	memset(rp, 0, sizeof (Sg_request));
+	rp->parentfp = sfp;
+	rp->header.duration = jiffies_to_msecs(jiffies);
+	list_add_tail(&rp->entry, &sfp->rq_list);
 	write_unlock_irqrestore(&sfp->rq_list_lock, iflags);
-	return resp;
+	return rp;
+out_unlock:
+	write_unlock_irqrestore(&sfp->rq_list_lock, iflags);
+	return NULL;
 }
 
 /* Return of 1 for found; 0 for not found */
 static int
 sg_remove_request(Sg_fd * sfp, Sg_request * srp)
 {
-	Sg_request *prev_rp;
-	Sg_request *rp;
 	unsigned long iflags;
 	int res = 0;
 
-	if ((!sfp) || (!srp) || (!sfp->headrp))
+	if (!sfp || !srp || list_empty(&sfp->rq_list))
 		return res;
 	write_lock_irqsave(&sfp->rq_list_lock, iflags);
-	prev_rp = sfp->headrp;
-	if (srp == prev_rp) {
-		sfp->headrp = prev_rp->nextrp;
-		prev_rp->parentfp = NULL;
+	if (!list_empty(&srp->entry)) {
+		list_del(&srp->entry);
+		srp->parentfp = NULL;
 		res = 1;
-	} else {
-		while ((rp = prev_rp->nextrp)) {
-			if (srp == rp) {
-				prev_rp->nextrp = rp->nextrp;
-				rp->parentfp = NULL;
-				res = 1;
-				break;
-			}
-			prev_rp = rp;
-		}
 	}
 	write_unlock_irqrestore(&sfp->rq_list_lock, iflags);
 	return res;
@@ -2137,13 +2144,12 @@ sg_add_sfp(Sg_device * sdp)
 
 	init_waitqueue_head(&sfp->read_wait);
 	rwlock_init(&sfp->rq_list_lock);
-
+	INIT_LIST_HEAD(&sfp->rq_list);
 	kref_init(&sfp->f_ref);
+	mutex_init(&sfp->f_mutex);
 	sfp->timeout = SG_DEFAULT_TIMEOUT;
 	sfp->timeout_user = SG_DEFAULT_TIMEOUT_USER;
 	sfp->force_packid = SG_DEF_FORCE_PACK_ID;
-	sfp->low_dma = (SG_DEF_FORCE_LOW_DMA == 0) ?
-	    sdp->device->host->unchecked_isa_dma : 1;
 	sfp->cmd_q = SG_DEF_COMMAND_Q;
 	sfp->keep_orphan = SG_DEF_KEEP_ORPHAN;
 	sfp->parentdp = sdp;
@@ -2177,10 +2183,18 @@ sg_remove_sfp_usercontext(struct work_struct *work)
 {
 	struct sg_fd *sfp = container_of(work, struct sg_fd, ew.work);
 	struct sg_device *sdp = sfp->parentdp;
+	Sg_request *srp;
+	unsigned long iflags;
 
 	/* Cleanup any responses which were never read(). */
-	while (sfp->headrp)
-		sg_finish_rem_req(sfp->headrp);
+	write_lock_irqsave(&sfp->rq_list_lock, iflags);
+	while (!list_empty(&sfp->rq_list)) {
+		srp = list_first_entry(&sfp->rq_list, Sg_request, entry);
+		sg_finish_rem_req(srp);
+		list_del(&srp->entry);
+		srp->parentfp = NULL;
+	}
+	write_unlock_irqrestore(&sfp->rq_list_lock, iflags);
 
 	if (sfp->reserve.bufflen > 0) {
 		SCSI_LOG_TIMEOUT(6, sg_printk(KERN_INFO, sdp,
@@ -2214,20 +2228,6 @@ sg_remove_sfp(struct kref *kref)
 	schedule_work(&sfp->ew.work);
 }
 
-static int
-sg_res_in_use(Sg_fd * sfp)
-{
-	const Sg_request *srp;
-	unsigned long iflags;
-
-	read_lock_irqsave(&sfp->rq_list_lock, iflags);
-	for (srp = sfp->headrp; srp; srp = srp->nextrp)
-		if (srp->res_used)
-			break;
-	read_unlock_irqrestore(&sfp->rq_list_lock, iflags);
-	return srp ? 1 : 0;
-}
-
 #ifdef CONFIG_SCSI_PROC_FS
 static int
 sg_idr_max_id(int id, void *p, void *data)
@@ -2597,7 +2597,7 @@ static int sg_proc_seq_show_devstrs(struct seq_file *s, void *v)
 /* must be called while holding sg_index_lock */
 static void sg_proc_debug_helper(struct seq_file *s, Sg_device * sdp)
 {
-	int k, m, new_interface, blen, usg;
+	int k, new_interface, blen, usg;
 	Sg_request *srp;
 	Sg_fd *fp;
 	const sg_io_hdr_t *hp;
@@ -2613,17 +2613,15 @@ static void sg_proc_debug_helper(struct seq_file *s, Sg_device * sdp)
 			   jiffies_to_msecs(fp->timeout),
 			   fp->reserve.bufflen,
 			   (int) fp->reserve.k_use_sg,
-			   (int) fp->low_dma);
+			   (int) sdp->device->host->unchecked_isa_dma);
 		seq_printf(s, "   cmd_q=%d f_packid=%d k_orphan=%d closed=0\n",
 			   (int) fp->cmd_q, (int) fp->force_packid,
 			   (int) fp->keep_orphan);
-		for (m = 0, srp = fp->headrp;
-				srp != NULL;
-				++m, srp = srp->nextrp) {
+		list_for_each_entry(srp, &fp->rq_list, entry) {
 			hp = &srp->header;
 			new_interface = (hp->interface_id == '\0') ? 0 : 1;
 			if (srp->res_used) {
-				if (new_interface && 
+				if (new_interface &&
 				    (SG_FLAG_MMAP_IO & hp->flags))
 					cp = "     mmap>> ";
 				else
@@ -2654,7 +2652,7 @@ static void sg_proc_debug_helper(struct seq_file *s, Sg_device * sdp)
 			seq_printf(s, "ms sgat=%d op=0x%02x\n", usg,
 				   (int) srp->data.cmd_opcode);
 		}
-		if (0 == m)
+		if (list_empty(&fp->rq_list))
 			seq_puts(s, "     No requests active\n");
 		read_unlock(&fp->rq_list_lock);
 	}
diff --git a/drivers/scsi/sgiwd93.c b/drivers/scsi/sgiwd93.c
index 6d215e2fb46d..71b4b91d2215 100644
--- a/drivers/scsi/sgiwd93.c
+++ b/drivers/scsi/sgiwd93.c
@@ -297,7 +297,7 @@ out:
 	return err;
 }
 
-static int __exit sgiwd93_remove(struct platform_device *pdev)
+static int sgiwd93_remove(struct platform_device *pdev)
 {
 	struct Scsi_Host *host = platform_get_drvdata(pdev);
 	struct ip22_hostdata *hdata = (struct ip22_hostdata *) host->hostdata;
diff --git a/drivers/scsi/sni_53c710.c b/drivers/scsi/sni_53c710.c
index 76278072147e..1f9a087daf69 100644
--- a/drivers/scsi/sni_53c710.c
+++ b/drivers/scsi/sni_53c710.c
@@ -117,7 +117,7 @@ static int snirm710_probe(struct platform_device *dev)
 	return -ENODEV;
 }
 
-static int __exit snirm710_driver_remove(struct platform_device *dev)
+static int snirm710_driver_remove(struct platform_device *dev)
 {
 	struct Scsi_Host *host = dev_get_drvdata(&dev->dev);
 	struct NCR_700_Host_Parameters *hostdata =
diff --git a/drivers/scsi/snic/snic_debugfs.c b/drivers/scsi/snic/snic_debugfs.c
index d30280326bde..269ddf791a73 100644
--- a/drivers/scsi/snic/snic_debugfs.c
+++ b/drivers/scsi/snic/snic_debugfs.c
@@ -548,7 +548,7 @@ snic_trc_debugfs_init(void)
 				 &snic_trc_fops);
 
 	if (!de) {
-		SNIC_ERR("Cann't create trace file.\n");
+		SNIC_ERR("Cannot create trace file.\n");
 
 		return ret;
 	}
diff --git a/drivers/scsi/stex.c b/drivers/scsi/stex.c
index 5b23175a584c..9b20643ab49d 100644
--- a/drivers/scsi/stex.c
+++ b/drivers/scsi/stex.c
@@ -26,6 +26,7 @@
 #include <linux/module.h>
 #include <linux/spinlock.h>
 #include <linux/ktime.h>
+#include <linux/reboot.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/byteorder.h>
@@ -38,9 +39,9 @@
 #include <scsi/scsi_eh.h>
 
 #define DRV_NAME "stex"
-#define ST_DRIVER_VERSION	"5.00.0000.01"
-#define ST_VER_MAJOR		5
-#define ST_VER_MINOR		00
+#define ST_DRIVER_VERSION	"6.02.0000.01"
+#define ST_VER_MAJOR		6
+#define ST_VER_MINOR		02
 #define ST_OEM				0000
 #define ST_BUILD_VER		01
 
@@ -64,6 +65,13 @@ enum {
 	YI2H_INT_C				= 0xa0,
 	YH2I_REQ				= 0xc0,
 	YH2I_REQ_HI				= 0xc4,
+	PSCRATCH0				= 0xb0,
+	PSCRATCH1				= 0xb4,
+	PSCRATCH2				= 0xb8,
+	PSCRATCH3				= 0xbc,
+	PSCRATCH4				= 0xc8,
+	MAILBOX_BASE			= 0x1000,
+	MAILBOX_HNDSHK_STS		= 0x0,
 
 	/* MU register value */
 	MU_INBOUND_DOORBELL_HANDSHAKE		= (1 << 0),
@@ -87,7 +95,7 @@ enum {
 	MU_STATE_STOP				= 5,
 	MU_STATE_NOCONNECT			= 6,
 
-	MU_MAX_DELAY				= 120,
+	MU_MAX_DELAY				= 50,
 	MU_HANDSHAKE_SIGNATURE			= 0x55aaaa55,
 	MU_HANDSHAKE_SIGNATURE_HALF		= 0x5a5a0000,
 	MU_HARD_RESET_WAIT			= 30000,
@@ -135,6 +143,7 @@ enum {
 	st_yosemite				= 2,
 	st_seq					= 3,
 	st_yel					= 4,
+	st_P3					= 5,
 
 	PASSTHRU_REQ_TYPE			= 0x00000001,
 	PASSTHRU_REQ_NO_WAKEUP			= 0x00000100,
@@ -339,6 +348,7 @@ struct st_hba {
 	u16 rq_size;
 	u16 sts_count;
 	u8  supports_pm;
+	int msi_lock;
 };
 
 struct st_card_info {
@@ -353,6 +363,12 @@ struct st_card_info {
 	u16 sts_count;
 };
 
+static int S6flag;
+static int stex_halt(struct notifier_block *nb, ulong event, void *buf);
+static struct notifier_block stex_notifier = {
+	stex_halt, NULL, 0
+};
+
 static int msi;
 module_param(msi, int, 0);
 MODULE_PARM_DESC(msi, "Enable Message Signaled Interrupts(0=off, 1=on)");
@@ -540,11 +556,15 @@ stex_ss_send_cmd(struct st_hba *hba, struct req_msg *req, u16 tag)
 
 	++hba->req_head;
 	hba->req_head %= hba->rq_count+1;
-
-	writel((addr >> 16) >> 16, hba->mmio_base + YH2I_REQ_HI);
-	readl(hba->mmio_base + YH2I_REQ_HI); /* flush */
-	writel(addr, hba->mmio_base + YH2I_REQ);
-	readl(hba->mmio_base + YH2I_REQ); /* flush */
+	if (hba->cardtype == st_P3) {
+		writel((addr >> 16) >> 16, hba->mmio_base + YH2I_REQ_HI);
+		writel(addr, hba->mmio_base + YH2I_REQ);
+	} else {
+		writel((addr >> 16) >> 16, hba->mmio_base + YH2I_REQ_HI);
+		readl(hba->mmio_base + YH2I_REQ_HI); /* flush */
+		writel(addr, hba->mmio_base + YH2I_REQ);
+		readl(hba->mmio_base + YH2I_REQ); /* flush */
+	}
 }
 
 static void return_abnormal_state(struct st_hba *hba, int status)
@@ -974,15 +994,31 @@ static irqreturn_t stex_ss_intr(int irq, void *__hba)
 
 	spin_lock_irqsave(hba->host->host_lock, flags);
 
-	data = readl(base + YI2H_INT);
-	if (data && data != 0xffffffff) {
-		/* clear the interrupt */
-		writel(data, base + YI2H_INT_C);
-		stex_ss_mu_intr(hba);
-		spin_unlock_irqrestore(hba->host->host_lock, flags);
-		if (unlikely(data & SS_I2H_REQUEST_RESET))
-			queue_work(hba->work_q, &hba->reset_work);
-		return IRQ_HANDLED;
+	if (hba->cardtype == st_yel) {
+		data = readl(base + YI2H_INT);
+		if (data && data != 0xffffffff) {
+			/* clear the interrupt */
+			writel(data, base + YI2H_INT_C);
+			stex_ss_mu_intr(hba);
+			spin_unlock_irqrestore(hba->host->host_lock, flags);
+			if (unlikely(data & SS_I2H_REQUEST_RESET))
+				queue_work(hba->work_q, &hba->reset_work);
+			return IRQ_HANDLED;
+		}
+	} else {
+		data = readl(base + PSCRATCH4);
+		if (data != 0xffffffff) {
+			if (data != 0) {
+				/* clear the interrupt */
+				writel(data, base + PSCRATCH1);
+				writel((1 << 22), base + YH2I_INT);
+			}
+			stex_ss_mu_intr(hba);
+			spin_unlock_irqrestore(hba->host->host_lock, flags);
+			if (unlikely(data & SS_I2H_REQUEST_RESET))
+				queue_work(hba->work_q, &hba->reset_work);
+			return IRQ_HANDLED;
+		}
 	}
 
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
@@ -1080,19 +1116,36 @@ static int stex_ss_handshake(struct st_hba *hba)
 	struct st_msg_header *msg_h;
 	struct handshake_frame *h;
 	__le32 *scratch;
-	u32 data, scratch_size;
+	u32 data, scratch_size, mailboxdata, operationaldata;
 	unsigned long before;
 	int ret = 0;
 
 	before = jiffies;
-	while ((readl(base + YIOA_STATUS) & SS_MU_OPERATIONAL) == 0) {
-		if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
-			printk(KERN_ERR DRV_NAME
-				"(%s): firmware not operational\n",
-				pci_name(hba->pdev));
-			return -1;
+
+	if (hba->cardtype == st_yel) {
+		operationaldata = readl(base + YIOA_STATUS);
+		while (operationaldata != SS_MU_OPERATIONAL) {
+			if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
+				printk(KERN_ERR DRV_NAME
+					"(%s): firmware not operational\n",
+					pci_name(hba->pdev));
+				return -1;
+			}
+			msleep(1);
+			operationaldata = readl(base + YIOA_STATUS);
+		}
+	} else {
+		operationaldata = readl(base + PSCRATCH3);
+		while (operationaldata != SS_MU_OPERATIONAL) {
+			if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
+				printk(KERN_ERR DRV_NAME
+					"(%s): firmware not operational\n",
+					pci_name(hba->pdev));
+				return -1;
+			}
+			msleep(1);
+			operationaldata = readl(base + PSCRATCH3);
 		}
-		msleep(1);
 	}
 
 	msg_h = (struct st_msg_header *)hba->dma_mem;
@@ -1111,30 +1164,60 @@ static int stex_ss_handshake(struct st_hba *hba)
 	scratch_size = (hba->sts_count+1)*sizeof(u32);
 	h->scratch_size = cpu_to_le32(scratch_size);
 
-	data = readl(base + YINT_EN);
-	data &= ~4;
-	writel(data, base + YINT_EN);
-	writel((hba->dma_handle >> 16) >> 16, base + YH2I_REQ_HI);
-	readl(base + YH2I_REQ_HI);
-	writel(hba->dma_handle, base + YH2I_REQ);
-	readl(base + YH2I_REQ); /* flush */
+	if (hba->cardtype == st_yel) {
+		data = readl(base + YINT_EN);
+		data &= ~4;
+		writel(data, base + YINT_EN);
+		writel((hba->dma_handle >> 16) >> 16, base + YH2I_REQ_HI);
+		readl(base + YH2I_REQ_HI);
+		writel(hba->dma_handle, base + YH2I_REQ);
+		readl(base + YH2I_REQ); /* flush */
+	} else {
+		data = readl(base + YINT_EN);
+		data &= ~(1 << 0);
+		data &= ~(1 << 2);
+		writel(data, base + YINT_EN);
+		if (hba->msi_lock == 0) {
+			/* P3 MSI Register cannot access twice */
+			writel((1 << 6), base + YH2I_INT);
+			hba->msi_lock  = 1;
+		}
+		writel((hba->dma_handle >> 16) >> 16, base + YH2I_REQ_HI);
+		writel(hba->dma_handle, base + YH2I_REQ);
+	}
 
-	scratch = hba->scratch;
 	before = jiffies;
-	while (!(le32_to_cpu(*scratch) & SS_STS_HANDSHAKE)) {
-		if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
-			printk(KERN_ERR DRV_NAME
-				"(%s): no signature after handshake frame\n",
-				pci_name(hba->pdev));
-			ret = -1;
-			break;
+	scratch = hba->scratch;
+	if (hba->cardtype == st_yel) {
+		while (!(le32_to_cpu(*scratch) & SS_STS_HANDSHAKE)) {
+			if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
+				printk(KERN_ERR DRV_NAME
+					"(%s): no signature after handshake frame\n",
+					pci_name(hba->pdev));
+				ret = -1;
+				break;
+			}
+			rmb();
+			msleep(1);
+		}
+	} else {
+		mailboxdata = readl(base + MAILBOX_BASE + MAILBOX_HNDSHK_STS);
+		while (mailboxdata != SS_STS_HANDSHAKE) {
+			if (time_after(jiffies, before + MU_MAX_DELAY * HZ)) {
+				printk(KERN_ERR DRV_NAME
+					"(%s): no signature after handshake frame\n",
+					pci_name(hba->pdev));
+				ret = -1;
+				break;
+			}
+			rmb();
+			msleep(1);
+			mailboxdata = readl(base + MAILBOX_BASE + MAILBOX_HNDSHK_STS);
 		}
-		rmb();
-		msleep(1);
 	}
-
 	memset(scratch, 0, scratch_size);
 	msg_h->flag = 0;
+
 	return ret;
 }
 
@@ -1144,8 +1227,10 @@ static int stex_handshake(struct st_hba *hba)
 	unsigned long flags;
 	unsigned int mu_status;
 
-	err = (hba->cardtype == st_yel) ?
-		stex_ss_handshake(hba) : stex_common_handshake(hba);
+	if (hba->cardtype == st_yel || hba->cardtype == st_P3)
+		err = stex_ss_handshake(hba);
+	else
+		err = stex_common_handshake(hba);
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	mu_status = hba->mu_status;
 	if (err == 0) {
@@ -1190,6 +1275,15 @@ static int stex_abort(struct scsi_cmnd *cmd)
 
 		writel(data, base + YI2H_INT_C);
 		stex_ss_mu_intr(hba);
+	} else if (hba->cardtype == st_P3) {
+		data = readl(base + PSCRATCH4);
+		if (data == 0xffffffff)
+			goto fail_out;
+		if (data != 0) {
+			writel(data, base + PSCRATCH1);
+			writel((1 << 22), base + YH2I_INT);
+		}
+		stex_ss_mu_intr(hba);
 	} else {
 		data = readl(base + ODBL);
 		if (data == 0 || data == 0xffffffff)
@@ -1197,7 +1291,6 @@ static int stex_abort(struct scsi_cmnd *cmd)
 
 		writel(data, base + ODBL);
 		readl(base + ODBL); /* flush */
-
 		stex_mu_intr(hba, data);
 	}
 	if (hba->wait_ccb == NULL) {
@@ -1293,6 +1386,12 @@ static void stex_ss_reset(struct st_hba *hba)
 	ssleep(5);
 }
 
+static void stex_p3_reset(struct st_hba *hba)
+{
+	writel(SS_H2I_INT_RESET, hba->mmio_base + YH2I_INT);
+	ssleep(5);
+}
+
 static int stex_do_reset(struct st_hba *hba)
 {
 	unsigned long flags;
@@ -1329,7 +1428,8 @@ static int stex_do_reset(struct st_hba *hba)
 		stex_hard_reset(hba);
 	else if (hba->cardtype == st_yel)
 		stex_ss_reset(hba);
-
+	else if (hba->cardtype == st_P3)
+		stex_p3_reset(hba);
 
 	return_abnormal_state(hba, DID_RESET);
 
@@ -1414,6 +1514,26 @@ static struct pci_device_id stex_pci_tbl[] = {
 	/* st_yel */
 	{ 0x105a, 0x8650, 0x1033, PCI_ANY_ID, 0, 0, st_yel },
 	{ 0x105a, 0x8760, PCI_ANY_ID, PCI_ANY_ID, 0, 0, st_yel },
+
+	/* st_P3, pluto */
+	{ PCI_VENDOR_ID_PROMISE, 0x8870, PCI_VENDOR_ID_PROMISE,
+		0x8870, 0, 0, st_P3 },
+	/* st_P3, p3 */
+	{ PCI_VENDOR_ID_PROMISE, 0x8870, PCI_VENDOR_ID_PROMISE,
+		0x4300, 0, 0, st_P3 },
+
+	/* st_P3, SymplyStor4E */
+	{ PCI_VENDOR_ID_PROMISE, 0x8871, PCI_VENDOR_ID_PROMISE,
+		0x4311, 0, 0, st_P3 },
+	/* st_P3, SymplyStor8E */
+	{ PCI_VENDOR_ID_PROMISE, 0x8871, PCI_VENDOR_ID_PROMISE,
+		0x4312, 0, 0, st_P3 },
+	/* st_P3, SymplyStor4 */
+	{ PCI_VENDOR_ID_PROMISE, 0x8871, PCI_VENDOR_ID_PROMISE,
+		0x4321, 0, 0, st_P3 },
+	/* st_P3, SymplyStor8 */
+	{ PCI_VENDOR_ID_PROMISE, 0x8871, PCI_VENDOR_ID_PROMISE,
+		0x4322, 0, 0, st_P3 },
 	{ }	/* terminate list */
 };
 
@@ -1482,6 +1602,19 @@ static struct st_card_info stex_card_info[] = {
 		.map_sg		= stex_ss_map_sg,
 		.send		= stex_ss_send_cmd,
 	},
+
+	/* st_P3 */
+	{
+		.max_id		= 129,
+		.max_lun	= 256,
+		.max_channel	= 0,
+		.rq_count	= 801,
+		.rq_size	= 512,
+		.sts_count	= 801,
+		.alloc_rq	= stex_ss_alloc_req,
+		.map_sg		= stex_ss_map_sg,
+		.send		= stex_ss_send_cmd,
+	},
 };
 
 static int stex_set_dma_mask(struct pci_dev * pdev)
@@ -1502,7 +1635,7 @@ static int stex_request_irq(struct st_hba *hba)
 	struct pci_dev *pdev = hba->pdev;
 	int status;
 
-	if (msi) {
+	if (msi || hba->cardtype == st_P3) {
 		status = pci_enable_msi(pdev);
 		if (status != 0)
 			printk(KERN_ERR DRV_NAME
@@ -1513,7 +1646,8 @@ static int stex_request_irq(struct st_hba *hba)
 	} else
 		hba->msi_enabled = 0;
 
-	status = request_irq(pdev->irq, hba->cardtype == st_yel ?
+	status = request_irq(pdev->irq,
+		(hba->cardtype == st_yel || hba->cardtype == st_P3) ?
 		stex_ss_intr : stex_intr, IRQF_SHARED, DRV_NAME, hba);
 
 	if (status != 0) {
@@ -1546,6 +1680,9 @@ static int stex_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	pci_set_master(pdev);
 
+	S6flag = 0;
+	register_reboot_notifier(&stex_notifier);
+
 	host = scsi_host_alloc(&driver_template, sizeof(struct st_hba));
 
 	if (!host) {
@@ -1597,12 +1734,12 @@ static int stex_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	case 0x4265:
 		break;
 	default:
-		if (hba->cardtype == st_yel)
+		if (hba->cardtype == st_yel || hba->cardtype == st_P3)
 			hba->supports_pm = 1;
 	}
 
 	sts_offset = scratch_offset = (ci->rq_count+1) * ci->rq_size;
-	if (hba->cardtype == st_yel)
+	if (hba->cardtype == st_yel || hba->cardtype == st_P3)
 		sts_offset += (ci->sts_count+1) * sizeof(u32);
 	cp_offset = sts_offset + (ci->sts_count+1) * sizeof(struct status_msg);
 	hba->dma_size = cp_offset + sizeof(struct st_frame);
@@ -1642,7 +1779,7 @@ static int stex_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		goto out_pci_free;
 	}
 
-	if (hba->cardtype == st_yel)
+	if (hba->cardtype == st_yel || hba->cardtype == st_P3)
 		hba->scratch = (__le32 *)(hba->dma_mem + scratch_offset);
 	hba->status_buffer = (struct status_msg *)(hba->dma_mem + sts_offset);
 	hba->copy_buffer = hba->dma_mem + cp_offset;
@@ -1653,8 +1790,9 @@ static int stex_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	hba->map_sg = ci->map_sg;
 	hba->send = ci->send;
 	hba->mu_status = MU_STATE_STARTING;
+	hba->msi_lock = 0;
 
-	if (hba->cardtype == st_yel)
+	if (hba->cardtype == st_yel || hba->cardtype == st_P3)
 		host->sg_tablesize = 38;
 	else
 		host->sg_tablesize = 32;
@@ -1736,28 +1874,29 @@ static void stex_hba_stop(struct st_hba *hba, int st_sleep_mic)
 
 	spin_lock_irqsave(hba->host->host_lock, flags);
 
-	if (hba->cardtype == st_yel && hba->supports_pm == 1)
-	{
-		if(st_sleep_mic == ST_NOTHANDLED)
-		{
+	if ((hba->cardtype == st_yel || hba->cardtype == st_P3) &&
+		hba->supports_pm == 1) {
+		if (st_sleep_mic == ST_NOTHANDLED) {
 			spin_unlock_irqrestore(hba->host->host_lock, flags);
 			return;
 		}
 	}
 	req = hba->alloc_rq(hba);
-	if (hba->cardtype == st_yel) {
+	if (hba->cardtype == st_yel || hba->cardtype == st_P3) {
 		msg_h = (struct st_msg_header *)req - 1;
 		memset(msg_h, 0, hba->rq_size);
 	} else
 		memset(req, 0, hba->rq_size);
 
-	if ((hba->cardtype == st_yosemite || hba->cardtype == st_yel)
+	if ((hba->cardtype == st_yosemite || hba->cardtype == st_yel
+		|| hba->cardtype == st_P3)
 		&& st_sleep_mic == ST_IGNORED) {
 		req->cdb[0] = MGT_CMD;
 		req->cdb[1] = MGT_CMD_SIGNATURE;
 		req->cdb[2] = CTLR_CONFIG_CMD;
 		req->cdb[3] = CTLR_SHUTDOWN;
-	} else if (hba->cardtype == st_yel && st_sleep_mic != ST_IGNORED) {
+	} else if ((hba->cardtype == st_yel || hba->cardtype == st_P3)
+		&& st_sleep_mic != ST_IGNORED) {
 		req->cdb[0] = MGT_CMD;
 		req->cdb[1] = MGT_CMD_SIGNATURE;
 		req->cdb[2] = CTLR_CONFIG_CMD;
@@ -1768,16 +1907,13 @@ static void stex_hba_stop(struct st_hba *hba, int st_sleep_mic)
 		req->cdb[1] = CTLR_POWER_STATE_CHANGE;
 		req->cdb[2] = CTLR_POWER_SAVING;
 	}
-
 	hba->ccb[tag].cmd = NULL;
 	hba->ccb[tag].sg_count = 0;
 	hba->ccb[tag].sense_bufflen = 0;
 	hba->ccb[tag].sense_buffer = NULL;
 	hba->ccb[tag].req_type = PASSTHRU_REQ_TYPE;
-
 	hba->send(hba, req, tag);
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
-
 	before = jiffies;
 	while (hba->ccb[tag].req_type & PASSTHRU_REQ_TYPE) {
 		if (time_after(jiffies, before + ST_INTERNAL_TIMEOUT * HZ)) {
@@ -1821,24 +1957,30 @@ static void stex_remove(struct pci_dev *pdev)
 	scsi_host_put(hba->host);
 
 	pci_disable_device(pdev);
+
+	unregister_reboot_notifier(&stex_notifier);
 }
 
 static void stex_shutdown(struct pci_dev *pdev)
 {
 	struct st_hba *hba = pci_get_drvdata(pdev);
 
-	if (hba->supports_pm == 0)
+	if (hba->supports_pm == 0) {
 		stex_hba_stop(hba, ST_IGNORED);
-	else
+	} else if (hba->supports_pm == 1 && S6flag) {
+		unregister_reboot_notifier(&stex_notifier);
+		stex_hba_stop(hba, ST_S6);
+	} else
 		stex_hba_stop(hba, ST_S5);
 }
 
-static int stex_choice_sleep_mic(pm_message_t state)
+static int stex_choice_sleep_mic(struct st_hba *hba, pm_message_t state)
 {
 	switch (state.event) {
 	case PM_EVENT_SUSPEND:
 		return ST_S3;
 	case PM_EVENT_HIBERNATE:
+		hba->msi_lock = 0;
 		return ST_S4;
 	default:
 		return ST_NOTHANDLED;
@@ -1849,8 +1991,9 @@ static int stex_suspend(struct pci_dev *pdev, pm_message_t state)
 {
 	struct st_hba *hba = pci_get_drvdata(pdev);
 
-	if (hba->cardtype == st_yel && hba->supports_pm == 1)
-		stex_hba_stop(hba, stex_choice_sleep_mic(state));
+	if ((hba->cardtype == st_yel || hba->cardtype == st_P3)
+		&& hba->supports_pm == 1)
+		stex_hba_stop(hba, stex_choice_sleep_mic(hba, state));
 	else
 		stex_hba_stop(hba, ST_IGNORED);
 	return 0;
@@ -1864,6 +2007,12 @@ static int stex_resume(struct pci_dev *pdev)
 	stex_handshake(hba);
 	return 0;
 }
+
+static int stex_halt(struct notifier_block *nb, unsigned long event, void *buf)
+{
+	S6flag = 1;
+	return NOTIFY_OK;
+}
 MODULE_DEVICE_TABLE(pci, stex_pci_tbl);
 
 static struct pci_driver stex_pci_driver = {
diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c
index 016639d7fef1..ae966dc3bbc5 100644
--- a/drivers/scsi/storvsc_drv.c
+++ b/drivers/scsi/storvsc_drv.c
@@ -476,6 +476,9 @@ struct storvsc_device {
 	 */
 	u64 node_name;
 	u64 port_name;
+#if IS_ENABLED(CONFIG_SCSI_FC_ATTRS)
+	struct fc_rport *rport;
+#endif
 };
 
 struct hv_host_device {
@@ -864,7 +867,7 @@ static int storvsc_channel_init(struct hv_device *device, bool is_fc)
 	 * We will however populate all the slots to evenly distribute
 	 * the load.
 	 */
-	stor_device->stor_chns = kzalloc(sizeof(void *) * num_possible_cpus(),
+	stor_device->stor_chns = kcalloc(num_possible_cpus(), sizeof(void *),
 					 GFP_KERNEL);
 	if (stor_device->stor_chns == NULL)
 		return -ENOMEM;
@@ -1189,8 +1192,6 @@ static void storvsc_on_channel_callback(void *context)
 			break;
 		}
 	} while (1);
-
-	return;
 }
 
 static int storvsc_connect_to_vsp(struct hv_device *device, u32 ring_size,
@@ -1823,19 +1824,27 @@ static int storvsc_probe(struct hv_device *device,
 		target = (device->dev_instance.b[5] << 8 |
 			 device->dev_instance.b[4]);
 		ret = scsi_add_device(host, 0, target, 0);
-		if (ret) {
-			scsi_remove_host(host);
-			goto err_out2;
-		}
+		if (ret)
+			goto err_out3;
 	}
 #if IS_ENABLED(CONFIG_SCSI_FC_ATTRS)
 	if (host->transportt == fc_transport_template) {
+		struct fc_rport_identifiers ids = {
+			.roles = FC_PORT_ROLE_FCP_DUMMY_INITIATOR,
+		};
+
 		fc_host_node_name(host) = stor_device->node_name;
 		fc_host_port_name(host) = stor_device->port_name;
+		stor_device->rport = fc_remote_port_add(host, 0, &ids);
+		if (!stor_device->rport)
+			goto err_out3;
 	}
 #endif
 	return 0;
 
+err_out3:
+	scsi_remove_host(host);
+
 err_out2:
 	/*
 	 * Once we have connected with the host, we would need to
@@ -1861,8 +1870,10 @@ static int storvsc_remove(struct hv_device *dev)
 	struct Scsi_Host *host = stor_device->host;
 
 #if IS_ENABLED(CONFIG_SCSI_FC_ATTRS)
-	if (host->transportt == fc_transport_template)
+	if (host->transportt == fc_transport_template) {
+		fc_remote_port_delete(stor_device->rport);
 		fc_remove_host(host);
+	}
 #endif
 	scsi_remove_host(host);
 	storvsc_dev_remove(dev);
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 096e95b911bd..abc7e87937cc 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -130,19 +130,12 @@ enum {
 	UFSHCD_UIC_DME_ERROR = (1 << 5), /* DME error */
 };
 
-/* Interrupt configuration options */
-enum {
-	UFSHCD_INT_DISABLE,
-	UFSHCD_INT_ENABLE,
-	UFSHCD_INT_CLEAR,
-};
-
 #define ufshcd_set_eh_in_progress(h) \
-	(h->eh_flags |= UFSHCD_EH_IN_PROGRESS)
+	((h)->eh_flags |= UFSHCD_EH_IN_PROGRESS)
 #define ufshcd_eh_in_progress(h) \
-	(h->eh_flags & UFSHCD_EH_IN_PROGRESS)
+	((h)->eh_flags & UFSHCD_EH_IN_PROGRESS)
 #define ufshcd_clear_eh_in_progress(h) \
-	(h->eh_flags &= ~UFSHCD_EH_IN_PROGRESS)
+	((h)->eh_flags &= ~UFSHCD_EH_IN_PROGRESS)
 
 #define ufshcd_set_ufs_dev_active(h) \
 	((h)->curr_dev_pwr_mode = UFS_ACTIVE_PWR_MODE)
@@ -540,15 +533,14 @@ static inline u32 ufshcd_get_intr_mask(struct ufs_hba *hba)
 	case UFSHCI_VERSION_10:
 		intr_mask = INTERRUPT_MASK_ALL_VER_10;
 		break;
-	/* allow fall through */
 	case UFSHCI_VERSION_11:
 	case UFSHCI_VERSION_20:
 		intr_mask = INTERRUPT_MASK_ALL_VER_11;
 		break;
-	/* allow fall through */
 	case UFSHCI_VERSION_21:
 	default:
 		intr_mask = INTERRUPT_MASK_ALL_VER_21;
+		break;
 	}
 
 	return intr_mask;
@@ -573,12 +565,12 @@ static inline u32 ufshcd_get_ufs_version(struct ufs_hba *hba)
  *			      the host controller
  * @hba: pointer to adapter instance
  *
- * Returns 1 if device present, 0 if no device detected
+ * Returns true if device present, false if no device detected
  */
-static inline int ufshcd_is_device_present(struct ufs_hba *hba)
+static inline bool ufshcd_is_device_present(struct ufs_hba *hba)
 {
 	return (ufshcd_readl(hba, REG_CONTROLLER_STATUS) &
-						DEVICE_PRESENT) ? 1 : 0;
+						DEVICE_PRESENT) ? true : false;
 }
 
 /**
@@ -668,16 +660,7 @@ static inline void ufshcd_outstanding_req_clear(struct ufs_hba *hba, int tag)
  */
 static inline int ufshcd_get_lists_status(u32 reg)
 {
-	/*
-	 * The mask 0xFF is for the following HCS register bits
-	 * Bit		Description
-	 *  0		Device Present
-	 *  1		UTRLRDY
-	 *  2		UTMRLRDY
-	 *  3		UCRDY
-	 * 4-7		reserved
-	 */
-	return ((reg & 0xFF) >> 1) ^ 0x07;
+	return !((reg & UFSHCD_STATUS_READY) == UFSHCD_STATUS_READY);
 }
 
 /**
@@ -820,11 +803,12 @@ static inline void ufshcd_hba_start(struct ufs_hba *hba)
  * ufshcd_is_hba_active - Get controller state
  * @hba: per adapter instance
  *
- * Returns zero if controller is active, 1 otherwise
+ * Returns false if controller is active, true otherwise
  */
-static inline int ufshcd_is_hba_active(struct ufs_hba *hba)
+static inline bool ufshcd_is_hba_active(struct ufs_hba *hba)
 {
-	return (ufshcd_readl(hba, REG_CONTROLLER_ENABLE) & 0x1) ? 0 : 1;
+	return (ufshcd_readl(hba, REG_CONTROLLER_ENABLE) & CONTROLLER_ENABLE)
+		? false : true;
 }
 
 static const char *ufschd_uic_link_state_to_string(
@@ -1478,7 +1462,7 @@ start:
 			break;
 		}
 		/*
-		 * If we here, it means gating work is either done or
+		 * If we are here, it means gating work is either done or
 		 * currently running. Hence, fall through to cancel gating
 		 * work and to enable clocks.
 		 */
@@ -3103,18 +3087,7 @@ static inline int ufshcd_read_power_desc(struct ufs_hba *hba,
 					 u8 *buf,
 					 u32 size)
 {
-	int err = 0;
-	int retries;
-
-	for (retries = QUERY_REQ_RETRIES; retries > 0; retries--) {
-		/* Read descriptor*/
-		err = ufshcd_read_desc(hba, QUERY_DESC_IDN_POWER, 0, buf, size);
-		if (!err)
-			break;
-		dev_dbg(hba->dev, "%s: error %d retrying\n", __func__, err);
-	}
-
-	return err;
+	return ufshcd_read_desc(hba, QUERY_DESC_IDN_POWER, 0, buf, size);
 }
 
 static int ufshcd_read_device_desc(struct ufs_hba *hba, u8 *buf, u32 size)
@@ -4272,24 +4245,16 @@ static void ufshcd_set_queue_depth(struct scsi_device *sdev)
 {
 	int ret = 0;
 	u8 lun_qdepth;
-	int retries;
 	struct ufs_hba *hba;
 
 	hba = shost_priv(sdev->host);
 
 	lun_qdepth = hba->nutrs;
-	for (retries = QUERY_REQ_RETRIES; retries > 0; retries--) {
-		/* Read descriptor*/
-		ret = ufshcd_read_unit_desc_param(hba,
-				  ufshcd_scsi_to_upiu_lun(sdev->lun),
-				  UNIT_DESC_PARAM_LU_Q_DEPTH,
-				  &lun_qdepth,
-				  sizeof(lun_qdepth));
-		if (!ret || ret == -ENOTSUPP)
-			break;
-
-		dev_dbg(hba->dev, "%s: error %d retrying\n", __func__, ret);
-	}
+	ret = ufshcd_read_unit_desc_param(hba,
+					  ufshcd_scsi_to_upiu_lun(sdev->lun),
+					  UNIT_DESC_PARAM_LU_Q_DEPTH,
+					  &lun_qdepth,
+					  sizeof(lun_qdepth));
 
 	/* Some WLUN doesn't support unit descriptor */
 	if (ret == -EOPNOTSUPP)
@@ -4717,7 +4682,7 @@ static int ufshcd_disable_ee(struct ufs_hba *hba, u16 mask)
 		goto out;
 
 	val = hba->ee_ctrl_mask & ~mask;
-	val &= 0xFFFF; /* 2 bytes */
+	val &= MASK_EE_STATUS;
 	err = ufshcd_query_attr_retry(hba, UPIU_QUERY_OPCODE_WRITE_ATTR,
 			QUERY_ATTR_IDN_EE_CONTROL, 0, 0, &val);
 	if (!err)
@@ -4745,7 +4710,7 @@ static int ufshcd_enable_ee(struct ufs_hba *hba, u16 mask)
 		goto out;
 
 	val = hba->ee_ctrl_mask | mask;
-	val &= 0xFFFF; /* 2 bytes */
+	val &= MASK_EE_STATUS;
 	err = ufshcd_query_attr_retry(hba, UPIU_QUERY_OPCODE_WRITE_ATTR,
 			QUERY_ATTR_IDN_EE_CONTROL, 0, 0, &val);
 	if (!err)
@@ -5960,24 +5925,6 @@ out:
 	return icc_level;
 }
 
-static int ufshcd_set_icc_levels_attr(struct ufs_hba *hba, u32 icc_level)
-{
-	int ret = 0;
-	int retries;
-
-	for (retries = QUERY_REQ_RETRIES; retries > 0; retries--) {
-		/* write attribute */
-		ret = ufshcd_query_attr(hba, UPIU_QUERY_OPCODE_WRITE_ATTR,
-			QUERY_ATTR_IDN_ACTIVE_ICC_LVL, 0, 0, &icc_level);
-		if (!ret)
-			break;
-
-		dev_dbg(hba->dev, "%s: failed with error %d\n", __func__, ret);
-	}
-
-	return ret;
-}
-
 static void ufshcd_init_icc_levels(struct ufs_hba *hba)
 {
 	int ret;
@@ -5998,8 +5945,9 @@ static void ufshcd_init_icc_levels(struct ufs_hba *hba)
 	dev_dbg(hba->dev, "%s: setting icc_level 0x%x",
 			__func__, hba->init_prefetch_data.icc_level);
 
-	ret = ufshcd_set_icc_levels_attr(hba,
-				 hba->init_prefetch_data.icc_level);
+	ret = ufshcd_query_attr_retry(hba, UPIU_QUERY_OPCODE_WRITE_ATTR,
+		QUERY_ATTR_IDN_ACTIVE_ICC_LVL, 0, 0,
+		&hba->init_prefetch_data.icc_level);
 
 	if (ret)
 		dev_err(hba->dev,
@@ -8000,7 +7948,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 		INIT_WORK(&hba->clk_scaling.resume_work,
 			  ufshcd_clk_scaling_resume_work);
 
-		snprintf(wq_name, ARRAY_SIZE(wq_name), "ufs_clkscaling_%d",
+		snprintf(wq_name, sizeof(wq_name), "ufs_clkscaling_%d",
 			 host->host_no);
 		hba->clk_scaling.workq = create_singlethread_workqueue(wq_name);
 
diff --git a/drivers/scsi/ufs/ufshci.h b/drivers/scsi/ufs/ufshci.h
index d14e9b965d1e..f60145d4a66e 100644
--- a/drivers/scsi/ufs/ufshci.h
+++ b/drivers/scsi/ufs/ufshci.h
@@ -48,6 +48,7 @@ enum {
 	REG_UFS_VERSION				= 0x08,
 	REG_CONTROLLER_DEV_ID			= 0x10,
 	REG_CONTROLLER_PROD_ID			= 0x14,
+	REG_AUTO_HIBERNATE_IDLE_TIMER		= 0x18,
 	REG_INTERRUPT_STATUS			= 0x20,
 	REG_INTERRUPT_ENABLE			= 0x24,
 	REG_CONTROLLER_STATUS			= 0x30,
@@ -159,6 +160,10 @@ enum {
 #define DEVICE_ERROR_INDICATOR			UFS_BIT(5)
 #define UIC_POWER_MODE_CHANGE_REQ_STATUS_MASK	UFS_MASK(0x7, 8)
 
+#define UFSHCD_STATUS_READY	(UTP_TRANSFER_REQ_LIST_READY |\
+				UTP_TASK_REQ_LIST_READY |\
+				UIC_COMMAND_READY)
+
 enum {
 	PWR_OK		= 0x0,
 	PWR_LOCAL	= 0x01,
@@ -171,6 +176,7 @@ enum {
 /* HCE - Host Controller Enable 34h */
 #define CONTROLLER_ENABLE	UFS_BIT(0)
 #define CONTROLLER_DISABLE	0x0
+#define CRYPTO_GENERAL_ENABLE	UFS_BIT(1)
 
 /* UECPA - Host UIC Error Code PHY Adapter Layer 38h */
 #define UIC_PHY_ADAPTER_LAYER_ERROR			UFS_BIT(31)
diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
index 939c47df73fa..a29d068b7696 100644
--- a/drivers/scsi/virtio_scsi.c
+++ b/drivers/scsi/virtio_scsi.c
@@ -29,6 +29,7 @@
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_tcq.h>
+#include <scsi/scsi_devinfo.h>
 #include <linux/seqlock.h>
 #include <linux/blk-mq-virtio.h>
 
@@ -705,6 +706,28 @@ static int virtscsi_device_reset(struct scsi_cmnd *sc)
 	return virtscsi_tmf(vscsi, cmd);
 }
 
+static int virtscsi_device_alloc(struct scsi_device *sdevice)
+{
+	/*
+	 * Passed through SCSI targets (e.g. with qemu's 'scsi-block')
+	 * may have transfer limits which come from the host SCSI
+	 * controller or something on the host side other than the
+	 * target itself.
+	 *
+	 * To make this work properly, the hypervisor can adjust the
+	 * target's VPD information to advertise these limits.  But
+	 * for that to work, the guest has to look at the VPD pages,
+	 * which we won't do by default if it is an SPC-2 device, even
+	 * if it does actually support it.
+	 *
+	 * So, set the blist to always try to read the VPD pages.
+	 */
+	sdevice->sdev_bflags = BLIST_TRY_VPD_PAGES;
+
+	return 0;
+}
+
+
 /**
  * virtscsi_change_queue_depth() - Change a virtscsi target's queue depth
  * @sdev:	Virtscsi target whose queue depth to change
@@ -783,6 +806,7 @@ static struct scsi_host_template virtscsi_host_template_single = {
 	.change_queue_depth = virtscsi_change_queue_depth,
 	.eh_abort_handler = virtscsi_abort,
 	.eh_device_reset_handler = virtscsi_device_reset,
+	.slave_alloc = virtscsi_device_alloc,
 
 	.can_queue = 1024,
 	.dma_boundary = UINT_MAX,
diff --git a/drivers/scsi/zalon.c b/drivers/scsi/zalon.c
index 97ccb0383539..b2cf1faa819d 100644
--- a/drivers/scsi/zalon.c
+++ b/drivers/scsi/zalon.c
@@ -167,7 +167,7 @@ static struct parisc_device_id zalon_tbl[] = {
 
 MODULE_DEVICE_TABLE(parisc, zalon_tbl);
 
-static int __exit zalon_remove(struct parisc_device *dev)
+static int zalon_remove(struct parisc_device *dev)
 {
 	struct Scsi_Host *host = dev_get_drvdata(&dev->dev);