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

Pull SCSI updates from James Bottomley:
 "This is mostly update of the usual drivers: aacraid, ufs, zfcp,
  NCR5380, lpfc, qla2xxx, smartpqi, hisi_sas, target, mpt3sas, pm80xx
  plus a whole load of minor updates and fixes.

  The major core changes are Al Viro's reworking of sg's handling of
  copy to/from user, Ming Lei's removal of the host busy counter to
  avoid contention in the multiqueue case and Damien Le Moal's fixing of
  residual tracking across error handling"

* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (251 commits)
  scsi: bnx2fc: timeout calculation invalid for bnx2fc_eh_abort()
  scsi: target: core: Fix a pr_debug() argument
  scsi: iscsi: Don't send data to unbound connection
  scsi: target: iscsi: Wait for all commands to finish before freeing a session
  scsi: target: core: Release SPC-2 reservations when closing a session
  scsi: target: core: Document target_cmd_size_check()
  scsi: bnx2i: fix potential use after free
  Revert "scsi: qla2xxx: Fix memory leak when sending I/O fails"
  scsi: NCR5380: Add disconnect_mask module parameter
  scsi: NCR5380: Unconditionally clear ICR after do_abort()
  scsi: NCR5380: Call scsi_set_resid() on command completion
  scsi: scsi_debug: num_tgts must be >= 0
  scsi: lpfc: use hdwq assigned cpu for allocation
  scsi: arcmsr: fix indentation issues
  scsi: qla4xxx: fix double free bug
  scsi: pm80xx: Modified the logic to collect fatal dump
  scsi: pm80xx: Tie the interrupt name to the module instance
  scsi: pm80xx: Controller fatal error through sysfs
  scsi: pm80xx: Do not request 12G sas speeds
  scsi: pm80xx: Cleanup command when a reset times out
  ...

1 files changed, 325 insertions, 66 deletions

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 614f78dddafe..c82b5792da98 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -87,6 +87,10 @@ static void lpfc_sli4_hba_handle_eqe(struct lpfc_hba *phba,
 				     struct lpfc_eqe *eqe);
 static bool lpfc_sli4_mbox_completions_pending(struct lpfc_hba *phba);
 static bool lpfc_sli4_process_missed_mbox_completions(struct lpfc_hba *phba);
+static struct lpfc_cqe *lpfc_sli4_cq_get(struct lpfc_queue *q);
+static void __lpfc_sli4_consume_cqe(struct lpfc_hba *phba,
+				    struct lpfc_queue *cq,
+				    struct lpfc_cqe *cqe);
 
 static IOCB_t *
 lpfc_get_iocb_from_iocbq(struct lpfc_iocbq *iocbq)
@@ -467,25 +471,52 @@ __lpfc_sli4_consume_eqe(struct lpfc_hba *phba, struct lpfc_queue *eq,
 }
 
 static void
-lpfc_sli4_eq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq)
+lpfc_sli4_eqcq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq)
 {
-	struct lpfc_eqe *eqe;
-	uint32_t count = 0;
+	struct lpfc_eqe *eqe = NULL;
+	u32 eq_count = 0, cq_count = 0;
+	struct lpfc_cqe *cqe = NULL;
+	struct lpfc_queue *cq = NULL, *childq = NULL;
+	int cqid = 0;
 
 	/* walk all the EQ entries and drop on the floor */
 	eqe = lpfc_sli4_eq_get(eq);
 	while (eqe) {
+		/* Get the reference to the corresponding CQ */
+		cqid = bf_get_le32(lpfc_eqe_resource_id, eqe);
+		cq = NULL;
+
+		list_for_each_entry(childq, &eq->child_list, list) {
+			if (childq->queue_id == cqid) {
+				cq = childq;
+				break;
+			}
+		}
+		/* If CQ is valid, iterate through it and drop all the CQEs */
+		if (cq) {
+			cqe = lpfc_sli4_cq_get(cq);
+			while (cqe) {
+				__lpfc_sli4_consume_cqe(phba, cq, cqe);
+				cq_count++;
+				cqe = lpfc_sli4_cq_get(cq);
+			}
+			/* Clear and re-arm the CQ */
+			phba->sli4_hba.sli4_write_cq_db(phba, cq, cq_count,
+			    LPFC_QUEUE_REARM);
+			cq_count = 0;
+		}
 		__lpfc_sli4_consume_eqe(phba, eq, eqe);
-		count++;
+		eq_count++;
 		eqe = lpfc_sli4_eq_get(eq);
 	}
 
 	/* Clear and re-arm the EQ */
-	phba->sli4_hba.sli4_write_eq_db(phba, eq, count, LPFC_QUEUE_REARM);
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, eq_count, LPFC_QUEUE_REARM);
 }
 
 static int
-lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq)
+lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq,
+		     uint8_t rearm)
 {
 	struct lpfc_eqe *eqe;
 	int count = 0, consumed = 0;
@@ -519,8 +550,8 @@ lpfc_sli4_process_eq(struct lpfc_hba *phba, struct lpfc_queue *eq)
 	eq->queue_claimed = 0;
 
 rearm_and_exit:
-	/* Always clear and re-arm the EQ */
-	phba->sli4_hba.sli4_write_eq_db(phba, eq, consumed, LPFC_QUEUE_REARM);
+	/* Always clear the EQ. */
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, consumed, rearm);
 
 	return count;
 }
@@ -2526,6 +2557,8 @@ lpfc_sli_def_mbox_cmpl(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 			} else {
 				__lpfc_sli_rpi_release(vport, ndlp);
 			}
+			if (vport->load_flag & FC_UNLOADING)
+				lpfc_nlp_put(ndlp);
 			pmb->ctx_ndlp = NULL;
 		}
 	}
@@ -2672,7 +2705,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 			lpfc_printf_log(phba, KERN_ERR, LOG_MBOX | LOG_SLI,
 					"(%d):0323 Unknown Mailbox command "
 					"x%x (x%x/x%x) Cmpl\n",
-					pmb->vport ? pmb->vport->vpi : 0,
+					pmb->vport ? pmb->vport->vpi :
+					LPFC_VPORT_UNKNOWN,
 					pmbox->mbxCommand,
 					lpfc_sli_config_mbox_subsys_get(phba,
 									pmb),
@@ -2693,7 +2727,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 					"(%d):0305 Mbox cmd cmpl "
 					"error - RETRYing Data: x%x "
 					"(x%x/x%x) x%x x%x x%x\n",
-					pmb->vport ? pmb->vport->vpi : 0,
+					pmb->vport ? pmb->vport->vpi :
+					LPFC_VPORT_UNKNOWN,
 					pmbox->mbxCommand,
 					lpfc_sli_config_mbox_subsys_get(phba,
 									pmb),
@@ -2701,7 +2736,8 @@ lpfc_sli_handle_mb_event(struct lpfc_hba *phba)
 									pmb),
 					pmbox->mbxStatus,
 					pmbox->un.varWords[0],
-					pmb->vport->port_state);
+					pmb->vport ? pmb->vport->port_state :
+					LPFC_VPORT_UNKNOWN);
 				pmbox->mbxStatus = 0;
 				pmbox->mbxOwner = OWN_HOST;
 				rc = lpfc_sli_issue_mbox(phba, pmb, MBX_NOWAIT);
@@ -6167,6 +6203,14 @@ lpfc_set_features(struct lpfc_hba *phba, LPFC_MBOXQ_t *mbox,
 		mbox->u.mqe.un.set_feature.feature = LPFC_SET_MDS_DIAGS;
 		mbox->u.mqe.un.set_feature.param_len = 8;
 		break;
+	case LPFC_SET_DUAL_DUMP:
+		bf_set(lpfc_mbx_set_feature_dd,
+		       &mbox->u.mqe.un.set_feature, LPFC_ENABLE_DUAL_DUMP);
+		bf_set(lpfc_mbx_set_feature_ddquery,
+		       &mbox->u.mqe.un.set_feature, 0);
+		mbox->u.mqe.un.set_feature.feature = LPFC_SET_DUAL_DUMP;
+		mbox->u.mqe.un.set_feature.param_len = 4;
+		break;
 	}
 
 	return;
@@ -6184,11 +6228,16 @@ lpfc_ras_stop_fwlog(struct lpfc_hba *phba)
 {
 	struct lpfc_ras_fwlog *ras_fwlog = &phba->ras_fwlog;
 
-	ras_fwlog->ras_active = false;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 
 	/* Disable FW logging to host memory */
 	writel(LPFC_CTL_PDEV_CTL_DDL_RAS,
 	       phba->sli4_hba.conf_regs_memmap_p + LPFC_CTL_PDEV_CTL_OFFSET);
+
+	/* Wait 10ms for firmware to stop using DMA buffer */
+	usleep_range(10 * 1000, 20 * 1000);
 }
 
 /**
@@ -6224,7 +6273,9 @@ lpfc_sli4_ras_dma_free(struct lpfc_hba *phba)
 		ras_fwlog->lwpd.virt = NULL;
 	}
 
-	ras_fwlog->ras_active = false;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 }
 
 /**
@@ -6326,7 +6377,9 @@ lpfc_sli4_ras_mbox_cmpl(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 		goto disable_ras;
 	}
 
-	ras_fwlog->ras_active = true;
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = ACTIVE;
+	spin_unlock_irq(&phba->hbalock);
 	mempool_free(pmb, phba->mbox_mem_pool);
 
 	return;
@@ -6358,6 +6411,10 @@ lpfc_sli4_ras_fwlog_init(struct lpfc_hba *phba,
 	uint32_t len = 0, fwlog_buffsize, fwlog_entry_count;
 	int rc = 0;
 
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = INACTIVE;
+	spin_unlock_irq(&phba->hbalock);
+
 	fwlog_buffsize = (LPFC_RAS_MIN_BUFF_POST_SIZE *
 			  phba->cfg_ras_fwlog_buffsize);
 	fwlog_entry_count = (fwlog_buffsize/LPFC_RAS_MAX_ENTRY_SIZE);
@@ -6417,6 +6474,9 @@ lpfc_sli4_ras_fwlog_init(struct lpfc_hba *phba,
 	mbx_fwlog->u.request.lwpd.addr_lo = putPaddrLow(ras_fwlog->lwpd.phys);
 	mbx_fwlog->u.request.lwpd.addr_hi = putPaddrHigh(ras_fwlog->lwpd.phys);
 
+	spin_lock_irq(&phba->hbalock);
+	ras_fwlog->state = REG_INPROGRESS;
+	spin_unlock_irq(&phba->hbalock);
 	mbox->vport = phba->pport;
 	mbox->mbox_cmpl = lpfc_sli4_ras_mbox_cmpl;
 
@@ -7148,7 +7208,7 @@ lpfc_post_rq_buffer(struct lpfc_hba *phba, struct lpfc_queue *hrq,
 int
 lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 {
-	int rc, i, cnt, len;
+	int rc, i, cnt, len, dd;
 	LPFC_MBOXQ_t *mboxq;
 	struct lpfc_mqe *mqe;
 	uint8_t *vpd;
@@ -7399,6 +7459,23 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 	phba->sli3_options |= (LPFC_SLI3_NPIV_ENABLED | LPFC_SLI3_HBQ_ENABLED);
 	spin_unlock_irq(&phba->hbalock);
 
+	/* Always try to enable dual dump feature if we can */
+	lpfc_set_features(phba, mboxq, LPFC_SET_DUAL_DUMP);
+	rc = lpfc_sli_issue_mbox(phba, mboxq, MBX_POLL);
+	dd = bf_get(lpfc_mbx_set_feature_dd, &mboxq->u.mqe.un.set_feature);
+	if ((rc == MBX_SUCCESS) && (dd == LPFC_ENABLE_DUAL_DUMP))
+		lpfc_printf_log(phba, KERN_ERR, LOG_SLI | LOG_INIT,
+				"6448 Dual Dump is enabled\n");
+	else
+		lpfc_printf_log(phba, KERN_INFO, LOG_SLI | LOG_INIT,
+				"6447 Dual Dump Mailbox x%x (x%x/x%x) failed, "
+				"rc:x%x dd:x%x\n",
+				bf_get(lpfc_mqe_command, &mboxq->u.mqe),
+				lpfc_sli_config_mbox_subsys_get(
+					phba, mboxq),
+				lpfc_sli_config_mbox_opcode_get(
+					phba, mboxq),
+				rc, dd);
 	/*
 	 * Allocate all resources (xri,rpi,vpi,vfi) now.  Subsequent
 	 * calls depends on these resources to complete port setup.
@@ -7523,9 +7600,11 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 		}
 		phba->sli4_hba.nvmet_xri_cnt = rc;
 
-		cnt = phba->cfg_iocb_cnt * 1024;
-		/* We need 1 iocbq for every SGL, for IO processing */
-		cnt += phba->sli4_hba.nvmet_xri_cnt;
+		/* We allocate an iocbq for every receive context SGL.
+		 * The additional allocation is for abort and ls handling.
+		 */
+		cnt = phba->sli4_hba.nvmet_xri_cnt +
+			phba->sli4_hba.max_cfg_param.max_xri;
 	} else {
 		/* update host common xri-sgl sizes and mappings */
 		rc = lpfc_sli4_io_sgl_update(phba);
@@ -7547,14 +7626,17 @@ lpfc_sli4_hba_setup(struct lpfc_hba *phba)
 			rc = -ENODEV;
 			goto out_destroy_queue;
 		}
-		cnt = phba->cfg_iocb_cnt * 1024;
+		/* Each lpfc_io_buf job structure has an iocbq element.
+		 * This cnt provides for abort, els, ct and ls requests.
+		 */
+		cnt = phba->sli4_hba.max_cfg_param.max_xri;
 	}
 
 	if (!phba->sli.iocbq_lookup) {
 		/* Initialize and populate the iocb list per host */
 		lpfc_printf_log(phba, KERN_INFO, LOG_INIT,
-				"2821 initialize iocb list %d total %d\n",
-				phba->cfg_iocb_cnt, cnt);
+				"2821 initialize iocb list with %d entries\n",
+				cnt);
 		rc = lpfc_init_iocb_list(phba, cnt);
 		if (rc) {
 			lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
@@ -7892,7 +7974,7 @@ lpfc_sli4_process_missed_mbox_completions(struct lpfc_hba *phba)
 
 	if (mbox_pending)
 		/* process and rearm the EQ */
-		lpfc_sli4_process_eq(phba, fpeq);
+		lpfc_sli4_process_eq(phba, fpeq, LPFC_QUEUE_REARM);
 	else
 		/* Always clear and re-arm the EQ */
 		sli4_hba->sli4_write_eq_db(phba, fpeq, 0, LPFC_QUEUE_REARM);
@@ -8964,7 +9046,8 @@ lpfc_mbox_api_table_setup(struct lpfc_hba *phba, uint8_t dev_grp)
  * @pring: Pointer to driver SLI ring object.
  * @piocb: Pointer to address of newly added command iocb.
  *
- * This function is called with hbalock held to add a command
+ * This function is called with hbalock held for SLI3 ports or
+ * the ring lock held for SLI4 ports to add a command
  * iocb to the txq when SLI layer cannot submit the command iocb
  * to the ring.
  **/
@@ -8972,7 +9055,10 @@ void
 __lpfc_sli_ringtx_put(struct lpfc_hba *phba, struct lpfc_sli_ring *pring,
 		    struct lpfc_iocbq *piocb)
 {
-	lockdep_assert_held(&phba->hbalock);
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		lockdep_assert_held(&pring->ring_lock);
+	else
+		lockdep_assert_held(&phba->hbalock);
 	/* Insert the caller's iocb in the txq tail for later processing. */
 	list_add_tail(&piocb->list, &pring->txq);
 }
@@ -9863,7 +9949,7 @@ lpfc_sli4_iocb2wqe(struct lpfc_hba *phba, struct lpfc_iocbq *iocbq,
  * __lpfc_sli_issue_iocb_s4 is used by other functions in the driver to issue
  * an iocb command to an HBA with SLI-4 interface spec.
  *
- * This function is called with hbalock held. The function will return success
+ * This function is called with ringlock held. The function will return success
  * after it successfully submit the iocb to firmware or after adding to the
  * txq.
  **/
@@ -10053,10 +10139,13 @@ lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number,
 		    struct lpfc_iocbq *piocb, uint32_t flag)
 {
 	struct lpfc_sli_ring *pring;
+	struct lpfc_queue *eq;
 	unsigned long iflags;
 	int rc;
 
 	if (phba->sli_rev == LPFC_SLI_REV4) {
+		eq = phba->sli4_hba.hdwq[piocb->hba_wqidx].hba_eq;
+
 		pring = lpfc_sli4_calc_ring(phba, piocb);
 		if (unlikely(pring == NULL))
 			return IOCB_ERROR;
@@ -10064,6 +10153,8 @@ lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number,
 		spin_lock_irqsave(&pring->ring_lock, iflags);
 		rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, flag);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(eq, LPFC_POLL_FASTPATH);
 	} else {
 		/* For now, SLI2/3 will still use hbalock */
 		spin_lock_irqsave(&phba->hbalock, iflags);
@@ -10678,14 +10769,14 @@ lpfc_sli_host_down(struct lpfc_vport *vport)
 				set_bit(LPFC_DATA_READY, &phba->data_flags);
 			}
 			prev_pring_flag = pring->flag;
-			spin_lock_irq(&pring->ring_lock);
+			spin_lock(&pring->ring_lock);
 			list_for_each_entry_safe(iocb, next_iocb,
 						 &pring->txq, list) {
 				if (iocb->vport != vport)
 					continue;
 				list_move_tail(&iocb->list, &completions);
 			}
-			spin_unlock_irq(&pring->ring_lock);
+			spin_unlock(&pring->ring_lock);
 			list_for_each_entry_safe(iocb, next_iocb,
 						 &pring->txcmplq, list) {
 				if (iocb->vport != vport)
@@ -11050,9 +11141,6 @@ lpfc_sli_abort_els_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 				irsp->ulpStatus, irsp->un.ulpWord[4]);
 
 		spin_unlock_irq(&phba->hbalock);
-		if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT &&
-		    irsp->un.ulpWord[4] == IOERR_SLI_ABORTED)
-			lpfc_sli_release_iocbq(phba, abort_iocb);
 	}
 release_iocb:
 	lpfc_sli_release_iocbq(phba, cmdiocb);
@@ -11736,7 +11824,10 @@ lpfc_sli_wake_iocb_wait(struct lpfc_hba *phba,
 		!(cmdiocbq->iocb_flag & LPFC_IO_LIBDFC)) {
 		lpfc_cmd = container_of(cmdiocbq, struct lpfc_io_buf,
 			cur_iocbq);
-		lpfc_cmd->exch_busy = rspiocbq->iocb_flag & LPFC_EXCHANGE_BUSY;
+		if (rspiocbq && (rspiocbq->iocb_flag & LPFC_EXCHANGE_BUSY))
+			lpfc_cmd->flags |= LPFC_SBUF_XBUSY;
+		else
+			lpfc_cmd->flags &= ~LPFC_SBUF_XBUSY;
 	}
 
 	pdone_q = cmdiocbq->context_un.wait_queue;
@@ -13158,13 +13249,19 @@ send_current_mbox:
 	phba->sli.sli_flag &= ~LPFC_SLI_MBOX_ACTIVE;
 	/* Setting active mailbox pointer need to be in sync to flag clear */
 	phba->sli.mbox_active = NULL;
+	if (bf_get(lpfc_trailer_consumed, mcqe))
+		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
 	spin_unlock_irqrestore(&phba->hbalock, iflags);
 	/* Wake up worker thread to post the next pending mailbox command */
 	lpfc_worker_wake_up(phba);
+	return workposted;
+
 out_no_mqe_complete:
+	spin_lock_irqsave(&phba->hbalock, iflags);
 	if (bf_get(lpfc_trailer_consumed, mcqe))
 		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
-	return workposted;
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
+	return false;
 }
 
 /**
@@ -13217,7 +13314,6 @@ lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq,
 	struct lpfc_sli_ring *pring = cq->pring;
 	int txq_cnt = 0;
 	int txcmplq_cnt = 0;
-	int fcp_txcmplq_cnt = 0;
 
 	/* Check for response status */
 	if (unlikely(bf_get(lpfc_wcqe_c_status, wcqe))) {
@@ -13239,9 +13335,8 @@ lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq,
 			txcmplq_cnt++;
 		lpfc_printf_log(phba, KERN_ERR, LOG_SLI,
 			"0387 NO IOCBQ data: txq_cnt=%d iocb_cnt=%d "
-			"fcp_txcmplq_cnt=%d, els_txcmplq_cnt=%d\n",
+			"els_txcmplq_cnt=%d\n",
 			txq_cnt, phba->iocb_cnt,
-			fcp_txcmplq_cnt,
 			txcmplq_cnt);
 		return false;
 	}
@@ -13592,6 +13687,7 @@ __lpfc_sli4_process_cq(struct lpfc_hba *phba, struct lpfc_queue *cq,
 			phba->sli4_hba.sli4_write_cq_db(phba, cq, consumed,
 						LPFC_QUEUE_NOARM);
 			consumed = 0;
+			cq->assoc_qp->q_flag |= HBA_EQ_DELAY_CHK;
 		}
 
 		if (count == LPFC_NVMET_CQ_NOTIFY)
@@ -14220,7 +14316,7 @@ lpfc_sli4_hba_intr_handler(int irq, void *dev_id)
 		spin_lock_irqsave(&phba->hbalock, iflag);
 		if (phba->link_state < LPFC_LINK_DOWN)
 			/* Flush, clear interrupt, and rearm the EQ */
-			lpfc_sli4_eq_flush(phba, fpeq);
+			lpfc_sli4_eqcq_flush(phba, fpeq);
 		spin_unlock_irqrestore(&phba->hbalock, iflag);
 		return IRQ_NONE;
 	}
@@ -14230,14 +14326,14 @@ lpfc_sli4_hba_intr_handler(int irq, void *dev_id)
 	fpeq->last_cpu = raw_smp_processor_id();
 
 	if (icnt > LPFC_EQD_ISR_TRIGGER &&
-	    phba->cfg_irq_chann == 1 &&
+	    fpeq->q_flag & HBA_EQ_DELAY_CHK &&
 	    phba->cfg_auto_imax &&
 	    fpeq->q_mode != LPFC_MAX_AUTO_EQ_DELAY &&
 	    phba->sli.sli_flag & LPFC_SLI_USE_EQDR)
 		lpfc_sli4_mod_hba_eq_delay(phba, fpeq, LPFC_MAX_AUTO_EQ_DELAY);
 
 	/* process and rearm the EQ */
-	ecount = lpfc_sli4_process_eq(phba, fpeq);
+	ecount = lpfc_sli4_process_eq(phba, fpeq, LPFC_QUEUE_REARM);
 
 	if (unlikely(ecount == 0)) {
 		fpeq->EQ_no_entry++;
@@ -14297,6 +14393,147 @@ lpfc_sli4_intr_handler(int irq, void *dev_id)
 	return (hba_handled == true) ? IRQ_HANDLED : IRQ_NONE;
 } /* lpfc_sli4_intr_handler */
 
+void lpfc_sli4_poll_hbtimer(struct timer_list *t)
+{
+	struct lpfc_hba *phba = from_timer(phba, t, cpuhp_poll_timer);
+	struct lpfc_queue *eq;
+	int i = 0;
+
+	rcu_read_lock();
+
+	list_for_each_entry_rcu(eq, &phba->poll_list, _poll_list)
+		i += lpfc_sli4_poll_eq(eq, LPFC_POLL_SLOWPATH);
+	if (!list_empty(&phba->poll_list))
+		mod_timer(&phba->cpuhp_poll_timer,
+			  jiffies + msecs_to_jiffies(LPFC_POLL_HB));
+
+	rcu_read_unlock();
+}
+
+inline int lpfc_sli4_poll_eq(struct lpfc_queue *eq, uint8_t path)
+{
+	struct lpfc_hba *phba = eq->phba;
+	int i = 0;
+
+	/*
+	 * Unlocking an irq is one of the entry point to check
+	 * for re-schedule, but we are good for io submission
+	 * path as midlayer does a get_cpu to glue us in. Flush
+	 * out the invalidate queue so we can see the updated
+	 * value for flag.
+	 */
+	smp_rmb();
+
+	if (READ_ONCE(eq->mode) == LPFC_EQ_POLL)
+		/* We will not likely get the completion for the caller
+		 * during this iteration but i guess that's fine.
+		 * Future io's coming on this eq should be able to
+		 * pick it up.  As for the case of single io's, they
+		 * will be handled through a sched from polling timer
+		 * function which is currently triggered every 1msec.
+		 */
+		i = lpfc_sli4_process_eq(phba, eq, LPFC_QUEUE_NOARM);
+
+	return i;
+}
+
+static inline void lpfc_sli4_add_to_poll_list(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	if (list_empty(&phba->poll_list)) {
+		timer_setup(&phba->cpuhp_poll_timer, lpfc_sli4_poll_hbtimer, 0);
+		/* kickstart slowpath processing for this eq */
+		mod_timer(&phba->cpuhp_poll_timer,
+			  jiffies + msecs_to_jiffies(LPFC_POLL_HB));
+	}
+
+	list_add_rcu(&eq->_poll_list, &phba->poll_list);
+	synchronize_rcu();
+}
+
+static inline void lpfc_sli4_remove_from_poll_list(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	/* Disable slowpath processing for this eq.  Kick start the eq
+	 * by RE-ARMING the eq's ASAP
+	 */
+	list_del_rcu(&eq->_poll_list);
+	synchronize_rcu();
+
+	if (list_empty(&phba->poll_list))
+		del_timer_sync(&phba->cpuhp_poll_timer);
+}
+
+void lpfc_sli4_cleanup_poll_list(struct lpfc_hba *phba)
+{
+	struct lpfc_queue *eq, *next;
+
+	list_for_each_entry_safe(eq, next, &phba->poll_list, _poll_list)
+		list_del(&eq->_poll_list);
+
+	INIT_LIST_HEAD(&phba->poll_list);
+	synchronize_rcu();
+}
+
+static inline void
+__lpfc_sli4_switch_eqmode(struct lpfc_queue *eq, uint8_t mode)
+{
+	if (mode == eq->mode)
+		return;
+	/*
+	 * currently this function is only called during a hotplug
+	 * event and the cpu on which this function is executing
+	 * is going offline.  By now the hotplug has instructed
+	 * the scheduler to remove this cpu from cpu active mask.
+	 * So we don't need to work about being put aside by the
+	 * scheduler for a high priority process.  Yes, the inte-
+	 * rrupts could come but they are known to retire ASAP.
+	 */
+
+	/* Disable polling in the fastpath */
+	WRITE_ONCE(eq->mode, mode);
+	/* flush out the store buffer */
+	smp_wmb();
+
+	/*
+	 * Add this eq to the polling list and start polling. For
+	 * a grace period both interrupt handler and poller will
+	 * try to process the eq _but_ that's fine.  We have a
+	 * synchronization mechanism in place (queue_claimed) to
+	 * deal with it.  This is just a draining phase for int-
+	 * errupt handler (not eq's) as we have guranteed through
+	 * barrier that all the CPUs have seen the new CQ_POLLED
+	 * state. which will effectively disable the REARMING of
+	 * the EQ.  The whole idea is eq's die off eventually as
+	 * we are not rearming EQ's anymore.
+	 */
+	mode ? lpfc_sli4_add_to_poll_list(eq) :
+	       lpfc_sli4_remove_from_poll_list(eq);
+}
+
+void lpfc_sli4_start_polling(struct lpfc_queue *eq)
+{
+	__lpfc_sli4_switch_eqmode(eq, LPFC_EQ_POLL);
+}
+
+void lpfc_sli4_stop_polling(struct lpfc_queue *eq)
+{
+	struct lpfc_hba *phba = eq->phba;
+
+	__lpfc_sli4_switch_eqmode(eq, LPFC_EQ_INTERRUPT);
+
+	/* Kick start for the pending io's in h/w.
+	 * Once we switch back to interrupt processing on a eq
+	 * the io path completion will only arm eq's when it
+	 * receives a completion.  But since eq's are in disa-
+	 * rmed state it doesn't receive a completion.  This
+	 * creates a deadlock scenaro.
+	 */
+	phba->sli4_hba.sli4_write_eq_db(phba, eq, 0, LPFC_QUEUE_REARM);
+}
+
 /**
  * lpfc_sli4_queue_free - free a queue structure and associated memory
  * @queue: The queue structure to free.
@@ -14371,6 +14608,7 @@ lpfc_sli4_queue_alloc(struct lpfc_hba *phba, uint32_t page_size,
 		return NULL;
 
 	INIT_LIST_HEAD(&queue->list);
+	INIT_LIST_HEAD(&queue->_poll_list);
 	INIT_LIST_HEAD(&queue->wq_list);
 	INIT_LIST_HEAD(&queue->wqfull_list);
 	INIT_LIST_HEAD(&queue->page_list);
@@ -18124,8 +18362,9 @@ lpfc_sli4_alloc_rpi(struct lpfc_hba *phba)
 		phba->sli4_hba.max_cfg_param.rpi_used++;
 		phba->sli4_hba.rpi_count++;
 	}
-	lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
-			"0001 rpi:%x max:%x lim:%x\n",
+	lpfc_printf_log(phba, KERN_INFO,
+			LOG_NODE | LOG_DISCOVERY,
+			"0001 Allocated rpi:x%x max:x%x lim:x%x\n",
 			(int) rpi, max_rpi, rpi_limit);
 
 	/*
@@ -18181,11 +18420,19 @@ lpfc_sli4_alloc_rpi(struct lpfc_hba *phba)
 static void
 __lpfc_sli4_free_rpi(struct lpfc_hba *phba, int rpi)
 {
+	/*
+	 * if the rpi value indicates a prior unreg has already
+	 * been done, skip the unreg.
+	 */
+	if (rpi == LPFC_RPI_ALLOC_ERROR)
+		return;
+
 	if (test_and_clear_bit(rpi, phba->sli4_hba.rpi_bmask)) {
 		phba->sli4_hba.rpi_count--;
 		phba->sli4_hba.max_cfg_param.rpi_used--;
 	} else {
-		lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
+		lpfc_printf_log(phba, KERN_INFO,
+				LOG_NODE | LOG_DISCOVERY,
 				"2016 rpi %x not inuse\n",
 				rpi);
 	}
@@ -19683,6 +19930,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 
@@ -19703,6 +19952,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 		}
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 
@@ -19731,6 +19982,8 @@ lpfc_sli4_issue_wqe(struct lpfc_hba *phba, struct lpfc_sli4_hdw_queue *qp,
 		}
 		lpfc_sli_ringtxcmpl_put(phba, pring, pwqe);
 		spin_unlock_irqrestore(&pring->ring_lock, iflags);
+
+		lpfc_sli4_poll_eq(qp->hba_eq, LPFC_POLL_FASTPATH);
 		return 0;
 	}
 	return WQE_ERROR;
@@ -20093,6 +20346,13 @@ void lpfc_release_io_buf(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_ncmd,
 	lpfc_ncmd->cur_iocbq.wqe_cmpl = NULL;
 	lpfc_ncmd->cur_iocbq.iocb_cmpl = NULL;
 
+	if (phba->cfg_xpsgl && !phba->nvmet_support &&
+	    !list_empty(&lpfc_ncmd->dma_sgl_xtra_list))
+		lpfc_put_sgl_per_hdwq(phba, lpfc_ncmd);
+
+	if (!list_empty(&lpfc_ncmd->dma_cmd_rsp_list))
+		lpfc_put_cmd_rsp_buf_per_hdwq(phba, lpfc_ncmd);
+
 	if (phba->cfg_xri_rebalancing) {
 		if (lpfc_ncmd->expedite) {
 			/* Return to expedite pool */
@@ -20157,13 +20417,6 @@ void lpfc_release_io_buf(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_ncmd,
 		spin_unlock_irqrestore(&qp->io_buf_list_put_lock,
 				       iflag);
 	}
-
-	if (phba->cfg_xpsgl && !phba->nvmet_support &&
-	    !list_empty(&lpfc_ncmd->dma_sgl_xtra_list))
-		lpfc_put_sgl_per_hdwq(phba, lpfc_ncmd);
-
-	if (!list_empty(&lpfc_ncmd->dma_cmd_rsp_list))
-		lpfc_put_cmd_rsp_buf_per_hdwq(phba, lpfc_ncmd);
 }
 
 /**
@@ -20399,8 +20652,9 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 	struct sli4_hybrid_sgl *allocated_sgl = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->sgl_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(buf_list))) {
 		/* break off 1 chunk from the sgl_list */
@@ -20412,9 +20666,9 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		}
 	} else {
 		/* allocate more */
-		spin_unlock_irq(&hdwq->hdwq_lock);
+		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(smp_processor_id()));
+				   cpu_to_node(hdwq->io_wq->chann));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8353 error kmalloc memory for HDWQ "
@@ -20434,7 +20688,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 			return NULL;
 		}
 
-		spin_lock_irq(&hdwq->hdwq_lock);
+		spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 		list_add_tail(&tmp->list_node, &lpfc_buf->dma_sgl_xtra_list);
 	}
 
@@ -20442,7 +20696,7 @@ lpfc_get_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 					struct sli4_hybrid_sgl,
 					list_node);
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 
 	return allocated_sgl;
 }
@@ -20466,8 +20720,9 @@ lpfc_put_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 	struct sli4_hybrid_sgl *tmp = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->sgl_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(&lpfc_buf->dma_sgl_xtra_list))) {
 		list_for_each_entry_safe(list_entry, tmp,
@@ -20480,7 +20735,7 @@ lpfc_put_sgl_per_hdwq(struct lpfc_hba *phba, struct lpfc_io_buf *lpfc_buf)
 		rc = -EINVAL;
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 	return rc;
 }
 
@@ -20501,8 +20756,9 @@ lpfc_free_sgl_per_hdwq(struct lpfc_hba *phba,
 	struct list_head *buf_list = &hdwq->sgl_list;
 	struct sli4_hybrid_sgl *list_entry = NULL;
 	struct sli4_hybrid_sgl *tmp = NULL;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	/* Free sgl pool */
 	list_for_each_entry_safe(list_entry, tmp,
@@ -20514,7 +20770,7 @@ lpfc_free_sgl_per_hdwq(struct lpfc_hba *phba,
 		kfree(list_entry);
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 }
 
 /**
@@ -20538,8 +20794,9 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct fcp_cmd_rsp_buf *allocated_buf = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(buf_list))) {
 		/* break off 1 chunk from the list */
@@ -20552,9 +20809,9 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		}
 	} else {
 		/* allocate more */
-		spin_unlock_irq(&hdwq->hdwq_lock);
+		spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 		tmp = kmalloc_node(sizeof(*tmp), GFP_ATOMIC,
-				   cpu_to_node(smp_processor_id()));
+				   cpu_to_node(hdwq->io_wq->chann));
 		if (!tmp) {
 			lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
 					"8355 error kmalloc memory for HDWQ "
@@ -20579,7 +20836,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		tmp->fcp_rsp = (struct fcp_rsp *)((uint8_t *)tmp->fcp_cmnd +
 				sizeof(struct fcp_cmnd));
 
-		spin_lock_irq(&hdwq->hdwq_lock);
+		spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 		list_add_tail(&tmp->list_node, &lpfc_buf->dma_cmd_rsp_list);
 	}
 
@@ -20587,7 +20844,7 @@ lpfc_get_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 					struct fcp_cmd_rsp_buf,
 					list_node);
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 
 	return allocated_buf;
 }
@@ -20612,8 +20869,9 @@ lpfc_put_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct fcp_cmd_rsp_buf *tmp = NULL;
 	struct lpfc_sli4_hdw_queue *hdwq = lpfc_buf->hdwq;
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	if (likely(!list_empty(&lpfc_buf->dma_cmd_rsp_list))) {
 		list_for_each_entry_safe(list_entry, tmp,
@@ -20626,7 +20884,7 @@ lpfc_put_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		rc = -EINVAL;
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 	return rc;
 }
 
@@ -20647,8 +20905,9 @@ lpfc_free_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 	struct list_head *buf_list = &hdwq->cmd_rsp_buf_list;
 	struct fcp_cmd_rsp_buf *list_entry = NULL;
 	struct fcp_cmd_rsp_buf *tmp = NULL;
+	unsigned long iflags;
 
-	spin_lock_irq(&hdwq->hdwq_lock);
+	spin_lock_irqsave(&hdwq->hdwq_lock, iflags);
 
 	/* Free cmd_rsp buf pool */
 	list_for_each_entry_safe(list_entry, tmp,
@@ -20661,5 +20920,5 @@ lpfc_free_cmd_rsp_buf_per_hdwq(struct lpfc_hba *phba,
 		kfree(list_entry);
 	}
 
-	spin_unlock_irq(&hdwq->hdwq_lock);
+	spin_unlock_irqrestore(&hdwq->hdwq_lock, iflags);
 }