[PATCH 02/17] lpfc: NVME Initiator: Base modifications Part A

James Smart jsmart2021 at gmail.com
Tue Jan 17 17:20:45 PST 2017


NVME Initiator: Base modifications

This is part A of parts A..F.

Part A is the bulk of the file list changed by the modifications as
most mods are small-ish. Changes may be for any aspect below.

*********

This set of patches (6 parts) adds base modifications for NVME initiator
support to the lpfc driver.

The base modifications consist of:
- Formal split of SLI3 rings from SLI-4 WQs (sometimes referred to as
  rings as well) as implementation now widely varies between the two.
- Addition of configuration modes:
   SCSI initiator only; NVME initiator only; NVME target only; and
   SCSI and NVME initiator.
   The configuration mode drives overall adapter configuration,
   offloads enabled, and resource splits.
   NVME support is only available on SLI-4 devices and newer fw.
- Implements the following based on configuration mode:
  - Exchange resources are split by protocol; Obviously, if only
     1 mode, then no split occurs. Default is 50/50. module attribute
     allows tuning.
  - Each protocol has it's own set of queues, but share interrupt
    vectors.
     SCSI:
       SLI3 devices have few queues and the original style of queue
         allocation remains.
       SLI4 devices piggy back on an "io-channel" concept that
         eventually needs to merge with scsi-mq/blk-mq support (it is
	 underway).  For now, the paradigm continues as it existed
	 prior. io channel allocates N msix and N WQs (N=4 default)
	 and either round robins or uses cpu # modulo N for scheduling.
	 A bunch of module parameters allow the configuration to be
	 tuned.
     NVME (initiator):
       Allocates an msix per cpu (or whatever pci_alloc_irq_vectors
         gets)
       Allocates a WQ per cpu, and maps the WQs to msix on a WQ #
         modulo msix vector count basis.
       Module parameters exist to cap/control the config if desired.
  - Each protocol has its own buffer and dma pools.

Unfortunately, it was very difficult to break out the above into
functional patches. A such, they are presented as a 6-patch set to
keep email size reasonable. All patches in the set must be applied to
create a functional unit.

Signed-off-by: Dick Kennedy <dick.kennedy at broadcom.com>
Signed-off-by: James Smart <james.smart at broadcom.com>
---
 drivers/scsi/lpfc/lpfc.h           |  77 ++++++++++++++--
 drivers/scsi/lpfc/lpfc_attr.h      |   1 +
 drivers/scsi/lpfc/lpfc_bsg.c       |  37 ++++++--
 drivers/scsi/lpfc/lpfc_crtn.h      |  44 ++++++++-
 drivers/scsi/lpfc/lpfc_disc.h      |   1 +
 drivers/scsi/lpfc/lpfc_els.c       |  56 ++++++++----
 drivers/scsi/lpfc/lpfc_hbadisc.c   | 148 +++++++++++++++++++-----------
 drivers/scsi/lpfc/lpfc_hw.h        |   5 +-
 drivers/scsi/lpfc/lpfc_hw4.h       | 183 +++++++++++++++++++++++++++++++++++--
 drivers/scsi/lpfc/lpfc_logmsg.h    |   4 +
 drivers/scsi/lpfc/lpfc_mbox.c      |  47 ++++++----
 drivers/scsi/lpfc/lpfc_mem.c       | 111 ++++++++++++++++------
 drivers/scsi/lpfc/lpfc_nportdisc.c |   9 +-
 drivers/scsi/lpfc/lpfc_nvme.h      |  87 ++++++++++++++++++
 drivers/scsi/lpfc/lpfc_scsi.c      | 140 +++++++++++++++++++---------
 drivers/scsi/lpfc/lpfc_scsi.h      |  18 ++--
 drivers/scsi/lpfc/lpfc_sli.h       |  37 ++++++--
 drivers/scsi/lpfc/lpfc_sli4.h      |  75 +++++++++++----
 drivers/scsi/lpfc/lpfc_version.h   |   1 +
 drivers/scsi/lpfc/lpfc_vport.c     |   3 +
 20 files changed, 854 insertions(+), 230 deletions(-)
 create mode 100644 drivers/scsi/lpfc/lpfc_nvme.h

diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h
index 6593b07..3fad85e 100644
--- a/drivers/scsi/lpfc/lpfc.h
+++ b/drivers/scsi/lpfc/lpfc.h
@@ -20,6 +20,7 @@
  *******************************************************************/
 
 #include <scsi/scsi_host.h>
+#include <linux/ktime.h>
 
 #if defined(CONFIG_DEBUG_FS) && !defined(CONFIG_SCSI_LPFC_DEBUG_FS)
 #define CONFIG_SCSI_LPFC_DEBUG_FS
@@ -53,6 +54,7 @@ struct lpfc_sli2_slim;
 #define LPFC_MAX_SG_SEG_CNT	4096	/* sg element count per scsi cmnd */
 #define LPFC_MAX_SGL_SEG_CNT	512	/* SGL element count per scsi cmnd */
 #define LPFC_MAX_BPL_SEG_CNT	4096	/* BPL element count per scsi cmnd */
+#define LPFC_MIN_NVME_SEG_CNT	254
 
 #define LPFC_MAX_SGE_SIZE       0x80000000 /* Maximum data allowed in a SGE */
 #define LPFC_IOCB_LIST_CNT	2250	/* list of IOCBs for fast-path usage. */
@@ -114,6 +116,13 @@ enum lpfc_polling_flags {
 	DISABLE_FCP_RING_INT    = 0x2
 };
 
+struct perf_prof {
+	uint16_t cmd_cpu[40];
+	uint16_t rsp_cpu[40];
+	uint16_t qh_cpu[40];
+	uint16_t wqidx[40];
+};
+
 /* Provide DMA memory definitions the driver uses per port instance. */
 struct lpfc_dmabuf {
 	struct list_head list;
@@ -131,10 +140,24 @@ struct lpfc_dma_pool {
 struct hbq_dmabuf {
 	struct lpfc_dmabuf hbuf;
 	struct lpfc_dmabuf dbuf;
-	uint32_t size;
+	uint16_t total_size;
+	uint16_t bytes_recv;
 	uint32_t tag;
 	struct lpfc_cq_event cq_event;
 	unsigned long time_stamp;
+	void *context;
+};
+
+struct rqb_dmabuf {
+	struct lpfc_dmabuf hbuf;
+	struct lpfc_dmabuf dbuf;
+	uint16_t total_size;
+	uint16_t bytes_recv;
+	void *context;
+	struct lpfc_iocbq *iocbq;
+	struct lpfc_sglq *sglq;
+	struct lpfc_queue *hrq;	  /* ptr to associated Header RQ */
+	struct lpfc_queue *drq;	  /* ptr to associated Data RQ */
 };
 
 /* Priority bit.  Set value to exceed low water mark in lpfc_mem. */
@@ -442,6 +465,11 @@ struct lpfc_vport {
 	uint16_t fdmi_num_disc;
 	uint32_t fdmi_hba_mask;
 	uint32_t fdmi_port_mask;
+
+	/* There is a single nvme instance per vport. */
+	struct nvme_fc_local_port *localport;
+	uint8_t  nvmei_support; /* driver supports NVME Initiator */
+	uint32_t last_fcp_wqidx;
 };
 
 struct hbq_s {
@@ -459,10 +487,9 @@ struct hbq_s {
 					       struct hbq_dmabuf *);
 };
 
-#define LPFC_MAX_HBQS  4
 /* this matches the position in the lpfc_hbq_defs array */
 #define LPFC_ELS_HBQ	0
-#define LPFC_EXTRA_HBQ	1
+#define LPFC_MAX_HBQS	1
 
 enum hba_temp_state {
 	HBA_NORMAL_TEMP,
@@ -652,6 +679,8 @@ struct lpfc_hba {
 					 * Firmware supports Forced Link Speed
 					 * capability
 					 */
+#define HBA_NVME_IOQ_FLUSH      0x80000 /* NVME IO queues flushed. */
+
 	uint32_t fcp_ring_in_use; /* When polling test if intr-hndlr active*/
 	struct lpfc_dmabuf slim2p;
 
@@ -700,6 +729,8 @@ struct lpfc_hba {
 	uint8_t  wwpn[8];
 	uint32_t RandomData[7];
 	uint8_t  fcp_embed_io;
+	uint8_t  nvme_support;	/* Firmware supports NVME */
+	uint8_t  nvmet_support;	/* driver supports NVMET */
 	uint8_t  mds_diags_support;
 
 	/* HBA Config Parameters */
@@ -725,6 +756,13 @@ struct lpfc_hba {
 	uint32_t cfg_fcp_imax;
 	uint32_t cfg_fcp_cpu_map;
 	uint32_t cfg_fcp_io_channel;
+	uint32_t cfg_suppress_rsp;
+	uint32_t cfg_nvme_oas;
+	uint32_t cfg_nvme_io_channel;
+	uint32_t cfg_fcp_max_hw_queue;
+	uint32_t cfg_nvme_max_hw_queue;
+	uint32_t cfg_nvme_posted_buf;
+	uint32_t cfg_nvme_enable_fb;
 	uint32_t cfg_total_seg_cnt;
 	uint32_t cfg_sg_seg_cnt;
 	uint32_t cfg_sg_dma_buf_size;
@@ -770,6 +808,12 @@ struct lpfc_hba {
 #define LPFC_FDMI_SUPPORT	1	/* FDMI supported? */
 	uint32_t cfg_enable_SmartSAN;
 	uint32_t cfg_enable_mds_diags;
+	uint32_t cfg_enable_fc4_type;
+	uint32_t cfg_xri_split;
+#define LPFC_ENABLE_FCP  1
+#define LPFC_ENABLE_NVME 2
+#define LPFC_ENABLE_BOTH 3
+	uint32_t io_channel;	/* max of fcp or nvme io channels */
 	lpfc_vpd_t vpd;		/* vital product data */
 
 	struct pci_dev *pcidev;
@@ -784,11 +828,11 @@ struct lpfc_hba {
 	unsigned long data_flags;
 
 	uint32_t hbq_in_use;		/* HBQs in use flag */
-	struct list_head rb_pend_list;  /* Received buffers to be processed */
 	uint32_t hbq_count;	        /* Count of configured HBQs */
 	struct hbq_s hbqs[LPFC_MAX_HBQS]; /* local copy of hbq indicies  */
 
-	atomic_t fcp_qidx;		/* next work queue to post work to */
+	atomic_t fcp_qidx;         /* next FCP WQ (RR Policy) */
+	atomic_t nvme_qidx;        /* next NVME WQ (RR Policy) */
 
 	phys_addr_t pci_bar0_map;     /* Physical address for PCI BAR0 */
 	phys_addr_t pci_bar1_map;     /* Physical address for PCI BAR1 */
@@ -843,9 +887,17 @@ struct lpfc_hba {
 	/*
 	 * stat  counters
 	 */
-	uint64_t fc4InputRequests;
-	uint64_t fc4OutputRequests;
-	uint64_t fc4ControlRequests;
+	uint64_t fc4ScsiInputRequests;
+	uint64_t fc4ScsiOutputRequests;
+	uint64_t fc4ScsiControlRequests;
+	uint64_t fc4ScsiIoCmpls;
+	uint64_t fc4NvmeInputRequests;
+	uint64_t fc4NvmeOutputRequests;
+	uint64_t fc4NvmeControlRequests;
+	uint64_t fc4NvmeIoCmpls;
+	uint64_t fc4NvmeLsRequests;
+	uint64_t fc4NvmeLsCmpls;
+
 	uint64_t bg_guard_err_cnt;
 	uint64_t bg_apptag_err_cnt;
 	uint64_t bg_reftag_err_cnt;
@@ -856,17 +908,23 @@ struct lpfc_hba {
 	struct list_head lpfc_scsi_buf_list_get;
 	struct list_head lpfc_scsi_buf_list_put;
 	uint32_t total_scsi_bufs;
+	spinlock_t nvme_buf_list_get_lock;  /* NVME buf alloc list lock */
+	spinlock_t nvme_buf_list_put_lock;  /* NVME buf free list lock */
+	struct list_head lpfc_nvme_buf_list_get;
+	struct list_head lpfc_nvme_buf_list_put;
+	uint32_t total_nvme_bufs;
 	struct list_head lpfc_iocb_list;
 	uint32_t total_iocbq_bufs;
 	struct list_head active_rrq_list;
 	spinlock_t hbalock;
 
 	/* pci_mem_pools */
-	struct pci_pool *lpfc_scsi_dma_buf_pool;
+	struct pci_pool *lpfc_sg_dma_buf_pool;
 	struct pci_pool *lpfc_mbuf_pool;
 	struct pci_pool *lpfc_hrb_pool;	/* header receive buffer pool */
 	struct pci_pool *lpfc_drb_pool; /* data receive buffer pool */
 	struct pci_pool *lpfc_hbq_pool;	/* SLI3 hbq buffer pool */
+	struct pci_pool *txrdy_payload_pool;
 	struct lpfc_dma_pool lpfc_mbuf_safety_pool;
 
 	mempool_t *mbox_mem_pool;
@@ -950,6 +1008,7 @@ struct lpfc_hba {
 	struct dentry *idiag_ctl_acc;
 	struct dentry *idiag_mbx_acc;
 	struct dentry *idiag_ext_acc;
+	uint8_t lpfc_idiag_last_eq;
 #endif
 
 	/* Used for deferred freeing of ELS data buffers */
diff --git a/drivers/scsi/lpfc/lpfc_attr.h b/drivers/scsi/lpfc/lpfc_attr.h
index b2bd28e..95c6b83 100644
--- a/drivers/scsi/lpfc/lpfc_attr.h
+++ b/drivers/scsi/lpfc/lpfc_attr.h
@@ -114,3 +114,4 @@ lpfc_vport_param_set(name, defval, minval, maxval)\
 lpfc_vport_param_store(name)\
 static DEVICE_ATTR(lpfc_##name, S_IRUGO | S_IWUSR,\
 		   lpfc_##name##_show, lpfc_##name##_store)
+
diff --git a/drivers/scsi/lpfc/lpfc_bsg.c b/drivers/scsi/lpfc/lpfc_bsg.c
index 7dca4d6..1e88b35 100644
--- a/drivers/scsi/lpfc/lpfc_bsg.c
+++ b/drivers/scsi/lpfc/lpfc_bsg.c
@@ -1704,6 +1704,7 @@ lpfc_bsg_diag_mode_enter(struct lpfc_hba *phba)
 	struct lpfc_vport **vports;
 	struct Scsi_Host *shost;
 	struct lpfc_sli *psli;
+	struct lpfc_queue *qp = NULL;
 	struct lpfc_sli_ring *pring;
 	int i = 0;
 
@@ -1711,9 +1712,6 @@ lpfc_bsg_diag_mode_enter(struct lpfc_hba *phba)
 	if (!psli)
 		return -ENODEV;
 
-	pring = &psli->ring[LPFC_FCP_RING];
-	if (!pring)
-		return -ENODEV;
 
 	if ((phba->link_state == LPFC_HBA_ERROR) ||
 	    (psli->sli_flag & LPFC_BLOCK_MGMT_IO) ||
@@ -1732,10 +1730,18 @@ lpfc_bsg_diag_mode_enter(struct lpfc_hba *phba)
 		scsi_block_requests(shost);
 	}
 
-	while (!list_empty(&pring->txcmplq)) {
-		if (i++ > 500)  /* wait up to 5 seconds */
+	if (phba->sli_rev != LPFC_SLI_REV4) {
+		pring = &psli->sli3_ring[LPFC_FCP_RING];
+		lpfc_emptyq_wait(phba, &pring->txcmplq, &phba->hbalock);
+		return 0;
+	}
+	list_for_each_entry(qp, &phba->sli4_hba.lpfc_wq_list, wq_list) {
+		pring = qp->pring;
+		if (!pring || (pring->ringno != LPFC_FCP_RING))
+			continue;
+		if (!lpfc_emptyq_wait(phba, &pring->txcmplq,
+				      &pring->ring_lock))
 			break;
-		msleep(10);
 	}
 	return 0;
 }
@@ -2704,7 +2710,7 @@ static int lpfcdiag_loop_get_xri(struct lpfc_hba *phba, uint16_t rpi,
  * @phba: Pointer to HBA context object
  *
  * This function allocates BSG_MBOX_SIZE (4KB) page size dma buffer and.
- * returns the pointer to the buffer.
+ * retruns the pointer to the buffer.
  **/
 static struct lpfc_dmabuf *
 lpfc_bsg_dma_page_alloc(struct lpfc_hba *phba)
@@ -2876,7 +2882,7 @@ static int lpfcdiag_loop_post_rxbufs(struct lpfc_hba *phba, uint16_t rxxri,
 			     size_t len)
 {
 	struct lpfc_sli *psli = &phba->sli;
-	struct lpfc_sli_ring *pring = &psli->ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
 	struct lpfc_iocbq *cmdiocbq;
 	IOCB_t *cmd = NULL;
 	struct list_head head, *curr, *next;
@@ -2890,6 +2896,11 @@ static int lpfcdiag_loop_post_rxbufs(struct lpfc_hba *phba, uint16_t rxxri,
 	int iocb_stat;
 	int i = 0;
 
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		pring = phba->sli4_hba.els_wq->pring;
+	else
+		pring = &psli->sli3_ring[LPFC_ELS_RING];
+
 	cmdiocbq = lpfc_sli_get_iocbq(phba);
 	rxbmp = kmalloc(sizeof(struct lpfc_dmabuf), GFP_KERNEL);
 	if (rxbmp != NULL) {
@@ -5258,7 +5269,7 @@ static int
 lpfc_forced_link_speed(struct bsg_job *job)
 {
 	struct Scsi_Host *shost = fc_bsg_to_shost(job);
-	struct lpfc_vport *vport = shost_priv(shost);
+	struct lpfc_vport *vport = (struct lpfc_vport *)shost->hostdata;
 	struct lpfc_hba *phba = vport->phba;
 	struct fc_bsg_reply *bsg_reply = job->reply;
 	struct forced_link_speed_support_reply *forced_reply;
@@ -5402,14 +5413,20 @@ lpfc_bsg_timeout(struct bsg_job *job)
 {
 	struct lpfc_vport *vport = shost_priv(fc_bsg_to_shost(job));
 	struct lpfc_hba *phba = vport->phba;
+	struct lpfc_sli *psli = &phba->sli;
 	struct lpfc_iocbq *cmdiocb;
-	struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
 	struct bsg_job_data *dd_data;
 	unsigned long flags;
 	int rc = 0;
 	LIST_HEAD(completions);
 	struct lpfc_iocbq *check_iocb, *next_iocb;
 
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		pring = phba->sli4_hba.els_wq->pring;
+	else
+		pring = &psli->sli3_ring[LPFC_ELS_RING];
+
 	/* if job's driver data is NULL, the command completed or is in the
 	 * the process of completing.  In this case, return status to request
 	 * so the timeout is retried.  This avoids double completion issues
diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h
index 309643a..4b7decb 100644
--- a/drivers/scsi/lpfc/lpfc_crtn.h
+++ b/drivers/scsi/lpfc/lpfc_crtn.h
@@ -21,6 +21,7 @@
 typedef int (*node_filter)(struct lpfc_nodelist *, void *);
 
 struct fc_rport;
+struct fc_frame_header;
 void lpfc_down_link(struct lpfc_hba *, LPFC_MBOXQ_t *);
 void lpfc_sli_read_link_ste(struct lpfc_hba *);
 void lpfc_dump_mem(struct lpfc_hba *, LPFC_MBOXQ_t *, uint16_t, uint16_t);
@@ -167,6 +168,8 @@ void lpfc_hb_timeout_handler(struct lpfc_hba *);
 void lpfc_ct_unsol_event(struct lpfc_hba *, struct lpfc_sli_ring *,
 			 struct lpfc_iocbq *);
 int lpfc_ct_handle_unsol_abort(struct lpfc_hba *, struct hbq_dmabuf *);
+int lpfc_issue_gidft(struct lpfc_vport *vport);
+int lpfc_get_gidft_type(struct lpfc_vport *vport, struct lpfc_iocbq *iocbq);
 int lpfc_ns_cmd(struct lpfc_vport *, int, uint8_t, uint32_t);
 int lpfc_fdmi_cmd(struct lpfc_vport *, struct lpfc_nodelist *, int, uint32_t);
 void lpfc_fdmi_num_disc_check(struct lpfc_vport *);
@@ -186,6 +189,8 @@ void lpfc_unblock_mgmt_io(struct lpfc_hba *);
 void lpfc_offline_prep(struct lpfc_hba *, int);
 void lpfc_offline(struct lpfc_hba *);
 void lpfc_reset_hba(struct lpfc_hba *);
+int lpfc_emptyq_wait(struct lpfc_hba *phba, struct list_head *hd,
+			spinlock_t *slock);
 
 int lpfc_fof_queue_create(struct lpfc_hba *);
 int lpfc_fof_queue_setup(struct lpfc_hba *);
@@ -193,7 +198,11 @@ int lpfc_fof_queue_destroy(struct lpfc_hba *);
 irqreturn_t lpfc_sli4_fof_intr_handler(int, void *);
 
 int lpfc_sli_setup(struct lpfc_hba *);
-int lpfc_sli_queue_setup(struct lpfc_hba *);
+int lpfc_sli4_setup(struct lpfc_hba *phba);
+void lpfc_sli_queue_init(struct lpfc_hba *phba);
+void lpfc_sli4_queue_init(struct lpfc_hba *phba);
+struct lpfc_sli_ring *lpfc_sli4_calc_ring(struct lpfc_hba *phba,
+					  struct lpfc_iocbq *iocbq);
 
 void lpfc_handle_eratt(struct lpfc_hba *);
 void lpfc_handle_latt(struct lpfc_hba *);
@@ -233,6 +242,11 @@ struct hbq_dmabuf *lpfc_sli4_rb_alloc(struct lpfc_hba *);
 void lpfc_sli4_rb_free(struct lpfc_hba *, struct hbq_dmabuf *);
 void lpfc_sli4_build_dflt_fcf_record(struct lpfc_hba *, struct fcf_record *,
 			uint16_t);
+int lpfc_sli4_rq_put(struct lpfc_queue *hq, struct lpfc_queue *dq,
+		     struct lpfc_rqe *hrqe, struct lpfc_rqe *drqe);
+int lpfc_post_rq_buffer(struct lpfc_hba *phba, struct lpfc_queue *hq,
+			struct lpfc_queue *dq, int count);
+int lpfc_free_rq_buffer(struct lpfc_hba *phba, struct lpfc_queue *hq);
 void lpfc_unregister_fcf(struct lpfc_hba *);
 void lpfc_unregister_fcf_rescan(struct lpfc_hba *);
 void lpfc_unregister_unused_fcf(struct lpfc_hba *);
@@ -287,6 +301,11 @@ void lpfc_sli_def_mbox_cmpl(struct lpfc_hba *, LPFC_MBOXQ_t *);
 void lpfc_sli4_unreg_rpi_cmpl_clr(struct lpfc_hba *, LPFC_MBOXQ_t *);
 int lpfc_sli_issue_iocb(struct lpfc_hba *, uint32_t,
 			struct lpfc_iocbq *, uint32_t);
+int lpfc_sli4_issue_wqe(struct lpfc_hba *phba, uint32_t rnum,
+			struct lpfc_iocbq *iocbq);
+struct lpfc_sglq *__lpfc_clear_active_sglq(struct lpfc_hba *phba, uint16_t xri);
+struct lpfc_sglq *__lpfc_sli_get_els_sglq(struct lpfc_hba *phba,
+					   struct lpfc_iocbq *iocbq);
 void lpfc_sli_pcimem_bcopy(void *, void *, uint32_t);
 void lpfc_sli_bemem_bcopy(void *, void *, uint32_t);
 void lpfc_sli_abort_iocb_ring(struct lpfc_hba *, struct lpfc_sli_ring *);
@@ -303,10 +322,10 @@ uint32_t lpfc_sli_get_buffer_tag(struct lpfc_hba *);
 struct lpfc_dmabuf * lpfc_sli_ring_taggedbuf_get(struct lpfc_hba *,
 			struct lpfc_sli_ring *, uint32_t );
 
-int lpfc_sli_hbq_count(void);
+int lpfc_sli_hbq_count(struct lpfc_hba *phba);
 int lpfc_sli_hbqbuf_add_hbqs(struct lpfc_hba *, uint32_t);
 void lpfc_sli_hbqbuf_free_all(struct lpfc_hba *);
-int lpfc_sli_hbq_size(void);
+int lpfc_sli_hbq_size(struct lpfc_hba *phba);
 int lpfc_sli_issue_abort_iotag(struct lpfc_hba *, struct lpfc_sli_ring *,
 			       struct lpfc_iocbq *);
 int lpfc_sli_sum_iocb(struct lpfc_vport *, uint16_t, uint64_t, lpfc_ctx_cmd);
@@ -338,6 +357,8 @@ void __lpfc_mbuf_free(struct lpfc_hba *, void *, dma_addr_t);
 void lpfc_mbuf_free(struct lpfc_hba *, void *, dma_addr_t);
 
 void lpfc_in_buf_free(struct lpfc_hba *, struct lpfc_dmabuf *);
+void lpfc_rq_buf_free(struct lpfc_hba *phba, struct lpfc_dmabuf *mp);
+
 /* Function prototypes. */
 const char* lpfc_info(struct Scsi_Host *);
 int lpfc_scan_finished(struct Scsi_Host *, unsigned long);
@@ -356,9 +377,11 @@ extern struct device_attribute *lpfc_hba_attrs[];
 extern struct device_attribute *lpfc_vport_attrs[];
 extern struct scsi_host_template lpfc_template;
 extern struct scsi_host_template lpfc_template_s3;
+extern struct scsi_host_template lpfc_template_nvme;
 extern struct scsi_host_template lpfc_vport_template;
 extern struct fc_function_template lpfc_transport_functions;
 extern struct fc_function_template lpfc_vport_transport_functions;
+extern int lpfc_new_scsi_buf(struct lpfc_vport *vport, int cnt);
 
 int  lpfc_vport_symbolic_node_name(struct lpfc_vport *, char *, size_t);
 int  lpfc_vport_symbolic_port_name(struct lpfc_vport *, char *,	size_t);
@@ -471,7 +494,9 @@ int lpfc_issue_unreg_vfi(struct lpfc_vport *);
 int lpfc_selective_reset(struct lpfc_hba *);
 int lpfc_sli4_read_config(struct lpfc_hba *);
 void lpfc_sli4_node_prep(struct lpfc_hba *);
-int lpfc_sli4_xri_sgl_update(struct lpfc_hba *);
+int lpfc_sli4_els_sgl_update(struct lpfc_hba *phba);
+int lpfc_sli4_scsi_sgl_update(struct lpfc_hba *phba);
+int lpfc_sli4_nvme_sgl_update(struct lpfc_hba *phba);
 void lpfc_free_sgl_list(struct lpfc_hba *, struct list_head *);
 uint32_t lpfc_sli_port_speed_get(struct lpfc_hba *);
 int lpfc_sli4_request_firmware_update(struct lpfc_hba *, uint8_t);
@@ -496,3 +521,14 @@ bool lpfc_find_next_oas_lun(struct lpfc_hba *, struct lpfc_name *,
 			    uint32_t *, uint32_t *);
 int lpfc_sli4_dump_page_a0(struct lpfc_hba *phba, struct lpfcMboxq *mbox);
 void lpfc_mbx_cmpl_rdp_page_a0(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb);
+struct lpfc_vport *lpfc_fc_frame_to_vport(struct lpfc_hba *phba,
+					  struct fc_frame_header *hdr,
+					  uint16_t fcfi, uint32_t did);
+
+/* SCSI Interfaces. */
+void lpfc_release_scsi_buf(struct lpfc_hba *phba, struct lpfc_scsi_buf *psb);
+struct lpfc_scsi_buf *lpfc_get_scsi_buf(struct lpfc_hba *phba,
+					struct lpfc_nodelist *ndlp);
+
+/* NVME interfaces. */
+void lpfc_nvme_mod_param_dep(struct lpfc_hba *phba);
diff --git a/drivers/scsi/lpfc/lpfc_disc.h b/drivers/scsi/lpfc/lpfc_disc.h
index 361f5b3..1d07a5f 100644
--- a/drivers/scsi/lpfc/lpfc_disc.h
+++ b/drivers/scsi/lpfc/lpfc_disc.h
@@ -133,6 +133,7 @@ struct lpfc_node_rrq {
 /* Defines for nlp_flag (uint32) */
 #define NLP_IGNR_REG_CMPL  0x00000001 /* Rcvd rscn before we cmpl reg login */
 #define NLP_REG_LOGIN_SEND 0x00000002   /* sent reglogin to adapter */
+#define NLP_SUPPRESS_RSP   0x00000010	/* Remote NPort supports suppress rsp */
 #define NLP_PLOGI_SND      0x00000020	/* sent PLOGI request for this entry */
 #define NLP_PRLI_SND       0x00000040	/* sent PRLI request for this entry */
 #define NLP_ADISC_SND      0x00000080	/* sent ADISC request for this entry */
diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c
index 27f0cbb..f61e1c2 100644
--- a/drivers/scsi/lpfc/lpfc_els.c
+++ b/drivers/scsi/lpfc/lpfc_els.c
@@ -1323,7 +1323,10 @@ lpfc_els_abort_flogi(struct lpfc_hba *phba)
 			"0201 Abort outstanding I/O on NPort x%x\n",
 			Fabric_DID);
 
-	pring = &phba->sli.ring[LPFC_ELS_RING];
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
 
 	/*
 	 * Check the txcmplq for an iocb that matches the nport the driver is
@@ -7153,7 +7156,11 @@ lpfc_els_timeout_handler(struct lpfc_vport *vport)
 
 	timeout = (uint32_t)(phba->fc_ratov << 1);
 
-	pring = &phba->sli.ring[LPFC_ELS_RING];
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
+
 	if ((phba->pport->load_flag & FC_UNLOADING))
 		return;
 	spin_lock_irq(&phba->hbalock);
@@ -7222,7 +7229,7 @@ lpfc_els_timeout_handler(struct lpfc_vport *vport)
 		spin_unlock_irq(&phba->hbalock);
 	}
 
-	if (!list_empty(&phba->sli.ring[LPFC_ELS_RING].txcmplq))
+	if (!list_empty(&pring->txcmplq))
 		if (!(phba->pport->load_flag & FC_UNLOADING))
 			mod_timer(&vport->els_tmofunc,
 				  jiffies + msecs_to_jiffies(1000 * timeout));
@@ -7253,7 +7260,7 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 {
 	LIST_HEAD(abort_list);
 	struct lpfc_hba  *phba = vport->phba;
-	struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
 	struct lpfc_iocbq *tmp_iocb, *piocb;
 	IOCB_t *cmd = NULL;
 
@@ -7265,8 +7272,13 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 	 * a working list and release the locks before calling the abort.
 	 */
 	spin_lock_irq(&phba->hbalock);
-	if (phba->sli_rev == LPFC_SLI_REV4)
+	if (phba->sli_rev != LPFC_SLI_REV4) {
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	} else {
+		pring = phba->sli4_hba.els_wq->pring;
 		spin_lock(&pring->ring_lock);
+	}
+
 
 	list_for_each_entry_safe(piocb, tmp_iocb, &pring->txcmplq, list) {
 		if (piocb->iocb_flag & LPFC_IO_LIBDFC)
@@ -9012,7 +9024,12 @@ void lpfc_fabric_abort_nport(struct lpfc_nodelist *ndlp)
 	LIST_HEAD(completions);
 	struct lpfc_hba  *phba = ndlp->phba;
 	struct lpfc_iocbq *tmp_iocb, *piocb;
-	struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
+
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
 
 	spin_lock_irq(&phba->hbalock);
 	list_for_each_entry_safe(piocb, tmp_iocb, &phba->fabric_iocb_list,
@@ -9068,13 +9085,13 @@ lpfc_sli4_vport_delete_els_xri_aborted(struct lpfc_vport *vport)
 	unsigned long iflag = 0;
 
 	spin_lock_irqsave(&phba->hbalock, iflag);
-	spin_lock(&phba->sli4_hba.abts_sgl_list_lock);
+	spin_lock(&phba->sli4_hba.sgl_list_lock);
 	list_for_each_entry_safe(sglq_entry, sglq_next,
 			&phba->sli4_hba.lpfc_abts_els_sgl_list, list) {
 		if (sglq_entry->ndlp && sglq_entry->ndlp->vport == vport)
 			sglq_entry->ndlp = NULL;
 	}
-	spin_unlock(&phba->sli4_hba.abts_sgl_list_lock);
+	spin_unlock(&phba->sli4_hba.sgl_list_lock);
 	spin_unlock_irqrestore(&phba->hbalock, iflag);
 	return;
 }
@@ -9098,22 +9115,25 @@ lpfc_sli4_els_xri_aborted(struct lpfc_hba *phba,
 	struct lpfc_sglq *sglq_entry = NULL, *sglq_next = NULL;
 	unsigned long iflag = 0;
 	struct lpfc_nodelist *ndlp;
-	struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
+
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
 
 	spin_lock_irqsave(&phba->hbalock, iflag);
-	spin_lock(&phba->sli4_hba.abts_sgl_list_lock);
+	spin_lock(&phba->sli4_hba.sgl_list_lock);
 	list_for_each_entry_safe(sglq_entry, sglq_next,
 			&phba->sli4_hba.lpfc_abts_els_sgl_list, list) {
 		if (sglq_entry->sli4_xritag == xri) {
 			list_del(&sglq_entry->list);
 			ndlp = sglq_entry->ndlp;
 			sglq_entry->ndlp = NULL;
-			spin_lock(&pring->ring_lock);
 			list_add_tail(&sglq_entry->list,
-				&phba->sli4_hba.lpfc_sgl_list);
+				&phba->sli4_hba.lpfc_els_sgl_list);
 			sglq_entry->state = SGL_FREED;
-			spin_unlock(&pring->ring_lock);
-			spin_unlock(&phba->sli4_hba.abts_sgl_list_lock);
+			spin_unlock(&phba->sli4_hba.sgl_list_lock);
 			spin_unlock_irqrestore(&phba->hbalock, iflag);
 			lpfc_set_rrq_active(phba, ndlp,
 				sglq_entry->sli4_lxritag,
@@ -9125,21 +9145,21 @@ lpfc_sli4_els_xri_aborted(struct lpfc_hba *phba,
 			return;
 		}
 	}
-	spin_unlock(&phba->sli4_hba.abts_sgl_list_lock);
+	spin_unlock(&phba->sli4_hba.sgl_list_lock);
 	lxri = lpfc_sli4_xri_inrange(phba, xri);
 	if (lxri == NO_XRI) {
 		spin_unlock_irqrestore(&phba->hbalock, iflag);
 		return;
 	}
-	spin_lock(&pring->ring_lock);
+	spin_lock(&phba->sli4_hba.sgl_list_lock);
 	sglq_entry = __lpfc_get_active_sglq(phba, lxri);
 	if (!sglq_entry || (sglq_entry->sli4_xritag != xri)) {
-		spin_unlock(&pring->ring_lock);
+		spin_unlock(&phba->sli4_hba.sgl_list_lock);
 		spin_unlock_irqrestore(&phba->hbalock, iflag);
 		return;
 	}
 	sglq_entry->state = SGL_XRI_ABORTED;
-	spin_unlock(&pring->ring_lock);
+	spin_unlock(&phba->sli4_hba.sgl_list_lock);
 	spin_unlock_irqrestore(&phba->hbalock, iflag);
 	return;
 }
diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index ed22393..4293b94 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -93,7 +93,7 @@ lpfc_terminate_rport_io(struct fc_rport *rport)
 
 	if (ndlp->nlp_sid != NLP_NO_SID) {
 		lpfc_sli_abort_iocb(ndlp->vport,
-			&phba->sli.ring[phba->sli.fcp_ring],
+			&phba->sli.sli3_ring[LPFC_FCP_RING],
 			ndlp->nlp_sid, 0, LPFC_CTX_TGT);
 	}
 }
@@ -247,8 +247,8 @@ lpfc_dev_loss_tmo_handler(struct lpfc_nodelist *ndlp)
 		if (ndlp->nlp_sid != NLP_NO_SID) {
 			/* flush the target */
 			lpfc_sli_abort_iocb(vport,
-					&phba->sli.ring[phba->sli.fcp_ring],
-					ndlp->nlp_sid, 0, LPFC_CTX_TGT);
+					    &phba->sli.sli3_ring[LPFC_FCP_RING],
+					    ndlp->nlp_sid, 0, LPFC_CTX_TGT);
 		}
 		put_node = rdata->pnode != NULL;
 		rdata->pnode = NULL;
@@ -283,7 +283,7 @@ lpfc_dev_loss_tmo_handler(struct lpfc_nodelist *ndlp)
 
 	if (ndlp->nlp_sid != NLP_NO_SID) {
 		warn_on = 1;
-		lpfc_sli_abort_iocb(vport, &phba->sli.ring[phba->sli.fcp_ring],
+		lpfc_sli_abort_iocb(vport, &phba->sli.sli3_ring[LPFC_FCP_RING],
 				    ndlp->nlp_sid, 0, LPFC_CTX_TGT);
 	}
 
@@ -495,11 +495,12 @@ lpfc_send_fastpath_evt(struct lpfc_hba *phba,
 		return;
 	}
 
-	fc_host_post_vendor_event(shost,
-		fc_get_event_number(),
-		evt_data_size,
-		evt_data,
-		LPFC_NL_VENDOR_ID);
+	if (phba->cfg_enable_fc4_type != LPFC_ENABLE_NVME)
+		fc_host_post_vendor_event(shost,
+			fc_get_event_number(),
+			evt_data_size,
+			evt_data,
+			LPFC_NL_VENDOR_ID);
 
 	lpfc_free_fast_evt(phba, fast_evt_data);
 	return;
@@ -682,7 +683,10 @@ lpfc_work_done(struct lpfc_hba *phba)
 		}
 	lpfc_destroy_vport_work_array(phba, vports);
 
-	pring = &phba->sli.ring[LPFC_ELS_RING];
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
 	status = (ha_copy & (HA_RXMASK  << (4*LPFC_ELS_RING)));
 	status >>= (4*LPFC_ELS_RING);
 	if ((status & HA_RXMASK) ||
@@ -894,11 +898,16 @@ lpfc_linkdown(struct lpfc_hba *phba)
 		spin_unlock_irq(shost->host_lock);
 	}
 	vports = lpfc_create_vport_work_array(phba);
-	if (vports != NULL)
+	if (vports != NULL) {
 		for (i = 0; i <= phba->max_vports && vports[i] != NULL; i++) {
 			/* Issue a LINK DOWN event to all nodes */
 			lpfc_linkdown_port(vports[i]);
+
+			vports[i]->fc_myDID = 0;
+
+			/* todo: init: revise localport nvme attributes */
 		}
+	}
 	lpfc_destroy_vport_work_array(phba, vports);
 	/* Clean up any firmware default rpi's */
 	mb = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
@@ -914,7 +923,6 @@ lpfc_linkdown(struct lpfc_hba *phba)
 
 	/* Setup myDID for link up if we are in pt2pt mode */
 	if (phba->pport->fc_flag & FC_PT2PT) {
-		phba->pport->fc_myDID = 0;
 		mb = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
 		if (mb) {
 			lpfc_config_link(phba, mb);
@@ -929,7 +937,6 @@ lpfc_linkdown(struct lpfc_hba *phba)
 		phba->pport->fc_flag &= ~(FC_PT2PT | FC_PT2PT_PLOGI);
 		spin_unlock_irq(shost->host_lock);
 	}
-
 	return 0;
 }
 
@@ -1016,7 +1023,7 @@ lpfc_linkup(struct lpfc_hba *phba)
  * This routine handles processing a CLEAR_LA mailbox
  * command upon completion. It is setup in the LPFC_MBOXQ
  * as the completion routine when the command is
- * handed off to the SLI layer.
+ * handed off to the SLI layer. SLI3 only.
  */
 static void
 lpfc_mbx_cmpl_clear_la(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
@@ -1028,9 +1035,8 @@ lpfc_mbx_cmpl_clear_la(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 	uint32_t control;
 
 	/* Since we don't do discovery right now, turn these off here */
-	psli->ring[psli->extra_ring].flag &= ~LPFC_STOP_IOCB_EVENT;
-	psli->ring[psli->fcp_ring].flag &= ~LPFC_STOP_IOCB_EVENT;
-	psli->ring[psli->next_ring].flag &= ~LPFC_STOP_IOCB_EVENT;
+	psli->sli3_ring[LPFC_EXTRA_RING].flag &= ~LPFC_STOP_IOCB_EVENT;
+	psli->sli3_ring[LPFC_FCP_RING].flag &= ~LPFC_STOP_IOCB_EVENT;
 
 	/* Check for error */
 	if ((mb->mbxStatus) && (mb->mbxStatus != 0x1601)) {
@@ -1751,7 +1757,7 @@ lpfc_sli4_new_fcf_random_select(struct lpfc_hba *phba, uint32_t fcf_cnt)
 	uint32_t rand_num;
 
 	/* Get 16-bit uniform random number */
-	rand_num = 0xFFFF & prandom_u32();
+	rand_num = (0xFFFF & prandom_u32());
 
 	/* Decision with probability 1/fcf_cnt */
 	if ((fcf_cnt * rand_num) < 0xFFFF)
@@ -3277,7 +3283,7 @@ lpfc_mbx_issue_link_down(struct lpfc_hba *phba)
  * This routine handles processing a READ_TOPOLOGY mailbox
  * command upon completion. It is setup in the LPFC_MBOXQ
  * as the completion routine when the command is
- * handed off to the SLI layer.
+ * handed off to the SLI layer. SLI4 only.
  */
 void
 lpfc_mbx_cmpl_read_topology(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
@@ -3289,7 +3295,12 @@ lpfc_mbx_cmpl_read_topology(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 	struct lpfc_dmabuf *mp = (struct lpfc_dmabuf *) (pmb->context1);
 
 	/* Unblock ELS traffic */
-	phba->sli.ring[LPFC_ELS_RING].flag &= ~LPFC_STOP_IOCB_EVENT;
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		phba->sli.sli3_ring[LPFC_ELS_RING].flag &=
+							 ~LPFC_STOP_IOCB_EVENT;
+	else
+		phba->sli4_hba.els_wq->pring->flag &= ~LPFC_STOP_IOCB_EVENT;
+
 	/* Check for error */
 	if (mb->mbxStatus) {
 		lpfc_printf_log(phba, KERN_INFO, LOG_LINK_EVENT,
@@ -3458,6 +3469,14 @@ lpfc_mbx_cmpl_reg_login(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 		spin_lock_irq(shost->host_lock);
 		ndlp->nlp_flag &= ~NLP_IGNR_REG_CMPL;
 		spin_unlock_irq(shost->host_lock);
+
+		/*
+		 * We cannot leave the RPI registered because
+		 * if we go thru discovery again for this ndlp
+		 * a subsequent REG_RPI will fail.
+		 */
+		ndlp->nlp_flag |= NLP_RPI_REGISTERED;
+		lpfc_unreg_rpi(vport, ndlp);
 	}
 
 	/* Call state machine */
@@ -3903,6 +3922,9 @@ lpfc_register_remote_port(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 	struct fc_rport_identifiers rport_ids;
 	struct lpfc_hba  *phba = vport->phba;
 
+	if (phba->cfg_enable_fc4_type == LPFC_ENABLE_NVME)
+		return;
+
 	/* Remote port has reappeared. Re-register w/ FC transport */
 	rport_ids.node_name = wwn_to_u64(ndlp->nlp_nodename.u.wwn);
 	rport_ids.port_name = wwn_to_u64(ndlp->nlp_portname.u.wwn);
@@ -3972,12 +3994,13 @@ static void
 lpfc_unregister_remote_port(struct lpfc_nodelist *ndlp)
 {
 	struct fc_rport *rport = ndlp->rport;
+	struct lpfc_vport *vport = ndlp->vport;
 
-	lpfc_debugfs_disc_trc(ndlp->vport, LPFC_DISC_TRC_RPORT,
+	lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_RPORT,
 		"rport delete:    did:x%x flg:x%x type x%x",
 		ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_type);
 
-	lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE,
+	lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
 			 "3184 rport unregister x%06x, rport %p\n",
 			 ndlp->nlp_DID, rport);
 
@@ -4394,7 +4417,6 @@ lpfc_check_sli_ndlp(struct lpfc_hba *phba,
 		    struct lpfc_iocbq *iocb,
 		    struct lpfc_nodelist *ndlp)
 {
-	struct lpfc_sli *psli = &phba->sli;
 	IOCB_t *icmd = &iocb->iocb;
 	struct lpfc_vport    *vport = ndlp->vport;
 
@@ -4413,9 +4435,7 @@ lpfc_check_sli_ndlp(struct lpfc_hba *phba,
 			if (iocb->context1 == (uint8_t *) ndlp)
 				return 1;
 		}
-	} else if (pring->ringno == psli->extra_ring) {
-
-	} else if (pring->ringno == psli->fcp_ring) {
+	} else if (pring->ringno == LPFC_FCP_RING) {
 		/* Skip match check if waiting to relogin to FCP target */
 		if ((ndlp->nlp_type & NLP_FCP_TARGET) &&
 		    (ndlp->nlp_flag & NLP_DELAY_TMO)) {
@@ -4424,8 +4444,6 @@ lpfc_check_sli_ndlp(struct lpfc_hba *phba,
 		if (icmd->ulpContext == (volatile ushort)ndlp->nlp_rpi) {
 			return 1;
 		}
-	} else if (pring->ringno == psli->next_ring) {
-
 	}
 	return 0;
 }
@@ -4441,6 +4459,7 @@ lpfc_no_rpi(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
 	struct lpfc_sli *psli;
 	struct lpfc_sli_ring *pring;
 	struct lpfc_iocbq *iocb, *next_iocb;
+	struct lpfc_queue *qp = NULL;
 	uint32_t i;
 
 	lpfc_fabric_abort_nport(ndlp);
@@ -4452,28 +4471,51 @@ lpfc_no_rpi(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
 	psli = &phba->sli;
 	if (ndlp->nlp_flag & NLP_RPI_REGISTERED) {
 		/* Now process each ring */
-		for (i = 0; i < psli->num_rings; i++) {
-			pring = &psli->ring[i];
-
-			spin_lock_irq(&phba->hbalock);
-			list_for_each_entry_safe(iocb, next_iocb, &pring->txq,
-						 list) {
-				/*
-				 * Check to see if iocb matches the nport we are
-				 * looking for
-				 */
+		if (phba->sli_rev != LPFC_SLI_REV4) {
+			for (i = 0; i < psli->num_rings; i++) {
+				pring = &psli->sli3_ring[i];
+				spin_lock_irq(&phba->hbalock);
+				list_for_each_entry_safe(iocb, next_iocb,
+							 &pring->txq, list) {
+					/*
+					 * Check to see if iocb matches the
+					 * nport we are looking for
+					 */
+					if ((lpfc_check_sli_ndlp(phba, pring,
+								 iocb, ndlp))) {
+						/* It matches, so deque and
+						 * call compl with an error
+						 */
+						list_move_tail(&iocb->list,
+							       &completions);
+					}
+				}
+				spin_unlock_irq(&phba->hbalock);
+			}
+			goto out;
+		}
+		spin_lock_irq(&phba->hbalock);
+		list_for_each_entry(qp, &phba->sli4_hba.lpfc_wq_list, wq_list) {
+			pring = qp->pring;
+			if (!pring)
+				continue;
+			spin_lock_irq(&pring->ring_lock);
+			list_for_each_entry_safe(iocb, next_iocb,
+						 &pring->txq, list) {
 				if ((lpfc_check_sli_ndlp(phba, pring, iocb,
 							 ndlp))) {
-					/* It matches, so deque and call compl
-					   with an error */
+					/* It matches, so deque and
+					 * call compl with an error
+					 */
 					list_move_tail(&iocb->list,
 						       &completions);
 				}
 			}
-			spin_unlock_irq(&phba->hbalock);
+			spin_unlock_irq(&pring->ring_lock);
 		}
+		spin_unlock_irq(&phba->hbalock);
 	}
-
+out:
 	/* Cancel all the IOCBs from the completions list */
 	lpfc_sli_cancel_iocbs(phba, &completions, IOSTAT_LOCAL_REJECT,
 			      IOERR_SLI_ABORTED);
@@ -5040,14 +5082,14 @@ lpfc_disc_list_loopmap(struct lpfc_vport *vport)
 	return;
 }
 
+/* SLI3 only */
 void
 lpfc_issue_clear_la(struct lpfc_hba *phba, struct lpfc_vport *vport)
 {
 	LPFC_MBOXQ_t *mbox;
 	struct lpfc_sli *psli = &phba->sli;
-	struct lpfc_sli_ring *extra_ring = &psli->ring[psli->extra_ring];
-	struct lpfc_sli_ring *fcp_ring   = &psli->ring[psli->fcp_ring];
-	struct lpfc_sli_ring *next_ring  = &psli->ring[psli->next_ring];
+	struct lpfc_sli_ring *extra_ring = &psli->sli3_ring[LPFC_EXTRA_RING];
+	struct lpfc_sli_ring *fcp_ring   = &psli->sli3_ring[LPFC_FCP_RING];
 	int  rc;
 
 	/*
@@ -5071,7 +5113,6 @@ lpfc_issue_clear_la(struct lpfc_hba *phba, struct lpfc_vport *vport)
 			lpfc_disc_flush_list(vport);
 			extra_ring->flag &= ~LPFC_STOP_IOCB_EVENT;
 			fcp_ring->flag &= ~LPFC_STOP_IOCB_EVENT;
-			next_ring->flag &= ~LPFC_STOP_IOCB_EVENT;
 			phba->link_state = LPFC_HBA_ERROR;
 		}
 	}
@@ -5207,7 +5248,10 @@ lpfc_free_tx(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
 	struct lpfc_sli_ring *pring;
 
 	psli = &phba->sli;
-	pring = &psli->ring[LPFC_ELS_RING];
+	if (phba->sli_rev != LPFC_SLI_REV4)
+		pring = &phba->sli.sli3_ring[LPFC_ELS_RING];
+	else
+		pring = phba->sli4_hba.els_wq->pring;
 
 	/* Error matching iocb on txq or txcmplq
 	 * First check the txq.
@@ -5523,12 +5567,14 @@ lpfc_disc_timeout_handler(struct lpfc_vport *vport)
 
 	if (clrlaerr) {
 		lpfc_disc_flush_list(vport);
-		psli->ring[(psli->extra_ring)].flag &= ~LPFC_STOP_IOCB_EVENT;
-		psli->ring[(psli->fcp_ring)].flag &= ~LPFC_STOP_IOCB_EVENT;
-		psli->ring[(psli->next_ring)].flag &= ~LPFC_STOP_IOCB_EVENT;
+		if (phba->sli_rev != LPFC_SLI_REV4) {
+			psli->sli3_ring[(LPFC_EXTRA_RING)].flag &=
+				~LPFC_STOP_IOCB_EVENT;
+			psli->sli3_ring[LPFC_FCP_RING].flag &=
+				~LPFC_STOP_IOCB_EVENT;
+		}
 		vport->port_state = LPFC_VPORT_READY;
 	}
-
 	return;
 }
 
diff --git a/drivers/scsi/lpfc/lpfc_hw.h b/drivers/scsi/lpfc/lpfc_hw.h
index 3b970d37..5d7416c 100644
--- a/drivers/scsi/lpfc/lpfc_hw.h
+++ b/drivers/scsi/lpfc/lpfc_hw.h
@@ -44,8 +44,6 @@
 #define LPFC_FCP_RING            0	/* ring 0 for FCP initiator commands */
 #define LPFC_EXTRA_RING          1	/* ring 1 for other protocols */
 #define LPFC_ELS_RING            2	/* ring 2 for ELS commands */
-#define LPFC_FCP_NEXT_RING       3
-#define LPFC_FCP_OAS_RING        3
 
 #define SLI2_IOCB_CMD_R0_ENTRIES    172	/* SLI-2 FCP command ring entries */
 #define SLI2_IOCB_RSP_R0_ENTRIES    134	/* SLI-2 FCP response ring entries */
@@ -1791,6 +1789,7 @@ typedef struct {		/* FireFly BIU registers */
 #define MBX_INIT_VFI        0xA3
 #define MBX_INIT_VPI        0xA4
 #define MBX_ACCESS_VDATA    0xA5
+#define MBX_REG_FCFI_MRQ    0xAF
 
 #define MBX_AUTH_PORT       0xF8
 #define MBX_SECURITY_MGMT   0xF9
@@ -4163,7 +4162,7 @@ typedef struct _IOCB {	/* IOCB structure */
 
 /* HBQ entries are 4 words each = 4k */
 #define LPFC_TOTAL_HBQ_SIZE (sizeof(struct lpfc_hbq_entry) *  \
-			     lpfc_sli_hbq_count())
+			     lpfc_sli_hbq_count(phba))
 
 struct lpfc_sli2_slim {
 	MAILBOX_t mbx;
diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h
index 5646699..7df854e 100644
--- a/drivers/scsi/lpfc/lpfc_hw4.h
+++ b/drivers/scsi/lpfc/lpfc_hw4.h
@@ -108,6 +108,7 @@ struct lpfc_sli_intf {
 #define LPFC_MAX_MQ_PAGE		8
 #define LPFC_MAX_WQ_PAGE_V0		4
 #define LPFC_MAX_WQ_PAGE		8
+#define LPFC_MAX_RQ_PAGE		8
 #define LPFC_MAX_CQ_PAGE		4
 #define LPFC_MAX_EQ_PAGE		8
 
@@ -198,7 +199,7 @@ struct lpfc_sli_intf {
 /* Configuration of Interrupts / sec for entire HBA port */
 #define LPFC_MIN_IMAX          5000
 #define LPFC_MAX_IMAX          5000000
-#define LPFC_DEF_IMAX          50000
+#define LPFC_DEF_IMAX          150000
 
 #define LPFC_MIN_CPU_MAP       0
 #define LPFC_MAX_CPU_MAP       2
@@ -348,6 +349,7 @@ struct lpfc_cqe {
 #define CQE_CODE_RECEIVE		0x4
 #define CQE_CODE_XRI_ABORTED		0x5
 #define CQE_CODE_RECEIVE_V1		0x9
+#define CQE_CODE_NVME_ERSP		0xd
 
 /*
  * Define mask value for xri_aborted and wcqe completed CQE extended status.
@@ -367,6 +369,9 @@ struct lpfc_wcqe_complete {
 #define lpfc_wcqe_c_hw_status_SHIFT	0
 #define lpfc_wcqe_c_hw_status_MASK	0x000000FF
 #define lpfc_wcqe_c_hw_status_WORD	word0
+#define lpfc_wcqe_c_ersp0_SHIFT		0
+#define lpfc_wcqe_c_ersp0_MASK		0x0000FFFF
+#define lpfc_wcqe_c_ersp0_WORD		word0
 	uint32_t total_data_placed;
 	uint32_t parameter;
 #define lpfc_wcqe_c_bg_edir_SHIFT	5
@@ -400,6 +405,9 @@ struct lpfc_wcqe_complete {
 #define lpfc_wcqe_c_code_SHIFT		lpfc_cqe_code_SHIFT
 #define lpfc_wcqe_c_code_MASK		lpfc_cqe_code_MASK
 #define lpfc_wcqe_c_code_WORD		lpfc_cqe_code_WORD
+#define lpfc_wcqe_c_sqhead_SHIFT	0
+#define lpfc_wcqe_c_sqhead_MASK		0x0000FFFF
+#define lpfc_wcqe_c_sqhead_WORD		word3
 };
 
 /* completion queue entry for wqe release */
@@ -1186,6 +1194,7 @@ struct lpfc_mbx_wq_create {
 #define lpfc_mbx_wq_create_page_size_SHIFT	0
 #define lpfc_mbx_wq_create_page_size_MASK	0x000000FF
 #define lpfc_mbx_wq_create_page_size_WORD	word1
+#define LPFC_WQ_PAGE_SIZE_4096	0x1
 #define lpfc_mbx_wq_create_wqe_size_SHIFT	8
 #define lpfc_mbx_wq_create_wqe_size_MASK	0x0000000F
 #define lpfc_mbx_wq_create_wqe_size_WORD	word1
@@ -1243,10 +1252,10 @@ struct rq_context {
 #define LPFC_RQ_RING_SIZE_1024		10	/* 1024 entries */
 #define LPFC_RQ_RING_SIZE_2048		11	/* 2048 entries */
 #define LPFC_RQ_RING_SIZE_4096		12	/* 4096 entries */
-#define lpfc_rq_context_rqe_count_1_SHIFT	16	/* Version 1 Only */
+#define lpfc_rq_context_rqe_count_1_SHIFT	16	/* Version 1-2 Only */
 #define lpfc_rq_context_rqe_count_1_MASK	0x0000FFFF
 #define lpfc_rq_context_rqe_count_1_WORD	word0
-#define lpfc_rq_context_rqe_size_SHIFT	8		/* Version 1 Only */
+#define lpfc_rq_context_rqe_size_SHIFT	8		/* Version 1-2 Only */
 #define lpfc_rq_context_rqe_size_MASK	0x0000000F
 #define lpfc_rq_context_rqe_size_WORD	word0
 #define LPFC_RQE_SIZE_8		2
@@ -1257,7 +1266,14 @@ struct rq_context {
 #define lpfc_rq_context_page_size_SHIFT	0		/* Version 1 Only */
 #define lpfc_rq_context_page_size_MASK	0x000000FF
 #define lpfc_rq_context_page_size_WORD	word0
-	uint32_t reserved1;
+#define	LPFC_RQ_PAGE_SIZE_4096	0x1
+	uint32_t word1;
+#define lpfc_rq_context_data_size_SHIFT	16		/* Version 2 Only */
+#define lpfc_rq_context_data_size_MASK	0x0000FFFF
+#define lpfc_rq_context_data_size_WORD	word1
+#define lpfc_rq_context_hdr_size_SHIFT	0		/* Version 2 Only */
+#define lpfc_rq_context_hdr_size_MASK	0x0000FFFF
+#define lpfc_rq_context_hdr_size_WORD	word1
 	uint32_t word2;
 #define lpfc_rq_context_cq_id_SHIFT	16
 #define lpfc_rq_context_cq_id_MASK	0x000003FF
@@ -1265,6 +1281,9 @@ struct rq_context {
 #define lpfc_rq_context_buf_size_SHIFT	0
 #define lpfc_rq_context_buf_size_MASK	0x0000FFFF
 #define lpfc_rq_context_buf_size_WORD	word2
+#define lpfc_rq_context_base_cq_SHIFT	0		/* Version 2 Only */
+#define lpfc_rq_context_base_cq_MASK	0x0000FFFF
+#define lpfc_rq_context_base_cq_WORD	word2
 	uint32_t buffer_size;				/* Version 1 Only */
 };
 
@@ -1286,10 +1305,65 @@ struct lpfc_mbx_rq_create {
 #define lpfc_mbx_rq_create_ulp_num_MASK		0x000000FF
 #define lpfc_mbx_rq_create_ulp_num_WORD		word0
 			struct rq_context context;
-			struct dma_address page[LPFC_MAX_WQ_PAGE];
+			struct dma_address page[LPFC_MAX_RQ_PAGE];
 		} request;
 		struct {
 			uint32_t word0;
+#define lpfc_mbx_rq_create_q_cnt_v2_SHIFT	16
+#define lpfc_mbx_rq_create_q_cnt_v2_MASK	0x0000FFFF
+#define lpfc_mbx_rq_create_q_cnt_v2_WORD	word0
+#define lpfc_mbx_rq_create_q_id_SHIFT		0
+#define lpfc_mbx_rq_create_q_id_MASK		0x0000FFFF
+#define lpfc_mbx_rq_create_q_id_WORD		word0
+			uint32_t doorbell_offset;
+			uint32_t word2;
+#define lpfc_mbx_rq_create_bar_set_SHIFT	0
+#define lpfc_mbx_rq_create_bar_set_MASK		0x0000FFFF
+#define lpfc_mbx_rq_create_bar_set_WORD		word2
+#define lpfc_mbx_rq_create_db_format_SHIFT	16
+#define lpfc_mbx_rq_create_db_format_MASK	0x0000FFFF
+#define lpfc_mbx_rq_create_db_format_WORD	word2
+		} response;
+	} u;
+};
+
+struct lpfc_mbx_rq_create_v2 {
+	union  lpfc_sli4_cfg_shdr cfg_shdr;
+	union {
+		struct {
+			uint32_t word0;
+#define lpfc_mbx_rq_create_num_pages_SHIFT	0
+#define lpfc_mbx_rq_create_num_pages_MASK	0x0000FFFF
+#define lpfc_mbx_rq_create_num_pages_WORD	word0
+#define lpfc_mbx_rq_create_rq_cnt_SHIFT		16
+#define lpfc_mbx_rq_create_rq_cnt_MASK		0x000000FF
+#define lpfc_mbx_rq_create_rq_cnt_WORD		word0
+#define lpfc_mbx_rq_create_dua_SHIFT		16
+#define lpfc_mbx_rq_create_dua_MASK		0x00000001
+#define lpfc_mbx_rq_create_dua_WORD		word0
+#define lpfc_mbx_rq_create_bqu_SHIFT		17
+#define lpfc_mbx_rq_create_bqu_MASK		0x00000001
+#define lpfc_mbx_rq_create_bqu_WORD		word0
+#define lpfc_mbx_rq_create_ulp_num_SHIFT	24
+#define lpfc_mbx_rq_create_ulp_num_MASK		0x000000FF
+#define lpfc_mbx_rq_create_ulp_num_WORD		word0
+#define lpfc_mbx_rq_create_dim_SHIFT		29
+#define lpfc_mbx_rq_create_dim_MASK		0x00000001
+#define lpfc_mbx_rq_create_dim_WORD		word0
+#define lpfc_mbx_rq_create_dfd_SHIFT		30
+#define lpfc_mbx_rq_create_dfd_MASK		0x00000001
+#define lpfc_mbx_rq_create_dfd_WORD		word0
+#define lpfc_mbx_rq_create_dnb_SHIFT		31
+#define lpfc_mbx_rq_create_dnb_MASK		0x00000001
+#define lpfc_mbx_rq_create_dnb_WORD		word0
+			struct rq_context context;
+			struct dma_address page[1];
+		} request;
+		struct {
+			uint32_t word0;
+#define lpfc_mbx_rq_create_q_cnt_v2_SHIFT	16
+#define lpfc_mbx_rq_create_q_cnt_v2_MASK	0x0000FFFF
+#define lpfc_mbx_rq_create_q_cnt_v2_WORD	word0
 #define lpfc_mbx_rq_create_q_id_SHIFT		0
 #define lpfc_mbx_rq_create_q_id_MASK		0x0000FFFF
 #define lpfc_mbx_rq_create_q_id_WORD		word0
@@ -2839,12 +2913,18 @@ struct lpfc_sli4_parameters {
 #define cfg_mqv_WORD				word6
 	uint32_t word7;
 	uint32_t word8;
+#define cfg_wqpcnt_SHIFT			0
+#define cfg_wqpcnt_MASK				0x0000000f
+#define cfg_wqpcnt_WORD				word8
 #define cfg_wqsize_SHIFT			8
 #define cfg_wqsize_MASK				0x0000000f
 #define cfg_wqsize_WORD				word8
 #define cfg_wqv_SHIFT				14
 #define cfg_wqv_MASK				0x00000003
 #define cfg_wqv_WORD				word8
+#define cfg_wqpsize_SHIFT			16
+#define cfg_wqpsize_MASK			0x000000ff
+#define cfg_wqpsize_WORD			word8
 	uint32_t word9;
 	uint32_t word10;
 #define cfg_rqv_SHIFT				14
@@ -2895,6 +2975,12 @@ struct lpfc_sli4_parameters {
 #define cfg_mds_diags_SHIFT			1
 #define cfg_mds_diags_MASK			0x00000001
 #define cfg_mds_diags_WORD			word19
+#define cfg_nvme_SHIFT				3
+#define cfg_nvme_MASK				0x00000001
+#define cfg_nvme_WORD				word19
+#define cfg_xib_SHIFT				4
+#define cfg_xib_MASK				0x00000001
+#define cfg_xib_WORD				word19
 };
 
 #define LPFC_SET_UE_RECOVERY		0x10
@@ -3298,6 +3384,7 @@ struct lpfc_mqe {
 		struct lpfc_mbx_cq_create cq_create;
 		struct lpfc_mbx_wq_create wq_create;
 		struct lpfc_mbx_rq_create rq_create;
+		struct lpfc_mbx_rq_create_v2 rq_create_v2;
 		struct lpfc_mbx_mq_destroy mq_destroy;
 		struct lpfc_mbx_eq_destroy eq_destroy;
 		struct lpfc_mbx_cq_destroy cq_destroy;
@@ -3657,6 +3744,9 @@ struct wqe_common {
 #define wqe_ebde_cnt_SHIFT    0
 #define wqe_ebde_cnt_MASK     0x0000000f
 #define wqe_ebde_cnt_WORD     word10
+#define wqe_nvme_SHIFT        4
+#define wqe_nvme_MASK         0x00000001
+#define wqe_nvme_WORD         word10
 #define wqe_oas_SHIFT         6
 #define wqe_oas_MASK          0x00000001
 #define wqe_oas_WORD          word10
@@ -3717,9 +3807,18 @@ struct wqe_common {
 #define LPFC_ELS_ID_FDISC	2
 #define LPFC_ELS_ID_LOGO	1
 #define LPFC_ELS_ID_DEFAULT	0
+#define wqe_irsp_SHIFT        4
+#define wqe_irsp_MASK         0x00000001
+#define wqe_irsp_WORD         word11
+#define wqe_sup_SHIFT         6
+#define wqe_sup_MASK          0x00000001
+#define wqe_sup_WORD          word11
 #define wqe_wqec_SHIFT        7
 #define wqe_wqec_MASK         0x00000001
 #define wqe_wqec_WORD         word11
+#define wqe_irsplen_SHIFT     8
+#define wqe_irsplen_MASK      0x0000000f
+#define wqe_irsplen_WORD      word11
 #define wqe_cqid_SHIFT        16
 #define wqe_cqid_MASK         0x0000ffff
 #define wqe_cqid_WORD         word11
@@ -3969,6 +4068,35 @@ struct fcp_icmnd64_wqe {
 	uint32_t rsvd_12_15[4];        /* word 12-15 */
 };
 
+struct fcp_trsp64_wqe {
+	struct ulp_bde64 bde;
+	uint32_t response_len;
+	uint32_t rsvd_4_5[2];
+	struct wqe_common wqe_com;      /* words 6-11 */
+	uint32_t rsvd_12_15[4];         /* word 12-15 */
+};
+
+struct fcp_tsend64_wqe {
+	struct ulp_bde64 bde;
+	uint32_t payload_offset_len;
+	uint32_t relative_offset;
+	uint32_t reserved;
+	struct wqe_common wqe_com;     /* words 6-11 */
+	uint32_t fcp_data_len;         /* word 12 */
+	uint32_t rsvd_13_15[3];        /* word 13-15 */
+};
+
+struct fcp_treceive64_wqe {
+	struct ulp_bde64 bde;
+	uint32_t payload_offset_len;
+	uint32_t relative_offset;
+	uint32_t reserved;
+	struct wqe_common wqe_com;     /* words 6-11 */
+	uint32_t fcp_data_len;         /* word 12 */
+	uint32_t rsvd_13_15[3];        /* word 13-15 */
+};
+#define TXRDY_PAYLOAD_LEN      12
+
 
 union lpfc_wqe {
 	uint32_t words[16];
@@ -3984,6 +4112,10 @@ union lpfc_wqe {
 	struct xmit_els_rsp64_wqe xmit_els_rsp;
 	struct els_request64_wqe els_req;
 	struct gen_req64_wqe gen_req;
+	struct fcp_trsp64_wqe fcp_trsp;
+	struct fcp_tsend64_wqe fcp_tsend;
+	struct fcp_treceive64_wqe fcp_treceive;
+
 };
 
 union lpfc_wqe128 {
@@ -3992,6 +4124,9 @@ union lpfc_wqe128 {
 	struct fcp_icmnd64_wqe fcp_icmd;
 	struct fcp_iread64_wqe fcp_iread;
 	struct fcp_iwrite64_wqe fcp_iwrite;
+	struct fcp_trsp64_wqe fcp_trsp;
+	struct fcp_tsend64_wqe fcp_tsend;
+	struct fcp_treceive64_wqe fcp_treceive;
 	struct xmit_seq64_wqe xmit_sequence;
 	struct gen_req64_wqe gen_req;
 };
@@ -4015,11 +4150,39 @@ struct lpfc_grp_hdr {
 	uint8_t revision[32];
 };
 
-#define FCP_COMMAND 0x0
-#define FCP_COMMAND_DATA_OUT 0x1
-#define ELS_COMMAND_NON_FIP 0xC
-#define ELS_COMMAND_FIP 0xD
-#define OTHER_COMMAND 0x8
+/* Defines for WQE command type */
+#define FCP_COMMAND		0x0
+#define NVME_READ_CMD		0x0
+#define FCP_COMMAND_DATA_OUT	0x1
+#define NVME_WRITE_CMD		0x1
+#define FCP_COMMAND_TRECEIVE	0x2
+#define FCP_COMMAND_TRSP	0x3
+#define FCP_COMMAND_TSEND	0x7
+#define OTHER_COMMAND		0x8
+#define ELS_COMMAND_NON_FIP	0xC
+#define ELS_COMMAND_FIP		0xD
+
+#define LPFC_NVME_EMBED_CMD	0x0
+#define LPFC_NVME_EMBED_WRITE	0x1
+#define LPFC_NVME_EMBED_READ	0x2
+
+/* WQE Commands */
+#define CMD_ABORT_XRI_WQE       0x0F
+#define CMD_XMIT_SEQUENCE64_WQE 0x82
+#define CMD_XMIT_BCAST64_WQE    0x84
+#define CMD_ELS_REQUEST64_WQE   0x8A
+#define CMD_XMIT_ELS_RSP64_WQE  0x95
+#define CMD_XMIT_BLS_RSP64_WQE  0x97
+#define CMD_FCP_IWRITE64_WQE    0x98
+#define CMD_FCP_IREAD64_WQE     0x9A
+#define CMD_FCP_ICMND64_WQE     0x9C
+#define CMD_FCP_TSEND64_WQE     0x9F
+#define CMD_FCP_TRECEIVE64_WQE  0xA1
+#define CMD_FCP_TRSP64_WQE      0xA3
+#define CMD_GEN_REQUEST64_WQE   0xC2
+
+#define CMD_WQE_MASK            0xff
+
 
 #define LPFC_FW_DUMP	1
 #define LPFC_FW_RESET	2
diff --git a/drivers/scsi/lpfc/lpfc_logmsg.h b/drivers/scsi/lpfc/lpfc_logmsg.h
index 2a4e5d2..3faf7a0 100644
--- a/drivers/scsi/lpfc/lpfc_logmsg.h
+++ b/drivers/scsi/lpfc/lpfc_logmsg.h
@@ -38,6 +38,10 @@
 #define LOG_FIP		0x00020000	/* FIP events */
 #define LOG_FCP_UNDER	0x00040000	/* FCP underruns errors */
 #define LOG_SCSI_CMD	0x00080000	/* ALL SCSI commands */
+#define LOG_NVME	0x00100000	/* NVME general events. */
+#define LOG_NVME_DISC   0x00200000      /* NVME Discovery/Connect events. */
+#define LOG_NVME_ABTS   0x00400000      /* NVME ABTS events. */
+#define LOG_NVME_IOERR  0x00800000      /* NVME IO Error events. */
 #define LOG_ALL_MSG	0xffffffff	/* LOG all messages */
 
 #define lpfc_printf_vlog(vport, level, mask, fmt, arg...) \
diff --git a/drivers/scsi/lpfc/lpfc_mbox.c b/drivers/scsi/lpfc/lpfc_mbox.c
index b234c50..e3258bf 100644
--- a/drivers/scsi/lpfc/lpfc_mbox.c
+++ b/drivers/scsi/lpfc/lpfc_mbox.c
@@ -954,7 +954,7 @@ lpfc_config_pcb_setup(struct lpfc_hba * phba)
 	pcbp->maxRing = (psli->num_rings - 1);
 
 	for (i = 0; i < psli->num_rings; i++) {
-		pring = &psli->ring[i];
+		pring = &psli->sli3_ring[i];
 
 		pring->sli.sli3.sizeCiocb =
 			phba->sli_rev == 3 ? SLI3_IOCB_CMD_SIZE :
@@ -1217,7 +1217,7 @@ lpfc_config_ring(struct lpfc_hba * phba, int ring, LPFC_MBOXQ_t * pmb)
 	mb->un.varCfgRing.recvNotify = 1;
 
 	psli = &phba->sli;
-	pring = &psli->ring[ring];
+	pring = &psli->sli3_ring[ring];
 	mb->un.varCfgRing.numMask = pring->num_mask;
 	mb->mbxCommand = MBX_CONFIG_RING;
 	mb->mbxOwner = OWN_HOST;
@@ -1291,7 +1291,7 @@ lpfc_config_port(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 			mb->un.varCfgPort.cdss = 1; /* Configure Security */
 		mb->un.varCfgPort.cerbm = 1; /* Request HBQs */
 		mb->un.varCfgPort.ccrp = 1; /* Command Ring Polling */
-		mb->un.varCfgPort.max_hbq = lpfc_sli_hbq_count();
+		mb->un.varCfgPort.max_hbq = lpfc_sli_hbq_count(phba);
 		if (phba->max_vpi && phba->cfg_enable_npiv &&
 		    phba->vpd.sli3Feat.cmv) {
 			mb->un.varCfgPort.max_vpi = LPFC_MAX_VPI;
@@ -2260,7 +2260,7 @@ lpfc_sli4_dump_cfg_rg23(struct lpfc_hba *phba, struct lpfcMboxq *mbox)
 	return 0;
 }
 
-static void
+void
 lpfc_mbx_cmpl_rdp_link_stat(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq)
 {
 	MAILBOX_t *mb;
@@ -2281,7 +2281,7 @@ lpfc_mbx_cmpl_rdp_link_stat(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq)
 	rdp_context->cmpl(phba, rdp_context, rc);
 }
 
-static void
+void
 lpfc_mbx_cmpl_rdp_page_a2(struct lpfc_hba *phba, LPFC_MBOXQ_t *mbox)
 {
 	struct lpfc_dmabuf *mp = (struct lpfc_dmabuf *) mbox->context1;
@@ -2328,7 +2328,7 @@ lpfc_mbx_cmpl_rdp_page_a0(struct lpfc_hba *phba, LPFC_MBOXQ_t *mbox)
 		goto error;
 
 	lpfc_sli_bemem_bcopy(mp->virt, &rdp_context->page_a0,
-				DMP_SFF_PAGE_A0_SIZE);
+			     DMP_SFF_PAGE_A0_SIZE);
 
 	memset(mbox, 0, sizeof(*mbox));
 
@@ -2341,13 +2341,13 @@ lpfc_mbx_cmpl_rdp_page_a0(struct lpfc_hba *phba, LPFC_MBOXQ_t *mbox)
 
 	bf_set(lpfc_mqe_command, &mbox->u.mqe, MBX_DUMP_MEMORY);
 	bf_set(lpfc_mbx_memory_dump_type3_type,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_LMSD);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_LMSD);
 	bf_set(lpfc_mbx_memory_dump_type3_link,
-		&mbox->u.mqe.un.mem_dump_type3, phba->sli4_hba.physical_port);
+	       &mbox->u.mqe.un.mem_dump_type3, phba->sli4_hba.physical_port);
 	bf_set(lpfc_mbx_memory_dump_type3_page_no,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_PAGE_A2);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_PAGE_A2);
 	bf_set(lpfc_mbx_memory_dump_type3_length,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_SFF_PAGE_A2_SIZE);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_SFF_PAGE_A2_SIZE);
 	mbox->u.mqe.un.mem_dump_type3.addr_lo = putPaddrLow(mp->phys);
 	mbox->u.mqe.un.mem_dump_type3.addr_hi = putPaddrHigh(mp->phys);
 
@@ -2400,13 +2400,13 @@ lpfc_sli4_dump_page_a0(struct lpfc_hba *phba, struct lpfcMboxq *mbox)
 	mbox->context1 = mp;
 
 	bf_set(lpfc_mbx_memory_dump_type3_type,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_LMSD);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_LMSD);
 	bf_set(lpfc_mbx_memory_dump_type3_link,
-		&mbox->u.mqe.un.mem_dump_type3, phba->sli4_hba.physical_port);
+	       &mbox->u.mqe.un.mem_dump_type3, phba->sli4_hba.physical_port);
 	bf_set(lpfc_mbx_memory_dump_type3_page_no,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_PAGE_A0);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_PAGE_A0);
 	bf_set(lpfc_mbx_memory_dump_type3_length,
-		&mbox->u.mqe.un.mem_dump_type3, DMP_SFF_PAGE_A0_SIZE);
+	       &mbox->u.mqe.un.mem_dump_type3, DMP_SFF_PAGE_A0_SIZE);
 	mbox->u.mqe.un.mem_dump_type3.addr_lo = putPaddrLow(mp->phys);
 	mbox->u.mqe.un.mem_dump_type3.addr_hi = putPaddrHigh(mp->phys);
 
@@ -2434,14 +2434,25 @@ lpfc_reg_fcfi(struct lpfc_hba *phba, struct lpfcMboxq *mbox)
 	memset(mbox, 0, sizeof(*mbox));
 	reg_fcfi = &mbox->u.mqe.un.reg_fcfi;
 	bf_set(lpfc_mqe_command, &mbox->u.mqe, MBX_REG_FCFI);
-	bf_set(lpfc_reg_fcfi_rq_id0, reg_fcfi, phba->sli4_hba.hdr_rq->queue_id);
-	bf_set(lpfc_reg_fcfi_rq_id1, reg_fcfi, REG_FCF_INVALID_QID);
+	if (phba->nvmet_support == 0) {
+		bf_set(lpfc_reg_fcfi_rq_id0, reg_fcfi,
+		       phba->sli4_hba.hdr_rq->queue_id);
+		/* Match everything - rq_id0 */
+		bf_set(lpfc_reg_fcfi_type_match0, reg_fcfi, 0);
+		bf_set(lpfc_reg_fcfi_type_mask0, reg_fcfi, 0);
+		bf_set(lpfc_reg_fcfi_rctl_match0, reg_fcfi, 0);
+		bf_set(lpfc_reg_fcfi_rctl_mask0, reg_fcfi, 0);
+
+		bf_set(lpfc_reg_fcfi_rq_id1, reg_fcfi, REG_FCF_INVALID_QID);
+
+		/* addr mode is bit wise inverted value of fcf addr_mode */
+		bf_set(lpfc_reg_fcfi_mam, reg_fcfi,
+		       (~phba->fcf.addr_mode) & 0x3);
+	}
 	bf_set(lpfc_reg_fcfi_rq_id2, reg_fcfi, REG_FCF_INVALID_QID);
 	bf_set(lpfc_reg_fcfi_rq_id3, reg_fcfi, REG_FCF_INVALID_QID);
 	bf_set(lpfc_reg_fcfi_info_index, reg_fcfi,
 	       phba->fcf.current_rec.fcf_indx);
-	/* reg_fcf addr mode is bit wise inverted value of fcf addr_mode */
-	bf_set(lpfc_reg_fcfi_mam, reg_fcfi, (~phba->fcf.addr_mode) & 0x3);
 	if (phba->fcf.current_rec.vlan_id != LPFC_FCOE_NULL_VID) {
 		bf_set(lpfc_reg_fcfi_vv, reg_fcfi, 1);
 		bf_set(lpfc_reg_fcfi_vlan_tag, reg_fcfi,
diff --git a/drivers/scsi/lpfc/lpfc_mem.c b/drivers/scsi/lpfc/lpfc_mem.c
index 3fa6533..0b2f309 100644
--- a/drivers/scsi/lpfc/lpfc_mem.c
+++ b/drivers/scsi/lpfc/lpfc_mem.c
@@ -24,10 +24,12 @@
 #include <linux/pci.h>
 #include <linux/interrupt.h>
 
+#include <scsi/scsi.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_transport_fc.h>
+#include <scsi/fc/fc_fs.h>
 
-#include <scsi/scsi.h>
+#include <linux/nvme-fc-driver.h>
 
 #include "lpfc_hw4.h"
 #include "lpfc_hw.h"
@@ -35,8 +37,9 @@
 #include "lpfc_sli4.h"
 #include "lpfc_nl.h"
 #include "lpfc_disc.h"
-#include "lpfc_scsi.h"
 #include "lpfc.h"
+#include "lpfc_scsi.h"
+#include "lpfc_nvme.h"
 #include "lpfc_crtn.h"
 #include "lpfc_logmsg.h"
 
@@ -66,7 +69,7 @@ lpfc_mem_alloc_active_rrq_pool_s4(struct lpfc_hba *phba) {
  * lpfc_mem_alloc - create and allocate all PCI and memory pools
  * @phba: HBA to allocate pools for
  *
- * Description: Creates and allocates PCI pools lpfc_scsi_dma_buf_pool,
+ * Description: Creates and allocates PCI pools lpfc_sg_dma_buf_pool,
  * lpfc_mbuf_pool, lpfc_hrb_pool.  Creates and allocates kmalloc-backed mempools
  * for LPFC_MBOXQ_t and lpfc_nodelist.  Also allocates the VPI bitmask.
  *
@@ -90,21 +93,23 @@ lpfc_mem_alloc(struct lpfc_hba *phba, int align)
 		else
 			i = SLI4_PAGE_SIZE;
 
-		phba->lpfc_scsi_dma_buf_pool =
-			pci_pool_create("lpfc_scsi_dma_buf_pool",
-				phba->pcidev,
-				phba->cfg_sg_dma_buf_size,
-				i,
-				0);
+		phba->lpfc_sg_dma_buf_pool =
+			pci_pool_create("lpfc_sg_dma_buf_pool",
+					phba->pcidev,
+					phba->cfg_sg_dma_buf_size,
+					i, 0);
+		if (!phba->lpfc_sg_dma_buf_pool)
+			goto fail;
+
 	} else {
-		phba->lpfc_scsi_dma_buf_pool =
-			pci_pool_create("lpfc_scsi_dma_buf_pool",
-				phba->pcidev, phba->cfg_sg_dma_buf_size,
-				align, 0);
-	}
+		phba->lpfc_sg_dma_buf_pool =
+			pci_pool_create("lpfc_sg_dma_buf_pool",
+					phba->pcidev, phba->cfg_sg_dma_buf_size,
+					align, 0);
 
-	if (!phba->lpfc_scsi_dma_buf_pool)
-		goto fail;
+		if (!phba->lpfc_sg_dma_buf_pool)
+			goto fail;
+	}
 
 	phba->lpfc_mbuf_pool = pci_pool_create("lpfc_mbuf_pool", phba->pcidev,
 							LPFC_BPL_SIZE,
@@ -170,12 +175,15 @@ lpfc_mem_alloc(struct lpfc_hba *phba, int align)
 					LPFC_DEVICE_DATA_POOL_SIZE,
 					sizeof(struct lpfc_device_data));
 		if (!phba->device_data_mem_pool)
-			goto fail_free_hrb_pool;
+			goto fail_free_drb_pool;
 	} else {
 		phba->device_data_mem_pool = NULL;
 	}
 
 	return 0;
+fail_free_drb_pool:
+	pci_pool_destroy(phba->lpfc_drb_pool);
+	phba->lpfc_drb_pool = NULL;
  fail_free_hrb_pool:
 	pci_pool_destroy(phba->lpfc_hrb_pool);
 	phba->lpfc_hrb_pool = NULL;
@@ -197,8 +205,8 @@ lpfc_mem_alloc(struct lpfc_hba *phba, int align)
 	pci_pool_destroy(phba->lpfc_mbuf_pool);
 	phba->lpfc_mbuf_pool = NULL;
  fail_free_dma_buf_pool:
-	pci_pool_destroy(phba->lpfc_scsi_dma_buf_pool);
-	phba->lpfc_scsi_dma_buf_pool = NULL;
+	pci_pool_destroy(phba->lpfc_sg_dma_buf_pool);
+	phba->lpfc_sg_dma_buf_pool = NULL;
  fail:
 	return -ENOMEM;
 }
@@ -227,6 +235,9 @@ lpfc_mem_free(struct lpfc_hba *phba)
 	if (phba->lpfc_hrb_pool)
 		pci_pool_destroy(phba->lpfc_hrb_pool);
 	phba->lpfc_hrb_pool = NULL;
+	if (phba->txrdy_payload_pool)
+		pci_pool_destroy(phba->txrdy_payload_pool);
+	phba->txrdy_payload_pool = NULL;
 
 	if (phba->lpfc_hbq_pool)
 		pci_pool_destroy(phba->lpfc_hbq_pool);
@@ -254,12 +265,14 @@ lpfc_mem_free(struct lpfc_hba *phba)
 			      pool->elements[i].phys);
 	kfree(pool->elements);
 
-	pci_pool_destroy(phba->lpfc_mbuf_pool);
+	if (phba->lpfc_mbuf_pool)
+		pci_pool_destroy(phba->lpfc_mbuf_pool);
 	phba->lpfc_mbuf_pool = NULL;
 
 	/* Free DMA buffer memory pool */
-	pci_pool_destroy(phba->lpfc_scsi_dma_buf_pool);
-	phba->lpfc_scsi_dma_buf_pool = NULL;
+	if (phba->lpfc_sg_dma_buf_pool)
+		pci_pool_destroy(phba->lpfc_sg_dma_buf_pool);
+	phba->lpfc_sg_dma_buf_pool = NULL;
 
 	/* Free Device Data memory pool */
 	if (phba->device_data_mem_pool) {
@@ -282,7 +295,7 @@ lpfc_mem_free(struct lpfc_hba *phba)
  * @phba: HBA to free memory for
  *
  * Description: Free memory from PCI and driver memory pools and also those
- * used : lpfc_scsi_dma_buf_pool, lpfc_mbuf_pool, lpfc_hrb_pool. Frees
+ * used : lpfc_sg_dma_buf_pool, lpfc_mbuf_pool, lpfc_hrb_pool. Frees
  * kmalloc-backed mempools for LPFC_MBOXQ_t and lpfc_nodelist. Also frees
  * the VPI bitmask.
  *
@@ -458,7 +471,7 @@ lpfc_els_hbq_alloc(struct lpfc_hba *phba)
 		kfree(hbqbp);
 		return NULL;
 	}
-	hbqbp->size = LPFC_BPL_SIZE;
+	hbqbp->total_size = LPFC_BPL_SIZE;
 	return hbqbp;
 }
 
@@ -518,7 +531,7 @@ lpfc_sli4_rb_alloc(struct lpfc_hba *phba)
 		kfree(dma_buf);
 		return NULL;
 	}
-	dma_buf->size = LPFC_BPL_SIZE;
+	dma_buf->total_size = LPFC_DATA_BUF_SIZE;
 	return dma_buf;
 }
 
@@ -540,7 +553,6 @@ lpfc_sli4_rb_free(struct lpfc_hba *phba, struct hbq_dmabuf *dmab)
 	pci_pool_free(phba->lpfc_hrb_pool, dmab->hbuf.virt, dmab->hbuf.phys);
 	pci_pool_free(phba->lpfc_drb_pool, dmab->dbuf.virt, dmab->dbuf.phys);
 	kfree(dmab);
-	return;
 }
 
 /**
@@ -565,13 +577,13 @@ lpfc_in_buf_free(struct lpfc_hba *phba, struct lpfc_dmabuf *mp)
 		return;
 
 	if (phba->sli3_options & LPFC_SLI3_HBQ_ENABLED) {
+		hbq_entry = container_of(mp, struct hbq_dmabuf, dbuf);
 		/* Check whether HBQ is still in use */
 		spin_lock_irqsave(&phba->hbalock, flags);
 		if (!phba->hbq_in_use) {
 			spin_unlock_irqrestore(&phba->hbalock, flags);
 			return;
 		}
-		hbq_entry = container_of(mp, struct hbq_dmabuf, dbuf);
 		list_del(&hbq_entry->dbuf.list);
 		if (hbq_entry->tag == -1) {
 			(phba->hbqs[LPFC_ELS_HBQ].hbq_free_buffer)
@@ -586,3 +598,48 @@ lpfc_in_buf_free(struct lpfc_hba *phba, struct lpfc_dmabuf *mp)
 	}
 	return;
 }
+
+/**
+ * lpfc_rq_buf_free - Free a RQ DMA buffer
+ * @phba: HBA buffer is associated with
+ * @mp: Buffer to free
+ *
+ * Description: Frees the given DMA buffer in the appropriate way given by
+ * reposting it to its associated RQ so it can be reused.
+ *
+ * Notes: Takes phba->hbalock.  Can be called with or without other locks held.
+ *
+ * Returns: None
+ **/
+void
+lpfc_rq_buf_free(struct lpfc_hba *phba, struct lpfc_dmabuf *mp)
+{
+	struct lpfc_rqb *rqbp;
+	struct lpfc_rqe hrqe;
+	struct lpfc_rqe drqe;
+	struct rqb_dmabuf *rqb_entry;
+	unsigned long flags;
+	int rc;
+
+	if (!mp)
+		return;
+
+	rqb_entry = container_of(mp, struct rqb_dmabuf, hbuf);
+	rqbp = rqb_entry->hrq->rqbp;
+
+	spin_lock_irqsave(&phba->hbalock, flags);
+	list_del(&rqb_entry->hbuf.list);
+	hrqe.address_lo = putPaddrLow(rqb_entry->hbuf.phys);
+	hrqe.address_hi = putPaddrHigh(rqb_entry->hbuf.phys);
+	drqe.address_lo = putPaddrLow(rqb_entry->dbuf.phys);
+	drqe.address_hi = putPaddrHigh(rqb_entry->dbuf.phys);
+	rc = lpfc_sli4_rq_put(rqb_entry->hrq, rqb_entry->drq, &hrqe, &drqe);
+	if (rc < 0) {
+		(rqbp->rqb_free_buffer)(phba, rqb_entry);
+	} else {
+		list_add_tail(&rqb_entry->hbuf.list, &rqbp->rqb_buffer_list);
+		rqbp->buffer_count++;
+	}
+
+	spin_unlock_irqrestore(&phba->hbalock, flags);
+}
diff --git a/drivers/scsi/lpfc/lpfc_nportdisc.c b/drivers/scsi/lpfc/lpfc_nportdisc.c
index 56a3df4..9643a59 100644
--- a/drivers/scsi/lpfc/lpfc_nportdisc.c
+++ b/drivers/scsi/lpfc/lpfc_nportdisc.c
@@ -205,9 +205,14 @@ lpfc_els_abort(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
 {
 	LIST_HEAD(abort_list);
 	struct lpfc_sli  *psli = &phba->sli;
-	struct lpfc_sli_ring *pring = &psli->ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring;
 	struct lpfc_iocbq *iocb, *next_iocb;
 
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		pring = phba->sli4_hba.els_wq->pring;
+	else
+		pring = &psli->sli3_ring[LPFC_ELS_RING];
+
 	/* Abort outstanding I/O on NPort <nlp_DID> */
 	lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_DISCOVERY,
 			 "2819 Abort outstanding I/O on NPort x%x "
@@ -2104,7 +2109,7 @@ lpfc_rcv_prlo_mapped_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 	struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *) arg;
 
 	/* flush the target */
-	lpfc_sli_abort_iocb(vport, &phba->sli.ring[phba->sli.fcp_ring],
+	lpfc_sli_abort_iocb(vport, &phba->sli.sli3_ring[LPFC_FCP_RING],
 			    ndlp->nlp_sid, 0, LPFC_CTX_TGT);
 
 	/* Treat like rcv logo */
diff --git a/drivers/scsi/lpfc/lpfc_nvme.h b/drivers/scsi/lpfc/lpfc_nvme.h
new file mode 100644
index 0000000..b5d9504
--- /dev/null
+++ b/drivers/scsi/lpfc/lpfc_nvme.h
@@ -0,0 +1,87 @@
+/*******************************************************************
+ * This file is part of the Emulex Linux Device Driver for         *
+ * Fibre Channel Host Bus Adapters.                                *
+ * Copyright (C) 2004-2016 Emulex.  All rights reserved.           *
+ * EMULEX and SLI are trademarks of Emulex.                        *
+ * www.emulex.com                                                  *
+ * Portions Copyright (C) 2004-2005 Christoph Hellwig              *
+ *                                                                 *
+ * This program is free software; you can redistribute it and/or   *
+ * modify it under the terms of version 2 of the GNU General       *
+ * Public License as published by the Free Software Foundation.    *
+ * This program is distributed in the hope that it will be useful. *
+ * ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND          *
+ * WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY,  *
+ * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE      *
+ * DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD *
+ * TO BE LEGALLY INVALID.  See the GNU General Public License for  *
+ * more details, a copy of which can be found in the file COPYING  *
+ * included with this package.                                     *
+ ********************************************************************/
+
+#define LPFC_NVME_MIN_SEGS		16
+#define LPFC_NVME_DEFAULT_SEGS		66	/* 256K IOs - 64 + 2 */
+#define LPFC_NVME_MAX_SEGS		510
+#define LPFC_NVME_MIN_POSTBUF		16
+#define LPFC_NVME_DEFAULT_POSTBUF	1024
+#define LPFC_NVME_MAX_POSTBUF		4096
+
+#define LPFC_NVME_ERSP_LEN		0x20
+
+/* Declare nvme-based local and remote port definitions. */
+struct lpfc_nvme_lport {
+	struct lpfc_vport *vport;
+	struct list_head rport_list;
+	struct completion lport_unreg_done;
+	/* Add sttats counters here */
+};
+
+struct lpfc_nvme_rport {
+	struct list_head list;
+	struct lpfc_nvme_lport *lport;
+	struct nvme_fc_remote_port *remoteport;
+	struct lpfc_nodelist *ndlp;
+	struct completion rport_unreg_done;
+};
+
+struct lpfc_nvme_buf {
+	struct list_head list;
+	struct nvmefc_fcp_req *nvmeCmd;
+	struct lpfc_nvme_rport *nrport;
+
+	uint32_t timeout;
+
+	uint16_t flags;  /* TBD convert exch_busy to flags */
+#define LPFC_SBUF_XBUSY         0x1     /* SLI4 hba reported XB on WCQE cmpl */
+	uint16_t exch_busy;     /* SLI4 hba reported XB on complete WCQE */
+	uint16_t status;	/* From IOCB Word 7- ulpStatus */
+	uint16_t cpu;
+	uint16_t qidx;
+	uint16_t sqid;
+	uint32_t result;	/* From IOCB Word 4. */
+
+	uint32_t   seg_cnt;	/* Number of scatter-gather segments returned by
+				 * dma_map_sg.  The driver needs this for calls
+				 * to dma_unmap_sg.
+				 */
+	dma_addr_t nonsg_phys;	/* Non scatter-gather physical address. */
+
+	/*
+	 * data and dma_handle are the kernel virtual and bus address of the
+	 * dma-able buffer containing the fcp_cmd, fcp_rsp and a scatter
+	 * gather bde list that supports the sg_tablesize value.
+	 */
+	void *data;
+	dma_addr_t dma_handle;
+
+	struct sli4_sge *nvme_sgl;
+	dma_addr_t dma_phys_sgl;
+
+	/* cur_iocbq has phys of the dma-able buffer.
+	 * Iotag is in here
+	 */
+	struct lpfc_iocbq cur_iocbq;
+
+	wait_queue_head_t *waitq;
+	unsigned long start_time;
+};
diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index 19d349f..c60dfc9 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -242,6 +242,7 @@ lpfc_update_stats(struct lpfc_hba *phba, struct  lpfc_scsi_buf *lpfc_cmd)
 	spin_unlock_irqrestore(shost->host_lock, flags);
 }
 
+
 /**
  * lpfc_rampdown_queue_depth - Post RAMP_DOWN_QUEUE event to worker thread
  * @phba: The Hba for which this call is being executed.
@@ -413,7 +414,7 @@ lpfc_new_scsi_buf_s3(struct lpfc_vport *vport, int num_to_alloc)
 		 * struct fcp_cmnd, struct fcp_rsp and the number of bde's
 		 * necessary to support the sg_tablesize.
 		 */
-		psb->data = pci_pool_zalloc(phba->lpfc_scsi_dma_buf_pool,
+		psb->data = pci_pool_zalloc(phba->lpfc_sg_dma_buf_pool,
 					GFP_KERNEL, &psb->dma_handle);
 		if (!psb->data) {
 			kfree(psb);
@@ -424,8 +425,8 @@ lpfc_new_scsi_buf_s3(struct lpfc_vport *vport, int num_to_alloc)
 		/* Allocate iotag for psb->cur_iocbq. */
 		iotag = lpfc_sli_next_iotag(phba, &psb->cur_iocbq);
 		if (iotag == 0) {
-			pci_pool_free(phba->lpfc_scsi_dma_buf_pool,
-					psb->data, psb->dma_handle);
+			pci_pool_free(phba->lpfc_sg_dma_buf_pool,
+				      psb->data, psb->dma_handle);
 			kfree(psb);
 			break;
 		}
@@ -522,6 +523,8 @@ lpfc_sli4_vport_delete_fcp_xri_aborted(struct lpfc_vport *vport)
 	struct lpfc_scsi_buf *psb, *next_psb;
 	unsigned long iflag = 0;
 
+	if (!(phba->cfg_enable_fc4_type & LPFC_ENABLE_FCP))
+		return;
 	spin_lock_irqsave(&phba->hbalock, iflag);
 	spin_lock(&phba->sli4_hba.abts_scsi_buf_list_lock);
 	list_for_each_entry_safe(psb, next_psb,
@@ -554,8 +557,10 @@ lpfc_sli4_fcp_xri_aborted(struct lpfc_hba *phba,
 	int i;
 	struct lpfc_nodelist *ndlp;
 	int rrq_empty = 0;
-	struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_ELS_RING];
+	struct lpfc_sli_ring *pring = phba->sli4_hba.els_wq->pring;
 
+	if (!(phba->cfg_enable_fc4_type & LPFC_ENABLE_FCP))
+		return;
 	spin_lock_irqsave(&phba->hbalock, iflag);
 	spin_lock(&phba->sli4_hba.abts_scsi_buf_list_lock);
 	list_for_each_entry_safe(psb, next_psb,
@@ -605,7 +610,7 @@ lpfc_sli4_fcp_xri_aborted(struct lpfc_hba *phba,
 }
 
 /**
- * lpfc_sli4_post_scsi_sgl_list - Post blocks of scsi buffer sgls from a list
+ * lpfc_sli4_post_scsi_sgl_list - Psot blocks of scsi buffer sgls from a list
  * @phba: pointer to lpfc hba data structure.
  * @post_sblist: pointer to the scsi buffer list.
  *
@@ -734,7 +739,7 @@ lpfc_sli4_post_scsi_sgl_list(struct lpfc_hba *phba,
 }
 
 /**
- * lpfc_sli4_repost_scsi_sgl_list - Repost all the allocated scsi buffer sgls
+ * lpfc_sli4_repost_scsi_sgl_list - Repsot all the allocated scsi buffer sgls
  * @phba: pointer to lpfc hba data structure.
  *
  * This routine walks the list of scsi buffers that have been allocated and
@@ -819,7 +824,7 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
 		 * for the struct fcp_cmnd, struct fcp_rsp and the number
 		 * of bde's necessary to support the sg_tablesize.
 		 */
-		psb->data = pci_pool_zalloc(phba->lpfc_scsi_dma_buf_pool,
+		psb->data = pci_pool_zalloc(phba->lpfc_sg_dma_buf_pool,
 						GFP_KERNEL, &psb->dma_handle);
 		if (!psb->data) {
 			kfree(psb);
@@ -832,7 +837,7 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
 		 */
 		if (phba->cfg_enable_bg  && (((unsigned long)(psb->data) &
 		    (unsigned long)(SLI4_PAGE_SIZE - 1)) != 0)) {
-			pci_pool_free(phba->lpfc_scsi_dma_buf_pool,
+			pci_pool_free(phba->lpfc_sg_dma_buf_pool,
 				      psb->data, psb->dma_handle);
 			kfree(psb);
 			break;
@@ -841,8 +846,8 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
 
 		lxri = lpfc_sli4_next_xritag(phba);
 		if (lxri == NO_XRI) {
-			pci_pool_free(phba->lpfc_scsi_dma_buf_pool,
-			      psb->data, psb->dma_handle);
+			pci_pool_free(phba->lpfc_sg_dma_buf_pool,
+				      psb->data, psb->dma_handle);
 			kfree(psb);
 			break;
 		}
@@ -850,11 +855,11 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
 		/* Allocate iotag for psb->cur_iocbq. */
 		iotag = lpfc_sli_next_iotag(phba, &psb->cur_iocbq);
 		if (iotag == 0) {
-			pci_pool_free(phba->lpfc_scsi_dma_buf_pool,
-				psb->data, psb->dma_handle);
+			pci_pool_free(phba->lpfc_sg_dma_buf_pool,
+				      psb->data, psb->dma_handle);
 			kfree(psb);
 			lpfc_printf_log(phba, KERN_ERR, LOG_FCP,
-					"3368 Failed to allocate IOTAG for"
+					"3368 Failed to allocated IOTAG for"
 					" XRI:0x%x\n", lxri);
 			lpfc_sli4_free_xri(phba, lxri);
 			break;
@@ -920,7 +925,7 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
 		phba->sli4_hba.scsi_xri_cnt++;
 		spin_unlock_irq(&phba->scsi_buf_list_get_lock);
 	}
-	lpfc_printf_log(phba, KERN_INFO, LOG_BG,
+	lpfc_printf_log(phba, KERN_INFO, LOG_BG | LOG_FCP,
 			"3021 Allocate %d out of %d requested new SCSI "
 			"buffers\n", bcnt, num_to_alloc);
 
@@ -946,7 +951,7 @@ lpfc_new_scsi_buf_s4(struct lpfc_vport *vport, int num_to_alloc)
  *   int - number of scsi buffers that were allocated.
  *   0 = failure, less than num_to_alloc is a partial failure.
  **/
-static inline int
+int
 lpfc_new_scsi_buf(struct lpfc_vport *vport, int num_to_alloc)
 {
 	return vport->phba->lpfc_new_scsi_buf(vport, num_to_alloc);
@@ -1045,7 +1050,7 @@ lpfc_get_scsi_buf_s4(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
  *   NULL - Error
  *   Pointer to lpfc_scsi_buf - Success
  **/
-static struct lpfc_scsi_buf*
+struct lpfc_scsi_buf*
 lpfc_get_scsi_buf(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp)
 {
 	return  phba->lpfc_get_scsi_buf(phba, ndlp);
@@ -1119,7 +1124,7 @@ lpfc_release_scsi_buf_s4(struct lpfc_hba *phba, struct lpfc_scsi_buf *psb)
  * This routine releases @psb scsi buffer by adding it to tail of @phba
  * lpfc_scsi_buf_list list.
  **/
-static void
+void
 lpfc_release_scsi_buf(struct lpfc_hba *phba, struct lpfc_scsi_buf *psb)
 {
 
@@ -1133,7 +1138,7 @@ lpfc_release_scsi_buf(struct lpfc_hba *phba, struct lpfc_scsi_buf *psb)
  *
  * This routine does the pci dma mapping for scatter-gather list of scsi cmnd
  * field of @lpfc_cmd for device with SLI-3 interface spec. This routine scans
- * through sg elements and format the bde. This routine also initializes all
+ * through sg elements and format the bdea. This routine also initializes all
  * IOCB fields which are dependent on scsi command request buffer.
  *
  * Return codes:
@@ -1266,16 +1271,13 @@ lpfc_scsi_prep_dma_buf_s3(struct lpfc_hba *phba, struct lpfc_scsi_buf *lpfc_cmd)
 
 #ifdef CONFIG_SCSI_LPFC_DEBUG_FS
 
-/* Return BG_ERR_INIT if error injection is detected by Initiator */
+/* Return if if error injection is detected by Initiator */
 #define BG_ERR_INIT	0x1
-/* Return BG_ERR_TGT if error injection is detected by Target */
+/* Return if if error injection is detected by Target */
 #define BG_ERR_TGT	0x2
-/* Return BG_ERR_SWAP if swapping CSUM<-->CRC is required for error injection */
+/* Return if if swapping CSUM<-->CRC is required for error injection */
 #define BG_ERR_SWAP	0x10
-/**
- * Return BG_ERR_CHECK if disabling Guard/Ref/App checking is required for
- * error injection
- **/
+/* Return if disabling Guard/Ref/App checking is required for error injection */
 #define BG_ERR_CHECK	0x20
 
 /**
@@ -3894,7 +3896,7 @@ int lpfc_sli4_scmd_to_wqidx_distr(struct lpfc_hba *phba,
 		}
 	}
 	chann = atomic_add_return(1, &phba->fcp_qidx);
-	chann = (chann % phba->cfg_fcp_io_channel);
+	chann = (chann % phba->cfg_fcp_max_hw_queue);
 	return chann;
 }
 
@@ -3925,6 +3927,8 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn,
 	struct Scsi_Host *shost;
 	uint32_t logit = LOG_FCP;
 
+	phba->fc4ScsiIoCmpls++;
+
 	/* Sanity check on return of outstanding command */
 	cmd = lpfc_cmd->pCmd;
 	if (!cmd)
@@ -3967,6 +3971,7 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn,
 		lpfc_cmd->prot_data_segment = NULL;
 	}
 #endif
+
 	if (pnode && NLP_CHK_NODE_ACT(pnode))
 		atomic_dec(&pnode->cmd_pending);
 
@@ -4241,19 +4246,19 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd,
 						vport->cfg_first_burst_size;
 			}
 			fcp_cmnd->fcpCntl3 = WRITE_DATA;
-			phba->fc4OutputRequests++;
+			phba->fc4ScsiOutputRequests++;
 		} else {
 			iocb_cmd->ulpCommand = CMD_FCP_IREAD64_CR;
 			iocb_cmd->ulpPU = PARM_READ_CHECK;
 			fcp_cmnd->fcpCntl3 = READ_DATA;
-			phba->fc4InputRequests++;
+			phba->fc4ScsiInputRequests++;
 		}
 	} else {
 		iocb_cmd->ulpCommand = CMD_FCP_ICMND64_CR;
 		iocb_cmd->un.fcpi.fcpi_parm = 0;
 		iocb_cmd->ulpPU = 0;
 		fcp_cmnd->fcpCntl3 = 0;
-		phba->fc4ControlRequests++;
+		phba->fc4ScsiControlRequests++;
 	}
 	if (phba->sli_rev == 3 &&
 	    !(phba->sli3_options & LPFC_SLI3_BG_ENABLED))
@@ -4467,7 +4472,7 @@ static __inline__ void lpfc_poll_rearm_timer(struct lpfc_hba * phba)
 	unsigned long  poll_tmo_expires =
 		(jiffies + msecs_to_jiffies(phba->cfg_poll_tmo));
 
-	if (!list_empty(&phba->sli.ring[LPFC_FCP_RING].txcmplq))
+	if (!list_empty(&phba->sli.sli3_ring[LPFC_FCP_RING].txcmplq))
 		mod_timer(&phba->fcp_poll_timer,
 			  poll_tmo_expires);
 }
@@ -4497,7 +4502,7 @@ void lpfc_poll_timeout(unsigned long ptr)
 
 	if (phba->cfg_poll & ENABLE_FCP_RING_POLLING) {
 		lpfc_sli_handle_fast_ring_event(phba,
-			&phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ);
+			&phba->sli.sli3_ring[LPFC_FCP_RING], HA_R0RE_REQ);
 
 		if (phba->cfg_poll & DISABLE_FCP_RING_INT)
 			lpfc_poll_rearm_timer(phba);
@@ -4561,7 +4566,7 @@ lpfc_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *cmnd)
 	if (lpfc_cmd == NULL) {
 		lpfc_rampdown_queue_depth(phba);
 
-		lpfc_printf_vlog(vport, KERN_INFO, LOG_MISC,
+		lpfc_printf_vlog(vport, KERN_INFO, LOG_FCP_ERROR,
 				 "0707 driver's buffer pool is empty, "
 				 "IO busied\n");
 		goto out_host_busy;
@@ -4634,9 +4639,10 @@ lpfc_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *cmnd)
 
 		goto out_host_busy_free_buf;
 	}
+
 	if (phba->cfg_poll & ENABLE_FCP_RING_POLLING) {
 		lpfc_sli_handle_fast_ring_event(phba,
-			&phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ);
+			&phba->sli.sli3_ring[LPFC_FCP_RING], HA_R0RE_REQ);
 
 		if (phba->cfg_poll & DISABLE_FCP_RING_INT)
 			lpfc_poll_rearm_timer(phba);
@@ -4681,7 +4687,7 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 	IOCB_t *cmd, *icmd;
 	int ret = SUCCESS, status = 0;
 	struct lpfc_sli_ring *pring_s4;
-	int ring_number, ret_val;
+	int ret_val;
 	unsigned long flags, iflags;
 	DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waitq);
 
@@ -4769,7 +4775,7 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 	icmd->ulpClass = cmd->ulpClass;
 
 	/* ABTS WQE must go to the same WQ as the WQE to be aborted */
-	abtsiocb->fcp_wqidx = iocb->fcp_wqidx;
+	abtsiocb->hba_wqidx = iocb->hba_wqidx;
 	abtsiocb->iocb_flag |= LPFC_USE_FCPWQIDX;
 	if (iocb->iocb_flag & LPFC_IO_FOF)
 		abtsiocb->iocb_flag |= LPFC_IO_FOF;
@@ -4782,8 +4788,11 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 	abtsiocb->iocb_cmpl = lpfc_sli_abort_fcp_cmpl;
 	abtsiocb->vport = vport;
 	if (phba->sli_rev == LPFC_SLI_REV4) {
-		ring_number = MAX_SLI3_CONFIGURED_RINGS + iocb->fcp_wqidx;
-		pring_s4 = &phba->sli.ring[ring_number];
+		pring_s4 = lpfc_sli4_calc_ring(phba, iocb);
+		if (pring_s4 == NULL) {
+			ret = FAILED;
+			goto out_unlock;
+		}
 		/* Note: both hbalock and ring_lock must be set here */
 		spin_lock_irqsave(&pring_s4->ring_lock, iflags);
 		ret_val = __lpfc_sli_issue_iocb(phba, pring_s4->ringno,
@@ -4805,7 +4814,7 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 
 	if (phba->cfg_poll & DISABLE_FCP_RING_INT)
 		lpfc_sli_handle_fast_ring_event(phba,
-			&phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ);
+			&phba->sli.sli3_ring[LPFC_FCP_RING], HA_R0RE_REQ);
 
 wait_for_cmpl:
 	lpfc_cmd->waitq = &waitq;
@@ -4959,11 +4968,11 @@ lpfc_send_taskmgmt(struct lpfc_vport *vport, struct scsi_cmnd *cmnd,
 	int status;
 
 	rdata = lpfc_rport_data_from_scsi_device(cmnd->device);
-	if (!rdata || !rdata->pnode || !NLP_CHK_NODE_ACT(rdata->pnode))
-		return FAILED;
 	pnode = rdata->pnode;
+	if (!pnode || !NLP_CHK_NODE_ACT(pnode))
+		return FAILED;
 
-	lpfc_cmd = lpfc_get_scsi_buf(phba, pnode);
+	lpfc_cmd = lpfc_get_scsi_buf(phba, rdata->pnode);
 	if (lpfc_cmd == NULL)
 		return FAILED;
 	lpfc_cmd->timeout = phba->cfg_task_mgmt_tmo;
@@ -4997,7 +5006,7 @@ lpfc_send_taskmgmt(struct lpfc_vport *vport, struct scsi_cmnd *cmnd,
 	if ((status != IOCB_SUCCESS) ||
 	    (iocbqrsp->iocb.ulpStatus != IOSTAT_SUCCESS)) {
 		if (status != IOCB_SUCCESS ||
-		    iocbqrsp->iocb.ulpStatus != IOSTAT_FCP_RSP_ERROR)
+			iocbqrsp->iocb.ulpStatus != IOSTAT_FCP_RSP_ERROR)
 			lpfc_printf_vlog(vport, KERN_ERR, LOG_FCP,
 					 "0727 TMF %s to TGT %d LUN %llu "
 					 "failed (%d, %d) iocb_flag x%x\n",
@@ -5105,7 +5114,7 @@ lpfc_reset_flush_io_context(struct lpfc_vport *vport, uint16_t tgt_id,
 	cnt = lpfc_sli_sum_iocb(vport, tgt_id, lun_id, context);
 	if (cnt)
 		lpfc_sli_abort_taskmgmt(vport,
-					&phba->sli.ring[phba->sli.fcp_ring],
+					&phba->sli.sli3_ring[LPFC_FCP_RING],
 					tgt_id, lun_id, context);
 	later = msecs_to_jiffies(2 * vport->cfg_devloss_tmo * 1000) + jiffies;
 	while (time_after(later, jiffies) && cnt) {
@@ -5531,10 +5540,9 @@ lpfc_slave_configure(struct scsi_device *sdev)
 	struct lpfc_hba   *phba = vport->phba;
 
 	scsi_change_queue_depth(sdev, vport->cfg_lun_queue_depth);
-
 	if (phba->cfg_poll & ENABLE_FCP_RING_POLLING) {
 		lpfc_sli_handle_fast_ring_event(phba,
-			&phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ);
+			&phba->sli.sli3_ring[LPFC_FCP_RING], HA_R0RE_REQ);
 		if (phba->cfg_poll & DISABLE_FCP_RING_INT)
 			lpfc_poll_rearm_timer(phba);
 	}
@@ -5898,6 +5906,48 @@ lpfc_disable_oas_lun(struct lpfc_hba *phba, struct lpfc_name *vport_wwpn,
 	return false;
 }
 
+static int
+lpfc_no_command(struct Scsi_Host *shost, struct scsi_cmnd *cmnd)
+{
+	return SCSI_MLQUEUE_HOST_BUSY;
+}
+
+static int
+lpfc_no_handler(struct scsi_cmnd *cmnd)
+{
+	return FAILED;
+}
+
+static int
+lpfc_no_slave(struct scsi_device *sdev)
+{
+	return -ENODEV;
+}
+
+struct scsi_host_template lpfc_template_nvme = {
+	.module			= THIS_MODULE,
+	.name			= LPFC_DRIVER_NAME,
+	.proc_name		= LPFC_DRIVER_NAME,
+	.info			= lpfc_info,
+	.queuecommand		= lpfc_no_command,
+	.eh_abort_handler	= lpfc_no_handler,
+	.eh_device_reset_handler = lpfc_no_handler,
+	.eh_target_reset_handler = lpfc_no_handler,
+	.eh_bus_reset_handler	= lpfc_no_handler,
+	.eh_host_reset_handler  = lpfc_no_handler,
+	.slave_alloc		= lpfc_no_slave,
+	.slave_configure	= lpfc_no_slave,
+	.scan_finished		= lpfc_scan_finished,
+	.this_id		= -1,
+	.sg_tablesize		= 1,
+	.cmd_per_lun		= 1,
+	.use_clustering		= ENABLE_CLUSTERING,
+	.shost_attrs		= lpfc_hba_attrs,
+	.max_sectors		= 0xFFFF,
+	.vendor_id		= LPFC_NL_VENDOR_ID,
+	.track_queue_depth	= 0,
+};
+
 struct scsi_host_template lpfc_template_s3 = {
 	.module			= THIS_MODULE,
 	.name			= LPFC_DRIVER_NAME,
diff --git a/drivers/scsi/lpfc/lpfc_scsi.h b/drivers/scsi/lpfc/lpfc_scsi.h
index 8cb80da..def0c0a5 100644
--- a/drivers/scsi/lpfc/lpfc_scsi.h
+++ b/drivers/scsi/lpfc/lpfc_scsi.h
@@ -135,6 +135,8 @@ struct lpfc_scsi_buf {
 
 	uint32_t timeout;
 
+	uint16_t flags;  /* TBD convert exch_busy to flags */
+#define LPFC_SBUF_XBUSY         0x1     /* SLI4 hba reported XB on WCQE cmpl */
 	uint16_t exch_busy;     /* SLI4 hba reported XB on complete WCQE */
 	uint16_t status;	/* From IOCB Word 7- ulpStatus */
 	uint32_t result;	/* From IOCB Word 4. */
@@ -164,6 +166,8 @@ struct lpfc_scsi_buf {
 	 * Iotag is in here
 	 */
 	struct lpfc_iocbq cur_iocbq;
+	uint16_t cpu;
+
 	wait_queue_head_t *waitq;
 	unsigned long start_time;
 
@@ -178,13 +182,15 @@ struct lpfc_scsi_buf {
 #endif
 };
 
-#define LPFC_SCSI_DMA_EXT_SIZE 264
-#define LPFC_BPL_SIZE          1024
-#define MDAC_DIRECT_CMD                  0x22
+#define LPFC_SCSI_DMA_EXT_SIZE	264
+#define LPFC_BPL_SIZE		1024
+#define MDAC_DIRECT_CMD		0x22
+
+#define FIND_FIRST_OAS_LUN	0
+#define NO_MORE_OAS_LUN		-1
+#define NOT_OAS_ENABLED_LUN	NO_MORE_OAS_LUN
 
-#define FIND_FIRST_OAS_LUN		 0
-#define NO_MORE_OAS_LUN			-1
-#define NOT_OAS_ENABLED_LUN		NO_MORE_OAS_LUN
+#define TXRDY_PAYLOAD_LEN	12
 
 int lpfc_sli4_scmd_to_wqidx_distr(struct lpfc_hba *phba,
 				  struct lpfc_scsi_buf *lpfc_cmd);
diff --git a/drivers/scsi/lpfc/lpfc_sli.h b/drivers/scsi/lpfc/lpfc_sli.h
index 74227a2..acb431f 100644
--- a/drivers/scsi/lpfc/lpfc_sli.h
+++ b/drivers/scsi/lpfc/lpfc_sli.h
@@ -54,9 +54,16 @@ struct lpfc_iocbq {
 	uint16_t iotag;         /* pre-assigned IO tag */
 	uint16_t sli4_lxritag;  /* logical pre-assigned XRI. */
 	uint16_t sli4_xritag;   /* pre-assigned XRI, (OXID) tag. */
+	uint16_t hba_wqidx;     /* index to HBA work queue */
 	struct lpfc_cq_event cq_event;
+	struct lpfc_wcqe_complete wcqe_cmpl;	/* WQE cmpl */
+	uint64_t isr_timestamp;
 
-	IOCB_t iocb;		/* IOCB cmd */
+	/* Be careful here */
+	union lpfc_wqe wqe;	/* WQE cmd */
+	IOCB_t iocb;		/* For IOCB cmd or if we want 128 byte WQE */
+
+	uint8_t rsvd2;
 	uint8_t priority;	/* OAS priority */
 	uint8_t retry;		/* retry counter for IOCB cmd - if needed */
 	uint32_t iocb_flag;
@@ -82,9 +89,12 @@ struct lpfc_iocbq {
 #define LPFC_IO_OAS		0x10000 /* OAS FCP IO */
 #define LPFC_IO_FOF		0x20000 /* FOF FCP IO */
 #define LPFC_IO_LOOPBACK	0x40000 /* Loopback IO */
+#define LPFC_PRLI_NVME_REQ	0x80000 /* This is an NVME PRLI. */
+#define LPFC_PRLI_FCP_REQ	0x100000 /* This is an NVME PRLI. */
+#define LPFC_IO_NVME	        0x200000 /* NVME FCP command */
+#define LPFC_IO_NVME_LS		0x400000 /* NVME LS command */
 
 	uint32_t drvrTimeout;	/* driver timeout in seconds */
-	uint32_t fcp_wqidx;	/* index to FCP work queue */
 	struct lpfc_vport *vport;/* virtual port pointer */
 	void *context1;		/* caller context information */
 	void *context2;		/* caller context information */
@@ -97,12 +107,14 @@ struct lpfc_iocbq {
 		struct lpfc_node_rrq *rrq;
 	} context_un;
 
-	void (*fabric_iocb_cmpl) (struct lpfc_hba *, struct lpfc_iocbq *,
+	void (*fabric_iocb_cmpl)(struct lpfc_hba *, struct lpfc_iocbq *,
 			   struct lpfc_iocbq *);
-	void (*wait_iocb_cmpl) (struct lpfc_hba *, struct lpfc_iocbq *,
+	void (*wait_iocb_cmpl)(struct lpfc_hba *, struct lpfc_iocbq *,
 			   struct lpfc_iocbq *);
-	void (*iocb_cmpl) (struct lpfc_hba *, struct lpfc_iocbq *,
+	void (*iocb_cmpl)(struct lpfc_hba *, struct lpfc_iocbq *,
 			   struct lpfc_iocbq *);
+	void (*wqe_cmpl)(struct lpfc_hba *, struct lpfc_iocbq *,
+			  struct lpfc_wcqe_complete *);
 };
 
 #define SLI_IOCB_RET_IOCB      1	/* Return IOCB if cmd ring full */
@@ -112,6 +124,14 @@ struct lpfc_iocbq {
 #define IOCB_ERROR          2
 #define IOCB_TIMEDOUT       3
 
+#define SLI_WQE_RET_WQE    1    /* Return WQE if cmd ring full */
+
+#define WQE_SUCCESS        0
+#define WQE_BUSY           1
+#define WQE_ERROR          2
+#define WQE_TIMEDOUT       3
+#define WQE_ABORTED        4
+
 #define LPFC_MBX_WAKE		1
 #define LPFC_MBX_IMED_UNREG	2
 
@@ -297,12 +317,9 @@ struct lpfc_sli {
 #define LPFC_BLOCK_MGMT_IO        0x800	/* Don't allow mgmt mbx or iocb cmds */
 #define LPFC_MENLO_MAINT          0x1000 /* need for menl fw download */
 #define LPFC_SLI_ASYNC_MBX_BLK    0x2000 /* Async mailbox is blocked */
+#define LPFC_SLI_SUPPRESS_RSP     0x4000 /* Suppress RSP feature is supported */
 
-	struct lpfc_sli_ring *ring;
-	int fcp_ring;		/* ring used for FCP initiator commands */
-	int next_ring;
-
-	int extra_ring;		/* extra ring used for other protocols */
+	struct lpfc_sli_ring *sli3_ring;
 
 	struct lpfc_sli_stat slistat;	/* SLI statistical info */
 	struct list_head mboxq;
diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h
index 0b88b570..0a56fc6 100644
--- a/drivers/scsi/lpfc/lpfc_sli4.h
+++ b/drivers/scsi/lpfc/lpfc_sli4.h
@@ -35,9 +35,10 @@
 #define LPFC_NEMBED_MBOX_SGL_CNT		254
 
 /* Multi-queue arrangement for FCP EQ/CQ/WQ tuples */
-#define LPFC_FCP_IO_CHAN_DEF       4
-#define LPFC_FCP_IO_CHAN_MIN       1
-#define LPFC_FCP_IO_CHAN_MAX       16
+#define LPFC_HBA_IO_CHAN_MIN	0
+#define LPFC_HBA_IO_CHAN_MAX	32
+#define LPFC_FCP_IO_CHAN_DEF	4
+#define LPFC_NVME_IO_CHAN_DEF	0
 
 /* Number of channels used for Flash Optimized Fabric (FOF) operations */
 
@@ -107,6 +108,8 @@ enum lpfc_sli4_queue_subtype {
 	LPFC_MBOX,
 	LPFC_FCP,
 	LPFC_ELS,
+	LPFC_NVME,
+	LPFC_NVME_LS,
 	LPFC_USOL
 };
 
@@ -125,25 +128,41 @@ union sli4_qe {
 	struct lpfc_rqe *rqe;
 };
 
+/* RQ buffer list */
+struct lpfc_rqb {
+	uint16_t entry_count;	  /* Current number of RQ slots */
+	uint16_t buffer_count;	  /* Current number of buffers posted */
+	struct list_head rqb_buffer_list;  /* buffers assigned to this HBQ */
+				  /* Callback for HBQ buffer allocation */
+	struct rqb_dmabuf *(*rqb_alloc_buffer)(struct lpfc_hba *);
+				  /* Callback for HBQ buffer free */
+	void               (*rqb_free_buffer)(struct lpfc_hba *,
+					       struct rqb_dmabuf *);
+};
+
 struct lpfc_queue {
 	struct list_head list;
+	struct list_head wq_list;
 	enum lpfc_sli4_queue_type type;
 	enum lpfc_sli4_queue_subtype subtype;
 	struct lpfc_hba *phba;
 	struct list_head child_list;
+	struct list_head page_list;
+	struct list_head sgl_list;
 	uint32_t entry_count;	/* Number of entries to support on the queue */
 	uint32_t entry_size;	/* Size of each queue entry. */
 	uint32_t entry_repost;	/* Count of entries before doorbell is rung */
 #define LPFC_QUEUE_MIN_REPOST	8
 	uint32_t queue_id;	/* Queue ID assigned by the hardware */
 	uint32_t assoc_qid;     /* Queue ID associated with, for CQ/WQ/MQ */
-	struct list_head page_list;
 	uint32_t page_count;	/* Number of pages allocated for this queue */
 	uint32_t host_index;	/* The host's index for putting or getting */
 	uint32_t hba_index;	/* The last known hba index for get or put */
 
 	struct lpfc_sli_ring *pring; /* ptr to io ring associated with q */
+	struct lpfc_rqb *rqbp;	/* ptr to RQ buffers */
 
+	uint16_t sgl_list_cnt;
 	uint16_t db_format;
 #define LPFC_DB_RING_FORMAT	0x01
 #define LPFC_DB_LIST_FORMAT	0x02
@@ -176,6 +195,8 @@ struct lpfc_queue {
 #define	RQ_buf_trunc		q_cnt_3
 #define	RQ_rcv_buf		q_cnt_4
 
+	uint64_t isr_timestamp;
+	struct lpfc_queue *assoc_qp;
 	union sli4_qe qe[1];	/* array to index entries (must be last) */
 };
 
@@ -338,6 +359,7 @@ struct lpfc_bmbx {
 #define LPFC_CQE_DEF_COUNT      1024
 #define LPFC_WQE_DEF_COUNT      256
 #define LPFC_WQE128_DEF_COUNT   128
+#define LPFC_WQE128_MAX_COUNT   256
 #define LPFC_MQE_DEF_COUNT      16
 #define LPFC_RQE_DEF_COUNT	512
 
@@ -379,10 +401,14 @@ struct lpfc_max_cfg_param {
 
 struct lpfc_hba;
 /* SLI4 HBA multi-fcp queue handler struct */
-struct lpfc_fcp_eq_hdl {
+struct lpfc_hba_eq_hdl {
 	uint32_t idx;
 	struct lpfc_hba *phba;
-	atomic_t fcp_eq_in_use;
+	atomic_t hba_eq_in_use;
+	struct cpumask *cpumask;
+	/* CPU affinitsed to or 0xffffffff if multiple */
+	uint32_t cpu;
+#define LPFC_MULTI_CPU_AFFINITY 0xffffffff
 };
 
 /* Port Capabilities for SLI4 Parameters */
@@ -427,6 +453,7 @@ struct lpfc_pc_sli4_params {
 	uint8_t wqsize;
 #define LPFC_WQ_SZ64_SUPPORT	1
 #define LPFC_WQ_SZ128_SUPPORT	2
+	uint8_t wqpcnt;
 };
 
 struct lpfc_iov {
@@ -445,7 +472,7 @@ struct lpfc_sli4_lnk_info {
 	uint8_t optic_state;
 };
 
-#define LPFC_SLI4_HANDLER_CNT		(LPFC_FCP_IO_CHAN_MAX+ \
+#define LPFC_SLI4_HANDLER_CNT		(LPFC_HBA_IO_CHAN_MAX+ \
 					 LPFC_FOF_IO_CHAN_NUM)
 #define LPFC_SLI4_HANDLER_NAME_SZ	16
 
@@ -517,21 +544,30 @@ struct lpfc_sli4_hba {
 	struct lpfc_pc_sli4_params pc_sli4_params;
 	struct msix_entry *msix_entries;
 	uint8_t handler_name[LPFC_SLI4_HANDLER_CNT][LPFC_SLI4_HANDLER_NAME_SZ];
-	struct lpfc_fcp_eq_hdl *fcp_eq_hdl; /* FCP per-WQ handle */
+	struct lpfc_hba_eq_hdl *hba_eq_hdl; /* HBA per-WQ handle */
 
 	/* Pointers to the constructed SLI4 queues */
-	struct lpfc_queue **hba_eq;/* Event queues for HBA */
-	struct lpfc_queue **fcp_cq;/* Fast-path FCP compl queue */
-	struct lpfc_queue **fcp_wq;/* Fast-path FCP work queue */
+	struct lpfc_queue **hba_eq;  /* Event queues for HBA */
+	struct lpfc_queue **fcp_cq;  /* Fast-path FCP compl queue */
+	struct lpfc_queue **nvme_cq; /* Fast-path NVME compl queue */
+	struct lpfc_queue **fcp_wq;  /* Fast-path FCP work queue */
+	struct lpfc_queue **nvme_wq; /* Fast-path NVME work queue */
 	uint16_t *fcp_cq_map;
+	uint16_t *nvme_cq_map;
+	struct list_head lpfc_wq_list;
 
 	struct lpfc_queue *mbx_cq; /* Slow-path mailbox complete queue */
 	struct lpfc_queue *els_cq; /* Slow-path ELS response complete queue */
+	struct lpfc_queue *nvmels_cq; /* NVME LS complete queue */
 	struct lpfc_queue *mbx_wq; /* Slow-path MBOX work queue */
 	struct lpfc_queue *els_wq; /* Slow-path ELS work queue */
+	struct lpfc_queue *nvmels_wq; /* NVME LS work queue */
 	struct lpfc_queue *hdr_rq; /* Slow-path Header Receive queue */
 	struct lpfc_queue *dat_rq; /* Slow-path Data Receive queue */
 
+	struct lpfc_name wwnn;
+	struct lpfc_name wwpn;
+
 	uint32_t fw_func_mode;	/* FW function protocol mode */
 	uint32_t ulp0_mode;	/* ULP0 protocol mode */
 	uint32_t ulp1_mode;	/* ULP1 protocol mode */
@@ -568,14 +604,17 @@ struct lpfc_sli4_hba {
 	uint16_t rpi_hdrs_in_use; /* must post rpi hdrs if set. */
 	uint16_t next_xri; /* last_xri - max_cfg_param.xri_base = used */
 	uint16_t next_rpi;
+	uint16_t nvme_xri_max;
+	uint16_t nvme_xri_cnt;
+	uint16_t nvme_xri_start;
 	uint16_t scsi_xri_max;
 	uint16_t scsi_xri_cnt;
-	uint16_t els_xri_cnt;
 	uint16_t scsi_xri_start;
-	struct list_head lpfc_free_sgl_list;
-	struct list_head lpfc_sgl_list;
+	uint16_t els_xri_cnt;
+	struct list_head lpfc_els_sgl_list;
 	struct list_head lpfc_abts_els_sgl_list;
 	struct list_head lpfc_abts_scsi_buf_list;
+	struct list_head lpfc_abts_nvme_buf_list;
 	struct lpfc_sglq **lpfc_sglq_active_list;
 	struct list_head lpfc_rpi_hdr_list;
 	unsigned long *rpi_bmask;
@@ -602,8 +641,9 @@ struct lpfc_sli4_hba {
 #define LPFC_SLI4_PPNAME_NON	0
 #define LPFC_SLI4_PPNAME_GET	1
 	struct lpfc_iov iov;
+	spinlock_t abts_nvme_buf_list_lock; /* list of aborted SCSI IOs */
 	spinlock_t abts_scsi_buf_list_lock; /* list of aborted SCSI IOs */
-	spinlock_t abts_sgl_list_lock; /* list of aborted els IOs */
+	spinlock_t sgl_list_lock; /* list of aborted els IOs */
 	uint32_t physical_port;
 
 	/* CPU to vector mapping information */
@@ -615,7 +655,7 @@ struct lpfc_sli4_hba {
 
 enum lpfc_sge_type {
 	GEN_BUFF_TYPE,
-	SCSI_BUFF_TYPE
+	SCSI_BUFF_TYPE,
 };
 
 enum lpfc_sgl_state {
@@ -694,7 +734,7 @@ struct lpfc_queue *lpfc_sli4_queue_alloc(struct lpfc_hba *, uint32_t,
 			uint32_t);
 void lpfc_sli4_queue_free(struct lpfc_queue *);
 int lpfc_eq_create(struct lpfc_hba *, struct lpfc_queue *, uint32_t);
-int lpfc_modify_fcp_eq_delay(struct lpfc_hba *, uint32_t);
+int lpfc_modify_hba_eq_delay(struct lpfc_hba *phba, uint32_t startq);
 int lpfc_cq_create(struct lpfc_hba *, struct lpfc_queue *,
 			struct lpfc_queue *, uint32_t, uint32_t);
 int32_t lpfc_mq_create(struct lpfc_hba *, struct lpfc_queue *,
@@ -746,6 +786,7 @@ int lpfc_sli4_brdreset(struct lpfc_hba *);
 int lpfc_sli4_add_fcf_record(struct lpfc_hba *, struct fcf_record *);
 void lpfc_sli_remove_dflt_fcf(struct lpfc_hba *);
 int lpfc_sli4_get_els_iocb_cnt(struct lpfc_hba *);
+int lpfc_sli4_get_iocb_cnt(struct lpfc_hba *phba);
 int lpfc_sli4_init_vpi(struct lpfc_vport *);
 uint32_t lpfc_sli4_cq_release(struct lpfc_queue *, bool);
 uint32_t lpfc_sli4_eq_release(struct lpfc_queue *, bool);
diff --git a/drivers/scsi/lpfc/lpfc_version.h b/drivers/scsi/lpfc/lpfc_version.h
index 0ee0623..31328a7 100644
--- a/drivers/scsi/lpfc/lpfc_version.h
+++ b/drivers/scsi/lpfc/lpfc_version.h
@@ -20,6 +20,7 @@
 
 #define LPFC_DRIVER_VERSION "11.2.0.4"
 #define LPFC_DRIVER_NAME		"lpfc"
+#define LPFCMGMT_NAME			"lpfcmgmt"
 
 /* Used for SLI 2/3 */
 #define LPFC_SP_DRIVER_HANDLER_NAME	"lpfc:sp"
diff --git a/drivers/scsi/lpfc/lpfc_vport.c b/drivers/scsi/lpfc/lpfc_vport.c
index e18bbc6..07c7c5f 100644
--- a/drivers/scsi/lpfc/lpfc_vport.c
+++ b/drivers/scsi/lpfc/lpfc_vport.c
@@ -33,6 +33,7 @@
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_transport_fc.h>
+
 #include "lpfc_hw4.h"
 #include "lpfc_hw.h"
 #include "lpfc_sli.h"
@@ -402,6 +403,8 @@ lpfc_vport_create(struct fc_vport *fc_vport, bool disable)
 		vport->fdmi_port_mask = phba->pport->fdmi_port_mask;
 	}
 
+	/* todo: init: register port with nvme */
+
 	/*
 	 * In SLI4, the vpi must be activated before it can be used
 	 * by the port.
-- 
2.5.0




More information about the Linux-nvme mailing list