scsi: lpfc: NVME Initiator: Base modifications

NVME Initiator: Base modifications

This patch adds base modifications for NVME initiator support.

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.
  - Pools and config parameters are separated per-protocol
  - 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.

I apologize for the size of the patch.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>

----
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

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 +756,9 @@ struct lpfc_hba {
 	uint32_t cfg_fcp_imax;
 	uint32_t cfg_fcp_cpu_map;
 	uint32_t cfg_fcp_io_channel;
+	uint32_t cfg_nvme_oas;
+	uint32_t cfg_nvme_io_channel;
+	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 +804,12 @@ 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_irqs;	/* number of irqs for io channels */
 	lpfc_vpd_t vpd;		/* vital product data */
 
 	struct pci_dev *pcidev;
@@ -784,11 +824,11 @@ 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 +883,17 @@ 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 +904,23 @@ 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;
@@ -1092,3 +1146,11 @@ lpfc_sli_read_hs(struct lpfc_hba *phba)
 
 	return 0;
 }
+
+static inline struct lpfc_sli_ring *
+lpfc_phba_elsring(struct lpfc_hba *phba)
+{
+	if (phba->sli_rev == LPFC_SLI_REV4)
+		return phba->sli4_hba.els_wq->pring;
+	return &phba->sli.sli3_ring[LPFC_ELS_RING];
+}

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;
 }
@@ -2875,8 +2881,7 @@ diag_cmd_data_alloc(struct lpfc_hba *phba,
 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 +2895,8 @@ static int lpfcdiag_loop_post_rxbufs(struct lpfc_hba *phba, uint16_t rxxri,
 	int iocb_stat;
 	int i = 0;
 
+	pring = lpfc_phba_elsring(phba);
+
 	cmdiocbq = lpfc_sli_get_iocbq(phba);
 	rxbmp = kmalloc(sizeof(struct lpfc_dmabuf), GFP_KERNEL);
 	if (rxbmp != NULL) {
@@ -5403,13 +5410,15 @@ 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_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;
 
+	pring = lpfc_phba_elsring(phba);
+
 	/* 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

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 +301,9 @@ 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);
 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 *);
@@ -356,6 +373,7 @@ 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;
@@ -471,7 +489,9 @@ 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 +516,6 @@ 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);
+
+/* NVME interfaces. */
+void lpfc_nvme_mod_param_dep(struct lpfc_hba *phba);

drivers/scsi/lpfc/lpfc_debugfs.c

@@ -484,20 +484,23 @@ lpfc_debugfs_dumpHostSlim_data(struct lpfc_hba *phba, char *buf, int size)
 		off += (8 * sizeof(uint32_t));
 	}
 
+	if (phba->sli_rev <= LPFC_SLI_REV3) {
 		for (i = 0; i < 4; i++) {
 			pgpp = &phba->port_gp[i];
-		pring = &psli->ring[i];
+			pring = &psli->sli3_ring[i];
 			len +=  snprintf(buf+len, size-len,
-				 "Ring %d: CMD GetInx:%d (Max:%d Next:%d "
-				 "Local:%d flg:x%x)  RSP PutInx:%d Max:%d\n",
-				 i, pgpp->cmdGetInx, pring->sli.sli3.numCiocb,
+					 "Ring %d: CMD GetInx:%d "
+					 "(Max:%d Next:%d "
+					 "Local:%d flg:x%x)  "
+					 "RSP PutInx:%d Max:%d\n",
+					 i, pgpp->cmdGetInx,
+					 pring->sli.sli3.numCiocb,
 					 pring->sli.sli3.next_cmdidx,
 					 pring->sli.sli3.local_getidx,
 					 pring->flag, pgpp->rspPutInx,
 					 pring->sli.sli3.numRiocb);
 		}
 
-	if (phba->sli_rev <= LPFC_SLI_REV3) {
 		word0 = readl(phba->HAregaddr);
 		word1 = readl(phba->CAregaddr);
 		word2 = readl(phba->HSregaddr);
@@ -535,6 +538,7 @@ lpfc_debugfs_nodelist_data(struct lpfc_vport *vport, char *buf, int size)
 
 	cnt = (LPFC_NODELIST_SIZE / LPFC_NODELIST_ENTRY_SIZE);
 
+	len += snprintf(buf+len, size-len, "\nFCP Nodelist Entries ...\n");
 	spin_lock_irq(shost->host_lock);
 	list_for_each_entry(ndlp, &vport->fc_nodes, nlp_listp) {
 		if (!cnt) {
@@ -2011,6 +2015,14 @@ lpfc_idiag_wqs_for_cq(struct lpfc_hba *phba, char *wqtype, char *pbuffer,
 		if (*len >= max_cnt)
 			return 1;
 	}
+	for (qidx = 0; qidx < phba->cfg_nvme_io_channel; qidx++) {
+		qp = phba->sli4_hba.nvme_wq[qidx];
+		if (qp->assoc_qid != cq_id)
+			continue;
+		*len = __lpfc_idiag_print_wq(qp, wqtype, pbuffer, *len);
+		if (*len >= max_cnt)
+			return 1;
+	}
 	return 0;
 }
 
@@ -2096,6 +2108,25 @@ lpfc_idiag_cqs_for_eq(struct lpfc_hba *phba, char *pbuffer,
 			return 1;
 	}
 
+	for (qidx = 0; qidx < phba->cfg_nvme_io_channel; qidx++) {
+		qp = phba->sli4_hba.nvme_cq[qidx];
+		if (qp->assoc_qid != eq_id)
+			continue;
+
+		*len = __lpfc_idiag_print_cq(qp, "NVME", pbuffer, *len);
+
+		/* Reset max counter */
+		qp->CQ_max_cqe = 0;
+
+		if (*len >= max_cnt)
+			return 1;
+
+		rc = lpfc_idiag_wqs_for_cq(phba, "NVME", pbuffer, len,
+				max_cnt, qp->queue_id);
+		if (rc)
+			return 1;
+	}
+
 	return 0;
 }
 
@@ -2162,21 +2193,21 @@ lpfc_idiag_queinfo_read(struct file *file, char __user *buf, size_t nbytes,
 	spin_lock_irq(&phba->hbalock);
 
 	/* Fast-path event queue */
-	if (phba->sli4_hba.hba_eq && phba->cfg_fcp_io_channel) {
+	if (phba->sli4_hba.hba_eq && phba->io_channel_irqs) {
 
 		x = phba->lpfc_idiag_last_eq;
-		if (phba->cfg_fof && (x >= phba->cfg_fcp_io_channel)) {
+		if (phba->cfg_fof && (x >= phba->io_channel_irqs)) {
 			phba->lpfc_idiag_last_eq = 0;
 			goto fof;
 		}
 		phba->lpfc_idiag_last_eq++;
-		if (phba->lpfc_idiag_last_eq >= phba->cfg_fcp_io_channel)
+		if (phba->lpfc_idiag_last_eq >= phba->io_channel_irqs)
 			if (phba->cfg_fof == 0)
 				phba->lpfc_idiag_last_eq = 0;
 
 		len += snprintf(pbuffer + len, LPFC_QUE_INFO_GET_BUF_SIZE - len,
 					"EQ %d out of %d HBA EQs\n",
-					x, phba->cfg_fcp_io_channel);
+					x, phba->io_channel_irqs);
 
 		/* Fast-path EQ */
 		qp = phba->sli4_hba.hba_eq[x];
@@ -2191,6 +2222,7 @@ lpfc_idiag_queinfo_read(struct file *file, char __user *buf, size_t nbytes,
 		if (len >= max_cnt)
 			goto too_big;
 
+		/* will dump both fcp and nvme cqs/wqs for the eq */
 		rc = lpfc_idiag_cqs_for_eq(phba, pbuffer, &len,
 			max_cnt, qp->queue_id);
 		if (rc)
@@ -2227,6 +2259,23 @@ lpfc_idiag_queinfo_read(struct file *file, char __user *buf, size_t nbytes,
 		if (len >= max_cnt)
 			goto too_big;
 
+		/* Slow-path NVME LS response CQ */
+		qp = phba->sli4_hba.nvmels_cq;
+		len = __lpfc_idiag_print_cq(qp, "NVME LS",
+						pbuffer, len);
+		/* Reset max counter */
+		if (qp)
+			qp->CQ_max_cqe = 0;
+		if (len >= max_cnt)
+			goto too_big;
+
+		/* Slow-path NVME LS WQ */
+		qp = phba->sli4_hba.nvmels_wq;
+		len = __lpfc_idiag_print_wq(qp, "NVME LS",
+						pbuffer, len);
+		if (len >= max_cnt)
+			goto too_big;
+
 		qp = phba->sli4_hba.hdr_rq;
 		len = __lpfc_idiag_print_rqpair(qp, phba->sli4_hba.dat_rq,
 				"RQpair", pbuffer, len);
@@ -2447,7 +2496,7 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf,
 	struct lpfc_hba *phba = (struct lpfc_hba *)debug->i_private;
 	uint32_t qidx, quetp, queid, index, count, offset, value;
 	uint32_t *pentry;
-	struct lpfc_queue *pque;
+	struct lpfc_queue *pque, *qp;
 	int rc;
 
 	/* This is a user write operation */
@@ -2483,19 +2532,15 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf,
 	case LPFC_IDIAG_EQ:
 		/* HBA event queue */
 		if (phba->sli4_hba.hba_eq) {
-			for (qidx = 0; qidx < phba->cfg_fcp_io_channel;
-				qidx++) {
-				if (phba->sli4_hba.hba_eq[qidx] &&
-				    phba->sli4_hba.hba_eq[qidx]->queue_id ==
-				    queid) {
+			for (qidx = 0; qidx < phba->io_channel_irqs; qidx++) {
+				qp = phba->sli4_hba.hba_eq[qidx];
+				if (qp && qp->queue_id == queid) {
 					/* Sanity check */
-					rc = lpfc_idiag_que_param_check(
-						phba->sli4_hba.hba_eq[qidx],
+					rc = lpfc_idiag_que_param_check(qp,
 						index, count);
 					if (rc)
 						goto error_out;
-					idiag.ptr_private =
-						phba->sli4_hba.hba_eq[qidx];
+					idiag.ptr_private = qp;
 					goto pass_check;
 				}
 			}
@@ -2525,24 +2570,32 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf,
 			idiag.ptr_private = phba->sli4_hba.els_cq;
 			goto pass_check;
 		}
+		/* NVME LS complete queue */
+		if (phba->sli4_hba.nvmels_cq &&
+		    phba->sli4_hba.nvmels_cq->queue_id == queid) {
+			/* Sanity check */
+			rc = lpfc_idiag_que_param_check(
+					phba->sli4_hba.nvmels_cq, index, count);
+			if (rc)
+				goto error_out;
+			idiag.ptr_private = phba->sli4_hba.nvmels_cq;
+			goto pass_check;
+		}
 		/* FCP complete queue */
 		if (phba->sli4_hba.fcp_cq) {
-			qidx = 0;
-			do {
-				if (phba->sli4_hba.fcp_cq[qidx] &&
-				    phba->sli4_hba.fcp_cq[qidx]->queue_id ==
-				    queid) {
+			for (qidx = 0; qidx < phba->cfg_fcp_io_channel;
+								qidx++) {
+				qp = phba->sli4_hba.fcp_cq[qidx];
+				if (qp && qp->queue_id == queid) {
 					/* Sanity check */
 					rc = lpfc_idiag_que_param_check(
-						phba->sli4_hba.fcp_cq[qidx],
-						index, count);
+						qp, index, count);
 					if (rc)
 						goto error_out;
-					idiag.ptr_private =
-						phba->sli4_hba.fcp_cq[qidx];
+					idiag.ptr_private = qp;
 					goto pass_check;
 				}
-			} while (++qidx < phba->cfg_fcp_io_channel);
+			}
 		}
 		goto error_out;
 		break;
@@ -2572,22 +2625,45 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf,
 			idiag.ptr_private = phba->sli4_hba.els_wq;
 			goto pass_check;
 		}
+		/* NVME LS work queue */
+		if (phba->sli4_hba.nvmels_wq &&
+		    phba->sli4_hba.nvmels_wq->queue_id == queid) {
+			/* Sanity check */
+			rc = lpfc_idiag_que_param_check(
+					phba->sli4_hba.nvmels_wq, index, count);
+			if (rc)
+				goto error_out;
+			idiag.ptr_private = phba->sli4_hba.nvmels_wq;
+			goto pass_check;
+		}
 		/* FCP work queue */
 		if (phba->sli4_hba.fcp_wq) {
 			for (qidx = 0; qidx < phba->cfg_fcp_io_channel;
 								qidx++) {
-				if (!phba->sli4_hba.fcp_wq[qidx])
-					continue;
-				if (phba->sli4_hba.fcp_wq[qidx]->queue_id ==
-				    queid) {
+				qp = phba->sli4_hba.fcp_wq[qidx];
+				if (qp && qp->queue_id == queid) {
 					/* Sanity check */
 					rc = lpfc_idiag_que_param_check(
-						phba->sli4_hba.fcp_wq[qidx],
-						index, count);
+						qp, index, count);
+					if (rc)
+						goto error_out;
+					idiag.ptr_private = qp;
+					goto pass_check;
+				}
+			}
+		}
+		/* NVME work queue */
+		if (phba->sli4_hba.nvme_wq) {
+			for (qidx = 0; qidx < phba->cfg_nvme_io_channel;
+								qidx++) {
+				qp = phba->sli4_hba.nvme_wq[qidx];
+				if (qp && qp->queue_id == queid) {
+					/* Sanity check */
+					rc = lpfc_idiag_que_param_check(
+						qp, index, count);
 					if (rc)
 						goto error_out;
-					idiag.ptr_private =
-						phba->sli4_hba.fcp_wq[qidx];
+					idiag.ptr_private = qp;
 					goto pass_check;
 				}
 			}
@@ -4562,10 +4638,14 @@ lpfc_debug_dump_all_queues(struct lpfc_hba *phba)
 	 */
 	lpfc_debug_dump_wq(phba, DUMP_MBX, 0);
 	lpfc_debug_dump_wq(phba, DUMP_ELS, 0);
+	lpfc_debug_dump_wq(phba, DUMP_NVMELS, 0);
 
 	for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++)
 		lpfc_debug_dump_wq(phba, DUMP_FCP, idx);
 
+	for (idx = 0; idx < phba->cfg_nvme_io_channel; idx++)
+		lpfc_debug_dump_wq(phba, DUMP_NVME, idx);
+
 	lpfc_debug_dump_hdr_rq(phba);
 	lpfc_debug_dump_dat_rq(phba);
 	/*
@@ -4573,13 +4653,17 @@ lpfc_debug_dump_all_queues(struct lpfc_hba *phba)
 	 */
 	lpfc_debug_dump_cq(phba, DUMP_MBX, 0);
 	lpfc_debug_dump_cq(phba, DUMP_ELS, 0);
+	lpfc_debug_dump_cq(phba, DUMP_NVMELS, 0);
 
 	for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++)
 		lpfc_debug_dump_cq(phba, DUMP_FCP, idx);
 
+	for (idx = 0; idx < phba->cfg_nvme_io_channel; idx++)
+		lpfc_debug_dump_cq(phba, DUMP_NVME, idx);
+
 	/*
 	 * Dump Event Queues (EQs)
 	 */
-	for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++)
+	for (idx = 0; idx < phba->io_channel_irqs; idx++)
 		lpfc_debug_dump_hba_eq(phba, idx);
 }

drivers/scsi/lpfc/lpfc_debugfs.h

@@ -44,8 +44,10 @@
 
 enum {
 	DUMP_FCP,
+	DUMP_NVME,
 	DUMP_MBX,
 	DUMP_ELS,
+	DUMP_NVMELS,
 };
 
 /*
@@ -364,11 +366,11 @@ lpfc_debug_dump_q(struct lpfc_queue *q)
 }
 
 /**
- * lpfc_debug_dump_wq - dump all entries from the fcp work queue
+ * lpfc_debug_dump_wq - dump all entries from the fcp or nvme work queue
  * @phba: Pointer to HBA context object.
- * @wqidx: Index to a FCP work queue.
+ * @wqidx: Index to a FCP or NVME work queue.
  *
- * This function dumps all entries from a FCP work queue specified
+ * This function dumps all entries from a FCP or NVME work queue specified
  * by the wqidx.
  **/
 static inline void
@@ -380,16 +382,22 @@ lpfc_debug_dump_wq(struct lpfc_hba *phba, int qtype, int wqidx)
 	if (qtype == DUMP_FCP) {
 		wq = phba->sli4_hba.fcp_wq[wqidx];
 		qtypestr = "FCP";
+	} else if (qtype == DUMP_NVME) {
+		wq = phba->sli4_hba.nvme_wq[wqidx];
+		qtypestr = "NVME";
 	} else if (qtype == DUMP_MBX) {
 		wq = phba->sli4_hba.mbx_wq;
 		qtypestr = "MBX";
 	} else if (qtype == DUMP_ELS) {
 		wq = phba->sli4_hba.els_wq;
 		qtypestr = "ELS";
+	} else if (qtype == DUMP_NVMELS) {
+		wq = phba->sli4_hba.nvmels_wq;
+		qtypestr = "NVMELS";
 	} else
 		return;
 
-	if (qtype == DUMP_FCP)
+	if (qtype == DUMP_FCP || qtype == DUMP_NVME)
 		pr_err("%s WQ: WQ[Idx:%d|Qid:%d]\n",
 			qtypestr, wqidx, wq->queue_id);
 	else
@@ -400,12 +408,12 @@ lpfc_debug_dump_wq(struct lpfc_hba *phba, int qtype, int wqidx)
 }
 
 /**
- * lpfc_debug_dump_cq - dump all entries from a fcp work queue's
+ * lpfc_debug_dump_cq - dump all entries from a fcp or nvme work queue's
  * cmpl queue
  * @phba: Pointer to HBA context object.
  * @wqidx: Index to a FCP work queue.
  *
- * This function dumps all entries from a FCP completion queue
+ * This function dumps all entries from a FCP or NVME completion queue
  * which is associated to the work queue specified by the @wqidx.
  **/
 static inline void
@@ -415,12 +423,16 @@ lpfc_debug_dump_cq(struct lpfc_hba *phba, int qtype, int wqidx)
 	char *qtypestr;
 	int eqidx;
 
-	/* fcp wq and cq are 1:1, thus same indexes */
+	/* fcp/nvme wq and cq are 1:1, thus same indexes */
 
 	if (qtype == DUMP_FCP) {
 		wq = phba->sli4_hba.fcp_wq[wqidx];
 		cq = phba->sli4_hba.fcp_cq[wqidx];
 		qtypestr = "FCP";
+	} else if (qtype == DUMP_NVME) {
+		wq = phba->sli4_hba.nvme_wq[wqidx];
+		cq = phba->sli4_hba.nvme_cq[wqidx];
+		qtypestr = "NVME";
 	} else if (qtype == DUMP_MBX) {
 		wq = phba->sli4_hba.mbx_wq;
 		cq = phba->sli4_hba.mbx_cq;
@@ -429,21 +441,25 @@ lpfc_debug_dump_cq(struct lpfc_hba *phba, int qtype, int wqidx)
 		wq = phba->sli4_hba.els_wq;
 		cq = phba->sli4_hba.els_cq;
 		qtypestr = "ELS";
+	} else if (qtype == DUMP_NVMELS) {
+		wq = phba->sli4_hba.nvmels_wq;
+		cq = phba->sli4_hba.nvmels_cq;
+		qtypestr = "NVMELS";
 	} else
 		return;
 
-	for (eqidx = 0; eqidx < phba->cfg_fcp_io_channel; eqidx++) {
+	for (eqidx = 0; eqidx < phba->io_channel_irqs; eqidx++) {
 		eq = phba->sli4_hba.hba_eq[eqidx];
 		if (cq->assoc_qid == eq->queue_id)
 			break;
 	}
-	if (eqidx == phba->cfg_fcp_io_channel) {
+	if (eqidx == phba->io_channel_irqs) {
 		pr_err("Couldn't find EQ for CQ. Using EQ[0]\n");
 		eqidx = 0;
 		eq = phba->sli4_hba.hba_eq[0];
 	}
 
-	if (qtype == DUMP_FCP)
+	if (qtype == DUMP_FCP || qtype == DUMP_NVME)
 		pr_err("%s CQ: WQ[Idx:%d|Qid%d]->CQ[Idx%d|Qid%d]"
 			"->EQ[Idx:%d|Qid:%d]:\n",
 			qtypestr, wqidx, wq->queue_id, wqidx, cq->queue_id,
@@ -527,11 +543,25 @@ lpfc_debug_dump_wq_by_id(struct lpfc_hba *phba, int qid)
 		return;
 	}
 
+	for (wq_idx = 0; wq_idx < phba->cfg_nvme_io_channel; wq_idx++)
+		if (phba->sli4_hba.nvme_wq[wq_idx]->queue_id == qid)
+			break;
+	if (wq_idx < phba->cfg_nvme_io_channel) {
+		pr_err("NVME WQ[Idx:%d|Qid:%d]\n", wq_idx, qid);
+		lpfc_debug_dump_q(phba->sli4_hba.nvme_wq[wq_idx]);
+		return;
+	}
+
 	if (phba->sli4_hba.els_wq->queue_id == qid) {
 		pr_err("ELS WQ[Qid:%d]\n", qid);
 		lpfc_debug_dump_q(phba->sli4_hba.els_wq);
 		return;
 	}
+
+	if (phba->sli4_hba.nvmels_wq->queue_id == qid) {
+		pr_err("NVME LS WQ[Qid:%d]\n", qid);
+		lpfc_debug_dump_q(phba->sli4_hba.nvmels_wq);
+	}
 }
 
 /**
@@ -596,12 +626,28 @@ lpfc_debug_dump_cq_by_id(struct lpfc_hba *phba, int qid)
 		return;
 	}
 
+	for (cq_idx = 0; cq_idx < phba->cfg_nvme_io_channel; cq_idx++)
+		if (phba->sli4_hba.nvme_cq[cq_idx]->queue_id == qid)
+			break;
+
+	if (cq_idx < phba->cfg_nvme_io_channel) {
+		pr_err("NVME CQ[Idx:%d|Qid:%d]\n", cq_idx, qid);
+		lpfc_debug_dump_q(phba->sli4_hba.nvme_cq[cq_idx]);
+		return;
+	}
+
 	if (phba->sli4_hba.els_cq->queue_id == qid) {
 		pr_err("ELS CQ[Qid:%d]\n", qid);
 		lpfc_debug_dump_q(phba->sli4_hba.els_cq);
 		return;
 	}
 
+	if (phba->sli4_hba.nvmels_cq->queue_id == qid) {
+		pr_err("NVME LS CQ[Qid:%d]\n", qid);
+		lpfc_debug_dump_q(phba->sli4_hba.nvmels_cq);
+		return;
+	}
+
 	if (phba->sli4_hba.mbx_cq->queue_id == qid) {
 		pr_err("MBX CQ[Qid:%d]\n", qid);
 		lpfc_debug_dump_q(phba->sli4_hba.mbx_cq);
@@ -621,17 +667,15 @@ lpfc_debug_dump_eq_by_id(struct lpfc_hba *phba, int qid)
 {
 	int eq_idx;
 
-	for (eq_idx = 0; eq_idx < phba->cfg_fcp_io_channel; eq_idx++) {
+	for (eq_idx = 0; eq_idx < phba->io_channel_irqs; eq_idx++)
 		if (phba->sli4_hba.hba_eq[eq_idx]->queue_id == qid)
 			break;
-	}
 
-	if (eq_idx < phba->cfg_fcp_io_channel) {
+	if (eq_idx < phba->io_channel_irqs) {
 		printk(KERN_ERR "FCP EQ[Idx:%d|Qid:%d]\n", eq_idx, qid);
 		lpfc_debug_dump_q(phba->sli4_hba.hba_eq[eq_idx]);
 		return;
 	}
-
 }
 
 void lpfc_debug_dump_all_queues(struct lpfc_hba *);

drivers/scsi/lpfc/lpfc_els.c

@@ -1323,7 +1323,7 @@ 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];
+	pring = lpfc_phba_elsring(phba);
 
 	/*
 	 * Check the txcmplq for an iocb that matches the nport the driver is
@@ -7155,7 +7155,8 @@ lpfc_els_timeout_handler(struct lpfc_vport *vport)
 
 	timeout = (uint32_t)(phba->fc_ratov << 1);
 
-	pring = &phba->sli.ring[LPFC_ELS_RING];
+	pring = lpfc_phba_elsring(phba);
+
 	if ((phba->pport->load_flag & FC_UNLOADING))
 		return;
 	spin_lock_irq(&phba->hbalock);
@@ -7224,7 +7225,7 @@ 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));
@@ -7255,7 +7256,7 @@ 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;
 
@@ -7267,6 +7268,7 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 	 * a working list and release the locks before calling the abort.
 	 */
 	spin_lock_irq(&phba->hbalock);
+	pring = lpfc_phba_elsring(phba);
 	if (phba->sli_rev == LPFC_SLI_REV4)
 		spin_lock(&pring->ring_lock);
 
@@ -9013,7 +9015,9 @@ 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;
+
+	pring = lpfc_phba_elsring(phba);
 
 	spin_lock_irq(&phba->hbalock);
 	list_for_each_entry_safe(piocb, tmp_iocb, &phba->fabric_iocb_list,
@@ -9069,13 +9073,13 @@ 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;
 }
@@ -9099,22 +9103,22 @@ 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;
+
+	pring = lpfc_phba_elsring(phba);
 
 	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,
@@ -9126,21 +9130,21 @@ 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;
 }

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

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 */
@@ -2841,12 +2849,18 @@ 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
@@ -2897,6 +2911,12 @@ 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
@@ -3659,6 +3679,9 @@ 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
@@ -4017,11 +4040,39 @@ struct lpfc_grp_hdr {
 	uint8_t revision[32];
 };
 
+/* 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 OTHER_COMMAND 0x8
+
+#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

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...) \

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;
@@ -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);
+	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,

drivers/scsi/lpfc/lpfc_nportdisc.c

@@ -204,10 +204,11 @@ int
 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;
 
+	pring = lpfc_phba_elsring(phba);
+
 	/* 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 +2105,7 @@ 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 */

drivers/scsi/lpfc/lpfc_nvme.h

+/*******************************************************************
+ * 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_NVMET_MIN_POSTBUF		16
+#define LPFC_NVMET_DEFAULT_POSTBUF	1024
+#define LPFC_NVMET_MAX_POSTBUF		4096
+#define LPFC_NVME_WQSIZE		256
+
+#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;
+};

drivers/scsi/lpfc/lpfc_vport.c

@@ -403,6 +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.