net: Add be2net driver.

Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

8 files changed, 4366 insertions, 0 deletions

diff --git a/drivers/net/benet/Kconfig b/drivers/net/benet/Kconfig
new file mode 100644
index 000000000000..c6934f179c09
--- /dev/null
+++ b/drivers/net/benet/Kconfig
@@ -0,0 +1,7 @@
+config BE2NET
+	tristate "ServerEngines' 10Gbps NIC - BladeEngine 2"
+	depends on PCI && INET
+	select INET_LRO
+	help
+	This driver implements the NIC functionality for ServerEngines'
+	10Gbps network adapter - BladeEngine 2.
diff --git a/drivers/net/benet/Makefile b/drivers/net/benet/Makefile
new file mode 100644
index 000000000000..a60cd8051135
--- /dev/null
+++ b/drivers/net/benet/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile to build the network driver for ServerEngine's BladeEngine.
+#
+
+obj-$(CONFIG_BE2NET) += be2net.o
+
+be2net-y :=  be_main.o be_cmds.o be_ethtool.o
diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
new file mode 100644
index 000000000000..63d593d53153
--- /dev/null
+++ b/drivers/net/benet/be.h
@@ -0,0 +1,327 @@
+/*
+ * Copyright (C) 2005 - 2009 ServerEngines
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.  The full GNU General
+ * Public License is included in this distribution in the file called COPYING.
+ *
+ * Contact Information:
+ * linux-drivers@serverengines.com
+ *
+ * ServerEngines
+ * 209 N. Fair Oaks Ave
+ * Sunnyvale, CA 94085
+ */
+
+#ifndef BE_H
+#define BE_H
+
+#include <linux/pci.h>
+#include <linux/etherdevice.h>
+#include <linux/version.h>
+#include <linux/delay.h>
+#include <net/tcp.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <linux/if_vlan.h>
+#include <linux/workqueue.h>
+#include <linux/interrupt.h>
+#include <linux/inet_lro.h>
+
+#include "be_hw.h"
+
+#define DRV_VER			"2.0.348"
+#define DRV_NAME		"be2net"
+#define BE_NAME			"ServerEngines BladeEngine2 10Gbps NIC"
+#define DRV_DESC		BE_NAME "Driver"
+
+/* Number of bytes of an RX frame that are copied to skb->data */
+#define BE_HDR_LEN 		64
+#define BE_MAX_JUMBO_FRAME_SIZE	9018
+#define BE_MIN_MTU		256
+
+#define BE_NUM_VLANS_SUPPORTED	64
+#define BE_MAX_EQD		96
+#define	BE_MAX_TX_FRAG_COUNT	30
+
+#define EVNT_Q_LEN		1024
+#define TX_Q_LEN		2048
+#define TX_CQ_LEN		1024
+#define RX_Q_LEN		1024	/* Does not support any other value */
+#define RX_CQ_LEN		1024
+#define MCC_Q_LEN		64	/* total size not to exceed 8 pages */
+#define MCC_CQ_LEN		256
+
+#define BE_NAPI_WEIGHT		64
+#define MAX_RX_POST 		BE_NAPI_WEIGHT /* Frags posted at a time */
+#define RX_FRAGS_REFILL_WM	(RX_Q_LEN - MAX_RX_POST)
+
+#define BE_MAX_LRO_DESCRIPTORS  16
+#define BE_MAX_FRAGS_PER_FRAME  16
+
+struct be_dma_mem {
+	void *va;
+	dma_addr_t dma;
+	u32 size;
+};
+
+struct be_queue_info {
+	struct be_dma_mem dma_mem;
+	u16 len;
+	u16 entry_size;	/* Size of an element in the queue */
+	u16 id;
+	u16 tail, head;
+	bool created;
+	atomic_t used;	/* Number of valid elements in the queue */
+};
+
+struct be_ctrl_info {
+	u8 __iomem *csr;
+	u8 __iomem *db;		/* Door Bell */
+	u8 __iomem *pcicfg;	/* PCI config space */
+	int pci_func;
+
+	/* Mbox used for cmd request/response */
+	spinlock_t cmd_lock;	/* For serializing cmds to BE card */
+	struct be_dma_mem mbox_mem;
+	/* Mbox mem is adjusted to align to 16 bytes. The allocated addr
+	 * is stored for freeing purpose */
+	struct be_dma_mem mbox_mem_alloced;
+};
+
+#include "be_cmds.h"
+
+struct be_drvr_stats {
+	u32 be_tx_reqs;		/* number of TX requests initiated */
+	u32 be_tx_stops;	/* number of times TX Q was stopped */
+	u32 be_fwd_reqs;	/* number of send reqs through forwarding i/f */
+	u32 be_tx_wrbs;		/* number of tx WRBs used */
+	u32 be_tx_events;	/* number of tx completion events  */
+	u32 be_tx_compl;	/* number of tx completion entries processed */
+	u64 be_tx_jiffies;
+	ulong be_tx_bytes;
+	ulong be_tx_bytes_prev;
+	u32 be_tx_rate;
+
+	u32 cache_barrier[16];
+
+	u32 be_ethrx_post_fail;/* number of ethrx buffer alloc failures */
+	u32 be_polls;		/* number of times NAPI called poll function */
+	u32 be_rx_events;	/* number of ucast rx completion events  */
+	u32 be_rx_compl;	/* number of rx completion entries processed */
+	u32 be_lro_hgram_data[8];	/* histogram of LRO data packets */
+	u32 be_lro_hgram_ack[8];	/* histogram of LRO ACKs */
+	u64 be_rx_jiffies;
+	ulong be_rx_bytes;
+	ulong be_rx_bytes_prev;
+	u32 be_rx_rate;
+	/* number of non ether type II frames dropped where
+	 * frame len > length field of Mac Hdr */
+	u32 be_802_3_dropped_frames;
+	/* number of non ether type II frames malformed where
+	 * in frame len < length field of Mac Hdr */
+	u32 be_802_3_malformed_frames;
+	u32 be_rxcp_err;	/* Num rx completion entries w/ err set. */
+	ulong rx_fps_jiffies;	/* jiffies at last FPS calc */
+	u32 be_rx_frags;
+	u32 be_prev_rx_frags;
+	u32 be_rx_fps;		/* Rx frags per second */
+};
+
+struct be_stats_obj {
+	struct be_drvr_stats drvr_stats;
+	struct net_device_stats net_stats;
+	struct be_dma_mem cmd;
+};
+
+struct be_eq_obj {
+	struct be_queue_info q;
+	char desc[32];
+
+	/* Adaptive interrupt coalescing (AIC) info */
+	bool enable_aic;
+	u16 min_eqd;		/* in usecs */
+	u16 max_eqd;		/* in usecs */
+	u16 cur_eqd;		/* in usecs */
+
+	struct napi_struct napi;
+};
+
+struct be_tx_obj {
+	struct be_queue_info q;
+	struct be_queue_info cq;
+	/* Remember the skbs that were transmitted */
+	struct sk_buff *sent_skb_list[TX_Q_LEN];
+};
+
+/* Struct to remember the pages posted for rx frags */
+struct be_rx_page_info {
+	struct page *page;
+	dma_addr_t bus;
+	u16 page_offset;
+	bool last_page_user;
+};
+
+struct be_rx_obj {
+	struct be_queue_info q;
+	struct be_queue_info cq;
+	struct be_rx_page_info page_info_tbl[RX_Q_LEN];
+	struct net_lro_mgr lro_mgr;
+	struct net_lro_desc lro_desc[BE_MAX_LRO_DESCRIPTORS];
+};
+
+#define BE_NUM_MSIX_VECTORS		2	/* 1 each for Tx and Rx */
+struct be_adapter {
+	struct pci_dev *pdev;
+	struct net_device *netdev;
+
+	/* Mbox, pci config, csr address information */
+	struct be_ctrl_info ctrl;
+
+	struct msix_entry msix_entries[BE_NUM_MSIX_VECTORS];
+	bool msix_enabled;
+	bool isr_registered;
+
+	/* TX Rings */
+	struct be_eq_obj tx_eq;
+	struct be_tx_obj tx_obj;
+
+	u32 cache_line_break[8];
+
+	/* Rx rings */
+	struct be_eq_obj rx_eq;
+	struct be_rx_obj rx_obj;
+	u32 big_page_size;	/* Compounded page size shared by rx wrbs */
+
+	struct vlan_group *vlan_grp;
+	u16 num_vlans;
+	u8 vlan_tag[VLAN_GROUP_ARRAY_LEN];
+
+	struct be_stats_obj stats;
+	/* Work queue used to perform periodic tasks like getting statistics */
+	struct delayed_work work;
+
+	/* Ethtool knobs and info */
+	bool rx_csum; 		/* BE card must perform rx-checksumming */
+	u32 max_rx_coal;
+	char fw_ver[FW_VER_LEN];
+	u32 if_handle;		/* Used to configure filtering */
+	u32 pmac_id;		/* MAC addr handle used by BE card */
+
+	struct be_link_info link;
+	u32 port_num;
+};
+
+extern struct ethtool_ops be_ethtool_ops;
+
+#define drvr_stats(adapter)		(&adapter->stats.drvr_stats)
+
+#define BE_SET_NETDEV_OPS(netdev, ops)	(netdev->netdev_ops = ops)
+
+static inline u32 MODULO(u16 val, u16 limit)
+{
+	BUG_ON(limit & (limit - 1));
+	return val & (limit - 1);
+}
+
+static inline void index_adv(u16 *index, u16 val, u16 limit)
+{
+	*index = MODULO((*index + val), limit);
+}
+
+static inline void index_inc(u16 *index, u16 limit)
+{
+	*index = MODULO((*index + 1), limit);
+}
+
+#define PAGE_SHIFT_4K		12
+#define PAGE_SIZE_4K		(1 << PAGE_SHIFT_4K)
+
+/* Returns number of pages spanned by the data starting at the given addr */
+#define PAGES_4K_SPANNED(_address, size) 				\
+		((u32)((((size_t)(_address) & (PAGE_SIZE_4K - 1)) + 	\
+			(size) + (PAGE_SIZE_4K - 1)) >> PAGE_SHIFT_4K))
+
+/* Byte offset into the page corresponding to given address */
+#define OFFSET_IN_PAGE(addr)						\
+		 ((size_t)(addr) & (PAGE_SIZE_4K-1))
+
+/* Returns bit offset within a DWORD of a bitfield */
+#define AMAP_BIT_OFFSET(_struct, field)  				\
+		(((size_t)&(((_struct *)0)->field))%32)
+
+/* Returns the bit mask of the field that is NOT shifted into location. */
+static inline u32 amap_mask(u32 bitsize)
+{
+	return (bitsize == 32 ? 0xFFFFFFFF : (1 << bitsize) - 1);
+}
+
+static inline void
+amap_set(void *ptr, u32 dw_offset, u32 mask, u32 offset, u32 value)
+{
+	u32 *dw = (u32 *) ptr + dw_offset;
+	*dw &= ~(mask << offset);
+	*dw |= (mask & value) << offset;
+}
+
+#define AMAP_SET_BITS(_struct, field, ptr, val)				\
+		amap_set(ptr,						\
+			offsetof(_struct, field)/32,			\
+			amap_mask(sizeof(((_struct *)0)->field)),	\
+			AMAP_BIT_OFFSET(_struct, field),		\
+			val)
+
+static inline u32 amap_get(void *ptr, u32 dw_offset, u32 mask, u32 offset)
+{
+	u32 *dw = (u32 *) ptr;
+	return mask & (*(dw + dw_offset) >> offset);
+}
+
+#define AMAP_GET_BITS(_struct, field, ptr)				\
+		amap_get(ptr,						\
+			offsetof(_struct, field)/32,			\
+			amap_mask(sizeof(((_struct *)0)->field)),	\
+			AMAP_BIT_OFFSET(_struct, field))
+
+#define be_dws_cpu_to_le(wrb, len)	swap_dws(wrb, len)
+#define be_dws_le_to_cpu(wrb, len)	swap_dws(wrb, len)
+static inline void swap_dws(void *wrb, int len)
+{
+#ifdef __BIG_ENDIAN
+	u32 *dw = wrb;
+	BUG_ON(len % 4);
+	do {
+		*dw = cpu_to_le32(*dw);
+		dw++;
+		len -= 4;
+	} while (len);
+#endif				/* __BIG_ENDIAN */
+}
+
+static inline u8 is_tcp_pkt(struct sk_buff *skb)
+{
+	u8 val = 0;
+
+	if (ip_hdr(skb)->version == 4)
+		val = (ip_hdr(skb)->protocol == IPPROTO_TCP);
+	else if (ip_hdr(skb)->version == 6)
+		val = (ipv6_hdr(skb)->nexthdr == NEXTHDR_TCP);
+
+	return val;
+}
+
+static inline u8 is_udp_pkt(struct sk_buff *skb)
+{
+	u8 val = 0;
+
+	if (ip_hdr(skb)->version == 4)
+		val = (ip_hdr(skb)->protocol == IPPROTO_UDP);
+	else if (ip_hdr(skb)->version == 6)
+		val = (ipv6_hdr(skb)->nexthdr == NEXTHDR_UDP);
+
+	return val;
+}
+
+#endif				/* BE_H */
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
new file mode 100644
index 000000000000..d444aed962bc
--- /dev/null
+++ b/drivers/net/benet/be_cmds.c
@@ -0,0 +1,861 @@
+/*
+ * Copyright (C) 2005 - 2009 ServerEngines
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.  The full GNU General
+ * Public License is included in this distribution in the file called COPYING.
+ *
+ * Contact Information:
+ * linux-drivers@serverengines.com
+ *
+ * ServerEngines
+ * 209 N. Fair Oaks Ave
+ * Sunnyvale, CA 94085
+ */
+
+#include "be.h"
+
+static int be_mbox_db_ready_wait(void __iomem *db)
+{
+	int cnt = 0, wait = 5;
+	u32 ready;
+
+	do {
+		ready = ioread32(db) & MPU_MAILBOX_DB_RDY_MASK;
+		if (ready)
+			break;
+
+		if (cnt > 200000) {
+			printk(KERN_WARNING DRV_NAME
+				": mbox_db poll timed out\n");
+			return -1;
+		}
+
+		if (cnt > 50)
+			wait = 200;
+		cnt += wait;
+		udelay(wait);
+	} while (true);
+
+	return 0;
+}
+
+/*
+ * Insert the mailbox address into the doorbell in two steps
+ */
+static int be_mbox_db_ring(struct be_ctrl_info *ctrl)
+{
+	int status;
+	u16 compl_status, extd_status;
+	u32 val = 0;
+	void __iomem *db = ctrl->db + MPU_MAILBOX_DB_OFFSET;
+	struct be_dma_mem *mbox_mem = &ctrl->mbox_mem;
+	struct be_mcc_mailbox *mbox = mbox_mem->va;
+	struct be_mcc_cq_entry *cqe = &mbox->cqe;
+
+	memset(cqe, 0, sizeof(*cqe));
+
+	val &= ~MPU_MAILBOX_DB_RDY_MASK;
+	val |= MPU_MAILBOX_DB_HI_MASK;
+	/* at bits 2 - 31 place mbox dma addr msb bits 34 - 63 */
+	val |= (upper_32_bits(mbox_mem->dma) >> 2) << 2;
+	iowrite32(val, db);
+
+	/* wait for ready to be set */
+	status = be_mbox_db_ready_wait(db);
+	if (status != 0)
+		return status;
+
+	val = 0;
+	val &= ~MPU_MAILBOX_DB_RDY_MASK;
+	val &= ~MPU_MAILBOX_DB_HI_MASK;
+	/* at bits 2 - 31 place mbox dma addr lsb bits 4 - 33 */
+	val |= (u32)(mbox_mem->dma >> 4) << 2;
+	iowrite32(val, db);
+
+	status = be_mbox_db_ready_wait(db);
+	if (status != 0)
+		return status;
+
+	/* compl entry has been made now */
+	be_dws_le_to_cpu(cqe, sizeof(*cqe));
+	if (!(cqe->flags & CQE_FLAGS_VALID_MASK)) {
+		printk(KERN_WARNING DRV_NAME ": ERROR invalid mbox compl\n");
+		return -1;
+	}
+
+	compl_status = (cqe->status >> CQE_STATUS_COMPL_SHIFT) &
+				CQE_STATUS_COMPL_MASK;
+	if (compl_status != MCC_STATUS_SUCCESS) {
+		extd_status = (cqe->status >> CQE_STATUS_EXTD_SHIFT) &
+				CQE_STATUS_EXTD_MASK;
+		printk(KERN_WARNING DRV_NAME
+			": ERROR in cmd compl. status(compl/extd)=%d/%d\n",
+			compl_status, extd_status);
+	}
+
+	return compl_status;
+}
+
+static int be_POST_stage_get(struct be_ctrl_info *ctrl, u16 *stage)
+{
+	u32 sem = ioread32(ctrl->csr + MPU_EP_SEMAPHORE_OFFSET);
+
+	*stage = sem & EP_SEMAPHORE_POST_STAGE_MASK;
+	if ((sem >> EP_SEMAPHORE_POST_ERR_SHIFT) & EP_SEMAPHORE_POST_ERR_MASK)
+		return -1;
+	else
+		return 0;
+}
+
+static int be_POST_stage_poll(struct be_ctrl_info *ctrl, u16 poll_stage)
+{
+	u16 stage, cnt, error;
+	for (cnt = 0; cnt < 5000; cnt++) {
+		error = be_POST_stage_get(ctrl, &stage);
+		if (error)
+			return -1;
+
+		if (stage == poll_stage)
+			break;
+		udelay(1000);
+	}
+	if (stage != poll_stage)
+		return -1;
+	return 0;
+}
+
+
+int be_cmd_POST(struct be_ctrl_info *ctrl)
+{
+	u16 stage, error;
+
+	error = be_POST_stage_get(ctrl, &stage);
+	if (error)
+		goto err;
+
+	if (stage == POST_STAGE_ARMFW_RDY)
+		return 0;
+
+	if (stage != POST_STAGE_AWAITING_HOST_RDY)
+		goto err;
+
+	/* On awaiting host rdy, reset and again poll on awaiting host rdy */
+	iowrite32(POST_STAGE_BE_RESET, ctrl->csr + MPU_EP_SEMAPHORE_OFFSET);
+	error = be_POST_stage_poll(ctrl, POST_STAGE_AWAITING_HOST_RDY);
+	if (error)
+		goto err;
+
+	/* Now kickoff POST and poll on armfw ready */
+	iowrite32(POST_STAGE_HOST_RDY, ctrl->csr + MPU_EP_SEMAPHORE_OFFSET);
+	error = be_POST_stage_poll(ctrl, POST_STAGE_ARMFW_RDY);
+	if (error)
+		goto err;
+
+	return 0;
+err:
+	printk(KERN_WARNING DRV_NAME ": ERROR, stage=%d\n", stage);
+	return -1;
+}
+
+static inline void *embedded_payload(struct be_mcc_wrb *wrb)
+{
+	return wrb->payload.embedded_payload;
+}
+
+static inline struct be_sge *nonembedded_sgl(struct be_mcc_wrb *wrb)
+{
+	return &wrb->payload.sgl[0];
+}
+
+/* Don't touch the hdr after it's prepared */
+static void be_wrb_hdr_prepare(struct be_mcc_wrb *wrb, int payload_len,
+				bool embedded, u8 sge_cnt)
+{
+	if (embedded)
+		wrb->embedded |= MCC_WRB_EMBEDDED_MASK;
+	else
+		wrb->embedded |= (sge_cnt & MCC_WRB_SGE_CNT_MASK) <<
+				MCC_WRB_SGE_CNT_SHIFT;
+	wrb->payload_length = payload_len;
+	be_dws_cpu_to_le(wrb, 20);
+}
+
+/* Don't touch the hdr after it's prepared */
+static void be_cmd_hdr_prepare(struct be_cmd_req_hdr *req_hdr,
+				u8 subsystem, u8 opcode, int cmd_len)
+{
+	req_hdr->opcode = opcode;
+	req_hdr->subsystem = subsystem;
+	req_hdr->request_length = cpu_to_le32(cmd_len - sizeof(*req_hdr));
+}
+
+static void be_cmd_page_addrs_prepare(struct phys_addr *pages, u32 max_pages,
+			struct be_dma_mem *mem)
+{
+	int i, buf_pages = min(PAGES_4K_SPANNED(mem->va, mem->size), max_pages);
+	u64 dma = (u64)mem->dma;
+
+	for (i = 0; i < buf_pages; i++) {
+		pages[i].lo = cpu_to_le32(dma & 0xFFFFFFFF);
+		pages[i].hi = cpu_to_le32(upper_32_bits(dma));
+		dma += PAGE_SIZE_4K;
+	}
+}
+
+/* Converts interrupt delay in microseconds to multiplier value */
+static u32 eq_delay_to_mult(u32 usec_delay)
+{
+#define MAX_INTR_RATE			651042
+	const u32 round = 10;
+	u32 multiplier;
+
+	if (usec_delay == 0)
+		multiplier = 0;
+	else {
+		u32 interrupt_rate = 1000000 / usec_delay;
+		/* Max delay, corresponding to the lowest interrupt rate */
+		if (interrupt_rate == 0)
+			multiplier = 1023;
+		else {
+			multiplier = (MAX_INTR_RATE - interrupt_rate) * round;
+			multiplier /= interrupt_rate;
+			/* Round the multiplier to the closest value.*/
+			multiplier = (multiplier + round/2) / round;
+			multiplier = min(multiplier, (u32)1023);
+		}
+	}
+	return multiplier;
+}
+
+static inline struct be_mcc_wrb *wrb_from_mbox(struct be_dma_mem *mbox_mem)
+{
+	return &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
+}
+
+int be_cmd_eq_create(struct be_ctrl_info *ctrl,
+		struct be_queue_info *eq, int eq_delay)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_eq_create *req = embedded_payload(wrb);
+	struct be_cmd_resp_eq_create *resp = embedded_payload(wrb);
+	struct be_dma_mem *q_mem = &eq->dma_mem;
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_EQ_CREATE, sizeof(*req));
+
+	req->num_pages =  cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
+
+	AMAP_SET_BITS(struct amap_eq_context, func, req->context,
+			ctrl->pci_func);
+	AMAP_SET_BITS(struct amap_eq_context, valid, req->context, 1);
+	/* 4byte eqe*/
+	AMAP_SET_BITS(struct amap_eq_context, size, req->context, 0);
+	AMAP_SET_BITS(struct amap_eq_context, count, req->context,
+			__ilog2_u32(eq->len/256));
+	AMAP_SET_BITS(struct amap_eq_context, delaymult, req->context,
+			eq_delay_to_mult(eq_delay));
+	be_dws_cpu_to_le(req->context, sizeof(req->context));
+
+	be_cmd_page_addrs_prepare(req->pages, ARRAY_SIZE(req->pages), q_mem);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		eq->id = le16_to_cpu(resp->eq_id);
+		eq->created = true;
+	}
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_mac_addr_query(struct be_ctrl_info *ctrl, u8 *mac_addr,
+			u8 type, bool permanent, u32 if_handle)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_mac_query *req = embedded_payload(wrb);
+	struct be_cmd_resp_mac_query *resp = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_MAC_QUERY, sizeof(*req));
+
+	req->type = type;
+	if (permanent) {
+		req->permanent = 1;
+	} else {
+		req->if_id = cpu_to_le16((u16)if_handle);
+		req->permanent = 0;
+	}
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status)
+		memcpy(mac_addr, resp->mac.addr, ETH_ALEN);
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_pmac_add(struct be_ctrl_info *ctrl, u8 *mac_addr,
+		u32 if_id, u32 *pmac_id)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_pmac_add *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_PMAC_ADD, sizeof(*req));
+
+	req->if_id = cpu_to_le32(if_id);
+	memcpy(req->mac_address, mac_addr, ETH_ALEN);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_pmac_add *resp = embedded_payload(wrb);
+		*pmac_id = le32_to_cpu(resp->pmac_id);
+	}
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_pmac_del(struct be_ctrl_info *ctrl, u32 if_id, u32 pmac_id)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_pmac_del *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_PMAC_DEL, sizeof(*req));
+
+	req->if_id = cpu_to_le32(if_id);
+	req->pmac_id = cpu_to_le32(pmac_id);
+
+	status = be_mbox_db_ring(ctrl);
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+int be_cmd_cq_create(struct be_ctrl_info *ctrl,
+		struct be_queue_info *cq, struct be_queue_info *eq,
+		bool sol_evts, bool no_delay, int coalesce_wm)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_cq_create *req = embedded_payload(wrb);
+	struct be_cmd_resp_cq_create *resp = embedded_payload(wrb);
+	struct be_dma_mem *q_mem = &cq->dma_mem;
+	void *ctxt = &req->context;
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_CQ_CREATE, sizeof(*req));
+
+	req->num_pages =  cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
+
+	AMAP_SET_BITS(struct amap_cq_context, coalescwm, ctxt, coalesce_wm);
+	AMAP_SET_BITS(struct amap_cq_context, nodelay, ctxt, no_delay);
+	AMAP_SET_BITS(struct amap_cq_context, count, ctxt,
+			__ilog2_u32(cq->len/256));
+	AMAP_SET_BITS(struct amap_cq_context, valid, ctxt, 1);
+	AMAP_SET_BITS(struct amap_cq_context, solevent, ctxt, sol_evts);
+	AMAP_SET_BITS(struct amap_cq_context, eventable, ctxt, 1);
+	AMAP_SET_BITS(struct amap_cq_context, eqid, ctxt, eq->id);
+	AMAP_SET_BITS(struct amap_cq_context, armed, ctxt, 0);
+	AMAP_SET_BITS(struct amap_cq_context, func, ctxt, ctrl->pci_func);
+	be_dws_cpu_to_le(ctxt, sizeof(req->context));
+
+	be_cmd_page_addrs_prepare(req->pages, ARRAY_SIZE(req->pages), q_mem);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		cq->id = le16_to_cpu(resp->cq_id);
+		cq->created = true;
+	}
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+int be_cmd_txq_create(struct be_ctrl_info *ctrl,
+			struct be_queue_info *txq,
+			struct be_queue_info *cq)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_eth_tx_create *req = embedded_payload(wrb);
+	struct be_dma_mem *q_mem = &txq->dma_mem;
+	void *ctxt = &req->context;
+	int status;
+	u32 len_encoded;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH, OPCODE_ETH_TX_CREATE,
+		sizeof(*req));
+
+	req->num_pages = PAGES_4K_SPANNED(q_mem->va, q_mem->size);
+	req->ulp_num = BE_ULP1_NUM;
+	req->type = BE_ETH_TX_RING_TYPE_STANDARD;
+
+	len_encoded = fls(txq->len); /* log2(len) + 1 */
+	if (len_encoded == 16)
+		len_encoded = 0;
+	AMAP_SET_BITS(struct amap_tx_context, tx_ring_size, ctxt, len_encoded);
+	AMAP_SET_BITS(struct amap_tx_context, pci_func_id, ctxt,
+			ctrl->pci_func);
+	AMAP_SET_BITS(struct amap_tx_context, ctx_valid, ctxt, 1);
+	AMAP_SET_BITS(struct amap_tx_context, cq_id_send, ctxt, cq->id);
+
+	be_dws_cpu_to_le(ctxt, sizeof(req->context));
+
+	be_cmd_page_addrs_prepare(req->pages, ARRAY_SIZE(req->pages), q_mem);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_eth_tx_create *resp = embedded_payload(wrb);
+		txq->id = le16_to_cpu(resp->cid);
+		txq->created = true;
+	}
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+int be_cmd_rxq_create(struct be_ctrl_info *ctrl,
+		struct be_queue_info *rxq, u16 cq_id, u16 frag_size,
+		u16 max_frame_size, u32 if_id, u32 rss)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_eth_rx_create *req = embedded_payload(wrb);
+	struct be_dma_mem *q_mem = &rxq->dma_mem;
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH, OPCODE_ETH_RX_CREATE,
+		sizeof(*req));
+
+	req->cq_id = cpu_to_le16(cq_id);
+	req->frag_size = fls(frag_size) - 1;
+	req->num_pages = 2;
+	be_cmd_page_addrs_prepare(req->pages, ARRAY_SIZE(req->pages), q_mem);
+	req->interface_id = cpu_to_le32(if_id);
+	req->max_frame_size = cpu_to_le16(max_frame_size);
+	req->rss_queue = cpu_to_le32(rss);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_eth_rx_create *resp = embedded_payload(wrb);
+		rxq->id = le16_to_cpu(resp->id);
+		rxq->created = true;
+	}
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+/* Generic destroyer function for all types of queues */
+int be_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q,
+		int queue_type)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_q_destroy *req = embedded_payload(wrb);
+	u8 subsys = 0, opcode = 0;
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+
+	memset(wrb, 0, sizeof(*wrb));
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	switch (queue_type) {
+	case QTYPE_EQ:
+		subsys = CMD_SUBSYSTEM_COMMON;
+		opcode = OPCODE_COMMON_EQ_DESTROY;
+		break;
+	case QTYPE_CQ:
+		subsys = CMD_SUBSYSTEM_COMMON;
+		opcode = OPCODE_COMMON_CQ_DESTROY;
+		break;
+	case QTYPE_TXQ:
+		subsys = CMD_SUBSYSTEM_ETH;
+		opcode = OPCODE_ETH_TX_DESTROY;
+		break;
+	case QTYPE_RXQ:
+		subsys = CMD_SUBSYSTEM_ETH;
+		opcode = OPCODE_ETH_RX_DESTROY;
+		break;
+	default:
+		printk(KERN_WARNING DRV_NAME ":bad Q type in Q destroy cmd\n");
+		status = -1;
+		goto err;
+	}
+	be_cmd_hdr_prepare(&req->hdr, subsys, opcode, sizeof(*req));
+	req->id = cpu_to_le16(q->id);
+
+	status = be_mbox_db_ring(ctrl);
+err:
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+/* Create an rx filtering policy configuration on an i/f */
+int be_cmd_if_create(struct be_ctrl_info *ctrl, u32 flags, u8 *mac,
+		bool pmac_invalid, u32 *if_handle, u32 *pmac_id)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_if_create *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_INTERFACE_CREATE, sizeof(*req));
+
+	req->capability_flags = cpu_to_le32(flags);
+	req->enable_flags = cpu_to_le32(flags);
+	if (!pmac_invalid)
+		memcpy(req->mac_addr, mac, ETH_ALEN);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_if_create *resp = embedded_payload(wrb);
+		*if_handle = le32_to_cpu(resp->interface_id);
+		if (!pmac_invalid)
+			*pmac_id = le32_to_cpu(resp->pmac_id);
+	}
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_if_destroy(struct be_ctrl_info *ctrl, u32 interface_id)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_if_destroy *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_INTERFACE_DESTROY, sizeof(*req));
+
+	req->interface_id = cpu_to_le32(interface_id);
+	status = be_mbox_db_ring(ctrl);
+
+	spin_unlock(&ctrl->cmd_lock);
+
+	return status;
+}
+
+/* Get stats is a non embedded command: the request is not embedded inside
+ * WRB but is a separate dma memory block
+ */
+int be_cmd_get_stats(struct be_ctrl_info *ctrl, struct be_dma_mem *nonemb_cmd)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_get_stats *req = nonemb_cmd->va;
+	struct be_sge *sge = nonembedded_sgl(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	memset(req, 0, sizeof(*req));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), false, 1);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH,
+		OPCODE_ETH_GET_STATISTICS, sizeof(*req));
+	sge->pa_hi = cpu_to_le32(upper_32_bits(nonemb_cmd->dma));
+	sge->pa_lo = cpu_to_le32(nonemb_cmd->dma & 0xFFFFFFFF);
+	sge->len = cpu_to_le32(nonemb_cmd->size);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_get_stats *resp = nonemb_cmd->va;
+		be_dws_le_to_cpu(&resp->hw_stats, sizeof(resp->hw_stats));
+	}
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
+			struct be_link_info *link)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_link_status *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_LINK_STATUS_QUERY, sizeof(*req));
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_link_status *resp = embedded_payload(wrb);
+		link->speed = resp->mac_speed;
+		link->duplex = resp->mac_duplex;
+		link->fault = resp->mac_fault;
+	} else {
+		link->speed = PHY_LINK_SPEED_ZERO;
+	}
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_get_fw_ver(struct be_ctrl_info *ctrl, char *fw_ver)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_get_fw_version *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_GET_FW_VERSION, sizeof(*req));
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_get_fw_version *resp = embedded_payload(wrb);
+		strncpy(fw_ver, resp->firmware_version_string, FW_VER_LEN);
+	}
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+/* set the EQ delay interval of an EQ to specified value */
+int be_cmd_modify_eqd(struct be_ctrl_info *ctrl, u32 eq_id, u32 eqd)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_modify_eq_delay *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_MODIFY_EQ_DELAY, sizeof(*req));
+
+	req->num_eq = cpu_to_le32(1);
+	req->delay[0].eq_id = cpu_to_le32(eq_id);
+	req->delay[0].phase = 0;
+	req->delay[0].delay_multiplier = cpu_to_le32(eqd);
+
+	status = be_mbox_db_ring(ctrl);
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_vlan_config(struct be_ctrl_info *ctrl, u32 if_id, u16 *vtag_array,
+			u32 num, bool untagged, bool promiscuous)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_vlan_config *req = embedded_payload(wrb);
+	int status;
+
+	spin_lock(&ctrl->cmd_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+		OPCODE_COMMON_NTWK_VLAN_CONFIG, sizeof(*req));
+
+	req->interface_id = if_id;
+	req->promiscuous = promiscuous;
+	req->untagged = untagged;
+	req->num_vlan = num;
+	if (!promiscuous) {
+		memcpy(req->normal_vlan, vtag_array,
+			req->num_vlan * sizeof(vtag_array[0]));
+	}
+
+	status = be_mbox_db_ring(ctrl);
+
+	spin_unlock(&ctrl->cmd_lock);
+	return status;
+}
+
+int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_promiscuous_con