net: Add be2net driver.

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

MAINTAINERS

@@ -3880,6 +3880,15 @@ L:	linux-ide@vger.kernel.org
 T:	git kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
 S:	Supported
 
+SERVER ENGINES 10Gbps NIC - BladeEngine 2 DRIVER
+P:	Sathya Perla
+M:	sathyap@serverengines.com
+P:      Subbu Seetharaman
+M:      subbus@serverengines.com
+L:      netdev@vger.kernel.org
+W:      http://www.serverengines.com
+S:      Supported
+
 SFC NETWORK DRIVER
 P:	Steve Hodgson
 P:	Ben Hutchings

drivers/net/Kconfig

@@ -2630,6 +2630,8 @@ config QLGE
 
 source "drivers/net/sfc/Kconfig"
 
+source "drivers/net/benet/Kconfig"
+
 endif # NETDEV_10000
 
 source "drivers/net/tokenring/Kconfig"

drivers/net/Makefile

@@ -22,6 +22,7 @@ obj-$(CONFIG_GIANFAR) += gianfar_driver.o
 obj-$(CONFIG_TEHUTI) += tehuti.o
 obj-$(CONFIG_ENIC) += enic/
 obj-$(CONFIG_JME) += jme.o
+obj-$(CONFIG_BE2NET) += benet/
 
 gianfar_driver-objs := gianfar.o \
 		gianfar_ethtool.o \

drivers/net/benet/Kconfig

+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.

drivers/net/benet/Makefile

+#
+# 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

drivers/net/benet/be.h

+/*
+ * 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 */

drivers/net/benet/be_hw.h

+/*
+ * 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
+ */
+
+/********* Mailbox door bell *************/
+/* Used for driver communication with the FW.
+ * The software must write this register twice to post any command. First,
+ * it writes the register with hi=1 and the upper bits of the physical address
+ * for the MAILBOX structure. Software must poll the ready bit until this
+ * is acknowledged. Then, sotware writes the register with hi=0 with the lower
+ * bits in the address. It must poll the ready bit until the command is
+ * complete. Upon completion, the MAILBOX will contain a valid completion
+ * queue entry.
+ */
+#define MPU_MAILBOX_DB_OFFSET	0x160
+#define MPU_MAILBOX_DB_RDY_MASK	0x1 	/* bit 0 */
+#define MPU_MAILBOX_DB_HI_MASK	0x2	/* bit 1 */
+
+#define MPU_EP_CONTROL 		0
+
+/********** MPU semphore ******************/
+#define MPU_EP_SEMAPHORE_OFFSET 	0xac
+#define EP_SEMAPHORE_POST_STAGE_MASK	0x0000FFFF
+#define EP_SEMAPHORE_POST_ERR_MASK	0x1
+#define EP_SEMAPHORE_POST_ERR_SHIFT	31
+/* MPU semphore POST stage values */
+#define POST_STAGE_AWAITING_HOST_RDY 	0x1 /* FW awaiting goahead from host */
+#define POST_STAGE_HOST_RDY 		0x2 /* Host has given go-ahed to FW */
+#define POST_STAGE_BE_RESET		0x3 /* Host wants to reset chip */
+#define POST_STAGE_ARMFW_RDY		0xc000	/* FW is done with POST */
+
+/********* Memory BAR register ************/
+#define PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET 	0xfc
+/* Host Interrupt Enable, if set interrupts are enabled although "PCI Interrupt
+ * Disable" may still globally block interrupts in addition to individual
+ * interrupt masks; a mechanism for the device driver to block all interrupts
+ * atomically without having to arbitrate for the PCI Interrupt Disable bit
+ * with the OS.
+ */
+#define MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK	(1 << 29) /* bit 29 */
+/* PCI physical function number */
+#define MEMBAR_CTRL_INT_CTRL_PFUNC_MASK  	0x7 	/* bits 26 - 28 */
+#define MEMBAR_CTRL_INT_CTRL_PFUNC_SHIFT	26
+
+/********* Event Q door bell *************/
+#define DB_EQ_OFFSET			DB_CQ_OFFSET
+#define DB_EQ_RING_ID_MASK		0x1FF	/* bits 0 - 8 */
+/* Clear the interrupt for this eq */
+#define DB_EQ_CLR_SHIFT			(9)	/* bit 9 */
+/* Must be 1 */
+#define DB_EQ_EVNT_SHIFT			(10)	/* bit 10 */
+/* Number of event entries processed */
+#define DB_EQ_NUM_POPPED_SHIFT		(16)	/* bits 16 - 28 */
+/* Rearm bit */
+#define DB_EQ_REARM_SHIFT		(29)	/* bit 29 */
+
+/********* Compl Q door bell *************/
+#define DB_CQ_OFFSET 			0x120
+#define DB_CQ_RING_ID_MASK		0x3FF	/* bits 0 - 9 */
+/* Number of event entries processed */
+#define DB_CQ_NUM_POPPED_SHIFT		(16) 	/* bits 16 - 28 */
+/* Rearm bit */
+#define DB_CQ_REARM_SHIFT		(29) 	/* bit 29 */
+
+/********** TX ULP door bell *************/
+#define DB_TXULP1_OFFSET		0x60
+#define DB_TXULP_RING_ID_MASK		0x7FF	/* bits 0 - 10 */
+/* Number of tx entries posted */
+#define DB_TXULP_NUM_POSTED_SHIFT	(16)	/* bits 16 - 29 */
+#define DB_TXULP_NUM_POSTED_MASK	0x3FFF	/* bits 16 - 29 */
+
+/********** RQ(erx) door bell ************/
+#define DB_RQ_OFFSET 			0x100
+#define DB_RQ_RING_ID_MASK		0x3FF	/* bits 0 - 9 */
+/* Number of rx frags posted */
+#define DB_RQ_NUM_POSTED_SHIFT		(24)	/* bits 24 - 31 */
+
+/*
+ * BE descriptors: host memory data structures whose formats
+ * are hardwired in BE silicon.
+ */
+/* Event Queue Descriptor */
+#define EQ_ENTRY_VALID_MASK 		0x1	/* bit 0 */
+#define EQ_ENTRY_RES_ID_MASK 		0xFFFF	/* bits 16 - 31 */
+#define EQ_ENTRY_RES_ID_SHIFT 		16
+struct be_eq_entry {
+	u32 evt;
+};
+
+/* TX Queue Descriptor */
+#define ETH_WRB_FRAG_LEN_MASK		0xFFFF
+struct be_eth_wrb {
+	u32 frag_pa_hi;		/* dword 0 */
+	u32 frag_pa_lo;		/* dword 1 */
+	u32 rsvd0;		/* dword 2 */
+	u32 frag_len;		/* dword 3: bits 0 - 15 */
+} __packed;
+
+/* Pseudo amap definition for eth_hdr_wrb in which each bit of the
+ * actual structure is defined as a byte : used to calculate
+ * offset/shift/mask of each field */
+struct amap_eth_hdr_wrb {
+	u8 rsvd0[32];		/* dword 0 */
+	u8 rsvd1[32];		/* dword 1 */
+	u8 complete;		/* dword 2 */
+	u8 event;
+	u8 crc;
+	u8 forward;
+	u8 ipsec;
+	u8 mgmt;
+	u8 ipcs;
+	u8 udpcs;
+	u8 tcpcs;
+	u8 lso;
+	u8 vlan;
+	u8 gso[2];
+	u8 num_wrb[5];
+	u8 lso_mss[14];
+	u8 len[16];		/* dword 3 */
+	u8 vlan_tag[16];
+} __packed;
+
+struct be_eth_hdr_wrb {
+	u32 dw[4];
+};
+
+/* TX Compl Queue Descriptor */
+
+/* Pseudo amap definition for eth_tx_compl in which each bit of the
+ * actual structure is defined as a byte: used to calculate
+ * offset/shift/mask of each field */
+struct amap_eth_tx_compl {
+	u8 wrb_index[16];	/* dword 0 */
+	u8 ct[2]; 		/* dword 0 */
+	u8 port[2];		/* dword 0 */
+	u8 rsvd0[8];		/* dword 0 */
+	u8 status[4];		/* dword 0 */
+	u8 user_bytes[16];	/* dword 1 */
+	u8 nwh_bytes[8];	/* dword 1 */
+	u8 lso;			/* dword 1 */
+	u8 cast_enc[2];		/* dword 1 */
+	u8 rsvd1[5];		/* dword 1 */
+	u8 rsvd2[32];		/* dword 2 */
+	u8 pkts[16];		/* dword 3 */
+	u8 ringid[11];		/* dword 3 */
+	u8 hash_val[4];		/* dword 3 */
+	u8 valid;		/* dword 3 */
+} __packed;
+
+struct be_eth_tx_compl {
+	u32 dw[4];
+};
+
+/* RX Queue Descriptor */
+struct be_eth_rx_d {
+	u32 fragpa_hi;
+	u32 fragpa_lo;
+};
+
+/* RX Compl Queue Descriptor */
+
+/* Pseudo amap definition for eth_rx_compl in which each bit of the
+ * actual structure is defined as a byte: used to calculate
+ * offset/shift/mask of each field */
+struct amap_eth_rx_compl {
+	u8 vlan_tag[16];	/* dword 0 */
+	u8 pktsize[14];		/* dword 0 */
+	u8 port;		/* dword 0 */
+	u8 ip_opt;		/* dword 0 */
+	u8 err;			/* dword 1 */
+	u8 rsshp;		/* dword 1 */
+	u8 ipf;			/* dword 1 */
+	u8 tcpf;		/* dword 1 */
+	u8 udpf;		/* dword 1 */
+	u8 ipcksm;		/* dword 1 */
+	u8 l4_cksm;		/* dword 1 */
+	u8 ip_version;		/* dword 1 */
+	u8 macdst[6];		/* dword 1 */
+	u8 vtp;			/* dword 1 */
+	u8 rsvd0;		/* dword 1 */
+	u8 fragndx[10];		/* dword 1 */
+	u8 ct[2];		/* dword 1 */
+	u8 sw;			/* dword 1 */
+	u8 numfrags[3];		/* dword 1 */
+	u8 rss_flush;		/* dword 2 */
+	u8 cast_enc[2];		/* dword 2 */
+	u8 qnq;			/* dword 2 */
+	u8 rss_bank;		/* dword 2 */
+	u8 rsvd1[23];		/* dword 2 */
+	u8 lro_pkt;		/* dword 2 */
+	u8 rsvd2[2];		/* dword 2 */
+	u8 valid;		/* dword 2 */
+	u8 rsshash[32];		/* dword 3 */
+} __packed;
+
+struct be_eth_rx_compl {
+	u32 dw[4];
+};