Merge branch 'liquidio-next'

Raghu Vatsavayi says:

====================
liquidio: updates and bug fixes

Please consider following patch series for liquidio bug fixes
and updates on top of net-next. Following patches should be
applied in the following order as some of them depend on
earlier patches in the series.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/cavium/liquidio/cn66xx_device.c

@@ -367,7 +367,8 @@ void lio_cn6xxx_enable_io_queues(struct octeon_device *oct)
 
 void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 {
-	u32 mask, i, loop = HZ;
+	int i;
+	u32 mask, loop = HZ;
 	u32 d32;
 
 	/* Reset the Enable bits for Input Queues. */
@@ -376,7 +377,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 	octeon_write_csr(oct, CN6XXX_SLI_PKT_INSTR_ENB, mask);
 
 	/* Wait until hardware indicates that the queues are out of reset. */
-	mask = oct->io_qmask.iq;
+	mask = (u32)oct->io_qmask.iq;
 	d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ);
 	while (((d32 & mask) != mask) && loop--) {
 		d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ);
@@ -384,8 +385,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 	}
 
 	/* Reset the doorbell register for each Input queue. */
-	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) {
-		if (!(oct->io_qmask.iq & (1UL << i)))
+	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.iq & (1ULL << i)))
 			continue;
 		octeon_write_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i), 0xFFFFFFFF);
 		d32 = octeon_read_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i));
@@ -398,7 +399,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 
 	/* Wait until hardware indicates that the queues are out of reset. */
 	loop = HZ;
-	mask = oct->io_qmask.oq;
+	mask = (u32)oct->io_qmask.oq;
 	d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ);
 	while (((d32 & mask) != mask) && loop--) {
 		d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ);
@@ -408,8 +409,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 
 	/* Reset the doorbell register for each Output queue. */
 	/* for (i = 0; i < oct->num_oqs; i++) { */
-	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) {
-		if (!(oct->io_qmask.oq & (1UL << i)))
+	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.oq & (1ULL << i)))
 			continue;
 		octeon_write_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i), 0xFFFFFFFF);
 		d32 = octeon_read_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i));
@@ -429,16 +430,16 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
 
 void lio_cn6xxx_reinit_regs(struct octeon_device *oct)
 {
-	u32 i;
+	int i;
 
-	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) {
-		if (!(oct->io_qmask.iq & (1UL << i)))
+	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.iq & (1ULL << i)))
 			continue;
 		oct->fn_list.setup_iq_regs(oct, i);
 	}
 
-	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) {
-		if (!(oct->io_qmask.oq & (1UL << i)))
+	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.oq & (1ULL << i)))
 			continue;
 		oct->fn_list.setup_oq_regs(oct, i);
 	}
@@ -450,8 +451,8 @@ void lio_cn6xxx_reinit_regs(struct octeon_device *oct)
 	oct->fn_list.enable_io_queues(oct);
 
 	/* for (i = 0; i < oct->num_oqs; i++) { */
-	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) {
-		if (!(oct->io_qmask.oq & (1UL << i)))
+	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
+		if (!(oct->io_qmask.oq & (1ULL << i)))
 			continue;
 		writel(oct->droq[i]->max_count, oct->droq[i]->pkts_credit_reg);
 	}
@@ -495,8 +496,7 @@ u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx)
 }
 
 u32
-lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)),
-			     struct octeon_instr_queue *iq)
+lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq)
 {
 	u32 new_idx = readl(iq->inst_cnt_reg);
 
@@ -557,7 +557,8 @@ lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64)
 int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct)
 {
 	struct octeon_droq *droq;
-	u32 oq_no, pkt_count, droq_time_mask, droq_mask, droq_int_enb;
+	int oq_no;
+	u32 pkt_count, droq_time_mask, droq_mask, droq_int_enb;
 	u32 droq_cnt_enb, droq_cnt_mask;
 
 	droq_cnt_enb = octeon_read_csr(oct, CN6XXX_SLI_PKT_CNT_INT_ENB);
@@ -573,8 +574,8 @@ int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct)
 	oct->droq_intr = 0;
 
 	/* for (oq_no = 0; oq_no < oct->num_oqs; oq_no++) { */
-	for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES; oq_no++) {
-		if (!(droq_mask & (1 << oq_no)))
+	for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); oq_no++) {
+		if (!(droq_mask & (1ULL << oq_no)))
 			continue;
 
 		droq = oct->droq[oq_no];

drivers/net/ethernet/cavium/liquidio/cn66xx_device.h

@@ -91,8 +91,7 @@ void lio_cn6xxx_bar1_idx_setup(struct octeon_device *oct, u64 core_addr,
 void lio_cn6xxx_bar1_idx_write(struct octeon_device *oct, u32 idx, u32 mask);
 u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx);
 u32
-lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)),
-			     struct octeon_instr_queue *iq);
+lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq);
 void lio_cn6xxx_enable_interrupt(void *chip);
 void lio_cn6xxx_disable_interrupt(void *chip);
 void cn6xxx_get_pcie_qlmport(struct octeon_device *oct);

drivers/net/ethernet/cavium/liquidio/liquidio_common.h

@@ -30,11 +30,10 @@
 
 #include "octeon_config.h"
 
-#define LIQUIDIO_VERSION        "1.1.9"
-#define LIQUIDIO_MAJOR_VERSION  1
-#define LIQUIDIO_MINOR_VERSION  1
-#define LIQUIDIO_MICRO_VERSION  9
-
+#define LIQUIDIO_BASE_VERSION   "1.4"
+#define LIQUIDIO_MICRO_VERSION  ".1"
+#define LIQUIDIO_PACKAGE ""
+#define LIQUIDIO_VERSION  "1.4.1"
 #define CONTROL_IQ 0
 /** Tag types used by Octeon cores in its work. */
 enum octeon_tag_type {
@@ -214,6 +213,10 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
 #define   OCTNET_CMD_VERBOSE_ENABLE   0x14
 #define   OCTNET_CMD_VERBOSE_DISABLE  0x15
 
+#define   OCTNET_CMD_ENABLE_VLAN_FILTER 0x16
+#define   OCTNET_CMD_ADD_VLAN_FILTER  0x17
+#define   OCTNET_CMD_DEL_VLAN_FILTER  0x18
+
 /* RX(packets coming from wire) Checksum verification flags */
 /* TCP/UDP csum */
 #define   CNNIC_L4SUM_VERIFIED             0x1
@@ -482,15 +485,15 @@ struct octeon_instr_irh {
 	u64 opcode:4;
 	u64 rflag:1;
 	u64 subcode:7;
-	u64 len:3;
-	u64 rid:13;
-	u64 reserved:4;
+	u64 vlan:12;
+	u64 priority:3;
+	u64 reserved:5;
 	u64 ossp:32;             /* opcode/subcode specific parameters */
 #else
 	u64 ossp:32;             /* opcode/subcode specific parameters */
-	u64 reserved:4;
-	u64 rid:13;
-	u64 len:3;
+	u64 reserved:5;
+	u64 priority:3;
+	u64 vlan:12;
 	u64 subcode:7;
 	u64 rflag:1;
 	u64 opcode:4;
@@ -517,28 +520,27 @@ union octeon_rh {
 	struct {
 		u64 opcode:4;
 		u64 subcode:8;
-		u64 len:3;       /** additional 64-bit words */
-		u64 rid:13;      /** request id in response to pkt sent by host */
-		u64 reserved:4;
-		u64 ossp:32;     /** opcode/subcode specific parameters */
+		u64 len:3;     /** additional 64-bit words */
+		u64 reserved:17;
+		u64 ossp:32;   /** opcode/subcode specific parameters */
 	} r;
 	struct {
 		u64 opcode:4;
 		u64 subcode:8;
-		u64 len:3;       /** additional 64-bit words */
-		u64 rid:13;      /** request id in response to pkt sent by host */
-		u64 extra:24;
-		u64 link:8;
+		u64 len:3;     /** additional 64-bit words */
+		u64 extra:28;
+		u64 vlan:12;
+		u64 priority:3;
 		u64 csum_verified:3;     /** checksum verified. */
 		u64 has_hwtstamp:1;      /** Has hardware timestamp. 1 = yes. */
 	} r_dh;
 	struct {
 		u64 opcode:4;
 		u64 subcode:8;
-		u64 len:3;       /** additional 64-bit words */
-		u64 rid:13;      /** request id in response to pkt sent by host */
+		u64 len:3;     /** additional 64-bit words */
+		u64 reserved:11;
 		u64 num_gmx_ports:8;
-		u64 max_nic_ports:8;
+		u64 max_nic_ports:10;
 		u64 app_cap_flags:4;
 		u64 app_mode:16;
 	} r_core_drv_init;
@@ -554,8 +556,7 @@ union octeon_rh {
 	u64 u64;
 	struct {
 		u64 ossp:32;  /** opcode/subcode specific parameters */
-		u64 reserved:4;
-		u64 rid:13;   /** req id in response to pkt sent by host */
+		u64 reserved:17;
 		u64 len:3;    /** additional 64-bit words */
 		u64 subcode:8;
 		u64 opcode:4;
@@ -563,9 +564,9 @@ union octeon_rh {
 	struct {
 		u64 has_hwtstamp:1;      /** 1 = has hwtstamp */
 		u64 csum_verified:3;     /** checksum verified. */
-		u64 link:8;
-		u64 extra:24;
-		u64 rid:13;   /** req id in response to pkt sent by host */
+		u64 priority:3;
+		u64 vlan:12;
+		u64 extra:28;
 		u64 len:3;    /** additional 64-bit words */
 		u64 subcode:8;
 		u64 opcode:4;
@@ -573,9 +574,9 @@ union octeon_rh {
 	struct {
 		u64 app_mode:16;
 		u64 app_cap_flags:4;
-		u64 max_nic_ports:8;
+		u64 max_nic_ports:10;
 		u64 num_gmx_ports:8;
-		u64 rid:13;
+		u64 reserved:11;
 		u64 len:3;       /** additional 64-bit words */
 		u64 subcode:8;
 		u64 opcode:4;
@@ -627,13 +628,13 @@ union oct_link_status {
 		u64 speed:16;
 		u64 link_up:1;
 		u64 autoneg:1;
-		u64 interface:4;
+		u64 if_mode:5;
 		u64 pause:1;
-		u64 reserved:17;
+		u64 reserved:16;
 #else
-		u64 reserved:17;
+		u64 reserved:16;
 		u64 pause:1;
-		u64 interface:4;
+		u64 if_mode:5;
 		u64 autoneg:1;
 		u64 link_up:1;
 		u64 speed:16;
@@ -710,6 +711,7 @@ struct liquidio_if_cfg_info {
 	u64 iqmask; /** mask for IQs enabled for  the port */
 	u64 oqmask; /** mask for OQs enabled for the port */
 	struct oct_link_info linfo; /** initial link information */
+	char   liquidio_firmware_version[32];
 };
 
 /** Stats for each NIC port in RX direction. */
@@ -734,10 +736,16 @@ struct nic_rx_stats {
 	u64 fw_err_pko;
 	u64 fw_err_link;
 	u64 fw_err_drop;
+
+	/* LRO */
 	u64 fw_lro_pkts;   /* Number of packets that are LROed      */
 	u64 fw_lro_octs;   /* Number of octets that are LROed       */
 	u64 fw_total_lro;  /* Number of LRO packets formed          */
 	u64 fw_lro_aborts; /* Number of times lRO of packet aborted */
+	u64 fw_lro_aborts_port;
+	u64 fw_lro_aborts_seq;
+	u64 fw_lro_aborts_tsval;
+	u64 fw_lro_aborts_timer;
 	/* intrmod: packet forward rate */
 	u64 fwd_rate;
 };
@@ -761,9 +769,13 @@ struct nic_tx_stats {
 	/* firmware stats */
 	u64 fw_total_sent;
 	u64 fw_total_fwd;
+	u64 fw_total_fwd_bytes;
 	u64 fw_err_pko;
 	u64 fw_err_link;
 	u64 fw_err_drop;
+	u64 fw_err_tso;
+	u64 fw_tso;		/* number of tso requests */
+	u64 fw_tso_fwd;		/* number of packets segmented in tso */
 };
 
 struct oct_link_stats {
@@ -794,23 +806,44 @@ struct oct_mdio_cmd {
 
 #define OCT_LINK_STATS_SIZE   (sizeof(struct oct_link_stats))
 
+/* intrmod: max. packet rate threshold */
+#define LIO_INTRMOD_MAXPKT_RATETHR	196608
+/* intrmod: min. packet rate threshold */
+#define LIO_INTRMOD_MINPKT_RATETHR	9216
+/* intrmod: max. packets to trigger interrupt */
+#define LIO_INTRMOD_RXMAXCNT_TRIGGER	384
+/* intrmod: min. packets to trigger interrupt */
+#define LIO_INTRMOD_RXMINCNT_TRIGGER	1
+/* intrmod: max. time to trigger interrupt */
+#define LIO_INTRMOD_RXMAXTMR_TRIGGER	128
+/* 66xx:intrmod: min. time to trigger interrupt
+ * (value of 1 is optimum for TCP_RR)
+ */
+#define LIO_INTRMOD_RXMINTMR_TRIGGER	1
+
+/* intrmod: max. packets to trigger interrupt */
+#define LIO_INTRMOD_TXMAXCNT_TRIGGER	64
+/* intrmod: min. packets to trigger interrupt */
+#define LIO_INTRMOD_TXMINCNT_TRIGGER	0
+
+/* intrmod: poll interval in seconds */
 #define LIO_INTRMOD_CHECK_INTERVAL  1
-#define LIO_INTRMOD_MAXPKT_RATETHR  196608 /* max pkt rate threshold */
-#define LIO_INTRMOD_MINPKT_RATETHR  9216   /* min pkt rate threshold */
-#define LIO_INTRMOD_MAXCNT_TRIGGER  384    /* max pkts to trigger interrupt */
-#define LIO_INTRMOD_MINCNT_TRIGGER  1      /* min pkts to trigger interrupt */
-#define LIO_INTRMOD_MAXTMR_TRIGGER  128    /* max time to trigger interrupt */
-#define LIO_INTRMOD_MINTMR_TRIGGER  32     /* min time to trigger interrupt */
 
 struct oct_intrmod_cfg {
-	u64 intrmod_enable;
-	u64 intrmod_check_intrvl;
-	u64 intrmod_maxpkt_ratethr;
-	u64 intrmod_minpkt_ratethr;
-	u64 intrmod_maxcnt_trigger;
-	u64 intrmod_maxtmr_trigger;
-	u64 intrmod_mincnt_trigger;
-	u64 intrmod_mintmr_trigger;
+	u64 rx_enable;
+	u64 tx_enable;
+	u64 check_intrvl;
+	u64 maxpkt_ratethr;
+	u64 minpkt_ratethr;
+	u64 rx_maxcnt_trigger;
+	u64 rx_mincnt_trigger;
+	u64 rx_maxtmr_trigger;
+	u64 rx_mintmr_trigger;
+	u64 tx_mincnt_trigger;
+	u64 tx_maxcnt_trigger;
+	u64 rx_frames;
+	u64 tx_frames;
+	u64 rx_usecs;
 };
 
 #define BASE_QUEUE_NOT_REQUESTED 65535

drivers/net/ethernet/cavium/liquidio/octeon_config.h

@@ -37,7 +37,7 @@
 /* Maximum octeon devices defined as MAX_OCTEON_NICIF to support
  * multiple(<= MAX_OCTEON_NICIF) Miniports
  */
-#define   MAX_OCTEON_NICIF             32
+#define   MAX_OCTEON_NICIF             128
 #define   MAX_OCTEON_DEVICES           MAX_OCTEON_NICIF
 #define   MAX_OCTEON_LINKS	       MAX_OCTEON_NICIF
 #define   MAX_OCTEON_MULTICAST_ADDR    32
@@ -135,7 +135,7 @@
 #define CFG_GET_IS_SLI_BP_ON(cfg)                ((cfg)->misc.enable_sli_oq_bp)
 
 /* Max IOQs per OCTEON Link */
-#define MAX_IOQS_PER_NICIF              32
+#define MAX_IOQS_PER_NICIF              64
 
 enum lio_card_type {
 	LIO_210SV = 0, /* Two port, 66xx */
@@ -416,9 +416,11 @@ struct octeon_config {
 #define DISPATCH_LIST_SIZE                      BIT(OPCODE_MASK_BITS)
 
 /* Maximum number of Octeon Instruction (command) queues */
-#define MAX_OCTEON_INSTR_QUEUES         CN6XXX_MAX_INPUT_QUEUES
+#define MAX_OCTEON_INSTR_QUEUES(oct)         CN6XXX_MAX_INPUT_QUEUES
+/* Maximum number of Octeon Output queues */
+#define MAX_OCTEON_OUTPUT_QUEUES(oct)         CN6XXX_MAX_OUTPUT_QUEUES
 
-/* Maximum number of Octeon Instruction (command) queues */
-#define MAX_OCTEON_OUTPUT_QUEUES        CN6XXX_MAX_OUTPUT_QUEUES
+#define MAX_POSSIBLE_OCTEON_INSTR_QUEUES       CN6XXX_MAX_INPUT_QUEUES
+#define MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES      CN6XXX_MAX_OUTPUT_QUEUES
 
 #endif /* __OCTEON_CONFIG_H__  */

drivers/net/ethernet/cavium/liquidio/octeon_device.c

@@ -549,17 +549,19 @@ static char *get_oct_app_string(u32 app_mode)
 	return oct_dev_app_str[CVM_DRV_INVALID_APP - CVM_DRV_APP_START];
 }
 
+u8 fbuf[4 * 1024 * 1024];
+
 int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 			     size_t size)
 {
 	int ret = 0;
-	u8 *p;
-	u8 *buffer;
+	u8 *p = fbuf;
 	u32 crc32_result;
 	u64 load_addr;
 	u32 image_len;
 	struct octeon_firmware_file_header *h;
-	u32 i;
+	u32 i, rem, base_len = strlen(LIQUIDIO_BASE_VERSION);
+	char *base;
 
 	if (size < sizeof(struct octeon_firmware_file_header)) {
 		dev_err(&oct->pci_dev->dev, "Firmware file too small (%d < %d).\n",
@@ -575,19 +577,26 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 		return -EINVAL;
 	}
 
-	crc32_result =
-		crc32(~0, data,
-		      sizeof(struct octeon_firmware_file_header) -
-		      sizeof(u32)) ^ ~0U;
+	crc32_result = crc32((unsigned int)~0, data,
+			     sizeof(struct octeon_firmware_file_header) -
+			     sizeof(u32)) ^ ~0U;
 	if (crc32_result != be32_to_cpu(h->crc32)) {
 		dev_err(&oct->pci_dev->dev, "Firmware CRC mismatch (0x%08x != 0x%08x).\n",
 			crc32_result, be32_to_cpu(h->crc32));
 		return -EINVAL;
 	}
 
-	if (memcmp(LIQUIDIO_VERSION, h->version, strlen(LIQUIDIO_VERSION))) {
-		dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s, got %s.\n",
-			LIQUIDIO_VERSION, h->version);
+	if (strncmp(LIQUIDIO_PACKAGE, h->version, strlen(LIQUIDIO_PACKAGE))) {
+		dev_err(&oct->pci_dev->dev, "Unmatched firmware package type. Expected %s, got %s.\n",
+			LIQUIDIO_PACKAGE, h->version);
+		return -EINVAL;
+	}
+
+	base = h->version + strlen(LIQUIDIO_PACKAGE);
+	ret = memcmp(LIQUIDIO_BASE_VERSION, base, base_len);
+	if (ret) {
+		dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s.x, got %s.\n",
+			LIQUIDIO_BASE_VERSION, base);
 		return -EINVAL;
 	}
 
@@ -601,56 +610,56 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data,
 	snprintf(oct->fw_info.liquidio_firmware_version, 32, "LIQUIDIO: %s",
 		 h->version);
 
-	buffer = kmemdup(data, size, GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	p = buffer + sizeof(struct octeon_firmware_file_header);
+	data += sizeof(struct octeon_firmware_file_header);
 
+	dev_info(&oct->pci_dev->dev, "%s: Loading %d images\n", __func__,
+		 be32_to_cpu(h->num_images));
 	/* load all images */
 	for (i = 0; i < be32_to_cpu(h->num_images); i++) {
 		load_addr = be64_to_cpu(h->desc[i].addr);
 		image_len = be32_to_cpu(h->desc[i].len);
 
-		/* validate the image */
-		crc32_result = crc32(~0, p, image_len) ^ ~0U;
-		if (crc32_result != be32_to_cpu(h->desc[i].crc32)) {
-			dev_err(&oct->pci_dev->dev,
-				"Firmware CRC mismatch in image %d (0x%08x != 0x%08x).\n",
-				i, crc32_result,
-				be32_to_cpu(h->desc[i].crc32));
-			ret = -EINVAL;
-			goto done_downloading;
-		}
+		dev_info(&oct->pci_dev->dev, "Loading firmware %d at %llx\n",
+			 image_len, load_addr);
 
-		/* download the image */
-		octeon_pci_write_core_mem(oct, load_addr, p, image_len);
+		/* Write in 4MB chunks*/
+		rem = image_len;
 
-		p += image_len;
-		dev_dbg(&oct->pci_dev->dev,
-			"Downloaded image %d (%d bytes) to address 0x%016llx\n",
-			i, image_len, load_addr);
+		while (rem) {
+			if (rem < (4 * 1024 * 1024))
+				size = rem;
+			else
+				size = 4 * 1024 * 1024;
+
+			memcpy(p, data, size);
+
+			/* download the image */
+			octeon_pci_write_core_mem(oct, load_addr, p, (u32)size);
+
+			data += size;
+			rem -= (u32)size;
+			load_addr += size;
+		}
 	}
+	dev_info(&oct->pci_dev->dev, "Writing boot command: %s\n",
+		 h->bootcmd);
 
 	/* Invoke the bootcmd */
 	ret = octeon_console_send_cmd(oct, h->bootcmd, 50);
 
-done_downloading:
-	kfree(buffer);
-
-	return ret;
+	return 0;
 }
 
 void octeon_free_device_mem(struct octeon_device *oct)
 {
 	u32 i;
 
-	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) {
+	for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
 		/* could check  mask as well */
 		vfree(oct->droq[i]);
 	}
 
-	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) {
+	for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
 		/* could check mask as well */
 		vfree(oct->instr_queue[i]);
 	}
@@ -734,7 +743,7 @@ struct octeon_device *octeon_allocate_device(u32 pci_id,
 	octeon_device[oct_idx] = oct;
 
 	oct->octeon_id = oct_idx;
-	snprintf((oct->device_name), sizeof(oct->device_name),
+	snprintf(oct->device_name, sizeof(oct->device_name),
 		 "LiquidIO%d", (oct->octeon_id));
 
 	return oct;
@@ -1157,8 +1166,8 @@ int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf)
 int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
 
 {
-	if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES) &&
-	    (oct->io_qmask.iq & (1UL << q_no)))
+	if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES(oct)) &&
+	    (oct->io_qmask.iq & (1ULL << q_no)))
 		return oct->instr_queue[q_no]->max_count;
 
 	return -1;
@@ -1166,8 +1175,8 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
 
 int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no)
 {
-	if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES) &&
-	    (oct->io_qmask.oq & (1UL << q_no)))
+	if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES(oct)) &&
+	    (oct->io_qmask.oq & (1ULL << q_no)))
 		return oct->droq[q_no]->max_count;
 	return -1;
 }
@@ -1258,10 +1267,10 @@ void lio_pci_writeq(struct octeon_device *oct,
 int octeon_mem_access_ok(struct octeon_device *oct)
 {
 	u64 access_okay = 0;
+	u64 lmc0_reset_ctl;
 
 	/* Check to make sure a DDR interface is enabled */
-	u64 lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL);
-
+	lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL);
 	access_okay = (lmc0_reset_ctl & CN6XXX_LMC0_RESET_CTL_DDR3RST_MASK);
 
 	return access_okay ? 0 : 1;
@@ -1275,9 +1284,6 @@ int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout)
 	if (!timeout)
 		return ret;
 
-	while (*timeout == 0)
-		schedule_timeout_uninterruptible(HZ / 10);
-
 	for (ms = 0; (ret != 0) && ((*timeout == 0) || (ms <= *timeout));
 	     ms += HZ / 10) {
 		ret = octeon_mem_access_ok(oct);

drivers/net/ethernet/cavium/liquidio/octeon_device.h

@@ -152,9 +152,9 @@ struct octeon_mmio {
 #define   MAX_OCTEON_MAPS    32
 
 struct octeon_io_enable {
-	u32 iq;
-	u32 oq;
-	u32 iq64B;
+	u64 iq;
+	u64 oq;
+	u64 iq64B;
 };
 
 struct octeon_reg_list {
@@ -204,8 +204,7 @@ struct octeon_fn_list {
 	void (*bar1_idx_setup)(struct octeon_device *, u64, u32, int);
 	void (*bar1_idx_write)(struct octeon_device *, u32, u32);
 	u32 (*bar1_idx_read)(struct octeon_device *, u32);
-	u32 (*update_iq_read_idx)(struct octeon_device *,
-				  struct octeon_instr_queue *);
+	u32 (*update_iq_read_idx)(struct octeon_instr_queue *);
 
 	void (*enable_oq_pkt_time_intr)(struct octeon_device *, u32);
 	void (*disable_oq_pkt_time_intr)(struct octeon_device *, u32);
@@ -267,6 +266,7 @@ struct octdev_props {
 	/* Each interface in the Octeon device has a network
 	 * device pointer (used for OS specific calls).
 	 */
+	int    napi_enabled;
 	int    gmxport;
 	struct net_device *netdev;
 };
@@ -325,7 +325,8 @@ struct octeon_device {
 	struct octeon_sc_buffer_pool	sc_buf_pool;
 
 	/** The input instruction queues */
-	struct octeon_instr_queue *instr_queue[MAX_OCTEON_INSTR_QUEUES];
+	struct octeon_instr_queue *instr_queue
+		[MAX_POSSIBLE_OCTEON_INSTR_QUEUES];
 
 	/** The doubly-linked list of instruction response */
 	struct octeon_response_list response_list[MAX_RESPONSE_LISTS];
@@ -333,7 +334,7 @@ struct octeon_device {
 	u32 num_oqs;
 
 	/** The DROQ output queues  */
-	struct octeon_droq *droq[MAX_OCTEON_OUTPUT_QUEUES];
+	struct octeon_droq *droq[MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES];
 
 	struct octeon_io_enable io_qmask;
 
@@ -382,15 +383,29 @@ struct octeon_device {
 
 	struct cavium_wq dma_comp_wq;
 
-	struct cavium_wq check_db_wq[MAX_OCTEON_INSTR_QUEUES];
+	/** Lock for dma response list */
+	spinlock_t cmd_resp_wqlock;
+	u32 cmd_resp_state;
+
+	struct cavium_wq check_db_wq[MAX_POSSIBLE_OCTEON_INSTR_QUEUES];
 
 	struct cavium_wk nic_poll_work;
 
 	struct cavium_wk console_poll_work[MAX_OCTEON_MAPS];
 
 	void *priv;
+
+	int rx_pause;
+	int tx_pause;
+
+	struct oct_link_stats link_stats; /*stastics from firmware*/
+
+	/* private flags to control driver-specific features through ethtool */
+	u32 priv_flags;
 };
 
+#define  OCT_DRV_ONLINE 1
+#define  OCT_DRV_OFFLINE 2
 #define  OCTEON_CN6XXX(oct)           ((oct->chip_id == OCTEON_CN66XX) || \
 				       (oct->chip_id == OCTEON_CN68XX))
 #define CHIP_FIELD(oct, TYPE, field)             \
@@ -647,4 +662,17 @@ void *oct_get_config_info(struct octeon_device *oct, u16 card_type);
  */
 struct octeon_config *octeon_get_conf(struct octeon_device *oct);
 
+/* LiquidIO driver pivate flags */
+enum {
+	OCT_PRIV_FLAG_TX_BYTES = 0, /* Tx interrupts by pending byte count */
+};
+
+static inline void lio_set_priv_flag(struct octeon_device *octdev, u32 flag,
+				     u32 val)
+{
+	if (val)
+		octdev->priv_flags |= (0x1 << flag);
+	else
+		octdev->priv_flags &= ~(0x1 << flag);
+}
 #endif

drivers/net/ethernet/cavium/liquidio/octeon_droq.c

@@ -337,7 +337,7 @@ int octeon_init_droq(struct octeon_device *oct,
 	/* For 56xx Pass1, this function won't be called, so no checks. */
 	oct->fn_list.setup_oq_regs(oct, q_no);
 
-	oct->io_qmask.oq |= (1 << q_no);
+	oct->io_qmask.oq |= (1ULL << q_no);
 
 	return 0;
 

drivers/net/ethernet/cavium/liquidio/octeon_iq.h

@@ -65,6 +65,10 @@ struct oct_iq_stats {
 	u64 tx_iq_busy;/**< Numof times this iq was found to be full. */
 	u64 tx_dropped;/**< Numof pkts dropped dueto xmitpath errors. */
 	u64 tx_tot_bytes;/**< Total count of bytes sento to network. */
+	u64 tx_gso;  /* count of tso */
+	u64 tx_dmamap_fail;
+	u64 tx_restart;
+	/*u64 tx_timeout_count;*/
 };
 
 #define OCT_IQ_STATS_SIZE   (sizeof(struct oct_iq_stats))
@@ -80,6 +84,12 @@ struct octeon_instr_queue {
 	/** A spinlock to protect access to the input ring.  */
 	spinlock_t lock;
 
+	/** A spinlock to protect while posting on the ring.  */
+	spinlock_t post_lock;
+
+	/** A spinlock to protect access to the input ring.*/
+	spinlock_t iq_flush_running_lock;
+
 	/** Flag that indicates if the queue uses 64 byte commands. */
 	u32 iqcmd_64B:1;
 
@@ -244,7 +254,7 @@ union octeon_instr_64B {
 
 /** The size of each buffer in soft command buffer pool
  */
-#define  SOFT_COMMAND_BUFFER_SIZE	1024
+#define  SOFT_COMMAND_BUFFER_SIZE	1536
 
 struct octeon_soft_command {
 	/** Soft command buffer info. */
@@ -282,7 +292,7 @@ struct octeon_soft_command {
 
 /** Maximum number of buffers to allocate into soft command buffer pool
  */
-#define  MAX_SOFT_COMMAND_BUFFERS	16
+#define  MAX_SOFT_COMMAND_BUFFERS	256
 
 /** Head of a soft command buffer pool.
  */
@@ -339,7 +349,7 @@ octeon_register_reqtype_free_fn(struct octeon_device *oct, int reqtype,
 
 int
 lio_process_iq_request_list(struct octeon_device *oct,
-			    struct octeon_instr_queue *iq);
+			    struct octeon_instr_queue *iq, u32 napi_budget);
 
 int octeon_send_command(struct octeon_device *oct, u32 iq_no,
 			u32 force_db, void *cmd, void *buf,
@@ -357,5 +367,7 @@ int octeon_send_soft_command(struct octeon_device *oct,
 int octeon_setup_iq(struct octeon_device *oct, int ifidx,
 		    int q_index, union oct_txpciq iq_no, u32 num_descs,
 		    void *app_ctx);
-
+int
+octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq,
+		u32 pending_thresh, u32 napi_budget);
 #endif				/* __OCTEON_IQ_H__ */

drivers/net/ethernet/cavium/liquidio/octeon_network.h

@@ -30,6 +30,17 @@
 #include <linux/dma-mapping.h>
 #include <linux/ptp_clock_kernel.h>
 
+struct oct_nic_stats_resp {
+	u64     rh;
+	struct oct_link_stats stats;
+	u64     status;
+};
+
+struct oct_nic_stats_ctrl {
+	struct completion complete;
+	struct net_device *netdev;
+};
+
 /** LiquidIO per-interface network private data */
 struct lio {
 	/** State of the interface. Rx/Tx happens only in the RUNNING state.  */

drivers/net/ethernet/cavium/liquidio/octeon_nic.c

@@ -171,20 +171,36 @@ octnet_send_nic_ctrl_pkt(struct octeon_device *oct,
 	int retval;
 	struct octeon_soft_command *sc = NULL;
 
+	spin_lock_bh(&oct->cmd_resp_wqlock);
+	/* Allow only rx ctrl command to stop traffic on the chip
+	 * during offline operations
+	 */
+	if ((oct->cmd_resp_state == OCT_DRV_OFFLINE) &&
+	    (nctrl->ncmd.s.cmd != OCTNET_CMD_RX_CTL)) {
+		spin_unlock_bh(&oct->cmd_resp_wqlock);
+		dev_err(&oct->pci_dev->dev,
+			"%s cmd:%d not processed since driver offline\n",
+			__func__, nctrl->ncmd.s.cmd);
+		return -1;
+	}
+
 	sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl);
 	if (!sc) {
 		dev_err(&oct->pci_dev->dev, "%s soft command alloc failed\n",
 			__func__);
+		spin_unlock_bh(&oct->cmd_resp_wqlock);
 		return -1;
 	}
 
 	retval = octeon_send_soft_command(oct, sc);
 	if (retval == IQ_SEND_FAILED) {
 		octeon_free_soft_command(oct, sc);
-		dev_err(&oct->pci_dev->dev, "%s soft command send failed status: %x\n",
-			__func__, retval);
+		dev_err(&oct->pci_dev->dev, "%s soft command:%d send failed status: %x\n",
+			__func__, nctrl->ncmd.s.cmd, retval);
+		spin_unlock_bh(&oct->cmd_resp_wqlock);
 		return -1;
 	}
 
+	spin_unlock_bh(&oct->cmd_resp_wqlock);
 	return retval;
 }

drivers/net/ethernet/cavium/liquidio/request_manager.c

@@ -51,7 +51,7 @@ struct iq_post_status {
 };
 
 static void check_db_timeout(struct work_struct *work);
-static void  __check_db_timeout(struct octeon_device *oct, unsigned long iq_no);
+static void  __check_db_timeout(struct octeon_device *oct, u64 iq_no);
 
 static void (*reqtype_free_fn[MAX_OCTEON_DEVICES][REQTYPE_LAST + 1]) (void *);
 
@@ -149,8 +149,11 @@ int octeon_init_instr_queue(struct octeon_device *oct,
 
 	/* Initialize the spinlock for this instruction queue */
 	spin_lock_init(&iq->lock);
+	spin_lock_init(&iq->post_lock);
 
-	oct->io_qmask.iq |= (1 << iq_no);
+	spin_lock_init(&iq->iq_flush_running_lock);
+
+	oct->io_qmask.iq |= (1ULL << iq_no);
 
 	/* Set the 32B/64B mode for each input queue */
 	oct->io_qmask.iq64B |= ((conf->instr_type == 64) << iq_no);
@@ -253,8 +256,8 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct)
 		instr_cnt = 0;
 
 		/*for (i = 0; i < oct->num_iqs; i++) {*/
-		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) {
-			if (!(oct->io_qmask.iq & (1UL << i)))
+		for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
+			if (!(oct->io_qmask.iq & (1ULL << i)))
 				continue;
 			pending =
 			    atomic_read(&oct->
@@ -391,13 +394,13 @@ __add_to_request_list(struct octeon_instr_queue *iq,
 
 int
 lio_process_iq_request_list(struct octeon_device *oct,
-			    struct octeon_instr_queue *iq)
+			    struct octeon_instr_queue *iq, u32 napi_budget)
 {
 	int reqtype;
 	void *buf;
 	u32 old = iq->flush_index;
 	u32 inst_count = 0;
-	unsigned pkts_compl = 0, bytes_compl = 0;
+	unsigned int pkts_compl = 0, bytes_compl = 0;
 	struct octeon_soft_command *sc;
 	struct octeon_instr_irh *irh;
 
@@ -457,6 +460,9 @@ lio_process_iq_request_list(struct octeon_device *oct,
  skip_this:
 		inst_count++;
 		INCR_INDEX_BY1(old, iq->max_count);
+
+		if ((napi_budget) && (inst_count >= napi_budget))
+			break;
 	}
 	if (bytes_compl)
 		octeon_report_tx_completion_to_bql(iq->app_ctx, pkts_compl,
@@ -466,38 +472,63 @@ lio_process_iq_request_list(struct octeon_device *oct,
 	return inst_count;
 }
 
-static inline void
-update_iq_indices(struct octeon_device *oct, struct octeon_instr_queue *iq)
+/* Can only be called from process context */
+int
+octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq,
+		u32 pending_thresh, u32 napi_budget)
 {
 	u32 inst_processed = 0;
+	u32 tot_inst_processed = 0;
+	int tx_done = 1;
 
-	/* Calculate how many commands Octeon has read and move the read index
-	 * accordingly.
-	 */
-	iq->octeon_read_index = oct->fn_list.update_iq_read_idx(oct, iq);
+	if (!spin_trylock(&iq->iq_flush_running_lock))
+		return tx_done;
 
-	/* Move the NORESPONSE requests to the per-device completion list. */
-	if (iq->flush_index != iq->octeon_read_index)
-		inst_processed = lio_process_iq_request_list(oct, iq);
+	spin_lock_bh(&iq->lock);
 
-	if (inst_processed) {
-		atomic_sub(inst_processed, &iq->instr_pending);
-		iq->stats.instr_processed += inst_processed;
-	}
-}
+	iq->octeon_read_index = oct->fn_list.update_iq_read_idx(iq);
 
-static void
-octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq,
-		u32 pending_thresh)
-{
 	if (atomic_read(&iq->instr_pending) >= (s32)pending_thresh) {
-		spin_lock_bh(&iq->lock);
-		update_iq_indices(oct, iq);
-		spin_unlock_bh(&iq->lock);
+		do {
+			/* Process any outstanding IQ packets. */
+			if (iq->flush_index == iq->octeon_read_index)
+				break;
+
+			if (napi_budget)
+				inst_processed = lio_process_iq_request_list
+					(oct, iq,
+					 napi_budget - tot_inst_processed);
+			else
+				inst_processed =
+					lio_process_iq_request_list(oct, iq, 0);
+
+			if (inst_processed) {
+				atomic_sub(inst_processed, &iq->instr_pending);
+				iq->stats.instr_processed += inst_processed;
+			}
+
+			tot_inst_processed += inst_processed;
+			inst_processed = 0;
+
+		} while (tot_inst_processed < napi_budget);
+
+		if (napi_budget && (tot_inst_processed >= napi_budget))
+			tx_done = 0;
 	}
+
+	iq->last_db_time = jiffies;
+
+	spin_unlock_bh(&iq->lock);
+
+	spin_unlock(&iq->iq_flush_running_lock);
+
+	return tx_done;
 }
 
-static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no)
+/* Process instruction queue after timeout.
+ * This routine gets called from a workqueue or when removing the module.
+ */
+static void __check_db_timeout(struct octeon_device *oct, u64 iq_no)
 {
 	struct octeon_instr_queue *iq;
 	u64 next_time;
@@ -508,24 +539,17 @@ static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no)
 	if (!iq)
 		return;
 
+	/* return immediately, if no work pending */
+	if (!atomic_read(&iq->instr_pending))
+		return;
 	/* If jiffies - last_db_time < db_timeout do nothing  */
 	next_time = iq->last_db_time + iq->db_timeout;
 	if (!time_after(jiffies, (unsigned long)next_time))
 		return;
 	iq->last_db_time = jiffies;
 
-	/* Get the lock and prevent tasklets. This routine gets called from
-	 * the poll thread. Instructions can now be posted in tasklet context
-	 */
-	spin_lock_bh(&iq->lock);
-	if (iq->fill_cnt != 0)
-		ring_doorbell(oct, iq);
-
-	spin_unlock_bh(&iq->lock);
-
 	/* Flush the instruction queue */
-	if (iq->do_auto_flush)
-		octeon_flush_iq(oct, iq, 1);
+	octeon_flush_iq(oct, iq, 1, 0);
 }
 
 /* Called by the Poll thread at regular intervals to check the instruction
@@ -550,7 +574,10 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no,
 	struct iq_post_status st;
 	struct octeon_instr_queue *iq = oct->instr_queue[iq_no];
 
-	spin_lock_bh(&iq->lock);
+	/* Get the lock and prevent other tasks and tx interrupt handler from
+	 * running.
+	 */
+	spin_lock_bh(&iq->post_lock);
 
 	st = __post_command2(oct, iq, force_db, cmd);
 
@@ -566,10 +593,13 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no,
 		INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_dropped, 1);
 	}
 
-	spin_unlock_bh(&iq->lock);
+	spin_unlock_bh(&iq->post_lock);
 
-	if (iq->do_auto_flush)
-		octeon_flush_iq(oct, iq, 2);
+	/* This is only done here to expedite packets being flushed
+	 * for cases where there are no IQ completion interrupts.
+	 */
+	/*if (iq->do_auto_flush)*/
+	/*	octeon_flush_iq(oct, iq, 2, 0);*/
 
 	return st.status;
 }

drivers/net/ethernet/cavium/liquidio/response_manager.c

@@ -54,6 +54,7 @@ int octeon_setup_response_list(struct octeon_device *oct)
 		spin_lock_init(&oct->response_list[i].lock);
 		atomic_set(&oct->response_list[i].pending_req_count, 0);
 	}
+	spin_lock_init(&oct->cmd_resp_wqlock);
 
 	oct->dma_comp_wq.wq = alloc_workqueue("dma-comp", WQ_MEM_RECLAIM, 0);
 	if (!oct->dma_comp_wq.wq) {
@@ -64,6 +65,7 @@ int octeon_setup_response_list(struct octeon_device *oct)
 	cwq = &oct->dma_comp_wq;
 	INIT_DELAYED_WORK(&cwq->wk.work, oct_poll_req_completion);
 	cwq->wk.ctxptr = oct;
+	oct->cmd_resp_state = OCT_DRV_ONLINE;
 	queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100));
 
 	return ret;