drivers/net/ethernet/marvell/mvpp2.c

/*
 * Driver for Marvell PPv2 network controller for Armada 375 SoC.
 *
 * Copyright (C) 2014 Marvell
 *
 * Marcin Wojtas <mw@semihalf.com>
 *
 * This file is licensed under the terms of the GNU General Public
 * License version 2. This program is licensed "as is" without any
 * warranty of any kind, whether express or implied.
 */

#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/platform_device.h>
#include <linux/skbuff.h>
#include <linux/inetdevice.h>
#include <linux/mbus.h>
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/cpumask.h>
#include <linux/of.h>
#include <linux/of_irq.h>
#include <linux/of_mdio.h>
#include <linux/of_net.h>
#include <linux/of_address.h>
#include <linux/of_device.h>
#include <linux/phy.h>
#include <linux/clk.h>
#include <linux/hrtimer.h>
#include <linux/ktime.h>
#include <uapi/linux/ppp_defs.h>
#include <net/ip.h>
#include <net/ipv6.h>

/* RX Fifo Registers */
#define MVPP2_RX_DATA_FIFO_SIZE_REG(port)	(0x00 + 4 * (port))
#define MVPP2_RX_ATTR_FIFO_SIZE_REG(port)	(0x20 + 4 * (port))
#define MVPP2_RX_MIN_PKT_SIZE_REG		0x60
#define MVPP2_RX_FIFO_INIT_REG			0x64

/* RX DMA Top Registers */
#define MVPP2_RX_CTRL_REG(port)			(0x140 + 4 * (port))
#define     MVPP2_RX_LOW_LATENCY_PKT_SIZE(s)	(((s) & 0xfff) << 16)
#define     MVPP2_RX_USE_PSEUDO_FOR_CSUM_MASK	BIT(31)
#define MVPP2_POOL_BUF_SIZE_REG(pool)		(0x180 + 4 * (pool))
#define     MVPP2_POOL_BUF_SIZE_OFFSET		5
#define MVPP2_RXQ_CONFIG_REG(rxq)		(0x800 + 4 * (rxq))
#define     MVPP2_SNOOP_PKT_SIZE_MASK		0x1ff
#define     MVPP2_SNOOP_BUF_HDR_MASK		BIT(9)
#define     MVPP2_RXQ_POOL_SHORT_OFFS		20
#define     MVPP21_RXQ_POOL_SHORT_MASK		0x700000
#define     MVPP22_RXQ_POOL_SHORT_MASK		0xf00000
#define     MVPP2_RXQ_POOL_LONG_OFFS		24
#define     MVPP21_RXQ_POOL_LONG_MASK		0x7000000
#define     MVPP22_RXQ_POOL_LONG_MASK		0xf000000
#define     MVPP2_RXQ_PACKET_OFFSET_OFFS	28
#define     MVPP2_RXQ_PACKET_OFFSET_MASK	0x70000000
#define     MVPP2_RXQ_DISABLE_MASK		BIT(31)

/* Parser Registers */
#define MVPP2_PRS_INIT_LOOKUP_REG		0x1000
#define     MVPP2_PRS_PORT_LU_MAX		0xf
#define     MVPP2_PRS_PORT_LU_MASK(port)	(0xff << ((port) * 4))
#define     MVPP2_PRS_PORT_LU_VAL(port, val)	((val) << ((port) * 4))
#define MVPP2_PRS_INIT_OFFS_REG(port)		(0x1004 + ((port) & 4))
#define     MVPP2_PRS_INIT_OFF_MASK(port)	(0x3f << (((port) % 4) * 8))
#define     MVPP2_PRS_INIT_OFF_VAL(port, val)	((val) << (((port) % 4) * 8))
#define MVPP2_PRS_MAX_LOOP_REG(port)		(0x100c + ((port) & 4))
#define     MVPP2_PRS_MAX_LOOP_MASK(port)	(0xff << (((port) % 4) * 8))
#define     MVPP2_PRS_MAX_LOOP_VAL(port, val)	((val) << (((port) % 4) * 8))
#define MVPP2_PRS_TCAM_IDX_REG			0x1100
#define MVPP2_PRS_TCAM_DATA_REG(idx)		(0x1104 + (idx) * 4)
#define     MVPP2_PRS_TCAM_INV_MASK		BIT(31)
#define MVPP2_PRS_SRAM_IDX_REG			0x1200
#define MVPP2_PRS_SRAM_DATA_REG(idx)		(0x1204 + (idx) * 4)
#define MVPP2_PRS_TCAM_CTRL_REG			0x1230
#define     MVPP2_PRS_TCAM_EN_MASK		BIT(0)

/* Classifier Registers */
#define MVPP2_CLS_MODE_REG			0x1800
#define     MVPP2_CLS_MODE_ACTIVE_MASK		BIT(0)
#define MVPP2_CLS_PORT_WAY_REG			0x1810
#define     MVPP2_CLS_PORT_WAY_MASK(port)	(1 << (port))
#define MVPP2_CLS_LKP_INDEX_REG			0x1814
#define     MVPP2_CLS_LKP_INDEX_WAY_OFFS	6
#define MVPP2_CLS_LKP_TBL_REG			0x1818
#define     MVPP2_CLS_LKP_TBL_RXQ_MASK		0xff
#define     MVPP2_CLS_LKP_TBL_LOOKUP_EN_MASK	BIT(25)
#define MVPP2_CLS_FLOW_INDEX_REG		0x1820
#define MVPP2_CLS_FLOW_TBL0_REG			0x1824
#define MVPP2_CLS_FLOW_TBL1_REG			0x1828
#define MVPP2_CLS_FLOW_TBL2_REG			0x182c
#define MVPP2_CLS_OVERSIZE_RXQ_LOW_REG(port)	(0x1980 + ((port) * 4))
#define     MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS	3
#define     MVPP2_CLS_OVERSIZE_RXQ_LOW_MASK	0x7
#define MVPP2_CLS_SWFWD_P2HQ_REG(port)		(0x19b0 + ((port) * 4))
#define MVPP2_CLS_SWFWD_PCTRL_REG		0x19d0
#define     MVPP2_CLS_SWFWD_PCTRL_MASK(port)	(1 << (port))

/* Descriptor Manager Top Registers */
#define MVPP2_RXQ_NUM_REG			0x2040
#define MVPP2_RXQ_DESC_ADDR_REG			0x2044
#define     MVPP22_DESC_ADDR_OFFS		8
#define MVPP2_RXQ_DESC_SIZE_REG			0x2048
#define     MVPP2_RXQ_DESC_SIZE_MASK		0x3ff0
#define MVPP2_RXQ_STATUS_UPDATE_REG(rxq)	(0x3000 + 4 * (rxq))
#define     MVPP2_RXQ_NUM_PROCESSED_OFFSET	0
#define     MVPP2_RXQ_NUM_NEW_OFFSET		16
#define MVPP2_RXQ_STATUS_REG(rxq)		(0x3400 + 4 * (rxq))
#define     MVPP2_RXQ_OCCUPIED_MASK		0x3fff
#define     MVPP2_RXQ_NON_OCCUPIED_OFFSET	16
#define     MVPP2_RXQ_NON_OCCUPIED_MASK		0x3fff0000
#define MVPP2_RXQ_THRESH_REG			0x204c
#define     MVPP2_OCCUPIED_THRESH_OFFSET	0
#define     MVPP2_OCCUPIED_THRESH_MASK		0x3fff
#define MVPP2_RXQ_INDEX_REG			0x2050
#define MVPP2_TXQ_NUM_REG			0x2080
#define MVPP2_TXQ_DESC_ADDR_REG			0x2084
#define MVPP2_TXQ_DESC_SIZE_REG			0x2088
#define     MVPP2_TXQ_DESC_SIZE_MASK		0x3ff0
#define MVPP2_TXQ_THRESH_REG			0x2094
#define	    MVPP2_TXQ_THRESH_OFFSET		16
#define	    MVPP2_TXQ_THRESH_MASK		0x3fff
#define MVPP2_AGGR_TXQ_UPDATE_REG		0x2090
#define MVPP2_TXQ_INDEX_REG			0x2098
#define MVPP2_TXQ_PREF_BUF_REG			0x209c
#define     MVPP2_PREF_BUF_PTR(desc)		((desc) & 0xfff)
#define     MVPP2_PREF_BUF_SIZE_4		(BIT(12) | BIT(13))
#define     MVPP2_PREF_BUF_SIZE_16		(BIT(12) | BIT(14))
#define     MVPP2_PREF_BUF_THRESH(val)		((val) << 17)
#define     MVPP2_TXQ_DRAIN_EN_MASK		BIT(31)
#define MVPP2_TXQ_PENDING_REG			0x20a0
#define     MVPP2_TXQ_PENDING_MASK		0x3fff
#define MVPP2_TXQ_INT_STATUS_REG		0x20a4
#define MVPP2_TXQ_SENT_REG(txq)			(0x3c00 + 4 * (txq))
#define     MVPP2_TRANSMITTED_COUNT_OFFSET	16
#define     MVPP2_TRANSMITTED_COUNT_MASK	0x3fff0000
#define MVPP2_TXQ_RSVD_REQ_REG			0x20b0
#define     MVPP2_TXQ_RSVD_REQ_Q_OFFSET		16
#define MVPP2_TXQ_RSVD_RSLT_REG			0x20b4
#define     MVPP2_TXQ_RSVD_RSLT_MASK		0x3fff
#define MVPP2_TXQ_RSVD_CLR_REG			0x20b8
#define     MVPP2_TXQ_RSVD_CLR_OFFSET		16
#define MVPP2_AGGR_TXQ_DESC_ADDR_REG(cpu)	(0x2100 + 4 * (cpu))
#define     MVPP22_AGGR_TXQ_DESC_ADDR_OFFS	8
#define MVPP2_AGGR_TXQ_DESC_SIZE_REG(cpu)	(0x2140 + 4 * (cpu))
#define     MVPP2_AGGR_TXQ_DESC_SIZE_MASK	0x3ff0
#define MVPP2_AGGR_TXQ_STATUS_REG(cpu)		(0x2180 + 4 * (cpu))
#define     MVPP2_AGGR_TXQ_PENDING_MASK		0x3fff
#define MVPP2_AGGR_TXQ_INDEX_REG(cpu)		(0x21c0 + 4 * (cpu))

/* MBUS bridge registers */
#define MVPP2_WIN_BASE(w)			(0x4000 + ((w) << 2))
#define MVPP2_WIN_SIZE(w)			(0x4020 + ((w) << 2))
#define MVPP2_WIN_REMAP(w)			(0x4040 + ((w) << 2))
#define MVPP2_BASE_ADDR_ENABLE			0x4060

/* AXI Bridge Registers */
#define MVPP22_AXI_BM_WR_ATTR_REG		0x4100
#define MVPP22_AXI_BM_RD_ATTR_REG		0x4104
#define MVPP22_AXI_AGGRQ_DESCR_RD_ATTR_REG	0x4110
#define MVPP22_AXI_TXQ_DESCR_WR_ATTR_REG	0x4114
#define MVPP22_AXI_TXQ_DESCR_RD_ATTR_REG	0x4118
#define MVPP22_AXI_RXQ_DESCR_WR_ATTR_REG	0x411c
#define MVPP22_AXI_RX_DATA_WR_ATTR_REG		0x4120
#define MVPP22_AXI_TX_DATA_RD_ATTR_REG		0x4130
#define MVPP22_AXI_RD_NORMAL_CODE_REG		0x4150
#define MVPP22_AXI_RD_SNOOP_CODE_REG		0x4154
#define MVPP22_AXI_WR_NORMAL_CODE_REG		0x4160
#define MVPP22_AXI_WR_SNOOP_CODE_REG		0x4164

/* Values for AXI Bridge registers */
#define MVPP22_AXI_ATTR_CACHE_OFFS		0
#define MVPP22_AXI_ATTR_DOMAIN_OFFS		12

#define MVPP22_AXI_CODE_CACHE_OFFS		0
#define MVPP22_AXI_CODE_DOMAIN_OFFS		4

#define MVPP22_AXI_CODE_CACHE_NON_CACHE		0x3
#define MVPP22_AXI_CODE_CACHE_WR_CACHE		0x7
#define MVPP22_AXI_CODE_CACHE_RD_CACHE		0xb

#define MVPP22_AXI_CODE_DOMAIN_OUTER_DOM	2
#define MVPP22_AXI_CODE_DOMAIN_SYSTEM		3

/* Interrupt Cause and Mask registers */
#define MVPP2_ISR_TX_THRESHOLD_REG(port)	(0x5140 + 4 * (port))
#define     MVPP2_MAX_ISR_TX_THRESHOLD		0xfffff0

#define MVPP2_ISR_RX_THRESHOLD_REG(rxq)		(0x5200 + 4 * (rxq))
#define     MVPP2_MAX_ISR_RX_THRESHOLD		0xfffff0
#define MVPP21_ISR_RXQ_GROUP_REG(port)		(0x5400 + 4 * (port))

#define MVPP22_ISR_RXQ_GROUP_INDEX_REG          0x5400
#define MVPP22_ISR_RXQ_GROUP_INDEX_SUBGROUP_MASK 0xf
#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_MASK   0x380
#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_OFFSET 7

#define MVPP22_ISR_RXQ_GROUP_INDEX_SUBGROUP_MASK 0xf
#define MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_MASK   0x380

#define MVPP22_ISR_RXQ_SUB_GROUP_CONFIG_REG     0x5404
#define MVPP22_ISR_RXQ_SUB_GROUP_STARTQ_MASK    0x1f
#define MVPP22_ISR_RXQ_SUB_GROUP_SIZE_MASK      0xf00
#define MVPP22_ISR_RXQ_SUB_GROUP_SIZE_OFFSET    8

#define MVPP2_ISR_ENABLE_REG(port)		(0x5420 + 4 * (port))
#define     MVPP2_ISR_ENABLE_INTERRUPT(mask)	((mask) & 0xffff)
#define     MVPP2_ISR_DISABLE_INTERRUPT(mask)	(((mask) << 16) & 0xffff0000)
#define MVPP2_ISR_RX_TX_CAUSE_REG(port)		(0x5480 + 4 * (port))
#define     MVPP2_CAUSE_RXQ_OCCUP_DESC_ALL_MASK	0xffff
#define     MVPP2_CAUSE_TXQ_OCCUP_DESC_ALL_MASK	0xff0000
#define     MVPP2_CAUSE_TXQ_OCCUP_DESC_ALL_OFFSET	16
#define     MVPP2_CAUSE_RX_FIFO_OVERRUN_MASK	BIT(24)
#define     MVPP2_CAUSE_FCS_ERR_MASK		BIT(25)
#define     MVPP2_CAUSE_TX_FIFO_UNDERRUN_MASK	BIT(26)
#define     MVPP2_CAUSE_TX_EXCEPTION_SUM_MASK	BIT(29)
#define     MVPP2_CAUSE_RX_EXCEPTION_SUM_MASK	BIT(30)
#define     MVPP2_CAUSE_MISC_SUM_MASK		BIT(31)
#define MVPP2_ISR_RX_TX_MASK_REG(port)		(0x54a0 + 4 * (port))
#define MVPP2_ISR_PON_RX_TX_MASK_REG		0x54bc
#define     MVPP2_PON_CAUSE_RXQ_OCCUP_DESC_ALL_MASK	0xffff
#define     MVPP2_PON_CAUSE_TXP_OCCUP_DESC_ALL_MASK	0x3fc00000
#define     MVPP2_PON_CAUSE_MISC_SUM_MASK		BIT(31)
#define MVPP2_ISR_MISC_CAUSE_REG		0x55b0

/* Buffer Manager registers */
#define MVPP2_BM_POOL_BASE_REG(pool)		(0x6000 + ((pool) * 4))
#define     MVPP2_BM_POOL_BASE_ADDR_MASK	0xfffff80
#define MVPP2_BM_POOL_SIZE_REG(pool)		(0x6040 + ((pool) * 4))
#define     MVPP2_BM_POOL_SIZE_MASK		0xfff0
#define MVPP2_BM_POOL_READ_PTR_REG(pool)	(0x6080 + ((pool) * 4))
#define     MVPP2_BM_POOL_GET_READ_PTR_MASK	0xfff0
#define MVPP2_BM_POOL_PTRS_NUM_REG(pool)	(0x60c0 + ((pool) * 4))
#define     MVPP2_BM_POOL_PTRS_NUM_MASK		0xfff0
#define MVPP2_BM_BPPI_READ_PTR_REG(pool)	(0x6100 + ((pool) * 4))
#define MVPP2_BM_BPPI_PTRS_NUM_REG(pool)	(0x6140 + ((pool) * 4))
#define     MVPP2_BM_BPPI_PTR_NUM_MASK		0x7ff
#define     MVPP2_BM_BPPI_PREFETCH_FULL_MASK	BIT(16)
#define MVPP2_BM_POOL_CTRL_REG(pool)		(0x6200 + ((pool) * 4))
#define     MVPP2_BM_START_MASK			BIT(0)
#define     MVPP2_BM_STOP_MASK			BIT(1)
#define     MVPP2_BM_STATE_MASK			BIT(4)
#define     MVPP2_BM_LOW_THRESH_OFFS		8
#define     MVPP2_BM_LOW_THRESH_MASK		0x7f00
#define     MVPP2_BM_LOW_THRESH_VALUE(val)	((val) << \
						MVPP2_BM_LOW_THRESH_OFFS)
#define     MVPP2_BM_HIGH_THRESH_OFFS		16
#define     MVPP2_BM_HIGH_THRESH_MASK		0x7f0000
#define     MVPP2_BM_HIGH_THRESH_VALUE(val)	((val) << \
						MVPP2_BM_HIGH_THRESH_OFFS)
#define MVPP2_BM_INTR_CAUSE_REG(pool)		(0x6240 + ((pool) * 4))
#define     MVPP2_BM_RELEASED_DELAY_MASK	BIT(0)
#define     MVPP2_BM_ALLOC_FAILED_MASK		BIT(1)
#define     MVPP2_BM_BPPE_EMPTY_MASK		BIT(2)
#define     MVPP2_BM_BPPE_FULL_MASK		BIT(3)
#define     MVPP2_BM_AVAILABLE_BP_LOW_MASK	BIT(4)
#define MVPP2_BM_INTR_MASK_REG(pool)		(0x6280 + ((pool) * 4))
#define MVPP2_BM_PHY_ALLOC_REG(pool)		(0x6400 + ((pool) * 4))
#define     MVPP2_BM_PHY_ALLOC_GRNTD_MASK	BIT(0)
#define MVPP2_BM_VIRT_ALLOC_REG			0x6440
#define MVPP22_BM_ADDR_HIGH_ALLOC		0x6444
#define     MVPP22_BM_ADDR_HIGH_PHYS_MASK	0xff
#define     MVPP22_BM_ADDR_HIGH_VIRT_MASK	0xff00
#define     MVPP22_BM_ADDR_HIGH_VIRT_SHIFT	8
#define MVPP2_BM_PHY_RLS_REG(pool)		(0x6480 + ((pool) * 4))
#define     MVPP2_BM_PHY_RLS_MC_BUFF_MASK	BIT(0)
#define     MVPP2_BM_PHY_RLS_PRIO_EN_MASK	BIT(1)
#define     MVPP2_BM_PHY_RLS_GRNTD_MASK		BIT(2)
#define MVPP2_BM_VIRT_RLS_REG			0x64c0
#define MVPP22_BM_ADDR_HIGH_RLS_REG		0x64c4
#define     MVPP22_BM_ADDR_HIGH_PHYS_RLS_MASK	0xff
#define	    MVPP22_BM_ADDR_HIGH_VIRT_RLS_MASK	0xff00
#define     MVPP22_BM_ADDR_HIGH_VIRT_RLS_SHIFT	8

/* TX Scheduler registers */
#define MVPP2_TXP_SCHED_PORT_INDEX_REG		0x8000
#define MVPP2_TXP_SCHED_Q_CMD_REG		0x8004
#define     MVPP2_TXP_SCHED_ENQ_MASK		0xff
#define     MVPP2_TXP_SCHED_DISQ_OFFSET		8
#define MVPP2_TXP_SCHED_CMD_1_REG		0x8010
#define MVPP2_TXP_SCHED_PERIOD_REG		0x8018
#define MVPP2_TXP_SCHED_MTU_REG			0x801c
#define     MVPP2_TXP_MTU_MAX			0x7FFFF
#define MVPP2_TXP_SCHED_REFILL_REG		0x8020
#define     MVPP2_TXP_REFILL_TOKENS_ALL_MASK	0x7ffff
#define     MVPP2_TXP_REFILL_PERIOD_ALL_MASK	0x3ff00000
#define     MVPP2_TXP_REFILL_PERIOD_MASK(v)	((v) << 20)
#define MVPP2_TXP_SCHED_TOKEN_SIZE_REG		0x8024
#define     MVPP2_TXP_TOKEN_SIZE_MAX		0xffffffff
#define MVPP2_TXQ_SCHED_REFILL_REG(q)		(0x8040 + ((q) << 2))
#define     MVPP2_TXQ_REFILL_TOKENS_ALL_MASK	0x7ffff
#define     MVPP2_TXQ_REFILL_PERIOD_ALL_MASK	0x3ff00000
#define     MVPP2_TXQ_REFILL_PERIOD_MASK(v)	((v) << 20)
#define MVPP2_TXQ_SCHED_TOKEN_SIZE_REG(q)	(0x8060 + ((q) << 2))
#define     MVPP2_TXQ_TOKEN_SIZE_MAX		0x7fffffff
#define MVPP2_TXQ_SCHED_TOKEN_CNTR_REG(q)	(0x8080 + ((q) << 2))
#define     MVPP2_TXQ_TOKEN_CNTR_MAX		0xffffffff

/* TX general registers */
#define MVPP2_TX_SNOOP_REG			0x8800
#define MVPP2_TX_PORT_FLUSH_REG			0x8810
#define     MVPP2_TX_PORT_FLUSH_MASK(port)	(1 << (port))

/* LMS registers */
#define MVPP2_SRC_ADDR_MIDDLE			0x24
#define MVPP2_SRC_ADDR_HIGH			0x28
#define MVPP2_PHY_AN_CFG0_REG			0x34
#define     MVPP2_PHY_AN_STOP_SMI0_MASK		BIT(7)
#define MVPP2_MNG_EXTENDED_GLOBAL_CTRL_REG	0x305c
#define     MVPP2_EXT_GLOBAL_CTRL_DEFAULT	0x27

/* Per-port registers */
#define MVPP2_GMAC_CTRL_0_REG			0x0
#define      MVPP2_GMAC_PORT_EN_MASK		BIT(0)
#define      MVPP2_GMAC_MAX_RX_SIZE_OFFS	2
#define      MVPP2_GMAC_MAX_RX_SIZE_MASK	0x7ffc
#define      MVPP2_GMAC_MIB_CNTR_EN_MASK	BIT(15)
#define MVPP2_GMAC_CTRL_1_REG			0x4
#define      MVPP2_GMAC_PERIODIC_XON_EN_MASK	BIT(1)
#define      MVPP2_GMAC_GMII_LB_EN_MASK		BIT(5)
#define      MVPP2_GMAC_PCS_LB_EN_BIT		6
#define      MVPP2_GMAC_PCS_LB_EN_MASK		BIT(6)
#define      MVPP2_GMAC_SA_LOW_OFFS		7
#define MVPP2_GMAC_CTRL_2_REG			0x8
#define      MVPP2_GMAC_INBAND_AN_MASK		BIT(0)
#define      MVPP2_GMAC_PCS_ENABLE_MASK		BIT(3)
#define      MVPP2_GMAC_PORT_RGMII_MASK		BIT(4)
#define      MVPP2_GMAC_PORT_RESET_MASK		BIT(6)
#define MVPP2_GMAC_AUTONEG_CONFIG		0xc
#define      MVPP2_GMAC_FORCE_LINK_DOWN		BIT(0)
#define      MVPP2_GMAC_FORCE_LINK_PASS		BIT(1)
#define      MVPP2_GMAC_CONFIG_MII_SPEED	BIT(5)
#define      MVPP2_GMAC_CONFIG_GMII_SPEED	BIT(6)
#define      MVPP2_GMAC_AN_SPEED_EN		BIT(7)
#define      MVPP2_GMAC_FC_ADV_EN		BIT(9)
#define      MVPP2_GMAC_CONFIG_FULL_DUPLEX	BIT(12)
#define      MVPP2_GMAC_AN_DUPLEX_EN		BIT(13)
#define MVPP2_GMAC_PORT_FIFO_CFG_1_REG		0x1c
#define      MVPP2_GMAC_TX_FIFO_MIN_TH_OFFS	6
#define      MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK	0x1fc0
#define      MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(v)	(((v) << 6) & \
					MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK)
#define MVPP22_GMAC_CTRL_4_REG			0x90
#define      MVPP22_CTRL4_EXT_PIN_GMII_SEL	BIT(0)
#define      MVPP22_CTRL4_DP_CLK_SEL		BIT(5)
#define      MVPP22_CTRL4_SYNC_BYPASS		BIT(6)
#define      MVPP22_CTRL4_QSGMII_BYPASS_ACTIVE	BIT(7)

/* Per-port XGMAC registers. PPv2.2 only, only for GOP port 0,
 * relative to port->base.
 */
#define MVPP22_XLG_CTRL0_REG			0x100
#define      MVPP22_XLG_CTRL0_PORT_EN		BIT(0)
#define      MVPP22_XLG_CTRL0_MAC_RESET_DIS	BIT(1)
#define      MVPP22_XLG_CTRL0_MIB_CNT_DIS	BIT(14)

#define MVPP22_XLG_CTRL3_REG			0x11c
#define      MVPP22_XLG_CTRL3_MACMODESELECT_MASK	(7 << 13)
#define      MVPP22_XLG_CTRL3_MACMODESELECT_GMAC	(0 << 13)
#define      MVPP22_XLG_CTRL3_MACMODESELECT_10G		(1 << 13)

/* SMI registers. PPv2.2 only, relative to priv->iface_base. */
#define MVPP22_SMI_MISC_CFG_REG			0x1204
#define      MVPP22_SMI_POLLING_EN		BIT(10)

#define MVPP22_GMAC_BASE(port)		(0x7000 + (port) * 0x1000 + 0xe00)

#define MVPP2_CAUSE_TXQ_SENT_DESC_ALL_MASK	0xff

/* Descriptor ring Macros */
#define MVPP2_QUEUE_NEXT_DESC(q, index) \
	(((index) < (q)->last_desc) ? ((index) + 1) : 0)

/* Various constants */

/* Coalescing */
#define MVPP2_TXDONE_COAL_PKTS_THRESH	15
#define MVPP2_TXDONE_HRTIMER_PERIOD_NS	1000000UL
#define MVPP2_TXDONE_COAL_USEC		1000
#define MVPP2_RX_COAL_PKTS		32
#define MVPP2_RX_COAL_USEC		100

/* The two bytes Marvell header. Either contains a special value used
 * by Marvell switches when a specific hardware mode is enabled (not
 * supported by this driver) or is filled automatically by zeroes on
 * the RX side. Those two bytes being at the front of the Ethernet
 * header, they allow to have the IP header aligned on a 4 bytes
 * boundary automatically: the hardware skips those two bytes on its
 * own.
 */
#define MVPP2_MH_SIZE			2
#define MVPP2_ETH_TYPE_LEN		2
#define MVPP2_PPPOE_HDR_SIZE		8
#define MVPP2_VLAN_TAG_LEN		4

/* Lbtd 802.3 type */
#define MVPP2_IP_LBDT_TYPE		0xfffa

#define MVPP2_TX_CSUM_MAX_SIZE		9800

/* Timeout constants */
#define MVPP2_TX_DISABLE_TIMEOUT_MSEC	1000
#define MVPP2_TX_PENDING_TIMEOUT_MSEC	1000

#define MVPP2_TX_MTU_MAX		0x7ffff

/* Maximum number of T-CONTs of PON port */
#define MVPP2_MAX_TCONT			16

/* Maximum number of supported ports */
#define MVPP2_MAX_PORTS			4

/* Maximum number of TXQs used by single port */
#define MVPP2_MAX_TXQ			8

/* Dfault number of RXQs in use */
#define MVPP2_DEFAULT_RXQ		4

/* Max number of Rx descriptors */
#define MVPP2_MAX_RXD			128

/* Max number of Tx descriptors */
#define MVPP2_MAX_TXD			1024

/* Amount of Tx descriptors that can be reserved at once by CPU */
#define MVPP2_CPU_DESC_CHUNK		64

/* Max number of Tx descriptors in each aggregated queue */
#define MVPP2_AGGR_TXQ_SIZE		256

/* Descriptor aligned size */
#define MVPP2_DESC_ALIGNED_SIZE		32

/* Descriptor alignment mask */
#define MVPP2_TX_DESC_ALIGN		(MVPP2_DESC_ALIGNED_SIZE - 1)

/* RX FIFO constants */
#define MVPP2_RX_FIFO_PORT_DATA_SIZE	0x2000
#define MVPP2_RX_FIFO_PORT_ATTR_SIZE	0x80
#define MVPP2_RX_FIFO_PORT_MIN_PKT	0x80

/* RX buffer constants */
#define MVPP2_SKB_SHINFO_SIZE \
	SKB_DATA_ALIGN(sizeof(struct skb_shared_info))

#define MVPP2_RX_PKT_SIZE(mtu) \
	ALIGN((mtu) + MVPP2_MH_SIZE + MVPP2_VLAN_TAG_LEN + \
	      ETH_HLEN + ETH_FCS_LEN, cache_line_size())

#define MVPP2_RX_BUF_SIZE(pkt_size)	((pkt_size) + NET_SKB_PAD)
#define MVPP2_RX_TOTAL_SIZE(buf_size)	((buf_size) + MVPP2_SKB_SHINFO_SIZE)
#define MVPP2_RX_MAX_PKT_SIZE(total_size) \
	((total_size) - NET_SKB_PAD - MVPP2_SKB_SHINFO_SIZE)

#define MVPP2_BIT_TO_BYTE(bit)		((bit) / 8)

/* IPv6 max L3 address size */
#define MVPP2_MAX_L3_ADDR_SIZE		16

/* Port flags */
#define MVPP2_F_LOOPBACK		BIT(0)

/* Marvell tag types */
enum mvpp2_tag_type {
	MVPP2_TAG_TYPE_NONE = 0,
	MVPP2_TAG_TYPE_MH   = 1,
	MVPP2_TAG_TYPE_DSA  = 2,
	MVPP2_TAG_TYPE_EDSA = 3,
	MVPP2_TAG_TYPE_VLAN = 4,
	MVPP2_TAG_TYPE_LAST = 5
};

/* Parser constants */
#define MVPP2_PRS_TCAM_SRAM_SIZE	256
#define MVPP2_PRS_TCAM_WORDS		6
#define MVPP2_PRS_SRAM_WORDS		4
#define MVPP2_PRS_FLOW_ID_SIZE		64
#define MVPP2_PRS_FLOW_ID_MASK		0x3f
#define MVPP2_PRS_TCAM_ENTRY_INVALID	1
#define MVPP2_PRS_TCAM_DSA_TAGGED_BIT	BIT(5)
#define MVPP2_PRS_IPV4_HEAD		0x40
#define MVPP2_PRS_IPV4_HEAD_MASK	0xf0
#define MVPP2_PRS_IPV4_MC		0xe0
#define MVPP2_PRS_IPV4_MC_MASK		0xf0
#define MVPP2_PRS_IPV4_BC_MASK		0xff
#define MVPP2_PRS_IPV4_IHL		0x5
#define MVPP2_PRS_IPV4_IHL_MASK		0xf
#define MVPP2_PRS_IPV6_MC		0xff
#define MVPP2_PRS_IPV6_MC_MASK		0xff
#define MVPP2_PRS_IPV6_HOP_MASK		0xff
#define MVPP2_PRS_TCAM_PROTO_MASK	0xff
#define MVPP2_PRS_TCAM_PROTO_MASK_L	0x3f
#define MVPP2_PRS_DBL_VLANS_MAX		100

/* Tcam structure:
 * - lookup ID - 4 bits
 * - port ID - 1 byte
 * - additional information - 1 byte
 * - header data - 8 bytes
 * The fields are represented by MVPP2_PRS_TCAM_DATA_REG(5)->(0).
 */
#define MVPP2_PRS_AI_BITS			8
#define MVPP2_PRS_PORT_MASK			0xff
#define MVPP2_PRS_LU_MASK			0xf
#define MVPP2_PRS_TCAM_DATA_BYTE(offs)		\
				    (((offs) - ((offs) % 2)) * 2 + ((offs) % 2))
#define MVPP2_PRS_TCAM_DATA_BYTE_EN(offs)	\
					      (((offs) * 2) - ((offs) % 2)  + 2)
#define MVPP2_PRS_TCAM_AI_BYTE			16
#define MVPP2_PRS_TCAM_PORT_BYTE		17
#define MVPP2_PRS_TCAM_LU_BYTE			20
#define MVPP2_PRS_TCAM_EN_OFFS(offs)		((offs) + 2)
#define MVPP2_PRS_TCAM_INV_WORD			5
/* Tcam entries ID */
#define MVPP2_PE_DROP_ALL		0
#define MVPP2_PE_FIRST_FREE_TID		1
#define MVPP2_PE_LAST_FREE_TID		(MVPP2_PRS_TCAM_SRAM_SIZE - 31)
#define MVPP2_PE_IP6_EXT_PROTO_UN	(MVPP2_PRS_TCAM_SRAM_SIZE - 30)
#define MVPP2_PE_MAC_MC_IP6		(MVPP2_PRS_TCAM_SRAM_SIZE - 29)
#define MVPP2_PE_IP6_ADDR_UN		(MVPP2_PRS_TCAM_SRAM_SIZE - 28)
#define MVPP2_PE_IP4_ADDR_UN		(MVPP2_PRS_TCAM_SRAM_SIZE - 27)
#define MVPP2_PE_LAST_DEFAULT_FLOW	(MVPP2_PRS_TCAM_SRAM_SIZE - 26)
#define MVPP2_PE_FIRST_DEFAULT_FLOW	(MVPP2_PRS_TCAM_SRAM_SIZE - 19)
#define MVPP2_PE_EDSA_TAGGED		(MVPP2_PRS_TCAM_SRAM_SIZE - 18)
#define MVPP2_PE_EDSA_UNTAGGED		(MVPP2_PRS_TCAM_SRAM_SIZE - 17)
#define MVPP2_PE_DSA_TAGGED		(MVPP2_PRS_TCAM_SRAM_SIZE - 16)
#define MVPP2_PE_DSA_UNTAGGED		(MVPP2_PRS_TCAM_SRAM_SIZE - 15)
#define MVPP2_PE_ETYPE_EDSA_TAGGED	(MVPP2_PRS_TCAM_SRAM_SIZE - 14)
#define MVPP2_PE_ETYPE_EDSA_UNTAGGED	(MVPP2_PRS_TCAM_SRAM_SIZE - 13)
#define MVPP2_PE_ETYPE_DSA_TAGGED	(MVPP2_PRS_TCAM_SRAM_SIZE - 12)
#define MVPP2_PE_ETYPE_DSA_UNTAGGED	(MVPP2_PRS_TCAM_SRAM_SIZE - 11)
#define MVPP2_PE_MH_DEFAULT		(MVPP2_PRS_TCAM_SRAM_SIZE - 10)
#define MVPP2_PE_DSA_DEFAULT		(MVPP2_PRS_TCAM_SRAM_SIZE - 9)
#define MVPP2_PE_IP6_PROTO_UN		(MVPP2_PRS_TCAM_SRAM_SIZE - 8)
#define MVPP2_PE_IP4_PROTO_UN		(MVPP2_PRS_TCAM_SRAM_SIZE - 7)
#define MVPP2_PE_ETH_TYPE_UN		(MVPP2_PRS_TCAM_SRAM_SIZE - 6)
#define MVPP2_PE_VLAN_DBL		(MVPP2_PRS_TCAM_SRAM_SIZE - 5)
#define MVPP2_PE_VLAN_NONE		(MVPP2_PRS_TCAM_SRAM_SIZE - 4)
#define MVPP2_PE_MAC_MC_ALL		(MVPP2_PRS_TCAM_SRAM_SIZE - 3)
#define MVPP2_PE_MAC_PROMISCUOUS	(MVPP2_PRS_TCAM_SRAM_SIZE - 2)
#define MVPP2_PE_MAC_NON_PROMISCUOUS	(MVPP2_PRS_TCAM_SRAM_SIZE - 1)

/* Sram structure
 * The fields are represented by MVPP2_PRS_TCAM_DATA_REG(3)->(0).
 */
#define MVPP2_PRS_SRAM_RI_OFFS			0
#define MVPP2_PRS_SRAM_RI_WORD			0
#define MVPP2_PRS_SRAM_RI_CTRL_OFFS		32
#define MVPP2_PRS_SRAM_RI_CTRL_WORD		1
#define MVPP2_PRS_SRAM_RI_CTRL_BITS		32
#define MVPP2_PRS_SRAM_SHIFT_OFFS		64
#define MVPP2_PRS_SRAM_SHIFT_SIGN_BIT		72
#define MVPP2_PRS_SRAM_UDF_OFFS			73
#define MVPP2_PRS_SRAM_UDF_BITS			8
#define MVPP2_PRS_SRAM_UDF_MASK			0xff
#define MVPP2_PRS_SRAM_UDF_SIGN_BIT		81
#define MVPP2_PRS_SRAM_UDF_TYPE_OFFS		82
#define MVPP2_PRS_SRAM_UDF_TYPE_MASK		0x7
#define MVPP2_PRS_SRAM_UDF_TYPE_L3		1
#define MVPP2_PRS_SRAM_UDF_TYPE_L4		4
#define MVPP2_PRS_SRAM_OP_SEL_SHIFT_OFFS	85
#define MVPP2_PRS_SRAM_OP_SEL_SHIFT_MASK	0x3
#define MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD		1
#define MVPP2_PRS_SRAM_OP_SEL_SHIFT_IP4_ADD	2
#define MVPP2_PRS_SRAM_OP_SEL_SHIFT_IP6_ADD	3
#define MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS		87
#define MVPP2_PRS_SRAM_OP_SEL_UDF_BITS		2
#define MVPP2_PRS_SRAM_OP_SEL_UDF_MASK		0x3
#define MVPP2_PRS_SRAM_OP_SEL_UDF_ADD		0
#define MVPP2_PRS_SRAM_OP_SEL_UDF_IP4_ADD	2
#define MVPP2_PRS_SRAM_OP_SEL_UDF_IP6_ADD	3
#define MVPP2_PRS_SRAM_OP_SEL_BASE_OFFS		89
#define MVPP2_PRS_SRAM_AI_OFFS			90
#define MVPP2_PRS_SRAM_AI_CTRL_OFFS		98
#define MVPP2_PRS_SRAM_AI_CTRL_BITS		8
#define MVPP2_PRS_SRAM_AI_MASK			0xff
#define MVPP2_PRS_SRAM_NEXT_LU_OFFS		106
#define MVPP2_PRS_SRAM_NEXT_LU_MASK		0xf
#define MVPP2_PRS_SRAM_LU_DONE_BIT		110
#define MVPP2_PRS_SRAM_LU_GEN_BIT		111

/* Sram result info bits assignment */
#define MVPP2_PRS_RI_MAC_ME_MASK		0x1
#define MVPP2_PRS_RI_DSA_MASK			0x2
#define MVPP2_PRS_RI_VLAN_MASK			(BIT(2) | BIT(3))
#define MVPP2_PRS_RI_VLAN_NONE			0x0
#define MVPP2_PRS_RI_VLAN_SINGLE		BIT(2)
#define MVPP2_PRS_RI_VLAN_DOUBLE		BIT(3)
#define MVPP2_PRS_RI_VLAN_TRIPLE		(BIT(2) | BIT(3))
#define MVPP2_PRS_RI_CPU_CODE_MASK		0x70
#define MVPP2_PRS_RI_CPU_CODE_RX_SPEC		BIT(4)
#define MVPP2_PRS_RI_L2_CAST_MASK		(BIT(9) | BIT(10))
#define MVPP2_PRS_RI_L2_UCAST			0x0
#define MVPP2_PRS_RI_L2_MCAST			BIT(9)
#define MVPP2_PRS_RI_L2_BCAST			BIT(10)
#define MVPP2_PRS_RI_PPPOE_MASK			0x800
#define MVPP2_PRS_RI_L3_PROTO_MASK		(BIT(12) | BIT(13) | BIT(14))
#define MVPP2_PRS_RI_L3_UN			0x0
#define MVPP2_PRS_RI_L3_IP4			BIT(12)
#define MVPP2_PRS_RI_L3_IP4_OPT			BIT(13)
#define MVPP2_PRS_RI_L3_IP4_OTHER		(BIT(12) | BIT(13))
#define MVPP2_PRS_RI_L3_IP6			BIT(14)
#define MVPP2_PRS_RI_L3_IP6_EXT			(BIT(12) | BIT(14))
#define MVPP2_PRS_RI_L3_ARP			(BIT(13) | BIT(14))
#define MVPP2_PRS_RI_L3_ADDR_MASK		(BIT(15) | BIT(16))
#define MVPP2_PRS_RI_L3_UCAST			0x0
#define MVPP2_PRS_RI_L3_MCAST			BIT(15)
#define MVPP2_PRS_RI_L3_BCAST			(BIT(15) | BIT(16))
#define MVPP2_PRS_RI_IP_FRAG_MASK		0x20000
#define MVPP2_PRS_RI_UDF3_MASK			0x300000
#define MVPP2_PRS_RI_UDF3_RX_SPECIAL		BIT(21)
#define MVPP2_PRS_RI_L4_PROTO_MASK		0x1c00000
#define MVPP2_PRS_RI_L4_TCP			BIT(22)
#define MVPP2_PRS_RI_L4_UDP			BIT(23)
#define MVPP2_PRS_RI_L4_OTHER			(BIT(22) | BIT(23))
#define MVPP2_PRS_RI_UDF7_MASK			0x60000000
#define MVPP2_PRS_RI_UDF7_IP6_LITE		BIT(29)
#define MVPP2_PRS_RI_DROP_MASK			0x80000000

/* Sram additional info bits assignment */
#define MVPP2_PRS_IPV4_DIP_AI_BIT		BIT(0)
#define MVPP2_PRS_IPV6_NO_EXT_AI_BIT		BIT(0)
#define MVPP2_PRS_IPV6_EXT_AI_BIT		BIT(1)
#define MVPP2_PRS_IPV6_EXT_AH_AI_BIT		BIT(2)
#define MVPP2_PRS_IPV6_EXT_AH_LEN_AI_BIT	BIT(3)
#define MVPP2_PRS_IPV6_EXT_AH_L4_AI_BIT		BIT(4)
#define MVPP2_PRS_SINGLE_VLAN_AI		0
#define MVPP2_PRS_DBL_VLAN_AI_BIT		BIT(7)

/* DSA/EDSA type */
#define MVPP2_PRS_TAGGED		true
#define MVPP2_PRS_UNTAGGED		false
#define MVPP2_PRS_EDSA			true
#define MVPP2_PRS_DSA			false

/* MAC entries, shadow udf */
enum mvpp2_prs_udf {
	MVPP2_PRS_UDF_MAC_DEF,
	MVPP2_PRS_UDF_MAC_RANGE,
	MVPP2_PRS_UDF_L2_DEF,
	MVPP2_PRS_UDF_L2_DEF_COPY,
	MVPP2_PRS_UDF_L2_USER,
};

/* Lookup ID */
enum mvpp2_prs_lookup {
	MVPP2_PRS_LU_MH,
	MVPP2_PRS_LU_MAC,
	MVPP2_PRS_LU_DSA,
	MVPP2_PRS_LU_VLAN,
	MVPP2_PRS_LU_L2,
	MVPP2_PRS_LU_PPPOE,
	MVPP2_PRS_LU_IP4,
	MVPP2_PRS_LU_IP6,
	MVPP2_PRS_LU_FLOWS,
	MVPP2_PRS_LU_LAST,
};

/* L3 cast enum */
enum mvpp2_prs_l3_cast {
	MVPP2_PRS_L3_UNI_CAST,
	MVPP2_PRS_L3_MULTI_CAST,
	MVPP2_PRS_L3_BROAD_CAST
};

/* Classifier constants */
#define MVPP2_CLS_FLOWS_TBL_SIZE	512
#define MVPP2_CLS_FLOWS_TBL_DATA_WORDS	3
#define MVPP2_CLS_LKP_TBL_SIZE		64

/* BM constants */
#define MVPP2_BM_POOLS_NUM		8
#define MVPP2_BM_LONG_BUF_NUM		1024
#define MVPP2_BM_SHORT_BUF_NUM		2048
#define MVPP2_BM_POOL_SIZE_MAX		(16*1024 - MVPP2_BM_POOL_PTR_ALIGN/4)
#define MVPP2_BM_POOL_PTR_ALIGN		128
#define MVPP2_BM_SWF_LONG_POOL(port)	((port > 2) ? 2 : port)
#define MVPP2_BM_SWF_SHORT_POOL		3

/* BM cookie (32 bits) definition */
#define MVPP2_BM_COOKIE_POOL_OFFS	8
#define MVPP2_BM_COOKIE_CPU_OFFS	24

/* BM short pool packet size
 * These value assure that for SWF the total number
 * of bytes allocated for each buffer will be 512
 */
#define MVPP2_BM_SHORT_PKT_SIZE		MVPP2_RX_MAX_PKT_SIZE(512)

#define MVPP21_ADDR_SPACE_SZ		0
#define MVPP22_ADDR_SPACE_SZ		SZ_64K

#define MVPP2_MAX_THREADS		8
#define MVPP2_MAX_QVECS			MVPP2_MAX_THREADS

enum mvpp2_bm_type {
	MVPP2_BM_FREE,
	MVPP2_BM_SWF_LONG,
	MVPP2_BM_SWF_SHORT
};

/* Definitions */

/* Shared Packet Processor resources */
struct mvpp2 {
	/* Shared registers' base addresses */
	void __iomem *lms_base;
	void __iomem *iface_base;

	/* On PPv2.2, each "software thread" can access the base
	 * register through a separate address space, each 64 KB apart
	 * from each other. Typically, such address spaces will be
	 * used per CPU.
	 */
	void __iomem *swth_base[MVPP2_MAX_THREADS];

	/* Common clocks */
	struct clk *pp_clk;
	struct clk *gop_clk;
	struct clk *mg_clk;

	/* List of pointers to port structures */
	struct mvpp2_port **port_list;

	/* Aggregated TXQs */
	struct mvpp2_tx_queue *aggr_txqs;

	/* BM pools */
	struct mvpp2_bm_pool *bm_pools;

	/* PRS shadow table */
	struct mvpp2_prs_shadow *prs_shadow;
	/* PRS auxiliary table for double vlan entries control */
	bool *prs_double_vlans;

	/* Tclk value */
	u32 tclk;

	/* HW version */
	enum { MVPP21, MVPP22 } hw_version;

	/* Maximum number of RXQs per port */
	unsigned int max_port_rxqs;
};

struct mvpp2_pcpu_stats {
	struct	u64_stats_sync syncp;
	u64	rx_packets;
	u64	rx_bytes;
	u64	tx_packets;
	u64	tx_bytes;
};

/* Per-CPU port control */
struct mvpp2_port_pcpu {
	struct hrtimer tx_done_timer;
	bool timer_scheduled;
	/* Tasklet for egress finalization */
	struct tasklet_struct tx_done_tasklet;
};

struct mvpp2_queue_vector {
	int irq;
	struct napi_struct napi;
	enum { MVPP2_QUEUE_VECTOR_SHARED, MVPP2_QUEUE_VECTOR_PRIVATE } type;
	int sw_thread_id;
	u16 sw_thread_mask;
	int first_rxq;
	int nrxqs;
	u32 pending_cause_rx;
	struct mvpp2_port *port;
};

struct mvpp2_port {
	u8 id;

	/* Index of the port from the "group of ports" complex point
	 * of view
	 */
	int gop_id;

	struct mvpp2 *priv;

	/* Per-port registers' base address */
	void __iomem *base;

	struct mvpp2_rx_queue **rxqs;
	unsigned int nrxqs;
	struct mvpp2_tx_queue **txqs;
	unsigned int ntxqs;
	struct net_device *dev;

	int pkt_size;

	/* Per-CPU port control */
	struct mvpp2_port_pcpu __percpu *pcpu;

	/* Flags */
	unsigned long flags;

	u16 tx_ring_size;
	u16 rx_ring_size;
	struct mvpp2_pcpu_stats __percpu *stats;

	phy_interface_t phy_interface;
	struct device_node *phy_node;
	unsigned int link;
	unsigned int duplex;
	unsigned int speed;

	struct mvpp2_bm_pool *pool_long;
	struct mvpp2_bm_pool *pool_short;

	/* Index of first port's physical RXQ */
	u8 first_rxq;

	struct mvpp2_queue_vector qvecs[MVPP2_MAX_QVECS];
	unsigned int nqvecs;
	bool has_tx_irqs;

	u32 tx_time_coal;
};

/* The mvpp2_tx_desc and mvpp2_rx_desc structures describe the
 * layout of the transmit and reception DMA descriptors, and their
 * layout is therefore defined by the hardware design
 */

#define MVPP2_TXD_L3_OFF_SHIFT		0
#define MVPP2_TXD_IP_HLEN_SHIFT		8
#define MVPP2_TXD_L4_CSUM_FRAG		BIT(13)
#define MVPP2_TXD_L4_CSUM_NOT		BIT(14)
#define MVPP2_TXD_IP_CSUM_DISABLE	BIT(15)
#define MVPP2_TXD_PADDING_DISABLE	BIT(23)
#define MVPP2_TXD_L4_UDP		BIT(24)
#define MVPP2_TXD_L3_IP6		BIT(26)
#define MVPP2_TXD_L_DESC		BIT(28)
#define MVPP2_TXD_F_DESC		BIT(29)

#define MVPP2_RXD_ERR_SUMMARY		BIT(15)
#define MVPP2_RXD_ERR_CODE_MASK		(BIT(13) | BIT(14))
#define MVPP2_RXD_ERR_CRC		0x0
#define MVPP2_RXD_ERR_OVERRUN		BIT(13)
#define MVPP2_RXD_ERR_RESOURCE		(BIT(13) | BIT(14))
#define MVPP2_RXD_BM_POOL_ID_OFFS	16
#define MVPP2_RXD_BM_POOL_ID_MASK	(BIT(16) | BIT(17) | BIT(18))
#define MVPP2_RXD_HWF_SYNC		BIT(21)
#define MVPP2_RXD_L4_CSUM_OK		BIT(22)
#define MVPP2_RXD_IP4_HEADER_ERR	BIT(24)
#define MVPP2_RXD_L4_TCP		BIT(25)
#define MVPP2_RXD_L4_UDP		BIT(26)
#define MVPP2_RXD_L3_IP4		BIT(28)
#define MVPP2_RXD_L3_IP6		BIT(30)
#define MVPP2_RXD_BUF_HDR		BIT(31)

/* HW TX descriptor for PPv2.1 */
struct mvpp21_tx_desc {
	u32 command;		/* Options used by HW for packet transmitting.*/
	u8  packet_offset;	/* the offset from the buffer beginning	*/
	u8  phys_txq;		/* destination queue ID			*/
	u16 data_size;		/* data size of transmitted packet in bytes */
	u32 buf_dma_addr;	/* physical addr of transmitted buffer	*/
	u32 buf_cookie;		/* cookie for access to TX buffer in tx path */
	u32 reserved1[3];	/* hw_cmd (for future use, BM, PON, PNC) */
	u32 reserved2;		/* reserved (for future use)		*/
};

/* HW RX descriptor for PPv2.1 */
struct mvpp21_rx_desc {
	u32 status;		/* info about received packet		*/
	u16 reserved1;		/* parser_info (for future use, PnC)	*/
	u16 data_size;		/* size of received packet in bytes	*/
	u32 buf_dma_addr;	/* physical address of the buffer	*/
	u32 buf_cookie;		/* cookie for access to RX buffer in rx path */
	u16 reserved2;		/* gem_port_id (for future use, PON)	*/
	u16 reserved3;		/* csum_l4 (for future use, PnC)	*/
	u8  reserved4;		/* bm_qset (for future use, BM)		*/
	u8  reserved5;
	u16 reserved6;		/* classify_info (for future use, PnC)	*/
	u32 reserved7;		/* flow_id (for future use, PnC) */
	u32 reserved8;
};

/* HW TX descriptor for PPv2.2 */
struct mvpp22_tx_desc {
	u32 command;
	u8  packet_offset;
	u8  phys_txq;
	u16 data_size;
	u64 reserved1;
	u64 buf_dma_addr_ptp;
	u64 buf_cookie_misc;
};

/* HW RX descriptor for PPv2.2 */
struct mvpp22_rx_desc {
	u32 status;
	u16 reserved1;
	u16 data_size;
	u32 reserved2;
	u32 reserved3;
	u64 buf_dma_addr_key_hash;
	u64 buf_cookie_misc;
};

/* Opaque type used by the driver to manipulate the HW TX and RX
 * descriptors
 */
struct mvpp2_tx_desc {
	union {
		struct mvpp21_tx_desc pp21;
		struct mvpp22_tx_desc pp22;
	};
};

struct mvpp2_rx_desc {
	union {
		struct mvpp21_rx_desc pp21;
		struct mvpp22_rx_desc pp22;
	};
};

struct mvpp2_txq_pcpu_buf {
	/* Transmitted SKB */
	struct sk_buff *skb;

	/* Physical address of transmitted buffer */
	dma_addr_t dma;

	/* Size transmitted */
	size_t size;
};

/* Per-CPU Tx queue control */
struct mvpp2_txq_pcpu {
	int cpu;

	/* Number of Tx DMA descriptors in the descriptor ring */
	int size;

	/* Number of currently used Tx DMA descriptor in the
	 * descriptor ring
	 */
	int count;

	/* Number of Tx DMA descriptors reserved for each CPU */
	int reserved_num;

	/* Infos about transmitted buffers */
	struct mvpp2_txq_pcpu_buf *buffs;

	/* Index of last TX DMA descriptor that was inserted */
	int txq_put_index;

	/* Index of the TX DMA descriptor to be cleaned up */
	int txq_get_index;
};

struct mvpp2_tx_queue {
	/* Physical number of this Tx queue */
	u8 id;

	/* Logical number of this Tx queue */
	u8 log_id;

	/* Number of Tx DMA descriptors in the descriptor ring */
	int size;

	/* Number of currently used Tx DMA descriptor in the descriptor ring */
	int count;

	/* Per-CPU control of physical Tx queues */
	struct mvpp2_txq_pcpu __percpu *pcpu;

	u32 done_pkts_coal;

	/* Virtual address of thex Tx DMA descriptors array */
	struct mvpp2_tx_desc *descs;

	/* DMA address of the Tx DMA descriptors array */
	dma_addr_t descs_dma;

	/* Index of the last Tx DMA descriptor */
	int last_desc;

	/* Index of the next Tx DMA descriptor to process */
	int next_desc_to_proc;
};

struct mvpp2_rx_queue {
	/* RX queue number, in the range 0-31 for physical RXQs */
	u8 id;

	/* Num of rx descriptors in the rx descriptor ring */
	int size;

	u32 pkts_coal;
	u32 time_coal;

	/* Virtual address of the RX DMA descriptors array */
	struct mvpp2_rx_desc *descs;

	/* DMA address of the RX DMA descriptors array */
	dma_addr_t descs_dma;

	/* Index of the last RX DMA descriptor */
	int last_desc;

	/* Index of the next RX DMA descriptor to process */
	int next_desc_to_proc;

	/* ID of port to which physical RXQ is mapped */
	int port;

	/* Port's logic RXQ number to which physical RXQ is mapped */
	int logic_rxq;
};

union mvpp2_prs_tcam_entry {
	u32 word[MVPP2_PRS_TCAM_WORDS];
	u8  byte[MVPP2_PRS_TCAM_WORDS * 4];
};

union mvpp2_prs_sram_entry {
	u32 word[MVPP2_PRS_SRAM_WORDS];
	u8  byte[MVPP2_PRS_SRAM_WORDS * 4];
};

struct mvpp2_prs_entry {
	u32 index;
	union mvpp2_prs_tcam_entry tcam;
	union mvpp2_prs_sram_entry sram;
};

struct mvpp2_prs_shadow {
	bool valid;
	bool finish;

	/* Lookup ID */
	int lu;

	/* User defined offset */
	int udf;

	/* Result info */
	u32 ri;
	u32 ri_mask;
};

struct mvpp2_cls_flow_entry {
	u32 index;
	u32 data[MVPP2_CLS_FLOWS_TBL_DATA_WORDS];
};

struct mvpp2_cls_lookup_entry {
	u32 lkpid;
	u32 way;
	u32 data;
};

struct mvpp2_bm_pool {
	/* Pool number in the range 0-7 */
	int id;
	enum mvpp2_bm_type type;

	/* Buffer Pointers Pool External (BPPE) size */
	int size;
	/* BPPE size in bytes */
	int size_bytes;
	/* Number of buffers for this pool */
	int buf_num;
	/* Pool buffer size */
	int buf_size;
	/* Packet size */
	int pkt_size;
	int frag_size;

	/* BPPE virtual base address */
	u32 *virt_addr;
	/* BPPE DMA base address */
	dma_addr_t dma_addr;

	/* Ports using BM pool */
	u32 port_map;
};

/* Queue modes */
#define MVPP2_QDIST_SINGLE_MODE	0
#define MVPP2_QDIST_MULTI_MODE	1

static int queue_mode = MVPP2_QDIST_SINGLE_MODE;

module_param(queue_mode, int, 0444);
MODULE_PARM_DESC(queue_mode, "Set queue_mode (single=0, multi=1)");

#define MVPP2_DRIVER_NAME "mvpp2"
#define MVPP2_DRIVER_VERSION "1.0"

/* Utility/helper methods */

static void mvpp2_write(struct mvpp2 *priv, u32 offset, u32 data)
{
	writel(data, priv->swth_base[0] + offset);
}

static u32 mvpp2_read(struct mvpp2 *priv, u32 offset)
{
	return readl(priv->swth_base[0] + offset);
}

/* These accessors should be used to access:
 *
 * - per-CPU registers, where each CPU has its own copy of the
 *   register.
 *
 *   MVPP2_BM_VIRT_ALLOC_REG
 *   MVPP2_BM_ADDR_HIGH_ALLOC
 *   MVPP22_BM_ADDR_HIGH_RLS_REG
 *   MVPP2_BM_VIRT_RLS_REG
 *   MVPP2_ISR_RX_TX_CAUSE_REG
 *   MVPP2_ISR_RX_TX_MASK_REG
 *   MVPP2_TXQ_NUM_REG
 *   MVPP2_AGGR_TXQ_UPDATE_REG
 *   MVPP2_TXQ_RSVD_REQ_REG
 *   MVPP2_TXQ_RSVD_RSLT_REG
 *   MVPP2_TXQ_SENT_REG
 *   MVPP2_RXQ_NUM_REG
 *
 * - global registers that must be accessed through a specific CPU
 *   window, because they are related to an access to a per-CPU
 *   register
 *
 *   MVPP2_BM_PHY_ALLOC_REG    (related to MVPP2_BM_VIRT_ALLOC_REG)
 *   MVPP2_BM_PHY_RLS_REG      (related to MVPP2_BM_VIRT_RLS_REG)
 *   MVPP2_RXQ_THRESH_REG      (related to MVPP2_RXQ_NUM_REG)
 *   MVPP2_RXQ_DESC_ADDR_REG   (related to MVPP2_RXQ_NUM_REG)
 *   MVPP2_RXQ_DESC_SIZE_REG   (related to MVPP2_RXQ_NUM_REG)
 *   MVPP2_RXQ_INDEX_REG       (related to MVPP2_RXQ_NUM_REG)
 *   MVPP2_TXQ_PENDING_REG     (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_DESC_ADDR_REG   (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_DESC_SIZE_REG   (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_INDEX_REG       (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_PENDING_REG     (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_PREF_BUF_REG    (related to MVPP2_TXQ_NUM_REG)
 *   MVPP2_TXQ_PREF_BUF_REG    (related to MVPP2_TXQ_NUM_REG)
 */
static void mvpp2_percpu_write(struct mvpp2 *priv, int cpu,
			       u32 offset, u32 data)
{
	writel(data, priv->swth_base[cpu] + offset);
}

static u32 mvpp2_percpu_read(struct mvpp2 *priv, int cpu,
			     u32 offset)
{
	return readl(priv->swth_base[cpu] + offset);
}

static dma_addr_t mvpp2_txdesc_dma_addr_get(struct mvpp2_port *port,
					    struct mvpp2_tx_desc *tx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return tx_desc->pp21.buf_dma_addr;
	else
		return tx_desc->pp22.buf_dma_addr_ptp & GENMASK_ULL(40, 0);
}

static void mvpp2_txdesc_dma_addr_set(struct mvpp2_port *port,
				      struct mvpp2_tx_desc *tx_desc,
				      dma_addr_t dma_addr)
{
	if (port->priv->hw_version == MVPP21) {
		tx_desc->pp21.buf_dma_addr = dma_addr;
	} else {
		u64 val = (u64)dma_addr;

		tx_desc->pp22.buf_dma_addr_ptp &= ~GENMASK_ULL(40, 0);
		tx_desc->pp22.buf_dma_addr_ptp |= val;
	}
}

static size_t mvpp2_txdesc_size_get(struct mvpp2_port *port,
				    struct mvpp2_tx_desc *tx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return tx_desc->pp21.data_size;
	else
		return tx_desc->pp22.data_size;
}

static void mvpp2_txdesc_size_set(struct mvpp2_port *port,
				  struct mvpp2_tx_desc *tx_desc,
				  size_t size)
{
	if (port->priv->hw_version == MVPP21)
		tx_desc->pp21.data_size = size;
	else
		tx_desc->pp22.data_size = size;
}

static void mvpp2_txdesc_txq_set(struct mvpp2_port *port,
				 struct mvpp2_tx_desc *tx_desc,
				 unsigned int txq)
{
	if (port->priv->hw_version == MVPP21)
		tx_desc->pp21.phys_txq = txq;
	else
		tx_desc->pp22.phys_txq = txq;
}

static void mvpp2_txdesc_cmd_set(struct mvpp2_port *port,
				 struct mvpp2_tx_desc *tx_desc,
				 unsigned int command)
{
	if (port->priv->hw_version == MVPP21)
		tx_desc->pp21.command = command;
	else
		tx_desc->pp22.command = command;
}

static void mvpp2_txdesc_offset_set(struct mvpp2_port *port,
				    struct mvpp2_tx_desc *tx_desc,
				    unsigned int offset)
{
	if (port->priv->hw_version == MVPP21)
		tx_desc->pp21.packet_offset = offset;
	else
		tx_desc->pp22.packet_offset = offset;
}

static unsigned int mvpp2_txdesc_offset_get(struct mvpp2_port *port,
					    struct mvpp2_tx_desc *tx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return tx_desc->pp21.packet_offset;
	else
		return tx_desc->pp22.packet_offset;
}

static dma_addr_t mvpp2_rxdesc_dma_addr_get(struct mvpp2_port *port,
					    struct mvpp2_rx_desc *rx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return rx_desc->pp21.buf_dma_addr;
	else
		return rx_desc->pp22.buf_dma_addr_key_hash & GENMASK_ULL(40, 0);
}

static unsigned long mvpp2_rxdesc_cookie_get(struct mvpp2_port *port,
					     struct mvpp2_rx_desc *rx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return rx_desc->pp21.buf_cookie;
	else
		return rx_desc->pp22.buf_cookie_misc & GENMASK_ULL(40, 0);
}

static size_t mvpp2_rxdesc_size_get(struct mvpp2_port *port,
				    struct mvpp2_rx_desc *rx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return rx_desc->pp21.data_size;
	else
		return rx_desc->pp22.data_size;
}

static u32 mvpp2_rxdesc_status_get(struct mvpp2_port *port,
				   struct mvpp2_rx_desc *rx_desc)
{
	if (port->priv->hw_version == MVPP21)
		return rx_desc->pp21.status;
	else
		return rx_desc->pp22.status;
}

static void mvpp2_txq_inc_get(struct mvpp2_txq_pcpu *txq_pcpu)
{
	txq_pcpu->txq_get_index++;
	if (txq_pcpu->txq_get_index == txq_pcpu->size)
		txq_pcpu->txq_get_index = 0;
}

static void mvpp2_txq_inc_put(struct mvpp2_port *port,
			      struct mvpp2_txq_pcpu *txq_pcpu,
			      struct sk_buff *skb,
			      struct mvpp2_tx_desc *tx_desc)
{
	struct mvpp2_txq_pcpu_buf *tx_buf =
		txq_pcpu->buffs + txq_pcpu->txq_put_index;
	tx_buf->skb = skb;
	tx_buf->size = mvpp2_txdesc_size_get(port, tx_desc);
	tx_buf->dma = mvpp2_txdesc_dma_addr_get(port, tx_desc) +
		mvpp2_txdesc_offset_get(port, tx_desc);
	txq_pcpu->txq_put_index++;
	if (txq_pcpu->txq_put_index == txq_pcpu->size)
		txq_pcpu->txq_put_index = 0;
}

/* Get number of physical egress port */
static inline int mvpp2_egress_port(struct mvpp2_port *port)
{
	return MVPP2_MAX_TCONT + port->id;
}

/* Get number of physical TXQ */
static inline int mvpp2_txq_phys(int port, int txq)
{
	return (MVPP2_MAX_TCONT + port) * MVPP2_MAX_TXQ + txq;
}

/* Parser configuration routines */

/* Update parser tcam and sram hw entries */
static int mvpp2_prs_hw_write(struct mvpp2 *priv, struct mvpp2_prs_entry *pe)
{
	int i;

	if (pe->index > MVPP2_PRS_TCAM_SRAM_SIZE - 1)
		return -EINVAL;

	/* Clear entry invalidation bit */
	pe->tcam.word[MVPP2_PRS_TCAM_INV_WORD] &= ~MVPP2_PRS_TCAM_INV_MASK;

	/* Write tcam index - indirect access */
	mvpp2_write(priv, MVPP2_PRS_TCAM_IDX_REG, pe->index);
	for (i = 0; i < MVPP2_PRS_TCAM_WORDS; i++)
		mvpp2_write(priv, MVPP2_PRS_TCAM_DATA_REG(i), pe->tcam.word[i]);

	/* Write sram index - indirect access */
	mvpp2_write(priv, MVPP2_PRS_SRAM_IDX_REG, pe->index);
	for (i = 0; i < MVPP2_PRS_SRAM_WORDS; i++)
		mvpp2_write(priv, MVPP2_PRS_SRAM_DATA_REG(i), pe->sram.word[i]);

	return 0;
}

/* Read tcam entry from hw */
static int mvpp2_prs_hw_read(struct mvpp2 *priv, struct mvpp2_prs_entry *pe)
{
	int i;

	if (pe->index > MVPP2_PRS_TCAM_SRAM_SIZE - 1)
		return -EINVAL;

	/* Write tcam index - indirect access */
	mvpp2_write(priv, MVPP2_PRS_TCAM_IDX_REG, pe->index);

	pe->tcam.word[MVPP2_PRS_TCAM_INV_WORD] = mvpp2_read(priv,
			      MVPP2_PRS_TCAM_DATA_REG(MVPP2_PRS_TCAM_INV_WORD));
	if (pe->tcam.word[MVPP2_PRS_TCAM_INV_WORD] & MVPP2_PRS_TCAM_INV_MASK)
		return MVPP2_PRS_TCAM_ENTRY_INVALID;

	for (i = 0; i < MVPP2_PRS_TCAM_WORDS; i++)
		pe->tcam.word[i] = mvpp2_read(priv, MVPP2_PRS_TCAM_DATA_REG(i));

	/* Write sram index - indirect access */
	mvpp2_write(priv, MVPP2_PRS_SRAM_IDX_REG, pe->index);
	for (i = 0; i < MVPP2_PRS_SRAM_WORDS; i++)
		pe->sram.word[i] = mvpp2_read(priv, MVPP2_PRS_SRAM_DATA_REG(i));

	return 0;
}

/* Invalidate tcam hw entry */
static void mvpp2_prs_hw_inv(struct mvpp2 *priv, int index)
{
	/* Write index - indirect access */
	mvpp2_write(priv, MVPP2_PRS_TCAM_IDX_REG, index);
	mvpp2_write(priv, MVPP2_PRS_TCAM_DATA_REG(MVPP2_PRS_TCAM_INV_WORD),
		    MVPP2_PRS_TCAM_INV_MASK);
}

/* Enable shadow table entry and set its lookup ID */
static void mvpp2_prs_shadow_set(struct mvpp2 *priv, int index, int lu)
{
	priv->prs_shadow[index].valid = true;
	priv->prs_shadow[index].lu = lu;
}

/* Update ri fields in shadow table entry */
static void mvpp2_prs_shadow_ri_set(struct mvpp2 *priv, int index,
				    unsigned int ri, unsigned int ri_mask)
{
	priv->prs_shadow[index].ri_mask = ri_mask;
	priv->prs_shadow[index].ri = ri;
}

/* Update lookup field in tcam sw entry */
static void mvpp2_prs_tcam_lu_set(struct mvpp2_prs_entry *pe, unsigned int lu)
{
	int enable_off = MVPP2_PRS_TCAM_EN_OFFS(MVPP2_PRS_TCAM_LU_BYTE);

	pe->tcam.byte[MVPP2_PRS_TCAM_LU_BYTE] = lu;
	pe->tcam.byte[enable_off] = MVPP2_PRS_LU_MASK;
}

/* Update mask for single port in tcam sw entry */
static void mvpp2_prs_tcam_port_set(struct mvpp2_prs_entry *pe,
				    unsigned int port, bool add)
{
	int enable_off = MVPP2_PRS_TCAM_EN_OFFS(MVPP2_PRS_TCAM_PORT_BYTE);

	if (add)
		pe->tcam.byte[enable_off] &= ~(1 << port);
	else
		pe->tcam.byte[enable_off] |= 1 << port;
}

/* Update port map in tcam sw entry */
static void mvpp2_prs_tcam_port_map_set(struct mvpp2_prs_entry *pe,
					unsigned int ports)
{
	unsigned char port_mask = MVPP2_PRS_PORT_MASK;
	int enable_off = MVPP2_PRS_TCAM_EN_OFFS(MVPP2_PRS_TCAM_PORT_BYTE);

	pe->tcam.byte[MVPP2_PRS_TCAM_PORT_BYTE] = 0;
	pe->tcam.byte[enable_off] &= ~port_mask;
	pe->tcam.byte[enable_off] |= ~ports & MVPP2_PRS_PORT_MASK;
}

/* Obtain port map from tcam sw entry */
static unsigned int mvpp2_prs_tcam_port_map_get(struct mvpp2_prs_entry *pe)
{
	int enable_off = MVPP2_PRS_TCAM_EN_OFFS(MVPP2_PRS_TCAM_PORT_BYTE);

	return ~(pe->tcam.byte[enable_off]) & MVPP2_PRS_PORT_MASK;
}

/* Set byte of data and its enable bits in tcam sw entry */
static void mvpp2_prs_tcam_data_byte_set(struct mvpp2_prs_entry *pe,
					 unsigned int offs, unsigned char byte,
					 unsigned char enable)
{
	pe->tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE(offs)] = byte;
	pe->tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE_EN(offs)] = enable;
}

/* Get byte of data and its enable bits from tcam sw entry */
static void mvpp2_prs_tcam_data_byte_get(struct mvpp2_prs_entry *pe,
					 unsigned int offs, unsigned char *byte,
					 unsigned char *enable)
{
	*byte = pe->tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE(offs)];
	*enable = pe->tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE_EN(offs)];
}

/* Compare tcam data bytes with a pattern */
static bool mvpp2_prs_tcam_data_cmp(struct mvpp2_prs_entry *pe, int offs,
				    u16 data)
{
	int off = MVPP2_PRS_TCAM_DATA_BYTE(offs);
	u16 tcam_data;

	tcam_data = (8 << pe->tcam.byte[off + 1]) | pe->tcam.byte[off];
	if (tcam_data != data)
		return false;
	return true;
}

/* Update ai bits in tcam sw entry */
static void mvpp2_prs_tcam_ai_update(struct mvpp2_prs_entry *pe,
				     unsigned int bits, unsigned int enable)
{
	int i, ai_idx = MVPP2_PRS_TCAM_AI_BYTE;

	for (i = 0; i < MVPP2_PRS_AI_BITS; i++) {

		if (!(enable & BIT(i)))
			continue;

		if (bits & BIT(i))
			pe->tcam.byte[ai_idx] |= 1 << i;
		else
			pe->tcam.byte[ai_idx] &= ~(1 << i);
	}

	pe->tcam.byte[MVPP2_PRS_TCAM_EN_OFFS(ai_idx)] |= enable;
}

/* Get ai bits from tcam sw entry */
static int mvpp2_prs_tcam_ai_get(struct mvpp2_prs_entry *pe)
{
	return pe->tcam.byte[MVPP2_PRS_TCAM_AI_BYTE];
}

/* Set ethertype in tcam sw entry */
static void mvpp2_prs_match_etype(struct mvpp2_prs_entry *pe, int offset,
				  unsigned short ethertype)
{
	mvpp2_prs_tcam_data_byte_set(pe, offset + 0, ethertype >> 8, 0xff);
	mvpp2_prs_tcam_data_byte_set(pe, offset + 1, ethertype & 0xff, 0xff);
}

/* Set bits in sram sw entry */
static void mvpp2_prs_sram_bits_set(struct mvpp2_prs_entry *pe, int bit_num,
				    int val)
{
	pe->sram.byte[MVPP2_BIT_TO_BYTE(bit_num)] |= (val << (bit_num % 8));
}

/* Clear bits in sram sw entry */
static void mvpp2_prs_sram_bits_clear(struct mvpp2_prs_entry *pe, int bit_num,
				      int val)
{
	pe->sram.byte[MVPP2_BIT_TO_BYTE(bit_num)] &= ~(val << (bit_num % 8));
}

/* Update ri bits in sram sw entry */
static void mvpp2_prs_sram_ri_update(struct mvpp2_prs_entry *pe,
				     unsigned int bits, unsigned int mask)
{
	unsigned int i;

	for (i = 0; i < MVPP2_PRS_SRAM_RI_CTRL_BITS; i++) {
		int ri_off = MVPP2_PRS_SRAM_RI_OFFS;

		if (!(mask & BIT(i)))
			continue;

		if (bits & BIT(i))
			mvpp2_prs_sram_bits_set(pe, ri_off + i, 1);
		else
			mvpp2_prs_sram_bits_clear(pe, ri_off + i, 1);

		mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_RI_CTRL_OFFS + i, 1);
	}
}

/* Obtain ri bits from sram sw entry */
static int mvpp2_prs_sram_ri_get(struct mvpp2_prs_entry *pe)
{
	return pe->sram.word[MVPP2_PRS_SRAM_RI_WORD];
}

/* Update ai bits in sram sw entry */
static void mvpp2_prs_sram_ai_update(struct mvpp2_prs_entry *pe,
				     unsigned int bits, unsigned int mask)
{
	unsigned int i;
	int ai_off = MVPP2_PRS_SRAM_AI_OFFS;

	for (i = 0; i < MVPP2_PRS_SRAM_AI_CTRL_BITS; i++) {

		if (!(mask & BIT(i)))
			continue;

		if (bits & BIT(i))
			mvpp2_prs_sram_bits_set(pe, ai_off + i, 1);
		else
			mvpp2_prs_sram_bits_clear(pe, ai_off + i, 1);

		mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_AI_CTRL_OFFS + i, 1);
	}
}

/* Read ai bits from sram sw entry */
static int mvpp2_prs_sram_ai_get(struct mvpp2_prs_entry *pe)
{
	u8 bits;
	int ai_off = MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_AI_OFFS);
	int ai_en_off = ai_off + 1;
	int ai_shift = MVPP2_PRS_SRAM_AI_OFFS % 8;

	bits = (pe->sram.byte[ai_off] >> ai_shift) |
	       (pe->sram.byte[ai_en_off] << (8 - ai_shift));

	return bits;
}

/* In sram sw entry set lookup ID field of the tcam key to be used in the next
 * lookup interation
 */
static void mvpp2_prs_sram_next_lu_set(struct mvpp2_prs_entry *pe,
				       unsigned int lu)
{
	int sram_next_off = MVPP2_PRS_SRAM_NEXT_LU_OFFS;

	mvpp2_prs_sram_bits_clear(pe, sram_next_off,
				  MVPP2_PRS_SRAM_NEXT_LU_MASK);
	mvpp2_prs_sram_bits_set(pe, sram_next_off, lu);
}

/* In the sram sw entry set sign and value of the next lookup offset
 * and the offset value generated to the classifier
 */
static void mvpp2_prs_sram_shift_set(struct mvpp2_prs_entry *pe, int shift,
				     unsigned int op)
{
	/* Set sign */
	if (shift < 0) {
		mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_SHIFT_SIGN_BIT, 1);
		shift = 0 - shift;
	} else {
		mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_SHIFT_SIGN_BIT, 1);
	}

	/* Set value */
	pe->sram.byte[MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_SHIFT_OFFS)] =
							   (unsigned char)shift;

	/* Reset and set operation */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_OP_SEL_SHIFT_OFFS,
				  MVPP2_PRS_SRAM_OP_SEL_SHIFT_MASK);
	mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_OP_SEL_SHIFT_OFFS, op);

	/* Set base offset as current */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_OP_SEL_BASE_OFFS, 1);
}

/* In the sram sw entry set sign and value of the user defined offset
 * generated to the classifier
 */
static void mvpp2_prs_sram_offset_set(struct mvpp2_prs_entry *pe,
				      unsigned int type, int offset,
				      unsigned int op)
{
	/* Set sign */
	if (offset < 0) {
		mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_UDF_SIGN_BIT, 1);
		offset = 0 - offset;
	} else {
		mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_UDF_SIGN_BIT, 1);
	}

	/* Set value */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_UDF_OFFS,
				  MVPP2_PRS_SRAM_UDF_MASK);
	mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_UDF_OFFS, offset);
	pe->sram.byte[MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_UDF_OFFS +
					MVPP2_PRS_SRAM_UDF_BITS)] &=
	      ~(MVPP2_PRS_SRAM_UDF_MASK >> (8 - (MVPP2_PRS_SRAM_UDF_OFFS % 8)));
	pe->sram.byte[MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_UDF_OFFS +
					MVPP2_PRS_SRAM_UDF_BITS)] |=
				(offset >> (8 - (MVPP2_PRS_SRAM_UDF_OFFS % 8)));

	/* Set offset type */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_UDF_TYPE_OFFS,
				  MVPP2_PRS_SRAM_UDF_TYPE_MASK);
	mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_UDF_TYPE_OFFS, type);

	/* Set offset operation */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_MASK);
	mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS, op);

	pe->sram.byte[MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS +
					MVPP2_PRS_SRAM_OP_SEL_UDF_BITS)] &=
					     ~(MVPP2_PRS_SRAM_OP_SEL_UDF_MASK >>
				    (8 - (MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS % 8)));

	pe->sram.byte[MVPP2_BIT_TO_BYTE(MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS +
					MVPP2_PRS_SRAM_OP_SEL_UDF_BITS)] |=
			     (op >> (8 - (MVPP2_PRS_SRAM_OP_SEL_UDF_OFFS % 8)));

	/* Set base offset as current */
	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_OP_SEL_BASE_OFFS, 1);
}

/* Find parser flow entry */
static struct mvpp2_prs_entry *mvpp2_prs_flow_find(struct mvpp2 *priv, int flow)
{
	struct mvpp2_prs_entry *pe;
	int tid;

	pe = kzalloc(sizeof(*pe), GFP_KERNEL);
	if (!pe)
		return NULL;
	mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_FLOWS);

	/* Go through the all entires with MVPP2_PRS_LU_FLOWS */
	for (tid = MVPP2_PRS_TCAM_SRAM_SIZE - 1; tid >= 0; tid--) {
		u8 bits;

		if (!priv->prs_shadow[tid].valid ||
		    priv->prs_shadow[tid].lu != MVPP2_PRS_LU_FLOWS)
			continue;

		pe->index = tid;
		mvpp2_prs_hw_read(priv, pe);
		bits = mvpp2_prs_sram_ai_get(pe);

		/* Sram store classification lookup ID in AI bits [5:0] */
		if ((bits & MVPP2_PRS_FLOW_ID_MASK) == flow)
			return pe;
	}
	kfree(pe);

	return NULL;
}

/* Return first free tcam index, seeking from start to end */
static int mvpp2_prs_tcam_first_free(struct mvpp2 *priv, unsigned char start,
				     unsigned char end)
{
	int tid;

	if (start > end)
		swap(start, end);

	if (end >= MVPP2_PRS_TCAM_SRAM_SIZE)
		end = MVPP2_PRS_TCAM_SRAM_SIZE - 1;

	for (tid = start; tid <= end; tid++) {
		if (!priv->prs_shadow[tid].valid)
			return tid;
	}

	return -EINVAL;
}

/* Enable/disable dropping all mac da's */
static void mvpp2_prs_mac_drop_all_set(struct mvpp2 *priv, int port, bool add)
{
	struct mvpp2_prs_entry pe;

	if (priv->prs_shadow[MVPP2_PE_DROP_ALL].valid) {
		/* Entry exist - update port only */
		pe.index = MVPP2_PE_DROP_ALL;
		mvpp2_prs_hw_read(priv, &pe);
	} else {
		/* Entry doesn't exist - create new */
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_MAC);
		pe.index = MVPP2_PE_DROP_ALL;

		/* Non-promiscuous mode for all ports - DROP unknown packets */
		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_DROP_MASK,
					 MVPP2_PRS_RI_DROP_MASK);

		mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
		mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);

		/* Update shadow table */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MAC);

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(&pe, 0);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(&pe, port, add);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Set port to promiscuous mode */
static void mvpp2_prs_mac_promisc_set(struct mvpp2 *priv, int port, bool add)
{
	struct mvpp2_prs_entry pe;

	/* Promiscuous mode - Accept unknown packets */

	if (priv->prs_shadow[MVPP2_PE_MAC_PROMISCUOUS].valid) {
		/* Entry exist - update port only */
		pe.index = MVPP2_PE_MAC_PROMISCUOUS;
		mvpp2_prs_hw_read(priv, &pe);
	} else {
		/* Entry doesn't exist - create new */
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_MAC);
		pe.index = MVPP2_PE_MAC_PROMISCUOUS;

		/* Continue - set next lookup */
		mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_DSA);

		/* Set result info bits */
		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L2_UCAST,
					 MVPP2_PRS_RI_L2_CAST_MASK);

		/* Shift to ethertype */
		mvpp2_prs_sram_shift_set(&pe, 2 * ETH_ALEN,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(&pe, 0);

		/* Update shadow table */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MAC);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(&pe, port, add);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Accept multicast */
static void mvpp2_prs_mac_multi_set(struct mvpp2 *priv, int port, int index,
				    bool add)
{
	struct mvpp2_prs_entry pe;
	unsigned char da_mc;

	/* Ethernet multicast address first byte is
	 * 0x01 for IPv4 and 0x33 for IPv6
	 */
	da_mc = (index == MVPP2_PE_MAC_MC_ALL) ? 0x01 : 0x33;

	if (priv->prs_shadow[index].valid) {
		/* Entry exist - update port only */
		pe.index = index;
		mvpp2_prs_hw_read(priv, &pe);
	} else {
		/* Entry doesn't exist - create new */
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_MAC);
		pe.index = index;

		/* Continue - set next lookup */
		mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_DSA);

		/* Set result info bits */
		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L2_MCAST,
					 MVPP2_PRS_RI_L2_CAST_MASK);

		/* Update tcam entry data first byte */
		mvpp2_prs_tcam_data_byte_set(&pe, 0, da_mc, 0xff);

		/* Shift to ethertype */
		mvpp2_prs_sram_shift_set(&pe, 2 * ETH_ALEN,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(&pe, 0);

		/* Update shadow table */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MAC);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(&pe, port, add);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Set entry for dsa packets */
static void mvpp2_prs_dsa_tag_set(struct mvpp2 *priv, int port, bool add,
				  bool tagged, bool extend)
{
	struct mvpp2_prs_entry pe;
	int tid, shift;

	if (extend) {
		tid = tagged ? MVPP2_PE_EDSA_TAGGED : MVPP2_PE_EDSA_UNTAGGED;
		shift = 8;
	} else {
		tid = tagged ? MVPP2_PE_DSA_TAGGED : MVPP2_PE_DSA_UNTAGGED;
		shift = 4;
	}

	if (priv->prs_shadow[tid].valid) {
		/* Entry exist - update port only */
		pe.index = tid;
		mvpp2_prs_hw_read(priv, &pe);
	} else {
		/* Entry doesn't exist - create new */
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_DSA);
		pe.index = tid;

		/* Shift 4 bytes if DSA tag or 8 bytes in case of EDSA tag*/
		mvpp2_prs_sram_shift_set(&pe, shift,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

		/* Update shadow table */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_DSA);

		if (tagged) {
			/* Set tagged bit in DSA tag */
			mvpp2_prs_tcam_data_byte_set(&pe, 0,
						     MVPP2_PRS_TCAM_DSA_TAGGED_BIT,
						     MVPP2_PRS_TCAM_DSA_TAGGED_BIT);
			/* Clear all ai bits for next iteration */
			mvpp2_prs_sram_ai_update(&pe, 0,
						 MVPP2_PRS_SRAM_AI_MASK);
			/* If packet is tagged continue check vlans */
			mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_VLAN);
		} else {
			/* Set result info bits to 'no vlans' */
			mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_VLAN_NONE,
						 MVPP2_PRS_RI_VLAN_MASK);
			mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_L2);
		}

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(&pe, 0);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(&pe, port, add);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Set entry for dsa ethertype */
static void mvpp2_prs_dsa_tag_ethertype_set(struct mvpp2 *priv, int port,
					    bool add, bool tagged, bool extend)
{
	struct mvpp2_prs_entry pe;
	int tid, shift, port_mask;

	if (extend) {
		tid = tagged ? MVPP2_PE_ETYPE_EDSA_TAGGED :
		      MVPP2_PE_ETYPE_EDSA_UNTAGGED;
		port_mask = 0;
		shift = 8;
	} else {
		tid = tagged ? MVPP2_PE_ETYPE_DSA_TAGGED :
		      MVPP2_PE_ETYPE_DSA_UNTAGGED;
		port_mask = MVPP2_PRS_PORT_MASK;
		shift = 4;
	}

	if (priv->prs_shadow[tid].valid) {
		/* Entry exist - update port only */
		pe.index = tid;
		mvpp2_prs_hw_read(priv, &pe);
	} else {
		/* Entry doesn't exist - create new */
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_DSA);
		pe.index = tid;

		/* Set ethertype */
		mvpp2_prs_match_etype(&pe, 0, ETH_P_EDSA);
		mvpp2_prs_match_etype(&pe, 2, 0);

		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_DSA_MASK,
					 MVPP2_PRS_RI_DSA_MASK);
		/* Shift ethertype + 2 byte reserved + tag*/
		mvpp2_prs_sram_shift_set(&pe, 2 + MVPP2_ETH_TYPE_LEN + shift,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

		/* Update shadow table */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_DSA);

		if (tagged) {
			/* Set tagged bit in DSA tag */
			mvpp2_prs_tcam_data_byte_set(&pe,
						     MVPP2_ETH_TYPE_LEN + 2 + 3,
						 MVPP2_PRS_TCAM_DSA_TAGGED_BIT,
						 MVPP2_PRS_TCAM_DSA_TAGGED_BIT);
			/* Clear all ai bits for next iteration */
			mvpp2_prs_sram_ai_update(&pe, 0,
						 MVPP2_PRS_SRAM_AI_MASK);
			/* If packet is tagged continue check vlans */
			mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_VLAN);
		} else {
			/* Set result info bits to 'no vlans' */
			mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_VLAN_NONE,
						 MVPP2_PRS_RI_VLAN_MASK);
			mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_L2);
		}
		/* Mask/unmask all ports, depending on dsa type */
		mvpp2_prs_tcam_port_map_set(&pe, port_mask);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(&pe, port, add);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Search for existing single/triple vlan entry */
static struct mvpp2_prs_entry *mvpp2_prs_vlan_find(struct mvpp2 *priv,
						   unsigned short tpid, int ai)
{
	struct mvpp2_prs_entry *pe;
	int tid;

	pe = kzalloc(sizeof(*pe), GFP_KERNEL);
	if (!pe)
		return NULL;
	mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_VLAN);

	/* Go through the all entries with MVPP2_PRS_LU_VLAN */
	for (tid = MVPP2_PE_FIRST_FREE_TID;
	     tid <= MVPP2_PE_LAST_FREE_TID; tid++) {
		unsigned int ri_bits, ai_bits;
		bool match;

		if (!priv->prs_shadow[tid].valid ||
		    priv->prs_shadow[tid].lu != MVPP2_PRS_LU_VLAN)
			continue;

		pe->index = tid;

		mvpp2_prs_hw_read(priv, pe);
		match = mvpp2_prs_tcam_data_cmp(pe, 0, swab16(tpid));
		if (!match)
			continue;

		/* Get vlan type */
		ri_bits = mvpp2_prs_sram_ri_get(pe);
		ri_bits &= MVPP2_PRS_RI_VLAN_MASK;

		/* Get current ai value from tcam */
		ai_bits = mvpp2_prs_tcam_ai_get(pe);
		/* Clear double vlan bit */
		ai_bits &= ~MVPP2_PRS_DBL_VLAN_AI_BIT;

		if (ai != ai_bits)
			continue;

		if (ri_bits == MVPP2_PRS_RI_VLAN_SINGLE ||
		    ri_bits == MVPP2_PRS_RI_VLAN_TRIPLE)
			return pe;
	}
	kfree(pe);

	return NULL;
}

/* Add/update single/triple vlan entry */
static int mvpp2_prs_vlan_add(struct mvpp2 *priv, unsigned short tpid, int ai,
			      unsigned int port_map)
{
	struct mvpp2_prs_entry *pe;
	int tid_aux, tid;
	int ret = 0;

	pe = mvpp2_prs_vlan_find(priv, tpid, ai);

	if (!pe) {
		/* Create new tcam entry */
		tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_LAST_FREE_TID,
						MVPP2_PE_FIRST_FREE_TID);
		if (tid < 0)
			return tid;

		pe = kzalloc(sizeof(*pe), GFP_KERNEL);
		if (!pe)
			return -ENOMEM;

		/* Get last double vlan tid */
		for (tid_aux = MVPP2_PE_LAST_FREE_TID;
		     tid_aux >= MVPP2_PE_FIRST_FREE_TID; tid_aux--) {
			unsigned int ri_bits;

			if (!priv->prs_shadow[tid_aux].valid ||
			    priv->prs_shadow[tid_aux].lu != MVPP2_PRS_LU_VLAN)
				continue;

			pe->index = tid_aux;
			mvpp2_prs_hw_read(priv, pe);
			ri_bits = mvpp2_prs_sram_ri_get(pe);
			if ((ri_bits & MVPP2_PRS_RI_VLAN_MASK) ==
			    MVPP2_PRS_RI_VLAN_DOUBLE)
				break;
		}

		if (tid <= tid_aux) {
			ret = -EINVAL;
			goto free_pe;
		}

		memset(pe, 0, sizeof(*pe));
		mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_VLAN);
		pe->index = tid;

		mvpp2_prs_match_etype(pe, 0, tpid);

		mvpp2_prs_sram_next_lu_set(pe, MVPP2_PRS_LU_L2);
		/* Shift 4 bytes - skip 1 vlan tag */
		mvpp2_prs_sram_shift_set(pe, MVPP2_VLAN_TAG_LEN,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
		/* Clear all ai bits for next iteration */
		mvpp2_prs_sram_ai_update(pe, 0, MVPP2_PRS_SRAM_AI_MASK);

		if (ai == MVPP2_PRS_SINGLE_VLAN_AI) {
			mvpp2_prs_sram_ri_update(pe, MVPP2_PRS_RI_VLAN_SINGLE,
						 MVPP2_PRS_RI_VLAN_MASK);
		} else {
			ai |= MVPP2_PRS_DBL_VLAN_AI_BIT;
			mvpp2_prs_sram_ri_update(pe, MVPP2_PRS_RI_VLAN_TRIPLE,
						 MVPP2_PRS_RI_VLAN_MASK);
		}
		mvpp2_prs_tcam_ai_update(pe, ai, MVPP2_PRS_SRAM_AI_MASK);

		mvpp2_prs_shadow_set(priv, pe->index, MVPP2_PRS_LU_VLAN);
	}
	/* Update ports' mask */
	mvpp2_prs_tcam_port_map_set(pe, port_map);

	mvpp2_prs_hw_write(priv, pe);
free_pe:
	kfree(pe);

	return ret;
}

/* Get first free double vlan ai number */
static int mvpp2_prs_double_vlan_ai_free_get(struct mvpp2 *priv)
{
	int i;

	for (i = 1; i < MVPP2_PRS_DBL_VLANS_MAX; i++) {
		if (!priv->prs_double_vlans[i])
			return i;
	}

	return -EINVAL;
}

/* Search for existing double vlan entry */
static struct mvpp2_prs_entry *mvpp2_prs_double_vlan_find(struct mvpp2 *priv,
							  unsigned short tpid1,
							  unsigned short tpid2)
{
	struct mvpp2_prs_entry *pe;
	int tid;

	pe = kzalloc(sizeof(*pe), GFP_KERNEL);
	if (!pe)
		return NULL;
	mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_VLAN);

	/* Go through the all entries with MVPP2_PRS_LU_VLAN */
	for (tid = MVPP2_PE_FIRST_FREE_TID;
	     tid <= MVPP2_PE_LAST_FREE_TID; tid++) {
		unsigned int ri_mask;
		bool match;

		if (!priv->prs_shadow[tid].valid ||
		    priv->prs_shadow[tid].lu != MVPP2_PRS_LU_VLAN)
			continue;

		pe->index = tid;
		mvpp2_prs_hw_read(priv, pe);

		match = mvpp2_prs_tcam_data_cmp(pe, 0, swab16(tpid1))
			&& mvpp2_prs_tcam_data_cmp(pe, 4, swab16(tpid2));

		if (!match)
			continue;

		ri_mask = mvpp2_prs_sram_ri_get(pe) & MVPP2_PRS_RI_VLAN_MASK;
		if (ri_mask == MVPP2_PRS_RI_VLAN_DOUBLE)
			return pe;
	}
	kfree(pe);

	return NULL;
}

/* Add or update double vlan entry */
static int mvpp2_prs_double_vlan_add(struct mvpp2 *priv, unsigned short tpid1,
				     unsigned short tpid2,
				     unsigned int port_map)
{
	struct mvpp2_prs_entry *pe;
	int tid_aux, tid, ai, ret = 0;

	pe = mvpp2_prs_double_vlan_find(priv, tpid1, tpid2);

	if (!pe) {
		/* Create new tcam entry */
		tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
				MVPP2_PE_LAST_FREE_TID);
		if (tid < 0)
			return tid;

		pe = kzalloc(sizeof(*pe), GFP_KERNEL);
		if (!pe)
			return -ENOMEM;

		/* Set ai value for new double vlan entry */
		ai = mvpp2_prs_double_vlan_ai_free_get(priv);
		if (ai < 0) {
			ret = ai;
			goto free_pe;
		}

		/* Get first single/triple vlan tid */
		for (tid_aux = MVPP2_PE_FIRST_FREE_TID;
		     tid_aux <= MVPP2_PE_LAST_FREE_TID; tid_aux++) {
			unsigned int ri_bits;

			if (!priv->prs_shadow[tid_aux].valid ||
			    priv->prs_shadow[tid_aux].lu != MVPP2_PRS_LU_VLAN)
				continue;

			pe->index = tid_aux;
			mvpp2_prs_hw_read(priv, pe);
			ri_bits = mvpp2_prs_sram_ri_get(pe);
			ri_bits &= MVPP2_PRS_RI_VLAN_MASK;
			if (ri_bits == MVPP2_PRS_RI_VLAN_SINGLE ||
			    ri_bits == MVPP2_PRS_RI_VLAN_TRIPLE)
				break;
		}

		if (tid >= tid_aux) {
			ret = -ERANGE;
			goto free_pe;
		}

		memset(pe, 0, sizeof(*pe));
		mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_VLAN);
		pe->index = tid;

		priv->prs_double_vlans[ai] = true;

		mvpp2_prs_match_etype(pe, 0, tpid1);
		mvpp2_prs_match_etype(pe, 4, tpid2);

		mvpp2_prs_sram_next_lu_set(pe, MVPP2_PRS_LU_VLAN);
		/* Shift 8 bytes - skip 2 vlan tags */
		mvpp2_prs_sram_shift_set(pe, 2 * MVPP2_VLAN_TAG_LEN,
					 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
		mvpp2_prs_sram_ri_update(pe, MVPP2_PRS_RI_VLAN_DOUBLE,
					 MVPP2_PRS_RI_VLAN_MASK);
		mvpp2_prs_sram_ai_update(pe, ai | MVPP2_PRS_DBL_VLAN_AI_BIT,
					 MVPP2_PRS_SRAM_AI_MASK);

		mvpp2_prs_shadow_set(priv, pe->index, MVPP2_PRS_LU_VLAN);
	}

	/* Update ports' mask */
	mvpp2_prs_tcam_port_map_set(pe, port_map);
	mvpp2_prs_hw_write(priv, pe);
free_pe:
	kfree(pe);
	return ret;
}

/* IPv4 header parsing for fragmentation and L4 offset */
static int mvpp2_prs_ip4_proto(struct mvpp2 *priv, unsigned short proto,
			       unsigned int ri, unsigned int ri_mask)
{
	struct mvpp2_prs_entry pe;
	int tid;

	if ((proto != IPPROTO_TCP) && (proto != IPPROTO_UDP) &&
	    (proto != IPPROTO_IGMP))
		return -EINVAL;

	/* Fragmented packet */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP4);
	pe.index = tid;

	/* Set next lu to IPv4 */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP4);
	mvpp2_prs_sram_shift_set(&pe, 12, MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	/* Set L4 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L4,
				  sizeof(struct iphdr) - 4,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);
	mvpp2_prs_sram_ai_update(&pe, MVPP2_PRS_IPV4_DIP_AI_BIT,
				 MVPP2_PRS_IPV4_DIP_AI_BIT);
	mvpp2_prs_sram_ri_update(&pe, ri | MVPP2_PRS_RI_IP_FRAG_MASK,
				 ri_mask | MVPP2_PRS_RI_IP_FRAG_MASK);

	mvpp2_prs_tcam_data_byte_set(&pe, 5, proto, MVPP2_PRS_TCAM_PROTO_MASK);
	mvpp2_prs_tcam_ai_update(&pe, 0, MVPP2_PRS_IPV4_DIP_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	/* Not fragmented packet */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	pe.index = tid;
	/* Clear ri before updating */
	pe.sram.word[MVPP2_PRS_SRAM_RI_WORD] = 0x0;
	pe.sram.word[MVPP2_PRS_SRAM_RI_CTRL_WORD] = 0x0;
	mvpp2_prs_sram_ri_update(&pe, ri, ri_mask);

	mvpp2_prs_tcam_data_byte_set(&pe, 2, 0x00, MVPP2_PRS_TCAM_PROTO_MASK_L);
	mvpp2_prs_tcam_data_byte_set(&pe, 3, 0x00, MVPP2_PRS_TCAM_PROTO_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* IPv4 L3 multicast or broadcast */
static int mvpp2_prs_ip4_cast(struct mvpp2 *priv, unsigned short l3_cast)
{
	struct mvpp2_prs_entry pe;
	int mask, tid;

	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP4);
	pe.index = tid;

	switch (l3_cast) {
	case MVPP2_PRS_L3_MULTI_CAST:
		mvpp2_prs_tcam_data_byte_set(&pe, 0, MVPP2_PRS_IPV4_MC,
					     MVPP2_PRS_IPV4_MC_MASK);
		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_MCAST,
					 MVPP2_PRS_RI_L3_ADDR_MASK);
		break;
	case  MVPP2_PRS_L3_BROAD_CAST:
		mask = MVPP2_PRS_IPV4_BC_MASK;
		mvpp2_prs_tcam_data_byte_set(&pe, 0, mask, mask);
		mvpp2_prs_tcam_data_byte_set(&pe, 1, mask, mask);
		mvpp2_prs_tcam_data_byte_set(&pe, 2, mask, mask);
		mvpp2_prs_tcam_data_byte_set(&pe, 3, mask, mask);
		mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_BCAST,
					 MVPP2_PRS_RI_L3_ADDR_MASK);
		break;
	default:
		return -EINVAL;
	}

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);

	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV4_DIP_AI_BIT,
				 MVPP2_PRS_IPV4_DIP_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Set entries for protocols over IPv6  */
static int mvpp2_prs_ip6_proto(struct mvpp2 *priv, unsigned short proto,
			       unsigned int ri, unsigned int ri_mask)
{
	struct mvpp2_prs_entry pe;
	int tid;

	if ((proto != IPPROTO_TCP) && (proto != IPPROTO_UDP) &&
	    (proto != IPPROTO_ICMPV6) && (proto != IPPROTO_IPIP))
		return -EINVAL;

	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = tid;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, ri, ri_mask);
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L4,
				  sizeof(struct ipv6hdr) - 6,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	mvpp2_prs_tcam_data_byte_set(&pe, 0, proto, MVPP2_PRS_TCAM_PROTO_MASK);
	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV6_NO_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Write HW */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP6);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* IPv6 L3 multicast entry */
static int mvpp2_prs_ip6_cast(struct mvpp2 *priv, unsigned short l3_cast)
{
	struct mvpp2_prs_entry pe;
	int tid;

	if (l3_cast != MVPP2_PRS_L3_MULTI_CAST)
		return -EINVAL;

	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = tid;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP6);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_MCAST,
				 MVPP2_PRS_RI_L3_ADDR_MASK);
	mvpp2_prs_sram_ai_update(&pe, MVPP2_PRS_IPV6_NO_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Shift back to IPv6 NH */
	mvpp2_prs_sram_shift_set(&pe, -18, MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

	mvpp2_prs_tcam_data_byte_set(&pe, 0, MVPP2_PRS_IPV6_MC,
				     MVPP2_PRS_IPV6_MC_MASK);
	mvpp2_prs_tcam_ai_update(&pe, 0, MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP6);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Parser per-port initialization */
static void mvpp2_prs_hw_port_init(struct mvpp2 *priv, int port, int lu_first,
				   int lu_max, int offset)
{
	u32 val;

	/* Set lookup ID */
	val = mvpp2_read(priv, MVPP2_PRS_INIT_LOOKUP_REG);
	val &= ~MVPP2_PRS_PORT_LU_MASK(port);
	val |=  MVPP2_PRS_PORT_LU_VAL(port, lu_first);
	mvpp2_write(priv, MVPP2_PRS_INIT_LOOKUP_REG, val);

	/* Set maximum number of loops for packet received from port */
	val = mvpp2_read(priv, MVPP2_PRS_MAX_LOOP_REG(port));
	val &= ~MVPP2_PRS_MAX_LOOP_MASK(port);
	val |= MVPP2_PRS_MAX_LOOP_VAL(port, lu_max);
	mvpp2_write(priv, MVPP2_PRS_MAX_LOOP_REG(port), val);

	/* Set initial offset for packet header extraction for the first
	 * searching loop
	 */
	val = mvpp2_read(priv, MVPP2_PRS_INIT_OFFS_REG(port));
	val &= ~MVPP2_PRS_INIT_OFF_MASK(port);
	val |= MVPP2_PRS_INIT_OFF_VAL(port, offset);
	mvpp2_write(priv, MVPP2_PRS_INIT_OFFS_REG(port), val);
}

/* Default flow entries initialization for all ports */
static void mvpp2_prs_def_flow_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int port;

	for (port = 0; port < MVPP2_MAX_PORTS; port++) {
		memset(&pe, 0, sizeof(pe));
		mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
		pe.index = MVPP2_PE_FIRST_DEFAULT_FLOW - port;

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(&pe, 0);

		/* Set flow ID*/
		mvpp2_prs_sram_ai_update(&pe, port, MVPP2_PRS_FLOW_ID_MASK);
		mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_DONE_BIT, 1);

		/* Update shadow table and hw entry */
		mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_FLOWS);
		mvpp2_prs_hw_write(priv, &pe);
	}
}

/* Set default entry for Marvell Header field */
static void mvpp2_prs_mh_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;

	memset(&pe, 0, sizeof(pe));

	pe.index = MVPP2_PE_MH_DEFAULT;
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_MH);
	mvpp2_prs_sram_shift_set(&pe, MVPP2_MH_SIZE,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_MAC);

	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MH);
	mvpp2_prs_hw_write(priv, &pe);
}

/* Set default entires (place holder) for promiscuous, non-promiscuous and
 * multicast MAC addresses
 */
static void mvpp2_prs_mac_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;

	memset(&pe, 0, sizeof(pe));

	/* Non-promiscuous mode for all ports - DROP unknown packets */
	pe.index = MVPP2_PE_MAC_NON_PROMISCUOUS;
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_MAC);

	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_DROP_MASK,
				 MVPP2_PRS_RI_DROP_MASK);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);

	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MAC);
	mvpp2_prs_hw_write(priv, &pe);

	/* place holders only - no ports */
	mvpp2_prs_mac_drop_all_set(priv, 0, false);
	mvpp2_prs_mac_promisc_set(priv, 0, false);
	mvpp2_prs_mac_multi_set(priv, MVPP2_PE_MAC_MC_ALL, 0, false);
	mvpp2_prs_mac_multi_set(priv, MVPP2_PE_MAC_MC_IP6, 0, false);
}

/* Set default entries for various types of dsa packets */
static void mvpp2_prs_dsa_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;

	/* None tagged EDSA entry - place holder */
	mvpp2_prs_dsa_tag_set(priv, 0, false, MVPP2_PRS_UNTAGGED,
			      MVPP2_PRS_EDSA);

	/* Tagged EDSA entry - place holder */
	mvpp2_prs_dsa_tag_set(priv, 0, false, MVPP2_PRS_TAGGED, MVPP2_PRS_EDSA);

	/* None tagged DSA entry - place holder */
	mvpp2_prs_dsa_tag_set(priv, 0, false, MVPP2_PRS_UNTAGGED,
			      MVPP2_PRS_DSA);

	/* Tagged DSA entry - place holder */
	mvpp2_prs_dsa_tag_set(priv, 0, false, MVPP2_PRS_TAGGED, MVPP2_PRS_DSA);

	/* None tagged EDSA ethertype entry - place holder*/
	mvpp2_prs_dsa_tag_ethertype_set(priv, 0, false,
					MVPP2_PRS_UNTAGGED, MVPP2_PRS_EDSA);

	/* Tagged EDSA ethertype entry - place holder*/
	mvpp2_prs_dsa_tag_ethertype_set(priv, 0, false,
					MVPP2_PRS_TAGGED, MVPP2_PRS_EDSA);

	/* None tagged DSA ethertype entry */
	mvpp2_prs_dsa_tag_ethertype_set(priv, 0, true,
					MVPP2_PRS_UNTAGGED, MVPP2_PRS_DSA);

	/* Tagged DSA ethertype entry */
	mvpp2_prs_dsa_tag_ethertype_set(priv, 0, true,
					MVPP2_PRS_TAGGED, MVPP2_PRS_DSA);

	/* Set default entry, in case DSA or EDSA tag not found */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_DSA);
	pe.index = MVPP2_PE_DSA_DEFAULT;
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_VLAN);

	/* Shift 0 bytes */
	mvpp2_prs_sram_shift_set(&pe, 0, MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_MAC);

	/* Clear all sram ai bits for next iteration */
	mvpp2_prs_sram_ai_update(&pe, 0, MVPP2_PRS_SRAM_AI_MASK);

	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	mvpp2_prs_hw_write(priv, &pe);
}

/* Match basic ethertypes */
static int mvpp2_prs_etype_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int tid;

	/* Ethertype: PPPoE */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, ETH_P_PPP_SES);

	mvpp2_prs_sram_shift_set(&pe, MVPP2_PPPOE_HDR_SIZE,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_PPPOE);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_PPPOE_MASK,
				 MVPP2_PRS_RI_PPPOE_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = false;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_PPPOE_MASK,
				MVPP2_PRS_RI_PPPOE_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Ethertype: ARP */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, ETH_P_ARP);

	/* Generate flow in the next iteration*/
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_ARP,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = true;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_L3_ARP,
				MVPP2_PRS_RI_L3_PROTO_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Ethertype: LBTD */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, MVPP2_IP_LBDT_TYPE);

	/* Generate flow in the next iteration*/
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_CPU_CODE_RX_SPEC |
				 MVPP2_PRS_RI_UDF3_RX_SPECIAL,
				 MVPP2_PRS_RI_CPU_CODE_MASK |
				 MVPP2_PRS_RI_UDF3_MASK);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = true;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_CPU_CODE_RX_SPEC |
				MVPP2_PRS_RI_UDF3_RX_SPECIAL,
				MVPP2_PRS_RI_CPU_CODE_MASK |
				MVPP2_PRS_RI_UDF3_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Ethertype: IPv4 without options */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, ETH_P_IP);
	mvpp2_prs_tcam_data_byte_set(&pe, MVPP2_ETH_TYPE_LEN,
				     MVPP2_PRS_IPV4_HEAD | MVPP2_PRS_IPV4_IHL,
				     MVPP2_PRS_IPV4_HEAD_MASK |
				     MVPP2_PRS_IPV4_IHL_MASK);

	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP4);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP4,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Skip eth_type + 4 bytes of IP header */
	mvpp2_prs_sram_shift_set(&pe, MVPP2_ETH_TYPE_LEN + 4,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = false;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_L3_IP4,
				MVPP2_PRS_RI_L3_PROTO_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Ethertype: IPv4 with options */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	pe.index = tid;

	/* Clear tcam data before updating */
	pe.tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE(MVPP2_ETH_TYPE_LEN)] = 0x0;
	pe.tcam.byte[MVPP2_PRS_TCAM_DATA_BYTE_EN(MVPP2_ETH_TYPE_LEN)] = 0x0;

	mvpp2_prs_tcam_data_byte_set(&pe, MVPP2_ETH_TYPE_LEN,
				     MVPP2_PRS_IPV4_HEAD,
				     MVPP2_PRS_IPV4_HEAD_MASK);

	/* Clear ri before updating */
	pe.sram.word[MVPP2_PRS_SRAM_RI_WORD] = 0x0;
	pe.sram.word[MVPP2_PRS_SRAM_RI_CTRL_WORD] = 0x0;
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP4_OPT,
				 MVPP2_PRS_RI_L3_PROTO_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = false;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_L3_IP4_OPT,
				MVPP2_PRS_RI_L3_PROTO_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Ethertype: IPv6 without options */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, ETH_P_IPV6);

	/* Skip DIP of IPV6 header */
	mvpp2_prs_sram_shift_set(&pe, MVPP2_ETH_TYPE_LEN + 8 +
				 MVPP2_MAX_L3_ADDR_SIZE,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP6);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP6,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = false;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_L3_IP6,
				MVPP2_PRS_RI_L3_PROTO_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	/* Default entry for MVPP2_PRS_LU_L2 - Unknown ethtype */
	memset(&pe, 0, sizeof(struct mvpp2_prs_entry));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_L2);
	pe.index = MVPP2_PE_ETH_TYPE_UN;

	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Generate flow in the next iteration*/
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_UN,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Set L3 offset even it's unknown L3 */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_L2);
	priv->prs_shadow[pe.index].udf = MVPP2_PRS_UDF_L2_DEF;
	priv->prs_shadow[pe.index].finish = true;
	mvpp2_prs_shadow_ri_set(priv, pe.index, MVPP2_PRS_RI_L3_UN,
				MVPP2_PRS_RI_L3_PROTO_MASK);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Configure vlan entries and detect up to 2 successive VLAN tags.
 * Possible options:
 * 0x8100, 0x88A8
 * 0x8100, 0x8100
 * 0x8100
 * 0x88A8
 */
static int mvpp2_prs_vlan_init(struct platform_device *pdev, struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int err;

	priv->prs_double_vlans = devm_kcalloc(&pdev->dev, sizeof(bool),
					      MVPP2_PRS_DBL_VLANS_MAX,
					      GFP_KERNEL);
	if (!priv->prs_double_vlans)
		return -ENOMEM;

	/* Double VLAN: 0x8100, 0x88A8 */
	err = mvpp2_prs_double_vlan_add(priv, ETH_P_8021Q, ETH_P_8021AD,
					MVPP2_PRS_PORT_MASK);
	if (err)
		return err;

	/* Double VLAN: 0x8100, 0x8100 */
	err = mvpp2_prs_double_vlan_add(priv, ETH_P_8021Q, ETH_P_8021Q,
					MVPP2_PRS_PORT_MASK);
	if (err)
		return err;

	/* Single VLAN: 0x88a8 */
	err = mvpp2_prs_vlan_add(priv, ETH_P_8021AD, MVPP2_PRS_SINGLE_VLAN_AI,
				 MVPP2_PRS_PORT_MASK);
	if (err)
		return err;

	/* Single VLAN: 0x8100 */
	err = mvpp2_prs_vlan_add(priv, ETH_P_8021Q, MVPP2_PRS_SINGLE_VLAN_AI,
				 MVPP2_PRS_PORT_MASK);
	if (err)
		return err;

	/* Set default double vlan entry */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_VLAN);
	pe.index = MVPP2_PE_VLAN_DBL;

	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_L2);
	/* Clear ai for next iterations */
	mvpp2_prs_sram_ai_update(&pe, 0, MVPP2_PRS_SRAM_AI_MASK);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_VLAN_DOUBLE,
				 MVPP2_PRS_RI_VLAN_MASK);

	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_DBL_VLAN_AI_BIT,
				 MVPP2_PRS_DBL_VLAN_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_VLAN);
	mvpp2_prs_hw_write(priv, &pe);

	/* Set default vlan none entry */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_VLAN);
	pe.index = MVPP2_PE_VLAN_NONE;

	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_L2);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_VLAN_NONE,
				 MVPP2_PRS_RI_VLAN_MASK);

	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_VLAN);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Set entries for PPPoE ethertype */
static int mvpp2_prs_pppoe_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int tid;

	/* IPv4 over PPPoE with options */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_PPPOE);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, PPP_IP);

	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP4);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP4_OPT,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Skip eth_type + 4 bytes of IP header */
	mvpp2_prs_sram_shift_set(&pe, MVPP2_ETH_TYPE_LEN + 4,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_PPPOE);
	mvpp2_prs_hw_write(priv, &pe);

	/* IPv4 over PPPoE without options */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	pe.index = tid;

	mvpp2_prs_tcam_data_byte_set(&pe, MVPP2_ETH_TYPE_LEN,
				     MVPP2_PRS_IPV4_HEAD | MVPP2_PRS_IPV4_IHL,
				     MVPP2_PRS_IPV4_HEAD_MASK |
				     MVPP2_PRS_IPV4_IHL_MASK);

	/* Clear ri before updating */
	pe.sram.word[MVPP2_PRS_SRAM_RI_WORD] = 0x0;
	pe.sram.word[MVPP2_PRS_SRAM_RI_CTRL_WORD] = 0x0;
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP4,
				 MVPP2_PRS_RI_L3_PROTO_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_PPPOE);
	mvpp2_prs_hw_write(priv, &pe);

	/* IPv6 over PPPoE */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_PPPOE);
	pe.index = tid;

	mvpp2_prs_match_etype(&pe, 0, PPP_IPV6);

	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP6);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_IP6,
				 MVPP2_PRS_RI_L3_PROTO_MASK);
	/* Skip eth_type + 4 bytes of IPv6 header */
	mvpp2_prs_sram_shift_set(&pe, MVPP2_ETH_TYPE_LEN + 4,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	/* Set L3 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_PPPOE);
	mvpp2_prs_hw_write(priv, &pe);

	/* Non-IP over PPPoE */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_PPPOE);
	pe.index = tid;

	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_UN,
				 MVPP2_PRS_RI_L3_PROTO_MASK);

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	/* Set L3 offset even if it's unknown L3 */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L3,
				  MVPP2_ETH_TYPE_LEN,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_PPPOE);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Initialize entries for IPv4 */
static int mvpp2_prs_ip4_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int err;

	/* Set entries for TCP, UDP and IGMP over IPv4 */
	err = mvpp2_prs_ip4_proto(priv, IPPROTO_TCP, MVPP2_PRS_RI_L4_TCP,
				  MVPP2_PRS_RI_L4_PROTO_MASK);
	if (err)
		return err;

	err = mvpp2_prs_ip4_proto(priv, IPPROTO_UDP, MVPP2_PRS_RI_L4_UDP,
				  MVPP2_PRS_RI_L4_PROTO_MASK);
	if (err)
		return err;

	err = mvpp2_prs_ip4_proto(priv, IPPROTO_IGMP,
				  MVPP2_PRS_RI_CPU_CODE_RX_SPEC |
				  MVPP2_PRS_RI_UDF3_RX_SPECIAL,
				  MVPP2_PRS_RI_CPU_CODE_MASK |
				  MVPP2_PRS_RI_UDF3_MASK);
	if (err)
		return err;

	/* IPv4 Broadcast */
	err = mvpp2_prs_ip4_cast(priv, MVPP2_PRS_L3_BROAD_CAST);
	if (err)
		return err;

	/* IPv4 Multicast */
	err = mvpp2_prs_ip4_cast(priv, MVPP2_PRS_L3_MULTI_CAST);
	if (err)
		return err;

	/* Default IPv4 entry for unknown protocols */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP4);
	pe.index = MVPP2_PE_IP4_PROTO_UN;

	/* Set next lu to IPv4 */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP4);
	mvpp2_prs_sram_shift_set(&pe, 12, MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);
	/* Set L4 offset */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L4,
				  sizeof(struct iphdr) - 4,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);
	mvpp2_prs_sram_ai_update(&pe, MVPP2_PRS_IPV4_DIP_AI_BIT,
				 MVPP2_PRS_IPV4_DIP_AI_BIT);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L4_OTHER,
				 MVPP2_PRS_RI_L4_PROTO_MASK);

	mvpp2_prs_tcam_ai_update(&pe, 0, MVPP2_PRS_IPV4_DIP_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	/* Default IPv4 entry for unicast address */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP4);
	pe.index = MVPP2_PE_IP4_ADDR_UN;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_UCAST,
				 MVPP2_PRS_RI_L3_ADDR_MASK);

	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV4_DIP_AI_BIT,
				 MVPP2_PRS_IPV4_DIP_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Initialize entries for IPv6 */
static int mvpp2_prs_ip6_init(struct mvpp2 *priv)
{
	struct mvpp2_prs_entry pe;
	int tid, err;

	/* Set entries for TCP, UDP and ICMP over IPv6 */
	err = mvpp2_prs_ip6_proto(priv, IPPROTO_TCP,
				  MVPP2_PRS_RI_L4_TCP,
				  MVPP2_PRS_RI_L4_PROTO_MASK);
	if (err)
		return err;

	err = mvpp2_prs_ip6_proto(priv, IPPROTO_UDP,
				  MVPP2_PRS_RI_L4_UDP,
				  MVPP2_PRS_RI_L4_PROTO_MASK);
	if (err)
		return err;

	err = mvpp2_prs_ip6_proto(priv, IPPROTO_ICMPV6,
				  MVPP2_PRS_RI_CPU_CODE_RX_SPEC |
				  MVPP2_PRS_RI_UDF3_RX_SPECIAL,
				  MVPP2_PRS_RI_CPU_CODE_MASK |
				  MVPP2_PRS_RI_UDF3_MASK);
	if (err)
		return err;

	/* IPv4 is the last header. This is similar case as 6-TCP or 17-UDP */
	/* Result Info: UDF7=1, DS lite */
	err = mvpp2_prs_ip6_proto(priv, IPPROTO_IPIP,
				  MVPP2_PRS_RI_UDF7_IP6_LITE,
				  MVPP2_PRS_RI_UDF7_MASK);
	if (err)
		return err;

	/* IPv6 multicast */
	err = mvpp2_prs_ip6_cast(priv, MVPP2_PRS_L3_MULTI_CAST);
	if (err)
		return err;

	/* Entry for checking hop limit */
	tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
					MVPP2_PE_LAST_FREE_TID);
	if (tid < 0)
		return tid;

	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = tid;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_UN |
				 MVPP2_PRS_RI_DROP_MASK,
				 MVPP2_PRS_RI_L3_PROTO_MASK |
				 MVPP2_PRS_RI_DROP_MASK);

	mvpp2_prs_tcam_data_byte_set(&pe, 1, 0x00, MVPP2_PRS_IPV6_HOP_MASK);
	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV6_NO_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_NO_EXT_AI_BIT);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	/* Default IPv6 entry for unknown protocols */
	memset(&pe, 0, sizeof(pe));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = MVPP2_PE_IP6_PROTO_UN;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L4_OTHER,
				 MVPP2_PRS_RI_L4_PROTO_MASK);
	/* Set L4 offset relatively to our current place */
	mvpp2_prs_sram_offset_set(&pe, MVPP2_PRS_SRAM_UDF_TYPE_L4,
				  sizeof(struct ipv6hdr) - 4,
				  MVPP2_PRS_SRAM_OP_SEL_UDF_ADD);

	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV6_NO_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	/* Default IPv6 entry for unknown ext protocols */
	memset(&pe, 0, sizeof(struct mvpp2_prs_entry));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = MVPP2_PE_IP6_EXT_PROTO_UN;

	/* Finished: go to flowid generation */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_FLOWS);
	mvpp2_prs_sram_bits_set(&pe, MVPP2_PRS_SRAM_LU_GEN_BIT, 1);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L4_OTHER,
				 MVPP2_PRS_RI_L4_PROTO_MASK);

	mvpp2_prs_tcam_ai_update(&pe, MVPP2_PRS_IPV6_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_EXT_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP4);
	mvpp2_prs_hw_write(priv, &pe);

	/* Default IPv6 entry for unicast address */
	memset(&pe, 0, sizeof(struct mvpp2_prs_entry));
	mvpp2_prs_tcam_lu_set(&pe, MVPP2_PRS_LU_IP6);
	pe.index = MVPP2_PE_IP6_ADDR_UN;

	/* Finished: go to IPv6 again */
	mvpp2_prs_sram_next_lu_set(&pe, MVPP2_PRS_LU_IP6);
	mvpp2_prs_sram_ri_update(&pe, MVPP2_PRS_RI_L3_UCAST,
				 MVPP2_PRS_RI_L3_ADDR_MASK);
	mvpp2_prs_sram_ai_update(&pe, MVPP2_PRS_IPV6_NO_EXT_AI_BIT,
				 MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Shift back to IPV6 NH */
	mvpp2_prs_sram_shift_set(&pe, -18, MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

	mvpp2_prs_tcam_ai_update(&pe, 0, MVPP2_PRS_IPV6_NO_EXT_AI_BIT);
	/* Unmask all ports */
	mvpp2_prs_tcam_port_map_set(&pe, MVPP2_PRS_PORT_MASK);

	/* Update shadow table and hw entry */
	mvpp2_prs_shadow_set(priv, pe.index, MVPP2_PRS_LU_IP6);
	mvpp2_prs_hw_write(priv, &pe);

	return 0;
}

/* Parser default initialization */
static int mvpp2_prs_default_init(struct platform_device *pdev,
				  struct mvpp2 *priv)
{
	int err, index, i;

	/* Enable tcam table */
	mvpp2_write(priv, MVPP2_PRS_TCAM_CTRL_REG, MVPP2_PRS_TCAM_EN_MASK);

	/* Clear all tcam and sram entries */
	for (index = 0; index < MVPP2_PRS_TCAM_SRAM_SIZE; index++) {
		mvpp2_write(priv, MVPP2_PRS_TCAM_IDX_REG, index);
		for (i = 0; i < MVPP2_PRS_TCAM_WORDS; i++)
			mvpp2_write(priv, MVPP2_PRS_TCAM_DATA_REG(i), 0);

		mvpp2_write(priv, MVPP2_PRS_SRAM_IDX_REG, index);
		for (i = 0; i < MVPP2_PRS_SRAM_WORDS; i++)
			mvpp2_write(priv, MVPP2_PRS_SRAM_DATA_REG(i), 0);
	}

	/* Invalidate all tcam entries */
	for (index = 0; index < MVPP2_PRS_TCAM_SRAM_SIZE; index++)
		mvpp2_prs_hw_inv(priv, index);

	priv->prs_shadow = devm_kcalloc(&pdev->dev, MVPP2_PRS_TCAM_SRAM_SIZE,
					sizeof(*priv->prs_shadow),
					GFP_KERNEL);
	if (!priv->prs_shadow)
		return -ENOMEM;

	/* Always start from lookup = 0 */
	for (index = 0; index < MVPP2_MAX_PORTS; index++)
		mvpp2_prs_hw_port_init(priv, index, MVPP2_PRS_LU_MH,
				       MVPP2_PRS_PORT_LU_MAX, 0);

	mvpp2_prs_def_flow_init(priv);

	mvpp2_prs_mh_init(priv);

	mvpp2_prs_mac_init(priv);

	mvpp2_prs_dsa_init(priv);

	err = mvpp2_prs_etype_init(priv);
	if (err)
		return err;

	err = mvpp2_prs_vlan_init(pdev, priv);
	if (err)
		return err;

	err = mvpp2_prs_pppoe_init(priv);
	if (err)
		return err;

	err = mvpp2_prs_ip6_init(priv);
	if (err)
		return err;

	err = mvpp2_prs_ip4_init(priv);
	if (err)
		return err;

	return 0;
}

/* Compare MAC DA with tcam entry data */
static bool mvpp2_prs_mac_range_equals(struct mvpp2_prs_entry *pe,
				       const u8 *da, unsigned char *mask)
{
	unsigned char tcam_byte, tcam_mask;
	int index;

	for (index = 0; index < ETH_ALEN; index++) {
		mvpp2_prs_tcam_data_byte_get(pe, index, &tcam_byte, &tcam_mask);
		if (tcam_mask != mask[index])
			return false;

		if ((tcam_mask & tcam_byte) != (da[index] & mask[index]))
			return false;
	}

	return true;
}

/* Find tcam entry with matched pair <MAC DA, port> */
static struct mvpp2_prs_entry *
mvpp2_prs_mac_da_range_find(struct mvpp2 *priv, int pmap, const u8 *da,
			    unsigned char *mask, int udf_type)
{
	struct mvpp2_prs_entry *pe;
	int tid;

	pe = kzalloc(sizeof(*pe), GFP_KERNEL);
	if (!pe)
		return NULL;
	mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_MAC);

	/* Go through the all entires with MVPP2_PRS_LU_MAC */
	for (tid = MVPP2_PE_FIRST_FREE_TID;
	     tid <= MVPP2_PE_LAST_FREE_TID; tid++) {
		unsigned int entry_pmap;

		if (!priv->prs_shadow[tid].valid ||
		    (priv->prs_shadow[tid].lu != MVPP2_PRS_LU_MAC) ||
		    (priv->prs_shadow[tid].udf != udf_type))
			continue;

		pe->index = tid;
		mvpp2_prs_hw_read(priv, pe);
		entry_pmap = mvpp2_prs_tcam_port_map_get(pe);

		if (mvpp2_prs_mac_range_equals(pe, da, mask) &&
		    entry_pmap == pmap)
			return pe;
	}
	kfree(pe);

	return NULL;
}

/* Update parser's mac da entry */
static int mvpp2_prs_mac_da_accept(struct mvpp2 *priv, int port,
				   const u8 *da, bool add)
{
	struct mvpp2_prs_entry *pe;
	unsigned int pmap, len, ri;
	unsigned char mask[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
	int tid;

	/* Scan TCAM and see if entry with this <MAC DA, port> already exist */
	pe = mvpp2_prs_mac_da_range_find(priv, (1 << port), da, mask,
					 MVPP2_PRS_UDF_MAC_DEF);

	/* No such entry */
	if (!pe) {
		if (!add)
			return 0;

		/* Create new TCAM entry */
		/* Find first range mac entry*/
		for (tid = MVPP2_PE_FIRST_FREE_TID;
		     tid <= MVPP2_PE_LAST_FREE_TID; tid++)
			if (priv->prs_shadow[tid].valid &&
			    (priv->prs_shadow[tid].lu == MVPP2_PRS_LU_MAC) &&
			    (priv->prs_shadow[tid].udf ==
						       MVPP2_PRS_UDF_MAC_RANGE))
				break;

		/* Go through the all entries from first to last */
		tid = mvpp2_prs_tcam_first_free(priv, MVPP2_PE_FIRST_FREE_TID,
						tid - 1);
		if (tid < 0)
			return tid;

		pe = kzalloc(sizeof(*pe), GFP_KERNEL);
		if (!pe)
			return -ENOMEM;
		mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_MAC);
		pe->index = tid;

		/* Mask all ports */
		mvpp2_prs_tcam_port_map_set(pe, 0);
	}

	/* Update port mask */
	mvpp2_prs_tcam_port_set(pe, port, add);

	/* Invalidate the entry if no ports are left enabled */
	pmap = mvpp2_prs_tcam_port_map_get(pe);
	if (pmap == 0) {
		if (add) {
			kfree(pe);
			return -EINVAL;
		}
		mvpp2_prs_hw_inv(priv, pe->index);
		priv->prs_shadow[pe->index].valid = false;
		kfree(pe);
		return 0;
	}

	/* Continue - set next lookup */
	mvpp2_prs_sram_next_lu_set(pe, MVPP2_PRS_LU_DSA);

	/* Set match on DA */
	len = ETH_ALEN;
	while (len--)
		mvpp2_prs_tcam_data_byte_set(pe, len, da[len], 0xff);

	/* Set result info bits */
	if (is_broadcast_ether_addr(da))
		ri = MVPP2_PRS_RI_L2_BCAST;
	else if (is_multicast_ether_addr(da))
		ri = MVPP2_PRS_RI_L2_MCAST;
	else
		ri = MVPP2_PRS_RI_L2_UCAST | MVPP2_PRS_RI_MAC_ME_MASK;

	mvpp2_prs_sram_ri_update(pe, ri, MVPP2_PRS_RI_L2_CAST_MASK |
				 MVPP2_PRS_RI_MAC_ME_MASK);
	mvpp2_prs_shadow_ri_set(priv, pe->index, ri, MVPP2_PRS_RI_L2_CAST_MASK |
				MVPP2_PRS_RI_MAC_ME_MASK);

	/* Shift to ethertype */
	mvpp2_prs_sram_shift_set(pe, 2 * ETH_ALEN,
				 MVPP2_PRS_SRAM_OP_SEL_SHIFT_ADD);

	/* Update shadow table and hw entry */
	priv->prs_shadow[pe->index].udf = MVPP2_PRS_UDF_MAC_DEF;
	mvpp2_prs_shadow_set(priv, pe->index, MVPP2_PRS_LU_MAC);
	mvpp2_prs_hw_write(priv, pe);

	kfree(pe);

	return 0;
}

static int mvpp2_prs_update_mac_da(struct net_device *dev, const u8 *da)
{
	struct mvpp2_port *port = netdev_priv(dev);
	int err;

	/* Remove old parser entry */
	err = mvpp2_prs_mac_da_accept(port->priv, port->id, dev->dev_addr,
				      false);
	if (err)
		return err;

	/* Add new parser entry */
	err = mvpp2_prs_mac_da_accept(port->priv, port->id, da, true);
	if (err)
		return err;

	/* Set addr in the device */
	ether_addr_copy(dev->dev_addr, da);

	return 0;
}

/* Delete all port's multicast simple (not range) entries */
static void mvpp2_prs_mcast_del_all(struct mvpp2 *priv, int port)
{
	struct mvpp2_prs_entry pe;
	int index, tid;

	for (tid = MVPP2_PE_FIRST_FREE_TID;
	     tid <= MVPP2_PE_LAST_FREE_TID; tid++) {
		unsigned char da[ETH_ALEN], da_mask[ETH_ALEN];

		if (!priv->prs_shadow[tid].valid ||
		    (priv->prs_shadow[tid].lu != MVPP2_PRS_LU_MAC) ||
		    (priv->prs_shadow[tid].udf != MVPP2_PRS_UDF_MAC_DEF))
			continue;

		/* Only simple mac entries */
		pe.index = tid;
		mvpp2_prs_hw_read(priv, &pe);

		/* Read mac addr from entry */
		for (index = 0; index < ETH_ALEN; index++)
			mvpp2_prs_tcam_data_byte_get(&pe, index, &da[index],
						     &da_mask[index]);

		if (is_multicast_ether_addr(da) && !is_broadcast_ether_addr(da))
			/* Delete this entry */
			mvpp2_prs_mac_da_accept(priv, port, da, false);
	}
}

static int mvpp2_prs_tag_mode_set(struct mvpp2 *priv, int port, int type)
{
	switch (type) {
	case MVPP2_TAG_TYPE_EDSA:
		/* Add port to EDSA entries */
		mvpp2_prs_dsa_tag_set(priv, port, true,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_EDSA);
		mvpp2_prs_dsa_tag_set(priv, port, true,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_EDSA);
		/* Remove port from DSA entries */
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_DSA);
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_DSA);
		break;

	case MVPP2_TAG_TYPE_DSA:
		/* Add port to DSA entries */
		mvpp2_prs_dsa_tag_set(priv, port, true,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_DSA);
		mvpp2_prs_dsa_tag_set(priv, port, true,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_DSA);
		/* Remove port from EDSA entries */
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_EDSA);
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_EDSA);
		break;

	case MVPP2_TAG_TYPE_MH:
	case MVPP2_TAG_TYPE_NONE:
		/* Remove port form EDSA and DSA entries */
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_DSA);
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_DSA);
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_TAGGED, MVPP2_PRS_EDSA);
		mvpp2_prs_dsa_tag_set(priv, port, false,
				      MVPP2_PRS_UNTAGGED, MVPP2_PRS_EDSA);
		break;

	default:
		if ((type < 0) || (type > MVPP2_TAG_TYPE_EDSA))
			return -EINVAL;
	}

	return 0;
}

/* Set prs flow for the port */
static int mvpp2_prs_def_flow(struct mvpp2_port *port)
{
	struct mvpp2_prs_entry *pe;
	int tid;

	pe = mvpp2_prs_flow_find(port->priv, port->id);

	/* Such entry not exist */
	if (!pe) {
		/* Go through the all entires from last to first */
		tid = mvpp2_prs_tcam_first_free(port->priv,
						MVPP2_PE_LAST_FREE_TID,
					       MVPP2_PE_FIRST_FREE_TID);
		if (tid < 0)
			return tid;

		pe = kzalloc(sizeof(*pe), GFP_KERNEL);
		if (!pe)
			return -ENOMEM;

		mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_FLOWS);
		pe->index = tid;

		/* Set flow ID*/
		mvpp2_prs_sram_ai_update(pe, port->id, MVPP2_PRS_FLOW_ID_MASK);
		mvpp2_prs_sram_bits_set(pe, MVPP2_PRS_SRAM_LU_DONE_BIT, 1);

		/* Update shadow table */
		mvpp2_prs_shadow_set(port->priv, pe->index, MVPP2_PRS_LU_FLOWS);
	}

	mvpp2_prs_tcam_port_map_set(pe, (1 << port->id));
	mvpp2_prs_hw_write(port->priv, pe);
	kfree(pe);

	return 0;
}

/* Classifier configuration routines */

/* Update classification flow table registers */
static void mvpp2_cls_flow_write(struct mvpp2 *priv,
				 struct mvpp2_cls_flow_entry *fe)
{
	mvpp2_write(priv, MVPP2_CLS_FLOW_INDEX_REG, fe->index);
	mvpp2_write(priv, MVPP2_CLS_FLOW_TBL0_REG,  fe->data[0]);
	mvpp2_write(priv, MVPP2_CLS_FLOW_TBL1_REG,  fe->data[1]);
	mvpp2_write(priv, MVPP2_CLS_FLOW_TBL2_REG,  fe->data[2]);
}

/* Update classification lookup table register */
static void mvpp2_cls_lookup_write(struct mvpp2 *priv,
				   struct mvpp2_cls_lookup_entry *le)
{
	u32 val;

	val = (le->way << MVPP2_CLS_LKP_INDEX_WAY_OFFS) | le->lkpid;
	mvpp2_write(priv, MVPP2_CLS_LKP_INDEX_REG, val);
	mvpp2_write(priv, MVPP2_CLS_LKP_TBL_REG, le->data);
}

/* Classifier default initialization */
static void mvpp2_cls_init(struct mvpp2 *priv)
{
	struct mvpp2_cls_lookup_entry le;
	struct mvpp2_cls_flow_entry fe;
	int index;

	/* Enable classifier */
	mvpp2_write(priv, MVPP2_CLS_MODE_REG, MVPP2_CLS_MODE_ACTIVE_MASK);

	/* Clear classifier flow table */
	memset(&fe.data, 0, sizeof(fe.data));
	for (index = 0; index < MVPP2_CLS_FLOWS_TBL_SIZE; index++) {
		fe.index = index;
		mvpp2_cls_flow_write(priv, &fe);
	}

	/* Clear classifier lookup table */
	le.data = 0;
	for (index = 0; index < MVPP2_CLS_LKP_TBL_SIZE; index++) {
		le.lkpid = index;
		le.way = 0;
		mvpp2_cls_lookup_write(priv, &le);

		le.way = 1;
		mvpp2_cls_lookup_write(priv, &le);
	}
}

static void mvpp2_cls_port_config(struct mvpp2_port *port)
{
	struct mvpp2_cls_lookup_entry le;
	u32 val;

	/* Set way for the port */
	val = mvpp2_read(port->priv, MVPP2_CLS_PORT_WAY_REG);
	val &= ~MVPP2_CLS_PORT_WAY_MASK(port->id);
	mvpp2_write(port->priv, MVPP2_CLS_PORT_WAY_REG, val);

	/* Pick the entry to be accessed in lookup ID decoding table
	 * according to the way and lkpid.
	 */
	le.lkpid = port->id;
	le.way = 0;
	le.data = 0;

	/* Set initial CPU queue for receiving packets */
	le.data &= ~MVPP2_CLS_LKP_TBL_RXQ_MASK;
	le.data |= port->first_rxq;

	/* Disable classification engines */
	le.data &= ~MVPP2_CLS_LKP_TBL_LOOKUP_EN_MASK;

	/* Update lookup ID table entry */
	mvpp2_cls_lookup_write(port->priv, &le);
}

/* Set CPU queue number for oversize packets */
static void mvpp2_cls_oversize_rxq_set(struct mvpp2_port *port)
{
	u32 val;

	mvpp2_write(port->priv, MVPP2_CLS_OVERSIZE_RXQ_LOW_REG(port->id),
		    port->first_rxq & MVPP2_CLS_OVERSIZE_RXQ_LOW_MASK);

	mvpp2_write(port->priv, MVPP2_CLS_SWFWD_P2HQ_REG(port->id),
		    (port->first_rxq >> MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS));

	val = mvpp2_read(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG);
	val |= MVPP2_CLS_SWFWD_PCTRL_MASK(port->id);
	mvpp2_write(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG, val);
}

static void *mvpp2_frag_alloc(const struct mvpp2_bm_pool *pool)
{
	if (likely(pool->frag_size <= PAGE_SIZE))
		return netdev_alloc_frag(pool->frag_size);
	else
		return kmalloc(pool->frag_size, GFP_ATOMIC);
}

static void mvpp2_frag_free(const struct mvpp2_bm_pool *pool, void *data)
{
	if (likely(pool->frag_size <= PAGE_SIZE))
		skb_free_frag(data);
	else
		kfree(data);
}

/* Buffer Manager configuration routines */

/* Create pool */
static int mvpp2_bm_pool_create(struct platform_device *pdev,
				struct mvpp2 *priv,
				struct mvpp2_bm_pool *bm_pool, int size)
{
	u32 val;

	/* Number of buffer pointers must be a multiple of 16, as per
	 * hardware constraints
	 */
	if (!IS_ALIGNED(size, 16))
		return -EINVAL;

	/* PPv2.1 needs 8 bytes per buffer pointer, PPv2.2 needs 16
	 * bytes per buffer pointer
	 */
	if (priv->hw_version == MVPP21)
		bm_pool->size_bytes = 2 * sizeof(u32) * size;
	else
		bm_pool->size_bytes = 2 * sizeof(u64) * size;

	bm_pool->virt_addr = dma_alloc_coherent(&pdev->dev, bm_pool->size_bytes,
						&bm_pool->dma_addr,
						GFP_KERNEL);
	if (!bm_pool->virt_addr)
		return -ENOMEM;

	if (!IS_ALIGNED((unsigned long)bm_pool->virt_addr,
			MVPP2_BM_POOL_PTR_ALIGN)) {
		dma_free_coherent(&pdev->dev, bm_pool->size_bytes,
				  bm_pool->virt_addr, bm_pool->dma_addr);
		dev_err(&pdev->dev, "BM pool %d is not %d bytes aligned\n",
			bm_pool->id, MVPP2_BM_POOL_PTR_ALIGN);
		return -ENOMEM;
	}

	mvpp2_write(priv, MVPP2_BM_POOL_BASE_REG(bm_pool->id),
		    lower_32_bits(bm_pool->dma_addr));
	mvpp2_write(priv, MVPP2_BM_POOL_SIZE_REG(bm_pool->id), size);

	val = mvpp2_read(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id));
	val |= MVPP2_BM_START_MASK;
	mvpp2_write(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id), val);

	bm_pool->type = MVPP2_BM_FREE;
	bm_pool->size = size;
	bm_pool->pkt_size = 0;
	bm_pool->buf_num = 0;

	return 0;
}

/* Set pool buffer size */
static void mvpp2_bm_pool_bufsize_set(struct mvpp2 *priv,
				      struct mvpp2_bm_pool *bm_pool,
				      int buf_size)
{
	u32 val;

	bm_pool->buf_size = buf_size;

	val = ALIGN(buf_size, 1 << MVPP2_POOL_BUF_SIZE_OFFSET);
	mvpp2_write(priv, MVPP2_POOL_BUF_SIZE_REG(bm_pool->id), val);
}

static void mvpp2_bm_bufs_get_addrs(struct device *dev, struct mvpp2 *priv,
				    struct mvpp2_bm_pool *bm_pool,
				    dma_addr_t *dma_addr,
				    phys_addr_t *phys_addr)
{
	int cpu = get_cpu();

	*dma_addr = mvpp2_percpu_read(priv, cpu,
				      MVPP2_BM_PHY_ALLOC_REG(bm_pool->id));
	*phys_addr = mvpp2_percpu_read(priv, cpu, MVPP2_BM_VIRT_ALLOC_REG);

	if (priv->hw_version == MVPP22) {
		u32 val;
		u32 dma_addr_highbits, phys_addr_highbits;

		val = mvpp2_percpu_read(priv, cpu, MVPP22_BM_ADDR_HIGH_ALLOC);
		dma_addr_highbits = (val & MVPP22_BM_ADDR_HIGH_PHYS_MASK);
		phys_addr_highbits = (val & MVPP22_BM_ADDR_HIGH_VIRT_MASK) >>
			MVPP22_BM_ADDR_HIGH_VIRT_SHIFT;

		if (sizeof(dma_addr_t) == 8)
			*dma_addr |= (u64)dma_addr_highbits << 32;

		if (sizeof(phys_addr_t) == 8)
			*phys_addr |= (u64)phys_addr_highbits << 32;
	}

	put_cpu();
}

/* Free all buffers from the pool */
static void mvpp2_bm_bufs_free(struct device *dev, struct mvpp2 *priv,
			       struct mvpp2_bm_pool *bm_pool)
{
	int i;

	for (i = 0; i < bm_pool->buf_num; i++) {
		dma_addr_t buf_dma_addr;
		phys_addr_t buf_phys_addr;
		void *data;

		mvpp2_bm_bufs_get_addrs(dev, priv, bm_pool,
					&buf_dma_addr, &buf_phys_addr);

		dma_unmap_single(dev, buf_dma_addr,
				 bm_pool->buf_size, DMA_FROM_DEVICE);

		data = (void *)phys_to_virt(buf_phys_addr);
		if (!data)
			break;

		mvpp2_frag_free(bm_pool, data);
	}

	/* Update BM driver with number of buffers removed from pool */
	bm_pool->buf_num -= i;
}

/* Cleanup pool */
static int mvpp2_bm_pool_destroy(struct platform_device *pdev,
				 struct mvpp2 *priv,
				 struct mvpp2_bm_pool *bm_pool)
{
	u32 val;

	mvpp2_bm_bufs_free(&pdev->dev, priv, bm_pool);
	if (bm_pool->buf_num) {
		WARN(1, "cannot free all buffers in pool %d\n", bm_pool->id);
		return 0;
	}

	val = mvpp2_read(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id));
	val |= MVPP2_BM_STOP_MASK;
	mvpp2_write(priv, MVPP2_BM_POOL_CTRL_REG(bm_pool->id), val);

	dma_free_coherent(&pdev->dev, bm_pool->size_bytes,
			  bm_pool->virt_addr,
			  bm_pool->dma_addr);
	return 0;
}

static int mvpp2_bm_pools_init(struct platform_device *pdev,
			       struct mvpp2 *priv)
{
	int i, err, size;
	struct mvpp2_bm_pool *bm_pool;

	/* Create all pools with maximum size */
	size = MVPP2_BM_POOL_SIZE_MAX;
	for (i = 0; i < MVPP2_BM_POOLS_NUM; i++) {
		bm_pool = &priv->bm_pools[i];
		bm_pool->id = i;
		err = mvpp2_bm_pool_create(pdev, priv, bm_pool, size);
		if (err)
			goto err_unroll_pools;
		mvpp2_bm_pool_bufsize_set(priv, bm_pool, 0);
	}
	return 0;

err_unroll_pools:
	dev_err(&pdev->dev, "failed to create BM pool %d, size %d\n", i, size);
	for (i = i - 1; i >= 0; i--)
		mvpp2_bm_pool_destroy(pdev, priv, &priv->bm_pools[i]);
	return err;
}

static int mvpp2_bm_init(struct platform_device *pdev, struct mvpp2 *priv)
{
	int i, err;

	for (i = 0; i < MVPP2_BM_POOLS_NUM; i++) {
		/* Mask BM all interrupts */
		mvpp2_write(priv, MVPP2_BM_INTR_MASK_REG(i), 0);
		/* Clear BM cause register */
		mvpp2_write(priv, MVPP2_BM_INTR_CAUSE_REG(i), 0);
	}

	/* Allocate and initialize BM pools */
	priv->bm_pools = devm_kcalloc(&pdev->dev, MVPP2_BM_POOLS_NUM,
				      sizeof(*priv->bm_pools), GFP_KERNEL);
	if (!priv->bm_pools)
		return -ENOMEM;

	err = mvpp2_bm_pools_init(pdev, priv);
	if (err < 0)
		return err;
	return 0;
}

/* Attach long pool to rxq */
static void mvpp2_rxq_long_pool_set(struct mvpp2_port *port,
				    int lrxq, int long_pool)
{
	u32 val, mask;
	int prxq;

	/* Get queue physical ID */
	prxq = port->rxqs[lrxq]->id;

	if (port->priv->hw_version == MVPP21)
		mask = MVPP21_RXQ_POOL_LONG_MASK;
	else
		mask = MVPP22_RXQ_POOL_LONG_MASK;

	val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
	val &= ~mask;
	val |= (long_pool << MVPP2_RXQ_POOL_LONG_OFFS) & mask;
	mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}

/* Attach short pool to rxq */
static void mvpp2_rxq_short_pool_set(struct mvpp2_port *port,
				     int lrxq, int short_pool)
{
	u32 val, mask;
	int prxq;

	/* Get queue physical ID */
	prxq = port->rxqs[lrxq]->id;

	if (port->priv->hw_version == MVPP21)
		mask = MVPP21_RXQ_POOL_SHORT_MASK;
	else
		mask = MVPP22_RXQ_POOL_SHORT_MASK;

	val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
	val &= ~mask;
	val |= (short_pool << MVPP2_RXQ_POOL_SHORT_OFFS) & mask;
	mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}

static void *mvpp2_buf_alloc(struct mvpp2_port *port,
			     struct mvpp2_bm_pool *bm_pool,
			     dma_addr_t *buf_dma_addr,
			     phys_addr_t *buf_phys_addr,
			     gfp_t gfp_mask)
{
	dma_addr_t dma_addr;
	void *data;

	data = mvpp2_frag_alloc(bm_pool);
	if (!data)
		return NULL;

	dma_addr = dma_map_single(port->dev->dev.parent, data,
				  MVPP2_RX_BUF_SIZE(bm_pool->pkt_size),
				  DMA_FROM_DEVICE);
	if (unlikely(dma_mapping_error(port->dev->dev.parent, dma_addr))) {
		mvpp2_frag_free(bm_pool, data);
		return NULL;
	}
	*buf_dma_addr = dma_addr;
	*buf_phys_addr = virt_to_phys(data);

	return data;
}

/* Release buffer to BM */
static inline void mvpp2_bm_pool_put(struct mvpp2_port *port, int pool,
				     dma_addr_t buf_dma_addr,
				     phys_addr_t buf_phys_addr)
{
	int cpu = get_cpu();

	if (port->priv->hw_version == MVPP22) {
		u32 val = 0;

		if (sizeof(dma_addr_t) == 8)
			val |= upper_32_bits(buf_dma_addr) &
				MVPP22_BM_ADDR_HIGH_PHYS_RLS_MASK;

		if (sizeof(phys_addr_t) == 8)
			val |= (upper_32_bits(buf_phys_addr)
				<< MVPP22_BM_ADDR_HIGH_VIRT_RLS_SHIFT) &
				MVPP22_BM_ADDR_HIGH_VIRT_RLS_MASK;

		mvpp2_percpu_write(port->priv, cpu,
				   MVPP22_BM_ADDR_HIGH_RLS_REG, val);
	}

	/* MVPP2_BM_VIRT_RLS_REG is not interpreted by HW, and simply
	 * returned in the "cookie" field of the RX
	 * descriptor. Instead of storing the virtual address, we
	 * store the physical address
	 */
	mvpp2_percpu_write(port->priv, cpu,
			   MVPP2_BM_VIRT_RLS_REG, buf_phys_addr);
	mvpp2_percpu_write(port->priv, cpu,
			   MVPP2_BM_PHY_RLS_REG(pool), buf_dma_addr);

	put_cpu();
}

/* Allocate buffers for the pool */
static int mvpp2_bm_bufs_add(struct mvpp2_port *port,
			     struct mvpp2_bm_pool *bm_pool, int buf_num)
{
	int i, buf_size, total_size;
	dma_addr_t dma_addr;
	phys_addr_t phys_addr;
	void *buf;

	buf_size = MVPP2_RX_BUF_SIZE(bm_pool->pkt_size);
	total_size = MVPP2_RX_TOTAL_SIZE(buf_size);

	if (buf_num < 0 ||
	    (buf_num + bm_pool->buf_num > bm_pool->size)) {
		netdev_err(port->dev,
			   "cannot allocate %d buffers for pool %d\n",
			   buf_num, bm_pool->id);
		return 0;
	}

	for (i = 0; i < buf_num; i++) {
		buf = mvpp2_buf_alloc(port, bm_pool, &dma_addr,
				      &phys_addr, GFP_KERNEL);
		if (!buf)
			break;

		mvpp2_bm_pool_put(port, bm_pool->id, dma_addr,
				  phys_addr);
	}

	/* Update BM driver with number of buffers added to pool */
	bm_pool->buf_num += i;

	netdev_dbg(port->dev,
		   "%s pool %d: pkt_size=%4d, buf_size=%4d, total_size=%4d\n",
		   bm_pool->type == MVPP2_BM_SWF_SHORT ? "short" : " long",
		   bm_pool->id, bm_pool->pkt_size, buf_size, total_size);

	netdev_dbg(port->dev,
		   "%s pool %d: %d of %d buffers added\n",
		   bm_pool->type == MVPP2_BM_SWF_SHORT ? "short" : " long",
		   bm_pool->id, i, buf_num);
	return i;
}

/* Notify the driver that BM pool is being used as specific type and return the
 * pool pointer on success
 */
static struct mvpp2_bm_pool *
mvpp2_bm_pool_use(struct mvpp2_port *port, int pool, enum mvpp2_bm_type type,
		  int pkt_size)
{
	struct mvpp2_bm_pool *new_pool = &port->priv->bm_pools[pool];
	int num;

	if (new_pool->type != MVPP2_BM_FREE && new_pool->type != type) {
		netdev_err(port->dev, "mixing pool types is forbidden\n");
		return NULL;
	}

	if (new_pool->type == MVPP2_BM_FREE)
		new_pool->type = type;

	/* Allocate buffers in case BM pool is used as long pool, but packet
	 * size doesn't match MTU or BM pool hasn't being used yet
	 */
	if (((type == MVPP2_BM_SWF_LONG) && (pkt_size > new_pool->pkt_size)) ||
	    (new_pool->pkt_size == 0)) {
		int pkts_num;

		/* Set default buffer number or free all the buffers in case
		 * the pool is not empty
		 */
		pkts_num = new_pool->buf_num;
		if (pkts_num == 0)
			pkts_num = type == MVPP2_BM_SWF_LONG ?
				   MVPP2_BM_LONG_BUF_NUM :
				   MVPP2_BM_SHORT_BUF_NUM;
		else
			mvpp2_bm_bufs_free(port->dev->dev.parent,
					   port->priv, new_pool);

		new_pool->pkt_size = pkt_size;
		new_pool->frag_size =
			SKB_DATA_ALIGN(MVPP2_RX_BUF_SIZE(pkt_size)) +
			MVPP2_SKB_SHINFO_SIZE;

		/* Allocate buffers for this pool */
		num = mvpp2_bm_bufs_add(port, new_pool, pkts_num);
		if (num != pkts_num) {
			WARN(1, "pool %d: %d of %d allocated\n",
			     new_pool->id, num, pkts_num);
			return NULL;
		}
	}

	mvpp2_bm_pool_bufsize_set(port->priv, new_pool,
				  MVPP2_RX_BUF_SIZE(new_pool->pkt_size));

	return new_pool;
}

/* Initialize pools for swf */
static int mvpp2_swf_bm_pool_init(struct mvpp2_port *port)
{
	int rxq;

	if (!port->pool_long) {
		port->pool_long =
		       mvpp2_bm_pool_use(port, MVPP2_BM_SWF_LONG_POOL(port->id),
					 MVPP2_BM_SWF_LONG,
					 port->pkt_size);
		if (!port->pool_long)
			return -ENOMEM;

		port->pool_long->port_map |= (1 << port->id);

		for (rxq = 0; rxq < port->nrxqs; rxq++)
			mvpp2_rxq_long_pool_set(port, rxq, port->pool_long->id);
	}

	if (!port->pool_short) {
		port->pool_short =
			mvpp2_bm_pool_use(port, MVPP2_BM_SWF_SHORT_POOL,
					  MVPP2_BM_SWF_SHORT,
					  MVPP2_BM_SHORT_PKT_SIZE);
		if (!port->pool_short)
			return -ENOMEM;

		port->pool_short->port_map |= (1 << port->id);

		for (rxq = 0; rxq < port->nrxqs; rxq++)
			mvpp2_rxq_short_pool_set(port, rxq,
						 port->pool_short->id);
	}

	return 0;
}

static int mvpp2_bm_update_mtu(struct net_device *dev, int mtu)
{
	struct mvpp2_port *port = netdev_priv(dev);
	struct mvpp2_bm_pool *port_pool = port->pool_long;
	int num, pkts_num = port_pool->buf_num;
	int pkt_size = MVPP2_RX_PKT_SIZE(mtu);

	/* Update BM pool with new buffer size */
	mvpp2_bm_bufs_free(dev->dev.parent, port->priv, port_pool);
	if (port_pool->buf_num) {
		WARN(1, "cannot free all buffers in pool %d\n", port_pool->id);
		return -EIO;
	}

	port_pool->pkt_size = pkt_size;
	port_pool->frag_size = SKB_DATA_ALIGN(MVPP2_RX_BUF_SIZE(pkt_size)) +
		MVPP2_SKB_SHINFO_SIZE;
	num = mvpp2_bm_bufs_add(port, port_pool, pkts_num);
	if (num != pkts_num) {
		WARN(1, "pool %d: %d of %d allocated\n",
		     port_pool->id, num, pkts_num);
		return -EIO;
	}

	mvpp2_bm_pool_bufsize_set(port->priv, port_pool,
				  MVPP2_RX_BUF_SIZE(port_pool->pkt_size));
	dev->mtu = mtu;
	netdev_update_features(dev);
	return 0;
}

static inline void mvpp2_interrupts_enable(struct mvpp2_port *port)
{
	int i, sw_thread_mask = 0;

	for (i = 0; i < port->nqvecs; i++)
		sw_thread_mask |= port->qvecs[i].sw_thread_mask;

	mvpp2_write(port->priv, MVPP2_ISR_ENABLE_REG(port->id),
		    MVPP2_ISR_ENABLE_INTERRUPT(sw_thread_mask));
}

static inline void mvpp2_interrupts_disable(struct mvpp2_port *port)
{
	int i, sw_thread_mask = 0;

	for (i = 0; i < port->nqvecs; i++)
		sw_thread_mask |= port->qvecs[i].sw_thread_mask;

	mvpp2_write(port->priv, MVPP2_ISR_ENABLE_REG(port->id),
		    MVPP2_ISR_DISABLE_INTERRUPT(sw_thread_mask));
}

static inline void mvpp2_qvec_interrupt_enable(struct mvpp2_queue_vector *qvec)
{
	struct mvpp2_port *port = qvec->port;

	mvpp2_write(port->priv, MVPP2_ISR_ENABLE_REG(port->id),
		    MVPP2_ISR_ENABLE_INTERRUPT(qvec->sw_thread_mask));
}

static inline void mvpp2_qvec_interrupt_disable(struct mvpp2_queue_vector *qvec)
{
	struct mvpp2_port *port = qvec->port;

	mvpp2_write(port->priv, MVPP2_ISR_ENABLE_REG(port->id),
		    MVPP2_ISR_DISABLE_INTERRUPT(qvec->sw_thread_mask));
}

/* Mask the current CPU's Rx/Tx interrupts
 * Called by on_each_cpu(), guaranteed to run with migration disabled,
 * using smp_processor_id() is OK.
 */
static void mvpp2_interrupts_mask(void *arg)
{
	struct mvpp2_port *port = arg;

	mvpp2_percpu_write(port->priv, smp_processor_id(),
			   MVPP2_ISR_RX_TX_MASK_REG(port->id), 0);
}

/* Unmask the current CPU's Rx/Tx interrupts.
 * Called by on_each_cpu(), guaranteed to run with migration disabled,
 * using smp_processor_id() is OK.
 */
static void mvpp2_interrupts_unmask(void *arg)
{
	struct mvpp2_port *port = arg;
	u32 val;

	val = MVPP2_CAUSE_MISC_SUM_MASK |
		MVPP2_CAUSE_RXQ_OCCUP_DESC_ALL_MASK;
	if (port->has_tx_irqs)
		val |= MVPP2_CAUSE_TXQ_OCCUP_DESC_ALL_MASK;

	mvpp2_percpu_write(port->priv, smp_processor_id(),
			   MVPP2_ISR_RX_TX_MASK_REG(port->id), val);
}

static void
mvpp2_shared_interrupt_mask_unmask(struct mvpp2_port *port, bool mask)
{
	u32 val;
	int i;

	if (port->priv->hw_version != MVPP22)
		return;

	if (mask)
		val = 0;
	else
		val = MVPP2_CAUSE_RXQ_OCCUP_DESC_ALL_MASK;

	for (i = 0; i < port->nqvecs; i++) {
		struct mvpp2_queue_vector *v = port->qvecs + i;

		if (v->type != MVPP2_QUEUE_VECTOR_SHARED)
			continue;

		mvpp2_percpu_write(port->priv, v->sw_thread_id,
				   MVPP2_ISR_RX_TX_MASK_REG(port->id), val);
	}
}

/* Port configuration routines */

static void mvpp22_port_mii_set(struct mvpp2_port *port)
{
	u32 val;

	/* Only GOP port 0 has an XLG MAC */
	if (port->gop_id == 0) {
		val = readl(port->base + MVPP22_XLG_CTRL3_REG);
		val &= ~MVPP22_XLG_CTRL3_MACMODESELECT_MASK;

		if (port->phy_interface == PHY_INTERFACE_MODE_XAUI ||
		    port->phy_interface == PHY_INTERFACE_MODE_10GKR)
			val |= MVPP22_XLG_CTRL3_MACMODESELECT_10G;
		else
			val |= MVPP22_XLG_CTRL3_MACMODESELECT_GMAC;

		writel(val, port->base + MVPP22_XLG_CTRL3_REG);
	}

	val = readl(port->base + MVPP22_GMAC_CTRL_4_REG);
	if (port->phy_interface == PHY_INTERFACE_MODE_RGMII)
		val |= MVPP22_CTRL4_EXT_PIN_GMII_SEL;
	else
		val &= ~MVPP22_CTRL4_EXT_PIN_GMII_SEL;
	val &= ~MVPP22_CTRL4_DP_CLK_SEL;
	val |= MVPP22_CTRL4_SYNC_BYPASS;
	val |= MVPP22_CTRL4_QSGMII_BYPASS_ACTIVE;
	writel(val, port->base + MVPP22_GMAC_CTRL_4_REG);
}

static void mvpp2_port_mii_set(struct mvpp2_port *port)
{
	u32 val;

	if (port->priv->hw_version == MVPP22)
		mvpp22_port_mii_set(port);

	val = readl(port->base + MVPP2_GMAC_CTRL_2_REG);

	switch (port->phy_interface) {
	case PHY_INTERFACE_MODE_SGMII:
		val |= MVPP2_GMAC_INBAND_AN_MASK;
		break;
	case PHY_INTERFACE_MODE_RGMII:
		val |= MVPP2_GMAC_PORT_RGMII_MASK;
	default:
		val &= ~MVPP2_GMAC_PCS_ENABLE_MASK;
	}

	writel(val, port->base + MVPP2_GMAC_CTRL_2_REG);
}

static void mvpp2_port_fc_adv_enable(struct mvpp2_port *port)
{
	u32 val;

	val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
	val |= MVPP2_GMAC_FC_ADV_EN;
	writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
}

static void mvpp2_port_enable(struct mvpp2_port *port)
{
	u32 val;

	/* Only GOP port 0 has an XLG MAC */
	if (port->gop_id == 0 &&
	    (port->phy_interface == PHY_INTERFACE_MODE_XAUI ||
	     port->phy_interface == PHY_INTERFACE_MODE_10GKR)) {
		val = readl(port->base + MVPP22_XLG_CTRL0_REG);
		val |= MVPP22_XLG_CTRL0_PORT_EN |
		       MVPP22_XLG_CTRL0_MAC_RESET_DIS;
		val &= ~MVPP22_XLG_CTRL0_MIB_CNT_DIS;
		writel(val, port->base + MVPP22_XLG_CTRL0_REG);
	} else {
		val = readl(port->base + MVPP2_GMAC_CTRL_0_REG);
		val |= MVPP2_GMAC_PORT_EN_MASK;
		val |= MVPP2_GMAC_MIB_CNTR_EN_MASK;
		writel(val, port->base + MVPP2_GMAC_CTRL_0_REG);
	}
}

static void mvpp2_port_disable(struct mvpp2_port *port)
{
	u32 val;

	/* Only GOP port 0 has an XLG MAC */
	if (port->gop_id == 0 &&
	    (port->phy_interface == PHY_INTERFACE_MODE_XAUI ||
	     port->phy_interface == PHY_INTERFACE_MODE_10GKR)) {
		val = readl(port->base + MVPP22_XLG_CTRL0_REG);
		val &= ~(MVPP22_XLG_CTRL0_PORT_EN |
			 MVPP22_XLG_CTRL0_MAC_RESET_DIS);
		writel(val, port->base + MVPP22_XLG_CTRL0_REG);
	} else {
		val = readl(port->base + MVPP2_GMAC_CTRL_0_REG);
		val &= ~(MVPP2_GMAC_PORT_EN_MASK);
		writel(val, port->base + MVPP2_GMAC_CTRL_0_REG);
	}
}

/* Set IEEE 802.3x Flow Control Xon Packet Transmission Mode */
static void mvpp2_port_periodic_xon_disable(struct mvpp2_port *port)
{
	u32 val;

	val = readl(port->base + MVPP2_GMAC_CTRL_1_REG) &
		    ~MVPP2_GMAC_PERIODIC_XON_EN_MASK;
	writel(val, port->base + MVPP2_GMAC_CTRL_1_REG);
}

/* Configure loopback port */
static void mvpp2_port_loopback_set(struct mvpp2_port *port)
{
	u32 val;

	val = readl(port->base + MVPP2_GMAC_CTRL_1_REG);

	if (port->speed == 1000)
		val |= MVPP2_GMAC_GMII_LB_EN_MASK;
	else
		val &= ~MVPP2_GMAC_GMII_LB_EN_MASK;

	if (port->phy_interface == PHY_INTERFACE_MODE_SGMII)
		val |= MVPP2_GMAC_PCS_LB_EN_MASK;
	else
		val &= ~MVPP2_GMAC_PCS_LB_EN_MASK;

	writel(val, port->base + MVPP2_GMAC_CTRL_1_REG);
}

static void mvpp2_port_reset(struct mvpp2_port *port)
{
	u32 val;

	val = readl(port->base + MVPP2_GMAC_CTRL_2_REG) &
		    ~MVPP2_GMAC_PORT_RESET_MASK;
	writel(val, port->base + MVPP2_GMAC_CTRL_2_REG);

	while (readl(port->base + MVPP2_GMAC_CTRL_2_REG) &
	       MVPP2_GMAC_PORT_RESET_MASK)
		continue;
}

/* Change maximum receive size of the port */
static inline void mvpp2_gmac_max_rx_size_set(struct mvpp2_port *port)
{
	u32 val;

	val = readl(port->base + MVPP2_GMAC_CTRL_0_REG);
	val &= ~MVPP2_GMAC_MAX_RX_SIZE_MASK;
	val |= (((port->pkt_size - MVPP2_MH_SIZE) / 2) <<
		    MVPP2_GMAC_MAX_RX_SIZE_OFFS);
	writel(val, port->base + MVPP2_GMAC_CTRL_0_REG);
}

/* Set defaults to the MVPP2 port */
static void mvpp2_defaults_set(struct mvpp2_port *port)
{
	int tx_port_num, val, queue, ptxq, lrxq;

	if (port->priv->hw_version == MVPP21) {
		/* Configure port to loopback if needed */
		if (port->flags & MVPP2_F_LOOPBACK)
			mvpp2_port_loopback_set(port);

		/* Update TX FIFO MIN Threshold */
		val = readl(port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
		val &= ~MVPP2_GMAC_TX_FIFO_MIN_TH_ALL_MASK;
		/* Min. TX threshold must be less than minimal packet length */
		val |= MVPP2_GMAC_TX_FIFO_MIN_TH_MASK(64 - 4 - 2);
		writel(val, port->base + MVPP2_GMAC_PORT_FIFO_CFG_1_REG);
	}

	/* Disable Legacy WRR, Disable EJP, Release from reset */
	tx_port_num = mvpp2_egress_port(port);
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PORT_INDEX_REG,
		    tx_port_num);
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_CMD_1_REG, 0);

	/* Close bandwidth for all queues */
	for (queue = 0; queue < MVPP2_MAX_TXQ; queue++) {
		ptxq = mvpp2_txq_phys(port->id, queue);
		mvpp2_write(port->priv,
			    MVPP2_TXQ_SCHED_TOKEN_CNTR_REG(ptxq), 0);
	}

	/* Set refill period to 1 usec, refill tokens
	 * and bucket size to maximum
	 */
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PERIOD_REG,
		    port->priv->tclk / USEC_PER_SEC);
	val = mvpp2_read(port->priv, MVPP2_TXP_SCHED_REFILL_REG);
	val &= ~MVPP2_TXP_REFILL_PERIOD_ALL_MASK;
	val |= MVPP2_TXP_REFILL_PERIOD_MASK(1);
	val |= MVPP2_TXP_REFILL_TOKENS_ALL_MASK;
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_REFILL_REG, val);
	val = MVPP2_TXP_TOKEN_SIZE_MAX;
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_TOKEN_SIZE_REG, val);

	/* Set MaximumLowLatencyPacketSize value to 256 */
	mvpp2_write(port->priv, MVPP2_RX_CTRL_REG(port->id),
		    MVPP2_RX_USE_PSEUDO_FOR_CSUM_MASK |
		    MVPP2_RX_LOW_LATENCY_PKT_SIZE(256));

	/* Enable Rx cache snoop */
	for (lrxq = 0; lrxq < port->nrxqs; lrxq++) {
		queue = port->rxqs[lrxq]->id;
		val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(queue));
		val |= MVPP2_SNOOP_PKT_SIZE_MASK |
			   MVPP2_SNOOP_BUF_HDR_MASK;
		mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(queue), val);
	}

	/* At default, mask all interrupts to all present cpus */
	mvpp2_interrupts_disable(port);
}

/* Enable/disable receiving packets */
static void mvpp2_ingress_enable(struct mvpp2_port *port)
{
	u32 val;
	int lrxq, queue;

	for (lrxq = 0; lrxq < port->nrxqs; lrxq++) {
		queue = port->rxqs[lrxq]->id;
		val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(queue));
		val &= ~MVPP2_RXQ_DISABLE_MASK;
		mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(queue), val);
	}
}

static void mvpp2_ingress_disable(struct mvpp2_port *port)
{
	u32 val;
	int lrxq, queue;

	for (lrxq = 0; lrxq < port->nrxqs; lrxq++) {
		queue = port->rxqs[lrxq]->id;
		val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(queue));
		val |= MVPP2_RXQ_DISABLE_MASK;
		mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(queue), val);
	}
}

/* Enable transmit via physical egress queue
 * - HW starts take descriptors from DRAM
 */
static void mvpp2_egress_enable(struct mvpp2_port *port)
{
	u32 qmap;
	int queue;
	int tx_port_num = mvpp2_egress_port(port);

	/* Enable all initialized TXs. */
	qmap = 0;
	for (queue = 0; queue < port->ntxqs; queue++) {
		struct mvpp2_tx_queue *txq = port->txqs[queue];

		if (txq->descs)
			qmap |= (1 << queue);
	}

	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PORT_INDEX_REG, tx_port_num);
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_Q_CMD_REG, qmap);
}

/* Disable transmit via physical egress queue
 * - HW doesn't take descriptors from DRAM
 */
static void mvpp2_egress_disable(struct mvpp2_port *port)
{
	u32 reg_data;
	int delay;
	int tx_port_num = mvpp2_egress_port(port);

	/* Issue stop command for active channels only */
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PORT_INDEX_REG, tx_port_num);
	reg_data = (mvpp2_read(port->priv, MVPP2_TXP_SCHED_Q_CMD_REG)) &
		    MVPP2_TXP_SCHED_ENQ_MASK;
	if (reg_data != 0)
		mvpp2_write(port->priv, MVPP2_TXP_SCHED_Q_CMD_REG,
			    (reg_data << MVPP2_TXP_SCHED_DISQ_OFFSET));

	/* Wait for all Tx activity to terminate. */
	delay = 0;
	do {
		if (delay >= MVPP2_TX_DISABLE_TIMEOUT_MSEC) {
			netdev_warn(port->dev,
				    "Tx stop timed out, status=0x%08x\n",
				    reg_data);
			break;
		}
		mdelay(1);
		delay++;

		/* Check port TX Command register that all
		 * Tx queues are stopped
		 */
		reg_data = mvpp2_read(port->priv, MVPP2_TXP_SCHED_Q_CMD_REG);
	} while (reg_data & MVPP2_TXP_SCHED_ENQ_MASK);
}

/* Rx descriptors helper methods */

/* Get number of Rx descriptors occupied by received packets */
static inline int
mvpp2_rxq_received(struct mvpp2_port *port, int rxq_id)
{
	u32 val = mvpp2_read(port->priv, MVPP2_RXQ_STATUS_REG(rxq_id));

	return val & MVPP2_RXQ_OCCUPIED_MASK;
}

/* Update Rx queue status with the number of occupied and available
 * Rx descriptor slots.
 */
static inline void
mvpp2_rxq_status_update(struct mvpp2_port *port, int rxq_id,
			int used_count, int free_count)
{
	/* Decrement the number of used descriptors and increment count
	 * increment the number of free descriptors.
	 */
	u32 val = used_count | (free_count << MVPP2_RXQ_NUM_NEW_OFFSET);

	mvpp2_write(port->priv, MVPP2_RXQ_STATUS_UPDATE_REG(rxq_id), val);
}

/* Get pointer to next RX descriptor to be processed by SW */
static inline struct mvpp2_rx_desc *
mvpp2_rxq_next_desc_get(struct mvpp2_rx_queue *rxq)
{
	int rx_desc = rxq->next_desc_to_proc;

	rxq->next_desc_to_proc = MVPP2_QUEUE_NEXT_DESC(rxq, rx_desc);
	prefetch(rxq->descs + rxq->next_desc_to_proc);
	return rxq->descs + rx_desc;
}

/* Set rx queue offset */
static void mvpp2_rxq_offset_set(struct mvpp2_port *port,
				 int prxq, int offset)
{
	u32 val;

	/* Convert offset from bytes to units of 32 bytes */
	offset = offset >> 5;

	val = mvpp2_read(port->priv, MVPP2_RXQ_CONFIG_REG(prxq));
	val &= ~MVPP2_RXQ_PACKET_OFFSET_MASK;

	/* Offset is in */
	val |= ((offset << MVPP2_RXQ_PACKET_OFFSET_OFFS) &
		    MVPP2_RXQ_PACKET_OFFSET_MASK);

	mvpp2_write(port->priv, MVPP2_RXQ_CONFIG_REG(prxq), val);
}

/* Tx descriptors helper methods */

/* Get pointer to next Tx descriptor to be processed (send) by HW */
static struct mvpp2_tx_desc *
mvpp2_txq_next_desc_get(struct mvpp2_tx_queue *txq)
{
	int tx_desc = txq->next_desc_to_proc;

	txq->next_desc_to_proc = MVPP2_QUEUE_NEXT_DESC(txq, tx_desc);
	return txq->descs + tx_desc;
}

/* Update HW with number of aggregated Tx descriptors to be sent
 *
 * Called only from mvpp2_tx(), so migration is disabled, using
 * smp_processor_id() is OK.
 */
static void mvpp2_aggr_txq_pend_desc_add(struct mvpp2_port *port, int pending)
{
	/* aggregated access - relevant TXQ number is written in TX desc */
	mvpp2_percpu_write(port->priv, smp_processor_id(),
			   MVPP2_AGGR_TXQ_UPDATE_REG, pending);
}


/* Check if there are enough free descriptors in aggregated txq.
 * If not, update the number of occupied descriptors and repeat the check.
 *
 * Called only from mvpp2_tx(), so migration is disabled, using
 * smp_processor_id() is OK.
 */
static int mvpp2_aggr_desc_num_check(struct mvpp2 *priv,
				     struct mvpp2_tx_queue *aggr_txq, int num)
{
	if ((aggr_txq->count + num) > aggr_txq->size) {
		/* Update number of occupied aggregated Tx descriptors */
		int cpu = smp_processor_id();
		u32 val = mvpp2_read(priv, MVPP2_AGGR_TXQ_STATUS_REG(cpu));

		aggr_txq->count = val & MVPP2_AGGR_TXQ_PENDING_MASK;
	}

	if ((aggr_txq->count + num) > aggr_txq->size)
		return -ENOMEM;

	return 0;
}

/* Reserved Tx descriptors allocation request
 *
 * Called only from mvpp2_txq_reserved_desc_num_proc(), itself called
 * only by mvpp2_tx(), so migration is disabled, using
 * smp_processor_id() is OK.
 */
static int mvpp2_txq_alloc_reserved_desc(struct mvpp2 *priv,
					 struct mvpp2_tx_queue *txq, int num)
{
	u32 val;
	int cpu = smp_processor_id();

	val = (txq->id << MVPP2_TXQ_RSVD_REQ_Q_OFFSET) | num;
	mvpp2_percpu_write(priv, cpu, MVPP2_TXQ_RSVD_REQ_REG, val);

	val = mvpp2_percpu_read(priv, cpu, MVPP2_TXQ_RSVD_RSLT_REG);

	return val & MVPP2_TXQ_RSVD_RSLT_MASK;
}

/* Check if there are enough reserved descriptors for transmission.
 * If not, request chunk of reserved descriptors and check again.
 */
static int mvpp2_txq_reserved_desc_num_proc(struct mvpp2 *priv,
					    struct mvpp2_tx_queue *txq,
					    struct mvpp2_txq_pcpu *txq_pcpu,
					    int num)
{
	int req, cpu, desc_count;

	if (txq_pcpu->reserved_num >= num)
		return 0;

	/* Not enough descriptors reserved! Update the reserved descriptor
	 * count and check again.
	 */

	desc_count = 0;
	/* Compute total of used descriptors */
	for_each_present_cpu(cpu) {
		struct mvpp2_txq_pcpu *txq_pcpu_aux;

		txq_pcpu_aux = per_cpu_ptr(txq->pcpu, cpu);
		desc_count += txq_pcpu_aux->count;
		desc_count += txq_pcpu_aux->reserved_num;
	}

	req = max(MVPP2_CPU_DESC_CHUNK, num - txq_pcpu->reserved_num);
	desc_count += req;

	if (desc_count >
	   (txq->size - (num_present_cpus() * MVPP2_CPU_DESC_CHUNK)))
		return -ENOMEM;

	txq_pcpu->reserved_num += mvpp2_txq_alloc_reserved_desc(priv, txq, req);

	/* OK, the descriptor cound has been updated: check again. */
	if (txq_pcpu->reserved_num < num)
		return -ENOMEM;
	return 0;
}

/* Release the last allocated Tx descriptor. Useful to handle DMA
 * mapping failures in the Tx path.
 */
static void mvpp2_txq_desc_put(struct mvpp2_tx_queue *txq)
{
	if (txq->next_desc_to_proc == 0)
		txq->next_desc_to_proc = txq->last_desc - 1;
	else
		txq->next_desc_to_proc--;
}

/* Set Tx descriptors fields relevant for CSUM calculation */
static u32 mvpp2_txq_desc_csum(int l3_offs, int l3_proto,
			       int ip_hdr_len, int l4_proto)
{
	u32 command;

	/* fields: L3_offset, IP_hdrlen, L3_type, G_IPv4_chk,
	 * G_L4_chk, L4_type required only for checksum calculation
	 */
	command = (l3_offs << MVPP2_TXD_L3_OFF_SHIFT);
	command |= (ip_hdr_len << MVPP2_TXD_IP_HLEN_SHIFT);
	command |= MVPP2_TXD_IP_CSUM_DISABLE;

	if (l3_proto == swab16(ETH_P_IP)) {
		command &= ~MVPP2_TXD_IP_CSUM_DISABLE;	/* enable IPv4 csum */
		command &= ~MVPP2_TXD_L3_IP6;		/* enable IPv4 */
	} else {
		command |= MVPP2_TXD_L3_IP6;		/* enable IPv6 */
	}

	if (l4_proto == IPPROTO_TCP) {
		command &= ~MVPP2_TXD_L4_UDP;		/* enable TCP */
		command &= ~MVPP2_TXD_L4_CSUM_FRAG;	/* generate L4 csum */
	} else if (l4_proto == IPPROTO_UDP) {
		command |= MVPP2_TXD_L4_UDP;		/* enable UDP */
		command &= ~MVPP2_TXD_L4_CSUM_FRAG;	/* generate L4 csum */
	} else {
		command |= MVPP2_TXD_L4_CSUM_NOT;
	}

	return command;
}

/* Get number of sent descriptors and decrement counter.
 * The number of sent descriptors is returned.
 * Per-CPU access
 *
 * Called only from mvpp2_txq_done(), called from mvpp2_tx()
 * (migration disabled) and from the TX completion tasklet (migration
 * disabled) so using smp_processor_id() is OK.
 */
static inline int mvpp2_txq_sent_desc_proc(struct mvpp2_port *port,
					   struct mvpp2_tx_queue *txq)
{
	u32 val;

	/* Reading status reg resets transmitted descriptor counter */
	val = mvpp2_percpu_read(port->priv, smp_processor_id(),
				MVPP2_TXQ_SENT_REG(txq->id));

	return (val & MVPP2_TRANSMITTED_COUNT_MASK) >>
		MVPP2_TRANSMITTED_COUNT_OFFSET;
}

/* Called through on_each_cpu(), so runs on all CPUs, with migration
 * disabled, therefore using smp_processor_id() is OK.
 */
static void mvpp2_txq_sent_counter_clear(void *arg)
{
	struct mvpp2_port *port = arg;
	int queue;

	for (queue = 0; queue < port->ntxqs; queue++) {
		int id = port->txqs[queue]->id;

		mvpp2_percpu_read(port->priv, smp_processor_id(),
				  MVPP2_TXQ_SENT_REG(id));
	}
}

/* Set max sizes for Tx queues */
static void mvpp2_txp_max_tx_size_set(struct mvpp2_port *port)
{
	u32	val, size, mtu;
	int	txq, tx_port_num;

	mtu = port->pkt_size * 8;
	if (mtu > MVPP2_TXP_MTU_MAX)
		mtu = MVPP2_TXP_MTU_MAX;

	/* WA for wrong Token bucket update: Set MTU value = 3*real MTU value */
	mtu = 3 * mtu;

	/* Indirect access to registers */
	tx_port_num = mvpp2_egress_port(port);
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PORT_INDEX_REG, tx_port_num);

	/* Set MTU */
	val = mvpp2_read(port->priv, MVPP2_TXP_SCHED_MTU_REG);
	val &= ~MVPP2_TXP_MTU_MAX;
	val |= mtu;
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_MTU_REG, val);

	/* TXP token size and all TXQs token size must be larger that MTU */
	val = mvpp2_read(port->priv, MVPP2_TXP_SCHED_TOKEN_SIZE_REG);
	size = val & MVPP2_TXP_TOKEN_SIZE_MAX;
	if (size < mtu) {
		size = mtu;
		val &= ~MVPP2_TXP_TOKEN_SIZE_MAX;
		val |= size;
		mvpp2_write(port->priv, MVPP2_TXP_SCHED_TOKEN_SIZE_REG, val);
	}

	for (txq = 0; txq < port->ntxqs; txq++) {
		val = mvpp2_read(port->priv,
				 MVPP2_TXQ_SCHED_TOKEN_SIZE_REG(txq));
		size = val & MVPP2_TXQ_TOKEN_SIZE_MAX;

		if (size < mtu) {
			size = mtu;
			val &= ~MVPP2_TXQ_TOKEN_SIZE_MAX;
			val |= size;
			mvpp2_write(port->priv,
				    MVPP2_TXQ_SCHED_TOKEN_SIZE_REG(txq),
				    val);
		}
	}
}

/* Set the number of packets that will be received before Rx interrupt
 * will be generated by HW.
 */
static void mvpp2_rx_pkts_coal_set(struct mvpp2_port *port,
				   struct mvpp2_rx_queue *rxq)
{
	int cpu = get_cpu();

	if (rxq->pkts_coal > MVPP2_OCCUPIED_THRESH_MASK)
		rxq->pkts_coal = MVPP2_OCCUPIED_THRESH_MASK;

	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_NUM_REG, rxq->id);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_THRESH_REG,
			   rxq->pkts_coal);

	put_cpu();
}

/* For some reason in the LSP this is done on each CPU. Why ? */
static void mvpp2_tx_pkts_coal_set(struct mvpp2_port *port,
				   struct mvpp2_tx_queue *txq)
{
	int cpu = get_cpu();
	u32 val;

	if (txq->done_pkts_coal > MVPP2_TXQ_THRESH_MASK)
		txq->done_pkts_coal = MVPP2_TXQ_THRESH_MASK;

	val = (txq->done_pkts_coal << MVPP2_TXQ_THRESH_OFFSET);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_NUM_REG, txq->id);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_THRESH_REG, val);

	put_cpu();
}

static u32 mvpp2_usec_to_cycles(u32 usec, unsigned long clk_hz)
{
	u64 tmp = (u64)clk_hz * usec;

	do_div(tmp, USEC_PER_SEC);

	return tmp > U32_MAX ? U32_MAX : tmp;
}

static u32 mvpp2_cycles_to_usec(u32 cycles, unsigned long clk_hz)
{
	u64 tmp = (u64)cycles * USEC_PER_SEC;

	do_div(tmp, clk_hz);

	return tmp > U32_MAX ? U32_MAX : tmp;
}

/* Set the time delay in usec before Rx interrupt */
static void mvpp2_rx_time_coal_set(struct mvpp2_port *port,
				   struct mvpp2_rx_queue *rxq)
{
	unsigned long freq = port->priv->tclk;
	u32 val = mvpp2_usec_to_cycles(rxq->time_coal, freq);

	if (val > MVPP2_MAX_ISR_RX_THRESHOLD) {
		rxq->time_coal =
			mvpp2_cycles_to_usec(MVPP2_MAX_ISR_RX_THRESHOLD, freq);

		/* re-evaluate to get actual register value */
		val = mvpp2_usec_to_cycles(rxq->time_coal, freq);
	}

	mvpp2_write(port->priv, MVPP2_ISR_RX_THRESHOLD_REG(rxq->id), val);
}

static void mvpp2_tx_time_coal_set(struct mvpp2_port *port)
{
	unsigned long freq = port->priv->tclk;
	u32 val = mvpp2_usec_to_cycles(port->tx_time_coal, freq);

	if (val > MVPP2_MAX_ISR_TX_THRESHOLD) {
		port->tx_time_coal =
			mvpp2_cycles_to_usec(MVPP2_MAX_ISR_TX_THRESHOLD, freq);

		/* re-evaluate to get actual register value */
		val = mvpp2_usec_to_cycles(port->tx_time_coal, freq);
	}

	mvpp2_write(port->priv, MVPP2_ISR_TX_THRESHOLD_REG(port->id), val);
}

/* Free Tx queue skbuffs */
static void mvpp2_txq_bufs_free(struct mvpp2_port *port,
				struct mvpp2_tx_queue *txq,
				struct mvpp2_txq_pcpu *txq_pcpu, int num)
{
	int i;

	for (i = 0; i < num; i++) {
		struct mvpp2_txq_pcpu_buf *tx_buf =
			txq_pcpu->buffs + txq_pcpu->txq_get_index;

		dma_unmap_single(port->dev->dev.parent, tx_buf->dma,
				 tx_buf->size, DMA_TO_DEVICE);
		if (tx_buf->skb)
			dev_kfree_skb_any(tx_buf->skb);

		mvpp2_txq_inc_get(txq_pcpu);
	}
}

static inline struct mvpp2_rx_queue *mvpp2_get_rx_queue(struct mvpp2_port *port,
							u32 cause)
{
	int queue = fls(cause) - 1;

	return port->rxqs[queue];
}

static inline struct mvpp2_tx_queue *mvpp2_get_tx_queue(struct mvpp2_port *port,
							u32 cause)
{
	int queue = fls(cause) - 1;

	return port->txqs[queue];
}

/* Handle end of transmission */
static void mvpp2_txq_done(struct mvpp2_port *port, struct mvpp2_tx_queue *txq,
			   struct mvpp2_txq_pcpu *txq_pcpu)
{
	struct netdev_queue *nq = netdev_get_tx_queue(port->dev, txq->log_id);
	int tx_done;

	if (txq_pcpu->cpu != smp_processor_id())
		netdev_err(port->dev, "wrong cpu on the end of Tx processing\n");

	tx_done = mvpp2_txq_sent_desc_proc(port, txq);
	if (!tx_done)
		return;
	mvpp2_txq_bufs_free(port, txq, txq_pcpu, tx_done);

	txq_pcpu->count -= tx_done;

	if (netif_tx_queue_stopped(nq))
		if (txq_pcpu->size - txq_pcpu->count >= MAX_SKB_FRAGS + 1)
			netif_tx_wake_queue(nq);
}

static unsigned int mvpp2_tx_done(struct mvpp2_port *port, u32 cause,
				  int cpu)
{
	struct mvpp2_tx_queue *txq;
	struct mvpp2_txq_pcpu *txq_pcpu;
	unsigned int tx_todo = 0;

	while (cause) {
		txq = mvpp2_get_tx_queue(port, cause);
		if (!txq)
			break;

		txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);

		if (txq_pcpu->count) {
			mvpp2_txq_done(port, txq, txq_pcpu);
			tx_todo += txq_pcpu->count;
		}

		cause &= ~(1 << txq->log_id);
	}
	return tx_todo;
}

/* Rx/Tx queue initialization/cleanup methods */

/* Allocate and initialize descriptors for aggr TXQ */
static int mvpp2_aggr_txq_init(struct platform_device *pdev,
			       struct mvpp2_tx_queue *aggr_txq,
			       int desc_num, int cpu,
			       struct mvpp2 *priv)
{
	u32 txq_dma;

	/* Allocate memory for TX descriptors */
	aggr_txq->descs = dma_alloc_coherent(&pdev->dev,
				desc_num * MVPP2_DESC_ALIGNED_SIZE,
				&aggr_txq->descs_dma, GFP_KERNEL);
	if (!aggr_txq->descs)
		return -ENOMEM;

	aggr_txq->last_desc = aggr_txq->size - 1;

	/* Aggr TXQ no reset WA */
	aggr_txq->next_desc_to_proc = mvpp2_read(priv,
						 MVPP2_AGGR_TXQ_INDEX_REG(cpu));

	/* Set Tx descriptors queue starting address indirect
	 * access
	 */
	if (priv->hw_version == MVPP21)
		txq_dma = aggr_txq->descs_dma;
	else
		txq_dma = aggr_txq->descs_dma >>
			MVPP22_AGGR_TXQ_DESC_ADDR_OFFS;

	mvpp2_write(priv, MVPP2_AGGR_TXQ_DESC_ADDR_REG(cpu), txq_dma);
	mvpp2_write(priv, MVPP2_AGGR_TXQ_DESC_SIZE_REG(cpu), desc_num);

	return 0;
}

/* Create a specified Rx queue */
static int mvpp2_rxq_init(struct mvpp2_port *port,
			  struct mvpp2_rx_queue *rxq)

{
	u32 rxq_dma;
	int cpu;

	rxq->size = port->rx_ring_size;

	/* Allocate memory for RX descriptors */
	rxq->descs = dma_alloc_coherent(port->dev->dev.parent,
					rxq->size * MVPP2_DESC_ALIGNED_SIZE,
					&rxq->descs_dma, GFP_KERNEL);
	if (!rxq->descs)
		return -ENOMEM;

	rxq->last_desc = rxq->size - 1;

	/* Zero occupied and non-occupied counters - direct access */
	mvpp2_write(port->priv, MVPP2_RXQ_STATUS_REG(rxq->id), 0);

	/* Set Rx descriptors queue starting address - indirect access */
	cpu = get_cpu();
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_NUM_REG, rxq->id);
	if (port->priv->hw_version == MVPP21)
		rxq_dma = rxq->descs_dma;
	else
		rxq_dma = rxq->descs_dma >> MVPP22_DESC_ADDR_OFFS;
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_DESC_ADDR_REG, rxq_dma);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_DESC_SIZE_REG, rxq->size);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_INDEX_REG, 0);
	put_cpu();

	/* Set Offset */
	mvpp2_rxq_offset_set(port, rxq->id, NET_SKB_PAD);

	/* Set coalescing pkts and time */
	mvpp2_rx_pkts_coal_set(port, rxq);
	mvpp2_rx_time_coal_set(port, rxq);

	/* Add number of descriptors ready for receiving packets */
	mvpp2_rxq_status_update(port, rxq->id, 0, rxq->size);

	return 0;
}

/* Push packets received by the RXQ to BM pool */
static void mvpp2_rxq_drop_pkts(struct mvpp2_port *port,
				struct mvpp2_rx_queue *rxq)
{
	int rx_received, i;

	rx_received = mvpp2_rxq_received(port, rxq->id);
	if (!rx_received)
		return;

	for (i = 0; i < rx_received; i++) {
		struct mvpp2_rx_desc *rx_desc = mvpp2_rxq_next_desc_get(rxq);
		u32 status = mvpp2_rxdesc_status_get(port, rx_desc);
		int pool;

		pool = (status & MVPP2_RXD_BM_POOL_ID_MASK) >>
			MVPP2_RXD_BM_POOL_ID_OFFS;

		mvpp2_bm_pool_put(port, pool,
				  mvpp2_rxdesc_dma_addr_get(port, rx_desc),
				  mvpp2_rxdesc_cookie_get(port, rx_desc));
	}
	mvpp2_rxq_status_update(port, rxq->id, rx_received, rx_received);
}

/* Cleanup Rx queue */
static void mvpp2_rxq_deinit(struct mvpp2_port *port,
			     struct mvpp2_rx_queue *rxq)
{
	int cpu;

	mvpp2_rxq_drop_pkts(port, rxq);

	if (rxq->descs)
		dma_free_coherent(port->dev->dev.parent,
				  rxq->size * MVPP2_DESC_ALIGNED_SIZE,
				  rxq->descs,
				  rxq->descs_dma);

	rxq->descs             = NULL;
	rxq->last_desc         = 0;
	rxq->next_desc_to_proc = 0;
	rxq->descs_dma         = 0;

	/* Clear Rx descriptors queue starting address and size;
	 * free descriptor number
	 */
	mvpp2_write(port->priv, MVPP2_RXQ_STATUS_REG(rxq->id), 0);
	cpu = get_cpu();
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_NUM_REG, rxq->id);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_DESC_ADDR_REG, 0);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_RXQ_DESC_SIZE_REG, 0);
	put_cpu();
}

/* Create and initialize a Tx queue */
static int mvpp2_txq_init(struct mvpp2_port *port,
			  struct mvpp2_tx_queue *txq)
{
	u32 val;
	int cpu, desc, desc_per_txq, tx_port_num;
	struct mvpp2_txq_pcpu *txq_pcpu;

	txq->size = port->tx_ring_size;

	/* Allocate memory for Tx descriptors */
	txq->descs = dma_alloc_coherent(port->dev->dev.parent,
				txq->size * MVPP2_DESC_ALIGNED_SIZE,
				&txq->descs_dma, GFP_KERNEL);
	if (!txq->descs)
		return -ENOMEM;

	txq->last_desc = txq->size - 1;

	/* Set Tx descriptors queue starting address - indirect access */
	cpu = get_cpu();
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_NUM_REG, txq->id);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_DESC_ADDR_REG,
			   txq->descs_dma);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_DESC_SIZE_REG,
			   txq->size & MVPP2_TXQ_DESC_SIZE_MASK);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_INDEX_REG, 0);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_RSVD_CLR_REG,
			   txq->id << MVPP2_TXQ_RSVD_CLR_OFFSET);
	val = mvpp2_percpu_read(port->priv, cpu, MVPP2_TXQ_PENDING_REG);
	val &= ~MVPP2_TXQ_PENDING_MASK;
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_PENDING_REG, val);

	/* Calculate base address in prefetch buffer. We reserve 16 descriptors
	 * for each existing TXQ.
	 * TCONTS for PON port must be continuous from 0 to MVPP2_MAX_TCONT
	 * GBE ports assumed to be continious from 0 to MVPP2_MAX_PORTS
	 */
	desc_per_txq = 16;
	desc = (port->id * MVPP2_MAX_TXQ * desc_per_txq) +
	       (txq->log_id * desc_per_txq);

	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_PREF_BUF_REG,
			   MVPP2_PREF_BUF_PTR(desc) | MVPP2_PREF_BUF_SIZE_16 |
			   MVPP2_PREF_BUF_THRESH(desc_per_txq / 2));
	put_cpu();

	/* WRR / EJP configuration - indirect access */
	tx_port_num = mvpp2_egress_port(port);
	mvpp2_write(port->priv, MVPP2_TXP_SCHED_PORT_INDEX_REG, tx_port_num);

	val = mvpp2_read(port->priv, MVPP2_TXQ_SCHED_REFILL_REG(txq->log_id));
	val &= ~MVPP2_TXQ_REFILL_PERIOD_ALL_MASK;
	val |= MVPP2_TXQ_REFILL_PERIOD_MASK(1);
	val |= MVPP2_TXQ_REFILL_TOKENS_ALL_MASK;
	mvpp2_write(port->priv, MVPP2_TXQ_SCHED_REFILL_REG(txq->log_id), val);

	val = MVPP2_TXQ_TOKEN_SIZE_MAX;
	mvpp2_write(port->priv, MVPP2_TXQ_SCHED_TOKEN_SIZE_REG(txq->log_id),
		    val);

	for_each_present_cpu(cpu) {
		txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);
		txq_pcpu->size = txq->size;
		txq_pcpu->buffs = kmalloc_array(txq_pcpu->size,
						sizeof(*txq_pcpu->buffs),
						GFP_KERNEL);
		if (!txq_pcpu->buffs)
			goto cleanup;

		txq_pcpu->count = 0;
		txq_pcpu->reserved_num = 0;
		txq_pcpu->txq_put_index = 0;
		txq_pcpu->txq_get_index = 0;
	}

	return 0;
cleanup:
	for_each_present_cpu(cpu) {
		txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);
		kfree(txq_pcpu->buffs);
	}

	dma_free_coherent(port->dev->dev.parent,
			  txq->size * MVPP2_DESC_ALIGNED_SIZE,
			  txq->descs, txq->descs_dma);

	return -ENOMEM;
}

/* Free allocated TXQ resources */
static void mvpp2_txq_deinit(struct mvpp2_port *port,
			     struct mvpp2_tx_queue *txq)
{
	struct mvpp2_txq_pcpu *txq_pcpu;
	int cpu;

	for_each_present_cpu(cpu) {
		txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);
		kfree(txq_pcpu->buffs);
	}

	if (txq->descs)
		dma_free_coherent(port->dev->dev.parent,
				  txq->size * MVPP2_DESC_ALIGNED_SIZE,
				  txq->descs, txq->descs_dma);

	txq->descs             = NULL;
	txq->last_desc         = 0;
	txq->next_desc_to_proc = 0;
	txq->descs_dma         = 0;

	/* Set minimum bandwidth for disabled TXQs */
	mvpp2_write(port->priv, MVPP2_TXQ_SCHED_TOKEN_CNTR_REG(txq->id), 0);

	/* Set Tx descriptors queue starting address and size */
	cpu = get_cpu();
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_NUM_REG, txq->id);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_DESC_ADDR_REG, 0);
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_DESC_SIZE_REG, 0);
	put_cpu();
}

/* Cleanup Tx ports */
static void mvpp2_txq_clean(struct mvpp2_port *port, struct mvpp2_tx_queue *txq)
{
	struct mvpp2_txq_pcpu *txq_pcpu;
	int delay, pending, cpu;
	u32 val;

	cpu = get_cpu();
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_NUM_REG, txq->id);
	val = mvpp2_percpu_read(port->priv, cpu, MVPP2_TXQ_PREF_BUF_REG);
	val |= MVPP2_TXQ_DRAIN_EN_MASK;
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_PREF_BUF_REG, val);

	/* The napi queue has been stopped so wait for all packets
	 * to be transmitted.
	 */
	delay = 0;
	do {
		if (delay >= MVPP2_TX_PENDING_TIMEOUT_MSEC) {
			netdev_warn(port->dev,
				    "port %d: cleaning queue %d timed out\n",
				    port->id, txq->log_id);
			break;
		}
		mdelay(1);
		delay++;

		pending = mvpp2_percpu_read(port->priv, cpu,
					    MVPP2_TXQ_PENDING_REG);
		pending &= MVPP2_TXQ_PENDING_MASK;
	} while (pending);

	val &= ~MVPP2_TXQ_DRAIN_EN_MASK;
	mvpp2_percpu_write(port->priv, cpu, MVPP2_TXQ_PREF_BUF_REG, val);
	put_cpu();

	for_each_present_cpu(cpu) {
		txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);

		/* Release all packets */
		mvpp2_txq_bufs_free(port, txq, txq_pcpu, txq_pcpu->count);

		/* Reset queue */
		txq_pcpu->count = 0;
		txq_pcpu->txq_put_index = 0;
		txq_pcpu->txq_get_index = 0;
	}
}

/* Cleanup all Tx queues */
static void mvpp2_cleanup_txqs(struct mvpp2_port *port)
{
	struct mvpp2_tx_queue *txq;
	int queue;
	u32 val;

	val = mvpp2_read(port->priv, MVPP2_TX_PORT_FLUSH_REG);

	/* Reset Tx ports and delete Tx queues */
	val |= MVPP2_TX_PORT_FLUSH_MASK(port->id);
	mvpp2_write(port->priv, MVPP2_TX_PORT_FLUSH_REG, val);

	for (queue = 0; queue < port->ntxqs; queue++) {
		txq = port->txqs[queue];
		mvpp2_txq_clean(port, txq);
		mvpp2_txq_deinit(port, txq);
	}

	on_each_cpu(mvpp2_txq_sent_counter_clear, port, 1);

	val &= ~MVPP2_TX_PORT_FLUSH_MASK(port->id);
	mvpp2_write(port->priv, MVPP2_TX_PORT_FLUSH_REG, val);
}

/* Cleanup all Rx queues */
static void mvpp2_cleanup_rxqs(struct mvpp2_port *port)
{
	int queue;

	for (queue = 0; queue < port->nrxqs; queue++)
		mvpp2_rxq_deinit(port, port->rxqs[queue]);
}

/* Init all Rx queues for port */
static int mvpp2_setup_rxqs(struct mvpp2_port *port)
{
	int queue, err;

	for (queue = 0; queue < port->nrxqs; queue++) {
		err = mvpp2_rxq_init(port, port->rxqs[queue]);
		if (err)
			goto err_cleanup;
	}
	return 0;

err_cleanup:
	mvpp2_cleanup_rxqs(port);
	return err;
}

/* Init all tx queues for port */
static int mvpp2_setup_txqs(struct mvpp2_port *port)
{
	struct mvpp2_tx_queue *txq;
	int queue, err;

	for (queue = 0; queue < port->ntxqs; queue++) {
		txq = port->txqs[queue];
		err = mvpp2_txq_init(port, txq);
		if (err)
			goto err_cleanup;
	}

	if (port->has_tx_irqs) {
		mvpp2_tx_time_coal_set(port);
		for (queue = 0; queue < port->ntxqs; queue++) {
			txq = port->txqs[queue];
			mvpp2_tx_pkts_coal_set(port, txq);
		}
	}

	on_each_cpu(mvpp2_txq_sent_counter_clear, port, 1);
	return 0;

err_cleanup:
	mvpp2_cleanup_txqs(port);
	return err;
}

/* The callback for per-port interrupt */
static irqreturn_t mvpp2_isr(int irq, void *dev_id)
{
	struct mvpp2_queue_vector *qv = dev_id;

	mvpp2_qvec_interrupt_disable(qv);

	napi_schedule(&qv->napi);

	return IRQ_HANDLED;
}

/* Adjust link */
static void mvpp2_link_event(struct net_device *dev)
{
	struct mvpp2_port *port = netdev_priv(dev);
	struct phy_device *phydev = dev->phydev;
	int status_change = 0;
	u32 val;

	if (phydev->link) {
		if ((port->speed != phydev->speed) ||
		    (port->duplex != phydev->duplex)) {
			u32 val;

			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
			val &= ~(MVPP2_GMAC_CONFIG_MII_SPEED |
				 MVPP2_GMAC_CONFIG_GMII_SPEED |
				 MVPP2_GMAC_CONFIG_FULL_DUPLEX |
				 MVPP2_GMAC_AN_SPEED_EN |
				 MVPP2_GMAC_AN_DUPLEX_EN);

			if (phydev->duplex)
				val |= MVPP2_GMAC_CONFIG_FULL_DUPLEX;

			if (phydev->speed == SPEED_1000)
				val |= MVPP2_GMAC_CONFIG_GMII_SPEED;
			else if (phydev->speed == SPEED_100)
				val |= MVPP2_GMAC_CONFIG_MII_SPEED;

			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);

			port->duplex = phydev->duplex;
			port->speed  = phydev->speed;
		}
	}

	if (phydev->link != port->link) {
		if (!phydev->link) {
			port->duplex = -1;
			port->speed = 0;
		}

		port->link = phydev->link;
		status_change = 1;
	}

	if (status_change) {
		if (phydev->link) {
			val = readl(port->base + MVPP2_GMAC_AUTONEG_CONFIG);
			val |= (MVPP2_GMAC_FORCE_LINK_PASS |
				MVPP2_GMAC_FORCE_LINK_DOWN);
			writel(val, port->base + MVPP2_GMAC_AUTONEG_CONFIG);
			mvpp2_egress_enable(port);
			mvpp2_ingress_enable(port);
		} else {
			mvpp2_ingress_disable(port);
			mvpp2_egress_disable(port);
		}
		phy_print_status(phydev);
	}
}

static void mvpp2_timer_set(struct mvpp2_port_pcpu *port_pcpu)
{
	ktime_t interval;

	if (!port_pcpu->timer_scheduled) {
		port_pcpu->timer_scheduled = true;
		interval = MVPP2_TXDONE_HRTIMER_PERIOD_NS;
		hrtimer_start(&port_pcpu->tx_done_timer, interval,
			      HRTIMER_MODE_REL_PINNED);
	}
}

static void mvpp2_tx_proc_cb(unsigned long data)
{
	struct net_device *dev = (struct net_device *)data;
	struct mvpp2_port *port = netdev_priv(dev);
	struct mvpp2_port_pcpu *port_pcpu = this_cpu_ptr(port->pcpu);
	unsigned int tx_todo, cause;

	if (!netif_running(dev))
		return;
	port_pcpu->timer_scheduled = false;

	/* Process all the Tx queues */
	cause = (1 << port->ntxqs) - 1;
	tx_todo = mvpp2_tx_done(port, cause, smp_processor_id());

	/* Set the timer in case not all the packets were processed */
	if (tx_todo)
		mvpp2_timer_set(port_pcpu);
}

static enum hrtimer_restart mvpp2_hr_timer_cb(struct hrtimer *timer)
{
	struct mvpp2_port_pcpu *port_pcpu = container_of(timer,
							 struct mvpp2_port_pcpu,
							 tx_done_timer);

	tasklet_schedule(&port_pcpu->tx_done_tasklet);

	return HRTIMER_NORESTART;
}

/* Main RX/TX processing routines */

/* Display more error info */
static void mvpp2_rx_error(struct mvpp2_port *port,
			   struct mvpp2_rx_desc *rx_desc)
{
	u32 status = mvpp2_rxdesc_status_get(port, rx_desc);
	size_t sz = mvpp2_rxdesc_size_get(port, rx_desc);

	switch (status & MVPP2_RXD_ERR_CODE_MASK) {
	case MVPP2_RXD_ERR_CRC:
		netdev_err(port->dev, "bad rx status %08x (crc error), size=%zu\n",
			   status, sz);
		break;
	case MVPP2_RXD_ERR_OVERRUN:
		netdev_err(port->dev, "bad rx status %08x (overrun error), size=%zu\n",
			   status, sz);
		break;
	case MVPP2_RXD_ERR_RESOURCE:
		netdev_err(port->dev, "bad rx status %08x (resource error), size=%zu\n",
			   status, sz);
		break;
	}
}

/* Handle RX checksum offload */
static void mvpp2_rx_csum(struct mvpp2_port *port, u32 status,
			  struct sk_buff *skb)
{
	if (((status & MVPP2_RXD_L3_IP4) &&
	     !(status & MVPP2_RXD_IP4_HEADER_ERR)) ||
	    (status & MVPP2_RXD_L3_IP6))
		if (((status & MVPP2_RXD_L4_UDP) ||
		     (status & MVPP2_RXD_L4_TCP)) &&
		     (status & MVPP2_RXD_L4_CSUM_OK)) {
			skb->csum = 0;
			skb->ip_summed = CHECKSUM_UNNECESSARY;
			return;
		}

	skb->ip_summed = CHECKSUM_NONE;
}

/* Reuse skb if possible, or allocate a new skb and add it to BM pool */
static int mvpp2_rx_refill(struct mvpp2_port *port,
			   struct mvpp2_bm_pool *bm_pool, int pool)
{
	dma_addr_t dma_addr;
	phys_addr_t phys_addr;
	void *buf;

	/* No recycle or too many buffers are in use, so allocate a new skb */
	buf = mvpp2_buf_alloc(port, bm_pool, &dma_addr, &phys_addr,
			      GFP_ATOMIC);
	if (!buf)
		return -ENOMEM;

	mvpp2_bm_pool_put(port, pool, dma_addr, phys_addr);

	return 0;
}

/* Handle tx checksum */
static u32 mvpp2_skb_tx_csum(struct mvpp2_port *port, struct sk_buff *skb)
{
	if (skb->ip_summed == CHECKSUM_PARTIAL) {
		int ip_hdr_len = 0;
		u8 l4_proto;

		if (skb->protocol == htons(ETH_P_IP)) {
			struct iphdr *ip4h = ip_hdr(skb);

			/* Calculate IPv4 checksum and L4 checksum */
			ip_hdr_len = ip4h->ihl;
			l4_proto = ip4h->protocol;
		} else if (skb->protocol == htons(ETH_P_IPV6)) {
			struct ipv6hdr *ip6h = ipv6_hdr(skb);

			/* Read l4_protocol from one of IPv6 extra headers */
			if (skb_network_header_len(skb) > 0)
				ip_hdr_len = (skb_network_header_len(skb) >> 2);
			l4_proto = ip6h->nexthdr;
		} else {
			return MVPP2_TXD_L4_CSUM_NOT;
		}

		return mvpp2_txq_desc_csum(skb_network_offset(skb),
				skb->protocol, ip_hdr_len, l4_proto);
	}

	return MVPP2_TXD_L4_CSUM_NOT | MVPP2_TXD_IP_CSUM_DISABLE;
}

/* Main rx processing */
static int mvpp2_rx(struct mvpp2_port *port, struct napi_struct *napi,
		    int rx_todo, struct mvpp2_rx_queue *rxq)
{
	struct net_device *dev = port->dev;
	int rx_received;
	int rx_done = 0;
	u32 rcvd_pkts = 0;
	u32 rcvd_bytes = 0;

	/* Get number of received packets and clamp the to-do */
	rx_received = mvpp2_rxq_received(port, rxq->id);
	if (rx_todo > rx_received)
		rx_todo = rx_received;

	while (rx_done < rx_todo) {
		struct mvpp2_rx_desc *rx_desc = mvpp2_rxq_next_desc_get(rxq);
		struct mvpp2_bm_pool *bm_pool;
		struct sk_buff *skb;
		unsigned int frag_size;
		dma_addr_t dma_addr;
		phys_addr_t phys_addr;
		u32 rx_status;
		int pool, rx_bytes, err;
		void *data;

		rx_done++;
		rx_status = mvpp2_rxdesc_status_get(port, rx_desc);
		rx_bytes = mvpp2_rxdesc_size_get(port, rx_desc);
		rx_bytes -= MVPP2_MH_SIZE;
		dma_addr = mvpp2_rxdesc_dma_addr_get(port, rx_desc);
		phys_addr = mvpp2_rxdesc_cookie_get(port, rx_desc);
		data = (void *)phys_to_virt(phys_addr);

		pool = (rx_status & MVPP2_RXD_BM_POOL_ID_MASK) >>
			MVPP2_RXD_BM_POOL_ID_OFFS;
		bm_pool = &port->priv->bm_pools[pool];

		/* In case of an error, release the requested buffer pointer
		 * to the Buffer Manager. This request process is controlled
		 * by the hardware, and the information about the buffer is
		 * comprised by the RX descriptor.
		 */
		if (rx_status & MVPP2_RXD_ERR_SUMMARY) {
err_drop_frame:
			dev->stats.rx_errors++;
			mvpp2_rx_error(port, rx_desc);
			/* Return the buffer to the pool */
			mvpp2_bm_pool_put(port, pool, dma_addr, phys_addr);
			continue;
		}

		if (bm_pool->frag_size > PAGE_SIZE)
			frag_size = 0;
		else
			frag_size = bm_pool->frag_size;

		skb = build_skb(data, frag_size);
		if (!skb) {
			netdev_warn(port->dev, "skb build failed\n");
			goto err_drop_frame;
		}

		err = mvpp2_rx_refill(port, bm_pool, pool);
		if (err) {
			netdev_err(port->dev, "failed to refill BM pools\n");
			goto err_drop_frame;
		}

		dma_unmap_single(dev->dev.parent, dma_addr,
				 bm_pool->buf_size, DMA_FROM_DEVICE);

		rcvd_pkts++;
		rcvd_bytes += rx_bytes;

		skb_reserve(skb, MVPP2_MH_SIZE + NET_SKB_PAD);
		skb_put(skb, rx_bytes);
		skb->protocol = eth_type_trans(skb, dev);
		mvpp2_rx_csum(port, rx_status, skb);

		napi_gro_receive(napi, skb);
	}

	if (rcvd_pkts) {
		struct mvpp2_pcpu_stats *stats = this_cpu_ptr(port->stats);

		u64_stats_update_begin(&stats->syncp);
		stats->rx_packets += rcvd_pkts;
		stats->rx_bytes   += rcvd_bytes;
		u64_stats_update_end(&stats->syncp);
	}

	/* Update Rx queue management counters */
	wmb();
	mvpp2_rxq_status_update(port, rxq->id, rx_done, rx_done);

	return rx_todo;
}

static inline void
tx_desc_unmap_put(struct mvpp2_port *port, struct mvpp2_tx_queue *txq,
		  struct mvpp2_tx_desc *desc)
{
	dma_addr_t buf_dma_addr =
		mvpp2_txdesc_dma_addr_get(port, desc);
	size_t buf_sz =
		mvpp2_txdesc_size_get(port, desc);
	dma_unmap_single(port->dev->dev.parent, buf_dma_addr,
			 buf_sz, DMA_TO_DEVICE);
	mvpp2_txq_desc_put(txq);
}

/* Handle tx fragmentation processing */
static int mvpp2_tx_frag_process(struct mvpp2_port *port, struct sk_buff *skb,
				 struct mvpp2_tx_queue *aggr_txq,
				 struct mvpp2_tx_queue *txq)
{
	struct mvpp2_txq_pcpu *txq_pcpu = this_cpu_ptr(txq->pcpu);
	struct mvpp2_tx_desc *tx_desc;
	int i;
	dma_addr_t buf_dma_addr;

	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
		void *addr = page_address(frag->page.p) + frag->page_offset;

		tx_desc = mvpp2_txq_next_desc_get(aggr_txq);
		mvpp2_txdesc_txq_set(port, tx_desc, txq->id);
		mvpp2_txdesc_size_set(port, tx_desc, frag->size);

		buf_dma_addr = dma_map_single(port->dev->dev.parent, addr,
					       frag->size,
					       DMA_TO_DEVICE);
		if (dma_mapping_error(port->dev->dev.parent, buf_dma_addr)) {
			mvpp2_txq_desc_put(txq);
			goto cleanup;
		}

		mvpp2_txdesc_offset_set(port, tx_desc,
					buf_dma_addr & MVPP2_TX_DESC_ALIGN);
		mvpp2_txdesc_dma_addr_set(port, tx_desc,
					  buf_dma_addr & ~MVPP2_TX_DESC_ALIGN);

		if (i == (skb_shinfo(skb)->nr_frags - 1)) {
			/* Last descriptor */
			mvpp2_txdesc_cmd_set(port, tx_desc,
					     MVPP2_TXD_L_DESC);
			mvpp2_txq_inc_put(port, txq_pcpu, skb, tx_desc);
		} else {
			/* Descriptor in the middle: Not First, Not Last */
			mvpp2_txdesc_cmd_set(port, tx_desc, 0);
			mvpp2_txq_inc_put(port, txq_pcpu, NULL, tx_desc);
		}
	}

	return 0;
cleanup:
	/* Release all descriptors that were used to map fragments of
	 * this packet, as well as the corresponding DMA mappings
	 */
	for (i = i - 1; i >= 0; i--) {
		tx_desc = txq->descs + i;
		tx_desc_unmap_put(port, txq, tx_desc);
	}

	return -ENOMEM;
}

/* Main tx processing */
static int mvpp2_tx(struct sk_buff *skb, struct net_device *dev)
{
	struct mvpp2_port *port = netdev_priv(dev);
	struct mvpp2_tx_queue *txq, *aggr_txq;
	struct mvpp2_txq_pcpu *txq_pcpu;
	struct mvpp2_tx_desc *tx_desc;
	dma_addr_t buf_dma_addr;
	int frags = 0;
	u16 txq_id;
	u32 tx_cmd;

	txq_id = skb_get_queue_mapping(skb);
	txq = port->txqs[txq_id];
	txq_pcpu = this_cpu_ptr(txq->pcpu);
	aggr_txq = &port->priv->aggr_txqs[smp_processor_id()];

	frags = skb_shinfo(skb)->nr_frags + 1;

	/* Check number of available descriptors */
	if (mvpp2_aggr_desc_num_check(port->priv, aggr_txq, frags) ||
	    mvpp2_txq_reserved_desc_num_proc(port->priv, txq,
					     txq_pcpu, frags)) {
		frags = 0;
		goto out;
	}

	/* Get a descriptor for the first part of the packet */
	tx_desc = mvpp2_txq_next_desc_get(aggr_txq);
	mvpp2_txdesc_txq_set(port, tx_desc, txq->id);
	mvpp2_txdesc_size_set(port, tx_desc, skb_headlen(skb));

	buf_dma_addr = dma_map_single(dev->dev.parent, skb->data,
				      skb_headlen(skb), DMA_TO_DEVICE);
	if (unlikely(dma_mapping_error(dev->dev.parent, buf_dma_addr))) {
		mvpp2_txq_desc_put(txq);
		frags = 0;
		goto out;
	}

	mvpp2_txdesc_offset_set(port, tx_desc,
				buf_dma_addr & MVPP2_TX_DESC_ALIGN);
	mvpp2_txdesc_dma_addr_set(port, tx_desc,
				  buf_dma_addr & ~MVPP2_TX_DESC_ALIGN);

	tx_cmd = mvpp2_skb_tx_csum(port, skb);

	if (frags == 1) {
		/* First and Last descriptor */
		tx_cmd |= MVPP2_TXD_F_DESC | MVPP2_TXD_L_DESC;
		mvpp2_txdesc_cmd_set(port, tx_desc, tx_cmd);
		mvpp2_txq_inc_put(port, txq_pcpu, skb, tx_desc);
	} else {
		/* First but not Last */
		tx_cmd |= MVPP2_TXD_F_DESC | MVPP2_TXD_PADDING_DISABLE;
		mvpp2_txdesc_cmd_set(port, tx_desc, tx_cmd);
		mvpp2_txq_inc_put(port, txq_pcpu, NULL, tx_desc);

		/* Continue with other skb fragments */
		if (mvpp2_tx_frag_process(port, skb, aggr_txq, txq)) {
			tx_desc_unmap_put(port, txq, tx_desc);
			frags = 0;
			goto out;
		}
	}

	txq_pcpu->reserved_num -= frags;
	txq_pcpu->count += frags;
	aggr_txq->count += frags;

	/* Enable transmit */
	wmb();
	mvpp2_aggr_txq_pend_desc_add(port, frags);

	if (txq_pcpu->size - txq_pcpu->count < MAX_SKB_FRAGS + 1) {
		struct netdev_queue *nq = netdev_get_tx_queue(dev, txq_id);

		netif_tx_stop_queue(nq);
	}
out:
	if (frags > 0) {
		struct mvpp2_pcpu_stats *stats = this_cpu_ptr(port->stats);

		u64_stats_update_begin(&stats->syncp);
		stats->tx_packets++;
		stats->tx_bytes += skb->len;
		u64_stats_update_end(&stats->syncp);
	} else {
		dev->stats.tx_dropped++;
		dev_kfree_skb_any(skb);
	}

	/* Finalize TX processing */
	if (txq_pcpu->count >= txq->done_pkts_coal)
		mvpp2_txq_done(port, txq, txq_pcpu);

	/* Set the timer in case not all frags were processed */
	if (!port->has_tx_irqs && txq_pcpu->count <= frags &&
	    txq_pcpu->count > 0) {
		struct mvpp2_port_pcpu *port_pcpu = this_cpu_ptr(port->pcpu);

		mvpp2_timer_set(port_pcpu);
	}

	return NETDEV_TX_OK;
}

static inline void mvpp2_cause_error(struct net_device *dev, int cause)
{
	if (cause & MVPP2_CAUSE_FCS_ERR_MASK)
		netdev_err(dev, "FCS error\n");
	if (cause & MVPP2_CAUSE_RX_FIFO_OVERRUN_MASK)
		netdev_err(dev, "rx fifo overrun error\n");
	if (cause & MVPP2_CAUSE_TX_FIFO_UNDERRUN_MASK)
		netdev_err(dev, "tx fifo underrun error\n");
}

static int mvpp2_poll(struct napi_struct *napi, int budget)
{
	u32 cause_rx_tx, cause_rx, cause_tx, cause_misc;
	int rx_done = 0;
	struct mvpp2_port *port = netdev_priv(napi->dev);
	struct mvpp2_queue_vector *qv;
	int cpu = smp_processor_id();

	qv = container_of(napi, struct mvpp2_queue_vector, napi);

	/* Rx/Tx cause register
	 *
	 * Bits 0-15: each bit indicates received packets on the Rx queue
	 * (bit 0 is for Rx queue 0).
	 *
	 * Bits 16-23: each bit indicates transmitted packets on the Tx queue
	 * (bit 16 is for Tx queue 0).
	 *
	 * Each CPU has its own Rx/Tx cause register
	 */
	cause_rx_tx = mvpp2_percpu_read(port->priv, qv->sw_thread_id,
					MVPP2_ISR_RX_TX_CAUSE_REG(port->id));

	cause_misc = cause_rx_tx & MVPP2_CAUSE_MISC_SUM_MASK;
	if (cause_misc) {
		mvpp2_cause_error(port->dev, cause_misc);

		/* Clear the cause register */
		mvpp2_write(port->priv, MVPP2_ISR_MISC_CAUSE_REG, 0);
		mvpp2_percpu_write(port->priv, cpu,
				   MVPP2_ISR_RX_TX_CAUSE_REG(port->id),
				   cause_rx_tx & ~MVPP2_CAUSE_MISC_SUM_MASK);
	}

	cause_tx = cause_rx_tx & MVPP2_CAUSE_TXQ_OCCUP_DESC_ALL_MASK;
	if (cause_tx) {
		cause_tx >>= MVPP2_CAUSE_TXQ_OCCUP_DESC_ALL_OFFSET;
		mvpp2_tx_done(port, cause_tx, qv->sw_thread_id);
	}

	/* Process RX packets */
	cause_rx = cause_rx_tx & MVPP2_CAUSE_RXQ_OCCUP_DESC_ALL_MASK;
	cause_rx <<= qv->first_rxq;
	cause_rx |= qv->pending_cause_rx;
	while (cause_rx && budget > 0) {
		int count;
		struct mvpp2_rx_queue *rxq;

		rxq = mvpp2_get_rx_queue(port, cause_rx);
		if (!rxq)
			break;

		count = mvpp2_rx(port, napi, budget, rxq);
		rx_done += count;
		budget -= count;
		if (budget > 0) {
			/* Clear the bit associated to this Rx queue
			 * so that next iteration will continue from
			 * the next Rx queue.
			 */
			cause_rx &= ~(1 << rxq->logic_rxq);
		}
	}

	if (budget > 0) {
		cause_rx = 0;
		napi_complete_done(napi, rx_done);

		mvpp2_qvec_interrupt_enable(qv);
	}
	qv->pending_cause_rx = cause_rx;
	return rx_done;
}

/* Set hw internals when starting port */
static void mvpp2_start_dev(struct mvpp2_port *port)
{
	struct net_device *ndev = port->dev;
	int i;

	mvpp2_gmac_max_rx_size_set(port);
	mvpp2_txp_max_tx_size_set(port);

	for (i = 0; i < port->nqvecs; i++)
		napi_enable(&port->qvecs[i].napi);

	/* Enable interrupts on all CPUs */
	mvpp2_interrupts_enable(port);

	mvpp2_port_enable(port);
	phy_start(ndev->phydev);
	netif_tx_start_all_queues(port->dev);
}

/* Set hw internals when stopping port */
static void mvpp2_stop_dev(struct mvpp2_port *port)
{
	struct net_device *ndev = port->dev;
	int i;

	/* Stop new packets from arriving to RXQs */
	mvpp2_ingress_disable(port);

	mdelay(10);

	/* Disable interrupts on all CPUs */
	mvpp2_interrupts_disable(port);

	for (i = 0; i < port->nqvecs; i++)
		napi_disable(&port->qvecs[i].napi);

	netif_carrier_off(port->dev);
	netif_tx_stop_all_queues(port->dev);

	mvpp2_egress_disable(port);
	mvpp2_port_disable(port);
	phy_stop(ndev->phydev);
}

static int mvpp2_check_ringparam_valid(struct net_device *dev,
				       struct ethtool_ringparam *ring)
{
	u16 new_rx_pending = ring->rx_pending;
	u16 new_tx_pending = ring->tx_pending;

	if (ring->rx_pending == 0 || ring->tx_pending == 0)
		return -EINVAL;

	if (ring->rx_pending > MVPP2_MAX_RXD)
		new_rx_pending = MVPP2_MAX_RXD;
	else if (!IS_ALIGNED(ring->rx_pending, 16))
		new_rx_pending = ALIGN(ring->rx_pending, 16);

	if (ring->tx_pending > MVPP2_MAX_TXD)
		new_tx_pending = MVPP2_MAX_TXD;
	else if (!IS_ALIGNED(ring->tx_pending, 32))
		new_tx_pending = ALIGN(ring->tx_pending, 32);

	if (ring->rx_pending != new_rx_pending) {
		netdev_info(dev, "illegal Rx ring size value %d, round to %d\n",
			    ring->rx_pending, new_rx_pending);
		ring->rx_pending = new_rx_pending;
	}

	if (ring->tx_pending != new_tx_pending) {
		netdev_info(dev, "illegal Tx ring size value %d, round to %d\n",
			    ring->tx_pending, new_tx_pending);
		ring->tx_pending = new_tx_pending;
	}

	return 0;
}

static void mvpp21_get_mac_address(struct mvpp2_port *port, unsigned char *addr)
{
	u32 mac_addr_l, mac_addr_m, mac_addr_h;

	mac_addr_l = readl(port->base + MVPP2_GMAC_CTRL_1_REG);
	mac_addr_m = readl(port->priv->lms_base + MVPP2_SRC_ADDR_MIDDLE);
	mac_addr_h = readl(port->priv->lms_base + MVPP2_SRC_ADDR_HIGH);
	addr[0] = (mac_addr_h >> 24) & 0xFF;
	addr[1] = (mac_addr_h >> 16) & 0xFF;
	addr[2] = (mac_addr_h >> 8) & 0xFF;
	addr[3] = mac_addr_h & 0xFF;
	addr[4] = mac_addr_m & 0xFF;
	addr[5] = (mac_addr_l >> MVPP2_GMAC_SA_LOW_OFFS) & 0xFF;
}

static int mvpp2_phy_connect(struct mvpp2_port *port)
{
	struct phy_device *phy_dev;

	phy_dev = of_phy_connect(port->dev, port->phy_node, mvpp2_link_event, 0,
				 port->phy_interface);
	if (!phy_dev) {
		netdev_err(port->dev, "cannot connect to phy\n");
		return -ENODEV;
	}
	phy_dev->supported &= PHY_GBIT_FEATURES;
	phy_dev->advertising = phy_dev->supported;

	port->link    = 0;
	port->duplex  = 0;
	port->speed   = 0;

	return 0;
}

static void mvpp2_phy_disconnect(struct mvpp2_port *port)
{
	struct net_device *ndev = port->dev;

	phy_disconnect(ndev->phydev);
}

static int mvpp2_irqs_init(struct mvpp2_port *port)
{
	int err, i;

	for (i = 0; i < port->nqvecs; i++) {
		struct mvpp2_queue_vector *qv = port->qvecs + i;

		err = request_irq(qv->irq, mvpp2_isr, 0, port->dev->name, qv);
		if (err)
			goto err;

		if (qv->type == MVPP2_QUEUE_VECTOR_PRIVATE)
			irq_set_affinity_hint(qv->irq,
					      cpumask_of(qv->sw_thread_id));
	}

	return 0;
err:
	for (i = 0; i < port->nqvecs; i++) {
		struct mvpp2_queue_vector *qv = port->qvecs + i;

		irq_set_affinity_hint(qv->irq, NULL);
		free_irq(qv->irq, qv);
	}

	return err;
}

static void mvpp2_irqs_deinit(struct mvpp2_port *port)
{
	int i;

	for (i = 0; i < port->nqvecs; i++) {
		struct mvpp2_queue_vector *qv = port->qvecs + i;

		irq_set_affinity_hint(qv->irq, NULL);
		free_irq(qv->irq, qv);
	}
}

static int mvpp2_open(struct net_device *dev)
{
	struct mvpp2_port *port = netdev_priv(dev);
	unsigned char mac_bcast[ETH_ALEN] = {
			0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
	int err;

	err = mvpp2_prs_mac_da_accept(port->priv, port->id, mac_bcast, true);
	if (err) {
		netdev_err(dev, "mvpp2_prs_mac_da_accept BC failed\n");
		return err;
	}
	err = mvpp2_prs_mac_da_accept(port->priv, port->id,
				      dev->dev_addr, true);
	if (err) {
		netdev_err(dev, "mvpp2_prs_mac_da_accept MC failed\n");
		return err;
	}
	err = mvpp2_prs_tag_mode_set(port->priv, port->id, MVPP2_TAG_TYPE_MH);
	if (err) {
		netdev_err(dev, "mvpp2_prs_tag_mode_set failed\n");
		return err;
	}
	err = mvpp2_prs_def_flow(port);
	if (err) {
		netdev_err(dev, "mvpp2_prs_def_flow failed\n");
		return err;
	}

	/* Allocate the Rx/Tx queues */
	err = mvpp2_setup_rxqs(port);
	if (err) {
		netdev_err(port->dev, "cannot allocate Rx queues\n");
		return err;
	}

	err = mvpp2_setup_txqs(port);
	if (err) {
		netdev_err(port->dev, "cannot allocate Tx queues\n");
		goto err_cleanup_rxqs;
	}

	err = mvpp2_irqs_init(port);
	if (err) {
		netdev_err(port->dev, "cannot init IRQs\n");
		goto err_cleanup_txqs;
	}

	/* In default link is down */
	netif_carrier_off(port->dev);

	err = mvpp2_phy_connect(port);
	if (err < 0)
		goto err_free_irq;

	/* Unmask interrupts on all CPUs */
	on_each_cpu(mvpp2_interrupts_unmask, port, 1);
	mvpp2_shared_interrupt_mask_unmask(port, false);

	mvpp2_start_dev(port);

	return 0;

err_free_irq:
	mvpp2_irqs_deinit(port);
err_cleanup_txqs:
	mvpp2_cleanup_txqs(port);
err_cleanup_rxqs:
	mvpp2_cleanup_rxqs(port);
	return err;
}

static int mvpp2_stop(struct net_device *dev)
{
	struct mvpp2_port *port = netdev_priv(dev);
	struct mvpp2_port_pcpu *port_pcpu;
	int cpu;

	mvpp2_stop_dev(port);
	mvpp2_phy_disconnect(port);

	/* Mask interrupts on all CPUs */
	on_each_cpu(mvpp2_interrupts_mask, port, 1);
	mvpp2_shared_interrupt_mask_unmask(port, true);

	mvpp2_irqs_deinit(port);
	if (!port->has_tx_irqs) {
		for_each_present_cpu(cpu) {
			port_pcpu = per_cpu_ptr(port->pcpu, cpu);

			hrtimer_cancel(&port_pcpu->tx_done_timer);
			port_pcpu->timer_scheduled = false;
			tasklet_kill(&port_pcpu->tx_done_tasklet);
		}
	}
	mvpp2_cleanup_rxqs(port);
	mvpp2_cleanup_txqs(port);

	return 0;
}

static void mvpp2_set_rx_mode(struct net_device *dev)
{
	struct mvpp2_port *port = netdev_priv(dev);
	struct mvpp2 *priv = port->priv;
	struct netdev_hw_addr *ha;
	int id = port->id;
	bool allmulti = dev->flags & IFF_ALLMULTI;

	mvpp2_prs_mac_promisc_set(priv, id, dev->flags & IFF_PROMISC);
	mvpp2_prs_mac_multi_set(priv, id, MVPP2_PE_MAC_MC_ALL, allmulti);
	mvpp2_prs_mac_multi_set(priv, id, MVPP2_PE_MAC_MC_IP6, allmulti);

	/* Remove all port->id's mcast enries */
	mvpp2_prs_mcast_del_all(priv, id);

	if (allmulti && !netdev_mc_empty(dev)) {
		netdev_for_each_mc_addr(ha, dev)
			mvpp2_prs_mac_da_accept(priv, id, ha->addr, true);
	}
}

static int mvpp2_set_mac_address(struct net_device *dev, void *p)
{
	struct mvpp2_port *port = netdev_priv(dev);
	const struct sockaddr *addr = p;
	int err;

	if (!is_valid_ether_addr(addr->sa_data)) {
		err = -EADDRNOTAVAIL;
		goto log_error;
	}

	if (!netif_running(dev)) {
		err = mvpp2_prs_update_mac_da(dev, addr->sa_data);
		if (!err)
			return 0;
		/* Reconfigure parser to accept the original MAC address */
		err = mvpp2_prs_update_mac_da(dev, dev->dev_addr);
		if (err)
			goto log_error;
	}

	mvpp2_stop_dev(port);

	err = mvpp2_prs_update_mac_da(dev, addr->sa_data);
	if (!err)
		goto out_start;

	/* Reconfigure parser accept the original MAC address */
	err = mvpp2_prs_update_mac_da(dev, dev->dev_addr);
	if (err)
		goto log_error;
out_start:
	mvpp2_start_dev(port);
	mvpp2_egress_enable(port);
	mvpp2_ingress_enable(port);
	return 0;
log_error:
	netdev_err(dev, "failed to change MAC address\n");
	return err;
}

static int mvpp2_change_mtu(struct net_device *dev, int mtu)
{
	struct mvpp2_port *port = netdev_priv(dev);
	int err;

	if (!IS_ALIGNED(MVPP2_RX_PKT_SIZE(mtu), 8)) {
		netdev_info(dev, "illegal MTU value %d, round to %d\n", mtu,
			    ALIGN(MVPP2_RX_PKT_SIZE(mtu), 8));
		mtu = ALIGN(MVPP2_RX_PKT_SIZE(mtu), 8);
	}

	if (!netif_running(dev)) {
		err = mvpp2_bm_update_mtu(dev, mtu);
		if (!err) {
			port->pkt_size =  MVPP2_RX_PKT_SIZE(mtu);
			return 0;
		}

		/* Reconfigure BM to the original MTU */
		err = mvpp2_bm_update_mtu(dev, dev->mtu);
		if (err)
			goto log_error;
	}

	mvpp2_stop_dev(port);

	err = mvpp2_bm_update_mtu(dev, mtu);
	if (!err) {
		port->pkt_size =  MVPP2_RX_PKT_SIZE(mtu);
		goto out_start;
	}

	/* Reconfigure BM to the original MTU */
	err = mvpp2_bm_update_mtu(dev, dev->mtu);
	if (err)
		goto log_error;

out_start:
	mvpp2_start_dev(port);
	mvpp2_egress_enable(port);
	mvpp2_ingress_enable(port);

	return 0;
log_error:
	netdev_err(dev, "failed to change MTU\n");
	return err;
}

static void
mvpp2_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
{
	struct mvpp2_port *port = netdev_priv(dev);
	unsigned int start;
	int cpu;

	for_each_possible_cpu(cpu) {
		struct mvpp2_pcpu_stats *cpu_stats;
		u64 rx_packets;
		u64 rx_bytes;
		u64 tx_packets;
		u64 tx_bytes;

		cpu_stats = per_cpu_ptr(port->stats, cpu);
		do {
			start = u64_stats_fetch_begin_irq(&cpu_stats->syncp);
			rx_packets = cpu_stats->rx_packets;
			rx_bytes   = cpu_stats->rx_bytes;
			tx_packets = cpu_stats->tx_packets;
			tx_bytes   = cpu_stats->tx_bytes;
		} while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start));

		stats->rx_packets += rx_packets;
		stats->rx_bytes   += rx_bytes;
		stats->tx_packets += tx_packets;
		stats->tx_bytes   += tx_bytes;
	}

	stats->rx_errors	= dev->stats.rx_errors;
	stats->rx_dropped	= dev->stats.rx_dropped;
	stats->tx_dropped	= dev->stats.tx_dropped;
}

static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
	int ret;

	if (!dev->phydev)
		return -ENOTSUPP;

	ret = phy_mii_ioctl(dev->phydev, ifr, cmd);
	if (!ret)
		mvpp2_link_event(dev);

	return ret;
}

/* Ethtool methods */

/* Set interrupt coalescing for ethtools */
static int mvpp2_ethtool_set_coalesce(struct net_device *dev,
				      struct ethtool_coalesce *c)
{
	struct mvpp2_port *port = netdev_priv(dev);
	int queue;

	for (queue = 0; queue < port->nrxqs; queue++) {
		struct mvpp2_rx_queue *rxq = port->rxqs[queue];

		rxq->time_coal = c->rx_coalesce_usecs;
		rxq->pkts_coal = c->rx_max_coalesced_frames;
		mvpp2_rx_pkts_coal_set(port, rxq);
		mvpp2_rx_time_coal_set(port, rxq);
	}

	if (port->has_tx_irqs) {
		port->tx_time_coal = c->tx_coalesce_usecs;
		mvpp2_tx_time_coal_set(port);
	}

	for (queue = 0; queue < port->ntxqs; queue++) {
		struct mvpp2_tx_queue *txq = port->txqs[queue];

		txq->done_pkts_coal = c->tx_max_coalesced_frames;

		if (port->has_tx_irqs)
			mvpp2_tx_pkts_coal_set(port, txq);
	}

	return 0;
}

/* get coalescing for ethtools */
static int mvpp2_ethtool_get_coalesce(struct net_device *dev,
				      struct ethtool_coalesce *c)
{
	struct mvpp2_port *port = netdev_priv(dev);

	c->rx_coalesce_usecs        = port->rxqs[0]->time_coal;
	c->rx_max_coalesced_frames  = port->rxqs[0]->pkts_coal;
	c->tx_max_coalesced_frames =  port->txqs[0]->done_pkts_coal;
	return 0;
}

static void mvpp2_ethtool_get_drvinfo(struct net_device *dev,
				      struct ethtool_drvinfo *drvinfo)
{
	strlcpy(drvinfo->driver, MVPP2_DRIVER_NAME,
		sizeof(drvinfo->driver));
	strlcpy(drvinfo->version, MVPP2_DRIVER_VERSION,
		sizeof(drvinfo->version));
	strlcpy(drvinfo->bus_info, dev_name(&dev->dev),
		sizeof(drvinfo->bus_info));
}

static void mvpp2_ethtool_get_ringparam(struct net_device *dev,
					struct ethtool_ringparam *ring)
{
	struct mvpp2_port *port = netdev_priv(dev);

	ring->rx_max_pending = MVPP2_MAX_RXD;
	ring->tx_max_pending = MVPP2_MAX_TXD;
	ring->rx_pending = port->rx_ring_size;
	ring->tx_pending = port->tx_ring_size;
}

static int mvpp2_ethtool_set_ringparam(struct net_device *dev,
				       struct ethtool_ringparam *ring)
{
	struct mvpp2_port *port = netdev_priv(dev);
	u16 prev_rx_ring_size = port->rx_ring_size;
	u16 prev_tx_ring_size = port->tx_ring_size;
	int err;

	err = mvpp2_check_ringparam_valid(dev, ring);
	if (err)
		return err;

	if (!netif_running(dev)) {
		port->rx_ring_size = ring->rx_pending;
		port->tx_ring_size = ring->tx_pending;
		return 0;
	}

	/* The interface is running, so we have to force a
	 * reallocation of the queues
	 */
	mvpp2_stop_dev(port);
	mvpp2_cleanup_rxqs(port);
	mvpp2_cleanup_txqs(port);

	port->rx_ring_size = ring->rx_pending;
	port->tx_ring_size = ring->tx_pending;

	err = mvpp2_setup_rxqs(port);
	if (err) {
		/* Reallocate Rx queues with the original ring size */
		port->rx_ring_size = prev_rx_ring_size;
		ring->rx_pending = prev_rx_ring_size;
		err = mvpp2_setup_rxqs(port);
		if (err)
			goto err_out;
	}
	err = mvpp2_setup_txqs(port);
	if (err) {
		/* Reallocate Tx queues with the original ring size */
		port->tx_ring_size = prev_tx_ring_size;
		ring->tx_pending = prev_tx_ring_size;
		err = mvpp2_setup_txqs(port);
		if (err)
			goto err_clean_rxqs;
	}

	mvpp2_start_dev(port);
	mvpp2_egress_enable(port);
	mvpp2_ingress_enable(port);

	return 0;

err_clean_rxqs:
	mvpp2_cleanup_rxqs(port);
err_out:
	netdev_err(dev, "failed to change ring parameters");
	return err;
}

/* Device ops */

static const struct net_device_ops mvpp2_netdev_ops = {
	.ndo_open		= mvpp2_open,
	.ndo_stop		= mvpp2_stop,
	.ndo_start_xmit		= mvpp2_tx,
	.ndo_set_rx_mode	= mvpp2_set_rx_mode,
	.ndo_set_mac_address	= mvpp2_set_mac_address,
	.ndo_change_mtu		= mvpp2_change_mtu,
	.ndo_get_stats64	= mvpp2_get_stats64,
	.ndo_do_ioctl		= mvpp2_ioctl,
};

static const struct ethtool_ops mvpp2_eth_tool_ops = {
	.nway_reset	= phy_ethtool_nway_reset,
	.get_link	= ethtool_op_get_link,
	.set_coalesce	= mvpp2_ethtool_set_coalesce,
	.get_coalesce	= mvpp2_ethtool_get_coalesce,
	.get_drvinfo	= mvpp2_ethtool_get_drvinfo,
	.get_ringparam	= mvpp2_ethtool_get_ringparam,
	.set_ringparam	= mvpp2_ethtool_set_ringparam,
	.get_link_ksettings = phy_ethtool_get_link_ksettings,
	.set_link_ksettings = phy_ethtool_set_link_ksettings,
};

/* Used for PPv2.1, or PPv2.2 with the old Device Tree binding that
 * had a single IRQ defined per-port.
 */
static int mvpp2_simple_queue_vectors_init(struct mvpp2_port *port,
					   struct device_node *port_node)
{
	struct mvpp2_queue_vector *v = &port->qvecs[0];

	v->first_rxq = 0;
	v->nrxqs = port->nrxqs;
	v->type = MVPP2_QUEUE_VECTOR_SHARED;
	v->sw_thread_id = 0;
	v->sw_thread_mask = *cpumask_bits(cpu_online_mask);
	v->port = port;
	v->irq = irq_of_parse_and_map(port_node, 0);
	if (v->irq <= 0)
		return -EINVAL;
	netif_napi_add(port->dev, &v->napi, mvpp2_poll,
		       NAPI_POLL_WEIGHT);

	port->nqvecs = 1;

	return 0;
}

static int mvpp2_multi_queue_vectors_init(struct mvpp2_port *port,
					  struct device_node *port_node)
{
	struct mvpp2_queue_vector *v;
	int i, ret;

	port->nqvecs = num_possible_cpus();
	if (queue_mode == MVPP2_QDIST_SINGLE_MODE)
		port->nqvecs += 1;

	for (i = 0; i < port->nqvecs; i++) {
		char irqname[16];

		v = port->qvecs + i;

		v->port = port;
		v->type = MVPP2_QUEUE_VECTOR_PRIVATE;
		v->sw_thread_id = i;
		v->sw_thread_mask = BIT(i);

		snprintf(irqname, sizeof(irqname), "tx-cpu%d", i);

		if (queue_mode == MVPP2_QDIST_MULTI_MODE) {
			v->first_rxq = i * MVPP2_DEFAULT_RXQ;
			v->nrxqs = MVPP2_DEFAULT_RXQ;
		} else if (queue_mode == MVPP2_QDIST_SINGLE_MODE &&
			   i == (port->nqvecs - 1)) {
			v->first_rxq = 0;
			v->nrxqs = port->nrxqs;
			v->type = MVPP2_QUEUE_VECTOR_SHARED;
			strncpy(irqname, "rx-shared", sizeof(irqname));
		}

		v->irq = of_irq_get_byname(port_node, irqname);
		if (v->irq <= 0) {
			ret = -EINVAL;
			goto err;
		}

		netif_napi_add(port->dev, &v->napi, mvpp2_poll,
			       NAPI_POLL_WEIGHT);
	}

	return 0;

err:
	for (i = 0; i < port->nqvecs; i++)
		irq_dispose_mapping(port->qvecs[i].irq);
	return ret;
}

static int mvpp2_queue_vectors_init(struct mvpp2_port *port,
				    struct device_node *port_node)
{
	if (port->has_tx_irqs)
		return mvpp2_multi_queue_vectors_init(port, port_node);
	else
		return mvpp2_simple_queue_vectors_init(port, port_node);
}

static void mvpp2_queue_vectors_deinit(struct mvpp2_port *port)
{
	int i;

	for (i = 0; i < port->nqvecs; i++)
		irq_dispose_mapping(port->qvecs[i].irq);
}

/* Configure Rx queue group interrupt for this port */
static void mvpp2_rx_irqs_setup(struct mvpp2_port *port)
{
	struct mvpp2 *priv = port->priv;
	u32 val;
	int i;

	if (priv->hw_version == MVPP21) {
		mvpp2_write(priv, MVPP21_ISR_RXQ_GROUP_REG(port->id),
			    port->nrxqs);
		return;
	}

	/* Handle the more complicated PPv2.2 case */
	for (i = 0; i < port->nqvecs; i++) {
		struct mvpp2_queue_vector *qv = port->qvecs + i;

		if (!qv->nrxqs)
			continue;

		val = qv->sw_thread_id;
		val |= port->id << MVPP22_ISR_RXQ_GROUP_INDEX_GROUP_OFFSET;
		mvpp2_write(priv, MVPP22_ISR_RXQ_GROUP_INDEX_REG, val);

		val = qv->first_rxq;
		val |= qv->nrxqs << MVPP22_ISR_RXQ_SUB_GROUP_SIZE_OFFSET;
		mvpp2_write(priv, MVPP22_ISR_RXQ_SUB_GROUP_CONFIG_REG, val);
	}
}

/* Initialize port HW */
static int mvpp2_port_init(struct mvpp2_port *port)
{
	struct device *dev = port->dev->dev.parent;
	struct mvpp2 *priv = port->priv;
	struct mvpp2_txq_pcpu *txq_pcpu;
	int queue, cpu, err;

	/* Checks for hardware constraints */
	if (port->first_rxq + port->nrxqs >
	    MVPP2_MAX_PORTS * priv->max_port_rxqs)
		return -EINVAL;

	if (port->nrxqs % 4 || (port->nrxqs > priv->max_port_rxqs) ||
	    (port->ntxqs > MVPP2_MAX_TXQ))
		return -EINVAL;

	/* Disable port */
	mvpp2_egress_disable(port);
	mvpp2_port_disable(port);

	port->tx_time_coal = MVPP2_TXDONE_COAL_USEC;

	port->txqs = devm_kcalloc(dev, port->ntxqs, sizeof(*port->txqs),
				  GFP_KERNEL);
	if (!port->txqs)
		return -ENOMEM;

	/* Associate physical Tx queues to this port and initialize.
	 * The mapping is predefined.
	 */
	for (queue = 0; queue < port->ntxqs; queue++) {
		int queue_phy_id = mvpp2_txq_phys(port->id, queue);
		struct mvpp2_tx_queue *txq;

		txq = devm_kzalloc(dev, sizeof(*txq), GFP_KERNEL);
		if (!txq) {
			err = -ENOMEM;
			goto err_free_percpu;
		}

		txq->pcpu = alloc_percpu(struct mvpp2_txq_pcpu);
		if (!txq->pcpu) {
			err = -ENOMEM;
			goto err_free_percpu;
		}

		txq->id = queue_phy_id;
		txq->log_id = queue;
		txq->done_pkts_coal = MVPP2_TXDONE_COAL_PKTS_THRESH;
		for_each_present_cpu(cpu) {
			txq_pcpu = per_cpu_ptr(txq->pcpu, cpu);
			txq_pcpu->cpu = cpu;
		}

		port->txqs[queue] = txq;
	}

	port->rxqs = devm_kcalloc(dev, port->nrxqs, sizeof(*port->rxqs),
				  GFP_KERNEL);
	if (!port->rxqs) {
		err = -ENOMEM;
		goto err_free_percpu;
	}

	/* Allocate and initialize Rx queue for this port */
	for (queue = 0; queue < port->nrxqs; queue++) {
		struct mvpp2_rx_queue *rxq;

		/* Map physical Rx queue to port's logical Rx queue */
		rxq = devm_kzalloc(dev, sizeof(*rxq), GFP_KERNEL);
		if (!rxq) {
			err = -ENOMEM;
			goto err_free_percpu;
		}
		/* Map this Rx queue to a physical queue */
		rxq->id = port->first_rxq + queue;
		rxq->port = port->id;
		rxq->logic_rxq = queue;

		port->rxqs[queue] = rxq;
	}

	mvpp2_rx_irqs_setup(port);

	/* Create Rx descriptor rings */
	for (queue = 0; queue < port->nrxqs; queue++) {
		struct mvpp2_rx_queue *rxq = port->rxqs[queue];

		rxq->size = port->rx_ring_size;
		rxq->pkts_coal = MVPP2_RX_COAL_PKTS;
		rxq->time_coal = MVPP2_RX_COAL_USEC;
	}

	mvpp2_ingress_disable(port);

	/* Port default configuration */
	mvpp2_defaults_set(port);

	/* Port's classifier configuration */
	mvpp2_cls_oversize_rxq_set(port);
	mvpp2_cls_port_config(port);

	/* Provide an initial Rx packet size */
	port->pkt_size = MVPP2_RX_PKT_SIZE(port->dev->mtu);

	/* Initialize pools for swf */
	err = mvpp2_swf_bm_pool_init(port);
	if (err)
		goto err_free_percpu;

	return 0;

err_free_percpu:
	for (queue = 0; queue < port->ntxqs; queue++) {
		if (!port->txqs[queue])
			continue;
		free_percpu(port->txqs[queue]->pcpu);
	}
	return err;
}

/* Checks if the port DT description has the TX interrupts
 * described. On PPv2.1, there are no such interrupts. On PPv2.2,
 * there are available, but we need to keep support for old DTs.
 */
static bool mvpp2_port_has_tx_irqs(struct mvpp2 *priv,
				   struct device_node *port_node)
{
	char *irqs[5] = { "rx-shared", "tx-cpu0", "tx-cpu1",
			  "tx-cpu2", "tx-cpu3" };
	int ret, i;

	if (priv->hw_version == MVPP21)
		return false;

	for (i = 0; i < 5; i++) {
		ret = of_property_match_string(port_node, "interrupt-names",
					       irqs[i]);
		if (ret < 0)
			return false;
	}

	return true;
}

/* Ports initialization */
static int mvpp2_port_probe(struct platform_device *pdev,
			    struct device_node *port_node,
			    struct mvpp2 *priv)
{
	struct device_node *phy_node;
	struct mvpp2_port *port;
	struct mvpp2_port_pcpu *port_pcpu;
	struct net_device *dev;
	struct resource *res;
	const char *dt_mac_addr;
	const char *mac_from;
	char hw_mac_addr[ETH_ALEN];
	unsigned int ntxqs, nrxqs;
	bool has_tx_irqs;
	u32 id;
	int features;
	int phy_mode;
	int err, i, cpu;

	has_tx_irqs = mvpp2_port_has_tx_irqs(priv, port_node);

	if (!has_tx_irqs)
		queue_mode = MVPP2_QDIST_SINGLE_MODE;

	ntxqs = MVPP2_MAX_TXQ;
	if (priv->hw_version == MVPP22 && queue_mode == MVPP2_QDIST_MULTI_MODE)
		nrxqs = MVPP2_DEFAULT_RXQ * num_possible_cpus();
	else
		nrxqs = MVPP2_DEFAULT_RXQ;

	dev = alloc_etherdev_mqs(sizeof(*port), ntxqs, nrxqs);
	if (!dev)
		return -ENOMEM;

	phy_node = of_parse_phandle(port_node, "phy", 0);
	if (!phy_node) {
		dev_err(&pdev->dev, "missing phy\n");
		err = -ENODEV;
		goto err_free_netdev;
	}

	phy_mode = of_get_phy_mode(port_node);
	if (phy_mode < 0) {
		dev_err(&pdev->dev, "incorrect phy mode\n");
		err = phy_mode;
		goto err_free_netdev;
	}

	if (of_property_read_u32(port_node, "port-id", &id)) {
		err = -EINVAL;
		dev_err(&pdev->dev, "missing port-id value\n");
		goto err_free_netdev;
	}

	dev->tx_queue_len = MVPP2_MAX_TXD;
	dev->watchdog_timeo = 5 * HZ;
	dev->netdev_ops = &mvpp2_netdev_ops;
	dev->ethtool_ops = &mvpp2_eth_tool_ops;

	port = netdev_priv(dev);
	port->dev = dev;
	port->ntxqs = ntxqs;
	port->nrxqs = nrxqs;
	port->priv = priv;
	port->has_tx_irqs = has_tx_irqs;

	err = mvpp2_queue_vectors_init(port, port_node);
	if (err)
		goto err_free_netdev;

	if (of_property_read_bool(port_node, "marvell,loopback"))
		port->flags |= MVPP2_F_LOOPBACK;

	port->id = id;
	if (priv->hw_version == MVPP21)
		port->first_rxq = port->id * port->nrxqs;
	else
		port->first_rxq = port->id * priv->max_port_rxqs;

	port->phy_node = phy_node;
	port->phy_interface = phy_mode;

	if (priv->hw_version == MVPP21) {
		res = platform_get_resource(pdev, IORESOURCE_MEM, 2 + id);
		port->base = devm_ioremap_resource(&pdev->dev, res);
		if (IS_ERR(port->base)) {
			err = PTR_ERR(port->base);
			goto err_deinit_qvecs;
		}
	} else {
		if (of_property_read_u32(port_node, "gop-port-id",
					 &port->gop_id)) {
			err = -EINVAL;
			dev_err(&pdev->dev, "missing gop-port-id value\n");
			goto err_deinit_qvecs;
		}

		port->base = priv->iface_base + MVPP22_GMAC_BASE(port->gop_id);
	}

	/* Alloc per-cpu stats */
	port->stats = netdev_alloc_pcpu_stats(struct mvpp2_pcpu_stats);
	if (!port->stats) {
		err = -ENOMEM;
		goto err_deinit_qvecs;
	}

	dt_mac_addr = of_get_mac_address(port_node);
	if (dt_mac_addr && is_valid_ether_addr(dt_mac_addr)) {
		mac_from = "device tree";
		ether_addr_copy(dev->dev_addr, dt_mac_addr);
	} else {
		if (priv->hw_version == MVPP21)
			mvpp21_get_mac_address(port, hw_mac_addr);
		if (is_valid_ether_addr(hw_mac_addr)) {
			mac_from = "hardware";
			ether_addr_copy(dev->dev_addr, hw_mac_addr);
		} else {
			mac_from = "random";
			eth_hw_addr_random(dev);
		}
	}

	port->tx_ring_size = MVPP2_MAX_TXD;
	port->rx_ring_size = MVPP2_MAX_RXD;
	SET_NETDEV_DEV(dev, &pdev->dev);

	err = mvpp2_port_init(port);
	if (err < 0) {
		dev_err(&pdev->dev, "failed to init port %d\n", id);
		goto err_free_stats;
	}

	mvpp2_port_mii_set(port);
	mvpp2_port_periodic_xon_disable(port);

	if (priv->hw_version == MVPP21)
		mvpp2_port_fc_adv_enable(port);

	mvpp2_port_reset(port);

	port->pcpu = alloc_percpu(struct mvpp2_port_pcpu);
	if (!port->pcpu) {
		err = -ENOMEM;
		goto err_free_txq_pcpu;
	}

	if (!port->has_tx_irqs) {
		for_each_present_cpu(cpu) {
			port_pcpu = per_cpu_ptr(port->pcpu, cpu);

			hrtimer_init(&port_pcpu->tx_done_timer, CLOCK_MONOTONIC,
				     HRTIMER_MODE_REL_PINNED);
			port_pcpu->tx_done_timer.function = mvpp2_hr_timer_cb;
			port_pcpu->timer_scheduled = false;

			tasklet_init(&port_pcpu->tx_done_tasklet,
				     mvpp2_tx_proc_cb,
				     (unsigned long)dev);
		}
	}

	features = NETIF_F_SG | NETIF_F_IP_CSUM;
	dev->features = features | NETIF_F_RXCSUM;
	dev->hw_features |= features | NETIF_F_RXCSUM | NETIF_F_GRO;
	dev->vlan_features |= features;

	/* MTU range: 68 - 9676 */
	dev->min_mtu = ETH_MIN_MTU;
	/* 9676 == 9700 - 20 and rounding to 8 */
	dev->max_mtu = 9676;

	err = register_netdev(dev);
	if (err < 0) {
		dev_err(&pdev->dev, "failed to register netdev\n");
		goto err_free_port_pcpu;
	}
	netdev_info(dev, "Using %s mac address %pM\n", mac_from, dev->dev_addr);

	priv->port_list[id] = port;
	return 0;

err_free_port_pcpu:
	free_percpu(port->pcpu);
err_free_txq_pcpu:
	for (i = 0; i < port->ntxqs; i++)
		free_percpu(port->txqs[i]->pcpu);
err_free_stats:
	free_percpu(port->stats);
err_deinit_qvecs:
	mvpp2_queue_vectors_deinit(port);
err_free_netdev:
	of_node_put(phy_node);
	free_netdev(dev);
	return err;
}

/* Ports removal routine */
static void mvpp2_port_remove(struct mvpp2_port *port)
{
	int i;

	unregister_netdev(port->dev);
	of_node_put(port->phy_node);
	free_percpu(port->pcpu);
	free_percpu(port->stats);
	for (i = 0; i < port->ntxqs; i++)
		free_percpu(port->txqs[i]->pcpu);
	mvpp2_queue_vectors_deinit(port);
	free_netdev(port->dev);
}

/* Initialize decoding windows */
static void mvpp2_conf_mbus_windows(const struct mbus_dram_target_info *dram,
				    struct mvpp2 *priv)
{
	u32 win_enable;
	int i;

	for (i = 0; i < 6; i++) {
		mvpp2_write(priv, MVPP2_WIN_BASE(i), 0);
		mvpp2_write(priv, MVPP2_WIN_SIZE(i), 0);

		if (i < 4)
			mvpp2_write(priv, MVPP2_WIN_REMAP(i), 0);
	}

	win_enable = 0;

	for (i = 0; i < dram->num_cs; i++) {
		const struct mbus_dram_window *cs = dram->cs + i;

		mvpp2_write(priv, MVPP2_WIN_BASE(i),
			    (cs->base & 0xffff0000) | (cs->mbus_attr << 8) |
			    dram->mbus_dram_target_id);

		mvpp2_write(priv, MVPP2_WIN_SIZE(i),
			    (cs->size - 1) & 0xffff0000);

		win_enable |= (1 << i);
	}

	mvpp2_write(priv, MVPP2_BASE_ADDR_ENABLE, win_enable);
}

/* Initialize Rx FIFO's */
static void mvpp2_rx_fifo_init(struct mvpp2 *priv)
{
	int port;

	for (port = 0; port < MVPP2_MAX_PORTS; port++) {
		mvpp2_write(priv, MVPP2_RX_DATA_FIFO_SIZE_REG(port),
			    MVPP2_RX_FIFO_PORT_DATA_SIZE);
		mvpp2_write(priv, MVPP2_RX_ATTR_FIFO_SIZE_REG(port),
			    MVPP2_RX_FIFO_PORT_ATTR_SIZE);
	}

	mvpp2_write(priv, MVPP2_RX_MIN_PKT_SIZE_REG,
		    MVPP2_RX_FIFO_PORT_MIN_PKT);
	mvpp2_write(priv, MVPP2_RX_FIFO_INIT_REG, 0x1);
}

static void mvpp2_axi_init(struct mvpp2 *priv)
{
	u32 val, rdval, wrval;

	mvpp2_write(priv, MVPP22_BM_ADDR_HIGH_RLS_REG, 0x0);

	/* AXI Bridge Configuration */

	rdval = MVPP22_AXI_CODE_CACHE_RD_CACHE
		<< MVPP22_AXI_ATTR_CACHE_OFFS;
	rdval |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
		<< MVPP22_AXI_ATTR_DOMAIN_OFFS;

	wrval = MVPP22_AXI_CODE_CACHE_WR_CACHE
		<< MVPP22_AXI_ATTR_CACHE_OFFS;
	wrval |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
		<< MVPP22_AXI_ATTR_DOMAIN_OFFS;

	/* BM */
	mvpp2_write(priv, MVPP22_AXI_BM_WR_ATTR_REG, wrval);
	mvpp2_write(priv, MVPP22_AXI_BM_RD_ATTR_REG, rdval);

	/* Descriptors */
	mvpp2_write(priv, MVPP22_AXI_AGGRQ_DESCR_RD_ATTR_REG, rdval);
	mvpp2_write(priv, MVPP22_AXI_TXQ_DESCR_WR_ATTR_REG, wrval);
	mvpp2_write(priv, MVPP22_AXI_TXQ_DESCR_RD_ATTR_REG, rdval);
	mvpp2_write(priv, MVPP22_AXI_RXQ_DESCR_WR_ATTR_REG, wrval);

	/* Buffer Data */
	mvpp2_write(priv, MVPP22_AXI_TX_DATA_RD_ATTR_REG, rdval);
	mvpp2_write(priv, MVPP22_AXI_RX_DATA_WR_ATTR_REG, wrval);

	val = MVPP22_AXI_CODE_CACHE_NON_CACHE
		<< MVPP22_AXI_CODE_CACHE_OFFS;
	val |= MVPP22_AXI_CODE_DOMAIN_SYSTEM
		<< MVPP22_AXI_CODE_DOMAIN_OFFS;
	mvpp2_write(priv, MVPP22_AXI_RD_NORMAL_CODE_REG, val);
	mvpp2_write(priv, MVPP22_AXI_WR_NORMAL_CODE_REG, val);

	val = MVPP22_AXI_CODE_CACHE_RD_CACHE
		<< MVPP22_AXI_CODE_CACHE_OFFS;
	val |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
		<< MVPP22_AXI_CODE_DOMAIN_OFFS;

	mvpp2_write(priv, MVPP22_AXI_RD_SNOOP_CODE_REG, val);

	val = MVPP22_AXI_CODE_CACHE_WR_CACHE
		<< MVPP22_AXI_CODE_CACHE_OFFS;
	val |= MVPP22_AXI_CODE_DOMAIN_OUTER_DOM
		<< MVPP22_AXI_CODE_DOMAIN_OFFS;

	mvpp2_write(priv, MVPP22_AXI_WR_SNOOP_CODE_REG, val);
}

/* Initialize network controller common part HW */
static int mvpp2_init(struct platform_device *pdev, struct mvpp2 *priv)
{
	const struct mbus_dram_target_info *dram_target_info;
	int err, i;
	u32 val;

	/* MBUS windows configuration */
	dram_target_info = mv_mbus_dram_info();
	if (dram_target_info)
		mvpp2_conf_mbus_windows(dram_target_info, priv);

	if (priv->hw_version == MVPP22)
		mvpp2_axi_init(priv);

	/* Disable HW PHY polling */
	if (priv->hw_version == MVPP21) {
		val = readl(priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
		val |= MVPP2_PHY_AN_STOP_SMI0_MASK;
		writel(val, priv->lms_base + MVPP2_PHY_AN_CFG0_REG);
	} else {
		val = readl(priv->iface_base + MVPP22_SMI_MISC_CFG_REG);
		val &= ~MVPP22_SMI_POLLING_EN;
		writel(val, priv->iface_base + MVPP22_SMI_MISC_CFG_REG);
	}

	/* Allocate and initialize aggregated TXQs */
	priv->aggr_txqs = devm_kcalloc(&pdev->dev, num_present_cpus(),
				       sizeof(*priv->aggr_txqs),
				       GFP_KERNEL);
	if (!priv->aggr_txqs)
		return -ENOMEM;

	for_each_present_cpu(i) {
		priv->aggr_txqs[i].id = i;
		priv->aggr_txqs[i].size = MVPP2_AGGR_TXQ_SIZE;
		err = mvpp2_aggr_txq_init(pdev, &priv->aggr_txqs[i],
					  MVPP2_AGGR_TXQ_SIZE, i, priv);
		if (err < 0)
			return err;
	}

	/* Rx Fifo Init */
	mvpp2_rx_fifo_init(priv);

	if (priv->hw_version == MVPP21)
		writel(MVPP2_EXT_GLOBAL_CTRL_DEFAULT,
		       priv->lms_base + MVPP2_MNG_EXTENDED_GLOBAL_CTRL_REG);

	/* Allow cache snoop when transmiting packets */
	mvpp2_write(priv, MVPP2_TX_SNOOP_REG, 0x1);

	/* Buffer Manager initialization */
	err = mvpp2_bm_init(pdev, priv);
	if (err < 0)
		return err;

	/* Parser default initialization */
	err = mvpp2_prs_default_init(pdev, priv);
	if (err < 0)
		return err;

	/* Classifier default initialization */
	mvpp2_cls_init(priv);

	return 0;
}

static int mvpp2_probe(struct platform_device *pdev)
{
	struct device_node *dn = pdev->dev.of_node;
	struct device_node *port_node;
	struct mvpp2 *priv;
	struct resource *res;
	void __iomem *base;
	int port_count, i;
	int err;

	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
	if (!priv)
		return -ENOMEM;

	priv->hw_version =
		(unsigned long)of_device_get_match_data(&pdev->dev);

	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
	base = devm_ioremap_resource(&pdev->dev, res);
	if (IS_ERR(base))
		return PTR_ERR(base);

	if (priv->hw_version == MVPP21) {
		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
		priv->lms_base = devm_ioremap_resource(&pdev->dev, res);
		if (IS_ERR(priv->lms_base))
			return PTR_ERR(priv->lms_base);
	} else {
		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
		priv->iface_base = devm_ioremap_resource(&pdev->dev, res);
		if (IS_ERR(priv->iface_base))
			return PTR_ERR(priv->iface_base);
	}

	for (i = 0; i < MVPP2_MAX_THREADS; i++) {
		u32 addr_space_sz;

		addr_space_sz = (priv->hw_version == MVPP21 ?
				 MVPP21_ADDR_SPACE_SZ : MVPP22_ADDR_SPACE_SZ);
		priv->swth_base[i] = base + i * addr_space_sz;
	}

	if (priv->hw_version == MVPP21)
		priv->max_port_rxqs = 8;
	else
		priv->max_port_rxqs = 32;

	priv->pp_clk = devm_clk_get(&pdev->dev, "pp_clk");
	if (IS_ERR(priv->pp_clk))
		return PTR_ERR(priv->pp_clk);
	err = clk_prepare_enable(priv->pp_clk);
	if (err < 0)
		return err;

	priv->gop_clk = devm_clk_get(&pdev->dev, "gop_clk");
	if (IS_ERR(priv->gop_clk)) {
		err = PTR_ERR(priv->gop_clk);
		goto err_pp_clk;
	}
	err = clk_prepare_enable(priv->gop_clk);
	if (err < 0)
		goto err_pp_clk;

	if (priv->hw_version == MVPP22) {
		priv->mg_clk = devm_clk_get(&pdev->dev, "mg_clk");
		if (IS_ERR(priv->mg_clk)) {
			err = PTR_ERR(priv->mg_clk);
			goto err_gop_clk;
		}

		err = clk_prepare_enable(priv->mg_clk);
		if (err < 0)
			goto err_gop_clk;
	}

	/* Get system's tclk rate */
	priv->tclk = clk_get_rate(priv->pp_clk);

	if (priv->hw_version == MVPP22) {
		err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(40));
		if (err)
			goto err_mg_clk;
		/* Sadly, the BM pools all share the same register to
		 * store the high 32 bits of their address. So they
		 * must all have the same high 32 bits, which forces
		 * us to restrict coherent memory to DMA_BIT_MASK(32).
		 */
		err = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
		if (err)
			goto err_mg_clk;
	}

	/* Initialize network controller */
	err = mvpp2_init(pdev, priv);
	if (err < 0) {
		dev_err(&pdev->dev, "failed to initialize controller\n");
		goto err_mg_clk;
	}

	port_count = of_get_available_child_count(dn);
	if (port_count == 0) {
		dev_err(&pdev->dev, "no ports enabled\n");
		err = -ENODEV;
		goto err_mg_clk;
	}

	priv->port_list = devm_kcalloc(&pdev->dev, port_count,
				       sizeof(*priv->port_list),
				       GFP_KERNEL);
	if (!priv->port_list) {
		err = -ENOMEM;
		goto err_mg_clk;
	}

	/* Initialize ports */
	for_each_available_child_of_node(dn, port_node) {
		err = mvpp2_port_probe(pdev, port_node, priv);
		if (err < 0)
			goto err_mg_clk;
	}

	platform_set_drvdata(pdev, priv);
	return 0;

err_mg_clk:
	if (priv->hw_version == MVPP22)
		clk_disable_unprepare(priv->mg_clk);
err_gop_clk:
	clk_disable_unprepare(priv->gop_clk);
err_pp_clk:
	clk_disable_unprepare(priv->pp_clk);
	return err;
}

static int mvpp2_remove(struct platform_device *pdev)
{
	struct mvpp2 *priv = platform_get_drvdata(pdev);
	struct device_node *dn = pdev->dev.of_node;
	struct device_node *port_node;
	int i = 0;

	for_each_available_child_of_node(dn, port_node) {
		if (priv->port_list[i])
			mvpp2_port_remove(priv->port_list[i]);
		i++;
	}

	for (i = 0; i < MVPP2_BM_POOLS_NUM; i++) {
		struct mvpp2_bm_pool *bm_pool = &priv->bm_pools[i];

		mvpp2_bm_pool_destroy(pdev, priv, bm_pool);
	}

	for_each_present_cpu(i) {
		struct mvpp2_tx_queue *aggr_txq = &priv->aggr_txqs[i];

		dma_free_coherent(&pdev->dev,
				  MVPP2_AGGR_TXQ_SIZE * MVPP2_DESC_ALIGNED_SIZE,
				  aggr_txq->descs,
				  aggr_txq->descs_dma);
	}

	clk_disable_unprepare(priv->mg_clk);
	clk_disable_unprepare(priv->pp_clk);
	clk_disable_unprepare(priv->gop_clk);

	return 0;
}

static const struct of_device_id mvpp2_match[] = {
	{
		.compatible = "marvell,armada-375-pp2",
		.data = (void *)MVPP21,
	},
	{
		.compatible = "marvell,armada-7k-pp22",
		.data = (void *)MVPP22,
	},
	{ }
};
MODULE_DEVICE_TABLE(of, mvpp2_match);

static struct platform_driver mvpp2_driver = {
	.probe = mvpp2_probe,
	.remove = mvpp2_remove,
	.driver = {
		.name = MVPP2_DRIVER_NAME,
		.of_match_table = mvpp2_match,
	},
};

module_platform_driver(mvpp2_driver);

MODULE_DESCRIPTION("Marvell PPv2 Ethernet Driver - www.marvell.com");
MODULE_AUTHOR("Marcin Wojtas <mw@semihalf.com>");
MODULE_LICENSE("GPL v2");