mtd: nand: add reworked Marvell NAND controller driver

Add marvell_nand driver which aims at replacing the existing pxa3xx_nand
driver.

The new driver intends to be easier to understand and follows the brand
new NAND framework rules by implementing hooks for every pattern the
controller might support and referencing them inside a parser object
that will be given to the core at each ->exec_op() call.

Raw accessors are implemented, useful to test/debug memory/filesystem
corruptions. Userspace binaries contained in the mtd-utils package may
now be used and their output trusted.

Most of the DT nodes using the old driver kept non-optimal timings from
the bootloader (even if there was some mechanisms to derive them if the
chip was ONFI compliant). The new default is to implement
->setup_data_interface() and follow the core's decision regarding the
chip.

Thanks to the improved timings, implementation of ONFI mode 5 support
(with EDO managed by adding a delay on data sampling), merging the
commands together and optimizing writes in the command registers, the
new driver may achieve faster throughputs in both directions.
Measurements show an improvement of about +23% read throughput and +24%
write throughput. These measurements have been done with an
Armada-385-DB-AP (4kiB NAND pages forced in 4-bit strength BCH ECC
correction) using the userspace tool 'flash_speed' from the MTD test
suite.

Besides these important topics, the new driver addresses several
unsolved known issues in the old driver which:
- did not work with ECC soft neither with ECC none ;
- relied on naked read/write (which is unchanged) while the NFCv1
  embedded in the pxa3xx platforms do not implement it, so several
  NAND commands did not actually ever work without any notice (like
  reading the ONFI PARAM_PAGE or SET/GET_FEATURES) ;
- wrote the OOB data correctly, but was not able to read it correctly
  past the first OOB data chunk ;
- did not retrieve ECC bytes ;
- used device tree bindings that did not allow more than one NAND chip,
  and did not allow to choose the correct chip select if not
  incrementing from 0. Plus, the Ready/Busy line used had to be 0.

Old device tree bindings are still supported but deprecated. A more
hierarchical view has to be used to keep the controller and the NAND
chip structures clearly separated both inside the device tree and also
in the driver code.

Signed-off-by: Miquel Raynal <miquel.raynal@free-electrons.com>
Tested-by: Sean Nyekjaer <sean.nyekjaer@prevas.dk>
Tested-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>

3 files changed, 2912 insertions, 0 deletions

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 859eb7790c46..e6b8c59f2c0d 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -315,6 +315,7 @@ config MTD_NAND_ATMEL
 
 config MTD_NAND_PXA3xx
 	tristate "NAND support on PXA3xx and Armada 370/XP"
+	depends on !MTD_NAND_MARVELL
 	depends on PXA3xx || ARCH_MMP || PLAT_ORION || ARCH_MVEBU
 	help
 
@@ -323,6 +324,18 @@ config MTD_NAND_PXA3xx
 	  platforms (XP, 370, 375, 38x, 39x) and 64-bit Armada
 	  platforms (7K, 8K) (NFCv2).
 
+config MTD_NAND_MARVELL
+	tristate "NAND controller support on Marvell boards"
+	depends on PXA3xx || ARCH_MMP || PLAT_ORION || ARCH_MVEBU || \
+		   COMPILE_TEST
+	depends on HAS_IOMEM
+	help
+	  This enables the NAND flash controller driver for Marvell boards,
+	  including:
+	  - PXA3xx processors (NFCv1)
+	  - 32-bit Armada platforms (XP, 37x, 38x, 39x) (NFCv2)
+	  - 64-bit Aramda platforms (7k, 8k) (NFCv2)
+
 config MTD_NAND_SLC_LPC32XX
 	tristate "NXP LPC32xx SLC Controller"
 	depends on ARCH_LPC32XX
diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
index 118a1349aad3..921634ba400c 100644
--- a/drivers/mtd/nand/Makefile
+++ b/drivers/mtd/nand/Makefile
@@ -32,6 +32,7 @@ obj-$(CONFIG_MTD_NAND_OMAP2) 		+= omap2_nand.o
 obj-$(CONFIG_MTD_NAND_OMAP_BCH_BUILD)	+= omap_elm.o
 obj-$(CONFIG_MTD_NAND_CM_X270)		+= cmx270_nand.o
 obj-$(CONFIG_MTD_NAND_PXA3xx)		+= pxa3xx_nand.o
+obj-$(CONFIG_MTD_NAND_MARVELL)		+= marvell_nand.o
 obj-$(CONFIG_MTD_NAND_TMIO)		+= tmio_nand.o
 obj-$(CONFIG_MTD_NAND_PLATFORM)		+= plat_nand.o
 obj-$(CONFIG_MTD_NAND_PASEMI)		+= pasemi_nand.o
diff --git a/drivers/mtd/nand/marvell_nand.c b/drivers/mtd/nand/marvell_nand.c
new file mode 100644
index 000000000000..b8fec6093b75
--- /dev/null
+++ b/drivers/mtd/nand/marvell_nand.c
@@ -0,0 +1,2898 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Marvell NAND flash controller driver
+ *
+ * Copyright (C) 2017 Marvell
+ * Author: Miquel RAYNAL <miquel.raynal@free-electrons.com>
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/clk.h>
+#include <linux/mtd/rawnand.h>
+#include <linux/of_platform.h>
+#include <linux/iopoll.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/mfd/syscon.h>
+#include <linux/regmap.h>
+#include <asm/unaligned.h>
+
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/dma/pxa-dma.h>
+#include <linux/platform_data/mtd-nand-pxa3xx.h>
+
+/* Data FIFO granularity, FIFO reads/writes must be a multiple of this length */
+#define FIFO_DEPTH		8
+#define FIFO_REP(x)		(x / sizeof(u32))
+#define BCH_SEQ_READS		(32 / FIFO_DEPTH)
+/* NFC does not support transfers of larger chunks at a time */
+#define MAX_CHUNK_SIZE		2112
+/* NFCv1 cannot read more that 7 bytes of ID */
+#define NFCV1_READID_LEN	7
+/* Polling is done at a pace of POLL_PERIOD us until POLL_TIMEOUT is reached */
+#define POLL_PERIOD		0
+#define POLL_TIMEOUT		100000
+/* Interrupt maximum wait period in ms */
+#define IRQ_TIMEOUT		1000
+/* Latency in clock cycles between SoC pins and NFC logic */
+#define MIN_RD_DEL_CNT		3
+/* Maximum number of contiguous address cycles */
+#define MAX_ADDRESS_CYC_NFCV1	5
+#define MAX_ADDRESS_CYC_NFCV2	7
+/* System control registers/bits to enable the NAND controller on some SoCs */
+#define GENCONF_SOC_DEVICE_MUX	0x208
+#define GENCONF_SOC_DEVICE_MUX_NFC_EN BIT(0)
+#define GENCONF_SOC_DEVICE_MUX_ECC_CLK_RST BIT(20)
+#define GENCONF_SOC_DEVICE_MUX_ECC_CORE_RST BIT(21)
+#define GENCONF_SOC_DEVICE_MUX_NFC_INT_EN BIT(25)
+#define GENCONF_CLK_GATING_CTRL	0x220
+#define GENCONF_CLK_GATING_CTRL_ND_GATE BIT(2)
+#define GENCONF_ND_CLK_CTRL	0x700
+#define GENCONF_ND_CLK_CTRL_EN	BIT(0)
+
+/* NAND controller data flash control register */
+#define NDCR			0x00
+#define NDCR_ALL_INT		GENMASK(11, 0)
+#define NDCR_CS1_CMDDM		BIT(7)
+#define NDCR_CS0_CMDDM		BIT(8)
+#define NDCR_RDYM		BIT(11)
+#define NDCR_ND_ARB_EN		BIT(12)
+#define NDCR_RA_START		BIT(15)
+#define NDCR_RD_ID_CNT(x)	(min_t(unsigned int, x, 0x7) << 16)
+#define NDCR_PAGE_SZ(x)		(x >= 2048 ? BIT(24) : 0)
+#define NDCR_DWIDTH_M		BIT(26)
+#define NDCR_DWIDTH_C		BIT(27)
+#define NDCR_ND_RUN		BIT(28)
+#define NDCR_DMA_EN		BIT(29)
+#define NDCR_ECC_EN		BIT(30)
+#define NDCR_SPARE_EN		BIT(31)
+#define NDCR_GENERIC_FIELDS_MASK (~(NDCR_RA_START | NDCR_PAGE_SZ(2048) | \
+				    NDCR_DWIDTH_M | NDCR_DWIDTH_C))
+
+/* NAND interface timing parameter 0 register */
+#define NDTR0			0x04
+#define NDTR0_TRP(x)		((min_t(unsigned int, x, 0xF) & 0x7) << 0)
+#define NDTR0_TRH(x)		(min_t(unsigned int, x, 0x7) << 3)
+#define NDTR0_ETRP(x)		((min_t(unsigned int, x, 0xF) & 0x8) << 3)
+#define NDTR0_SEL_NRE_EDGE	BIT(7)
+#define NDTR0_TWP(x)		(min_t(unsigned int, x, 0x7) << 8)
+#define NDTR0_TWH(x)		(min_t(unsigned int, x, 0x7) << 11)
+#define NDTR0_TCS(x)		(min_t(unsigned int, x, 0x7) << 16)
+#define NDTR0_TCH(x)		(min_t(unsigned int, x, 0x7) << 19)
+#define NDTR0_RD_CNT_DEL(x)	(min_t(unsigned int, x, 0xF) << 22)
+#define NDTR0_SELCNTR		BIT(26)
+#define NDTR0_TADL(x)		(min_t(unsigned int, x, 0x1F) << 27)
+
+/* NAND interface timing parameter 1 register */
+#define NDTR1			0x0C
+#define NDTR1_TAR(x)		(min_t(unsigned int, x, 0xF) << 0)
+#define NDTR1_TWHR(x)		(min_t(unsigned int, x, 0xF) << 4)
+#define NDTR1_TRHW(x)		(min_t(unsigned int, x / 16, 0x3) << 8)
+#define NDTR1_PRESCALE		BIT(14)
+#define NDTR1_WAIT_MODE		BIT(15)
+#define NDTR1_TR(x)		(min_t(unsigned int, x, 0xFFFF) << 16)
+
+/* NAND controller status register */
+#define NDSR			0x14
+#define NDSR_WRCMDREQ		BIT(0)
+#define NDSR_RDDREQ		BIT(1)
+#define NDSR_WRDREQ		BIT(2)
+#define NDSR_CORERR		BIT(3)
+#define NDSR_UNCERR		BIT(4)
+#define NDSR_CMDD(cs)		BIT(8 - cs)
+#define NDSR_RDY(rb)		BIT(11 + rb)
+#define NDSR_ERRCNT(x)		((x >> 16) & 0x1F)
+
+/* NAND ECC control register */
+#define NDECCCTRL		0x28
+#define NDECCCTRL_BCH_EN	BIT(0)
+
+/* NAND controller data buffer register */
+#define NDDB			0x40
+
+/* NAND controller command buffer 0 register */
+#define NDCB0			0x48
+#define NDCB0_CMD1(x)		((x & 0xFF) << 0)
+#define NDCB0_CMD2(x)		((x & 0xFF) << 8)
+#define NDCB0_ADDR_CYC(x)	((x & 0x7) << 16)
+#define NDCB0_ADDR_GET_NUM_CYC(x) (((x) >> 16) & 0x7)
+#define NDCB0_DBC		BIT(19)
+#define NDCB0_CMD_TYPE(x)	((x & 0x7) << 21)
+#define NDCB0_CSEL		BIT(24)
+#define NDCB0_RDY_BYP		BIT(27)
+#define NDCB0_LEN_OVRD		BIT(28)
+#define NDCB0_CMD_XTYPE(x)	((x & 0x7) << 29)
+
+/* NAND controller command buffer 1 register */
+#define NDCB1			0x4C
+#define NDCB1_COLS(x)		((x & 0xFFFF) << 0)
+#define NDCB1_ADDRS_PAGE(x)	(x << 16)
+
+/* NAND controller command buffer 2 register */
+#define NDCB2			0x50
+#define NDCB2_ADDR5_PAGE(x)	(((x >> 16) & 0xFF) << 0)
+#define NDCB2_ADDR5_CYC(x)	((x & 0xFF) << 0)
+
+/* NAND controller command buffer 3 register */
+#define NDCB3			0x54
+#define NDCB3_ADDR6_CYC(x)	((x & 0xFF) << 16)
+#define NDCB3_ADDR7_CYC(x)	((x & 0xFF) << 24)
+
+/* NAND controller command buffer 0 register 'type' and 'xtype' fields */
+#define TYPE_READ		0
+#define TYPE_WRITE		1
+#define TYPE_ERASE		2
+#define TYPE_READ_ID		3
+#define TYPE_STATUS		4
+#define TYPE_RESET		5
+#define TYPE_NAKED_CMD		6
+#define TYPE_NAKED_ADDR		7
+#define TYPE_MASK		7
+#define XTYPE_MONOLITHIC_RW	0
+#define XTYPE_LAST_NAKED_RW	1
+#define XTYPE_FINAL_COMMAND	3
+#define XTYPE_READ		4
+#define XTYPE_WRITE_DISPATCH	4
+#define XTYPE_NAKED_RW		5
+#define XTYPE_COMMAND_DISPATCH	6
+#define XTYPE_MASK		7
+
+/**
+ * Marvell ECC engine works differently than the others, in order to limit the
+ * size of the IP, hardware engineers chose to set a fixed strength at 16 bits
+ * per subpage, and depending on a the desired strength needed by the NAND chip,
+ * a particular layout mixing data/spare/ecc is defined, with a possible last
+ * chunk smaller that the others.
+ *
+ * @writesize:		Full page size on which the layout applies
+ * @chunk:		Desired ECC chunk size on which the layout applies
+ * @strength:		Desired ECC strength (per chunk size bytes) on which the
+ *			layout applies
+ * @nchunks:		Total number of chunks
+ * @full_chunk_cnt:	Number of full-sized chunks, which is the number of
+ *			repetitions of the pattern:
+ *			(data_bytes + spare_bytes + ecc_bytes).
+ * @data_bytes:		Number of data bytes per chunk
+ * @spare_bytes:	Number of spare bytes per chunk
+ * @ecc_bytes:		Number of ecc bytes per chunk
+ * @last_data_bytes:	Number of data bytes in the last chunk
+ * @last_spare_bytes:	Number of spare bytes in the last chunk
+ * @last_ecc_bytes:	Number of ecc bytes in the last chunk
+ */
+struct marvell_hw_ecc_layout {
+	/* Constraints */
+	int writesize;
+	int chunk;
+	int strength;
+	/* Corresponding layout */
+	int nchunks;
+	int full_chunk_cnt;
+	int data_bytes;
+	int spare_bytes;
+	int ecc_bytes;
+	int last_data_bytes;
+	int last_spare_bytes;
+	int last_ecc_bytes;
+};
+
+#define MARVELL_LAYOUT(ws, dc, ds, nc, fcc, db, sb, eb, ldb, lsb, leb)	\
+	{								\
+		.writesize = ws,					\
+		.chunk = dc,						\
+		.strength = ds,						\
+		.nchunks = nc,						\
+		.full_chunk_cnt = fcc,					\
+		.data_bytes = db,					\
+		.spare_bytes = sb,					\
+		.ecc_bytes = eb,					\
+		.last_data_bytes = ldb,					\
+		.last_spare_bytes = lsb,				\
+		.last_ecc_bytes = leb,					\
+	}
+
+/* Layouts explained in AN-379_Marvell_SoC_NFC_ECC */
+static const struct marvell_hw_ecc_layout marvell_nfc_layouts[] = {
+	MARVELL_LAYOUT(  512,   512,  1,  1,  1,  512,  8,  8,  0,  0,  0),
+	MARVELL_LAYOUT( 2048,   512,  1,  1,  1, 2048, 40, 24,  0,  0,  0),
+	MARVELL_LAYOUT( 2048,   512,  4,  1,  1, 2048, 32, 30,  0,  0,  0),
+	MARVELL_LAYOUT( 4096,   512,  4,  2,  2, 2048, 32, 30,  0,  0,  0),
+	MARVELL_LAYOUT( 4096,   512,  8,  5,  4, 1024,  0, 30,  0, 64, 30),
+};
+
+/**
+ * The Nand Flash Controller has up to 4 CE and 2 RB pins. The CE selection
+ * is made by a field in NDCB0 register, and in another field in NDCB2 register.
+ * The datasheet describes the logic with an error: ADDR5 field is once
+ * declared at the beginning of NDCB2, and another time at its end. Because the
+ * ADDR5 field of NDCB2 may be used by other bytes, it would be more logical
+ * to use the last bit of this field instead of the first ones.
+ *
+ * @cs:			Wanted CE lane.
+ * @ndcb0_csel:		Value of the NDCB0 register with or without the flag
+ *			selecting the wanted CE lane. This is set once when
+ *			the Device Tree is probed.
+ * @rb:			Ready/Busy pin for the flash chip
+ */
+struct marvell_nand_chip_sel {
+	unsigned int cs;
+	u32 ndcb0_csel;
+	unsigned int rb;
+};
+
+/**
+ * NAND chip structure: stores NAND chip device related information
+ *
+ * @chip:		Base NAND chip structure
+ * @node:		Used to store NAND chips into a list
+ * @layout		NAND layout when using hardware ECC
+ * @ndcr:		Controller register value for this NAND chip
+ * @ndtr0:		Timing registers 0 value for this NAND chip
+ * @ndtr1:		Timing registers 1 value for this NAND chip
+ * @selected_die:	Current active CS
+ * @nsels:		Number of CS lines required by the NAND chip
+ * @sels:		Array of CS lines descriptions
+ */
+struct marvell_nand_chip {
+	struct nand_chip chip;
+	struct list_head node;
+	const struct marvell_hw_ecc_layout *layout;
+	u32 ndcr;
+	u32 ndtr0;
+	u32 ndtr1;
+	int addr_cyc;
+	int selected_die;
+	unsigned int nsels;
+	struct marvell_nand_chip_sel sels[0];
+};
+
+static inline struct marvell_nand_chip *to_marvell_nand(struct nand_chip *chip)
+{
+	return container_of(chip, struct marvell_nand_chip, chip);
+}
+
+static inline struct marvell_nand_chip_sel *to_nand_sel(struct marvell_nand_chip
+							*nand)
+{
+	return &nand->sels[nand->selected_die];
+}
+
+/**
+ * NAND controller capabilities for distinction between compatible strings
+ *
+ * @max_cs_nb:		Number of Chip Select lines available
+ * @max_rb_nb:		Number of Ready/Busy lines available
+ * @need_system_controller: Indicates if the SoC needs to have access to the
+ *                      system controller (ie. to enable the NAND controller)
+ * @legacy_of_bindings:	Indicates if DT parsing must be done using the old
+ *			fashion way
+ * @is_nfcv2:		NFCv2 has numerous enhancements compared to NFCv1, ie.
+ *			BCH error detection and correction algorithm,
+ *			NDCB3 register has been added
+ * @use_dma:		Use dma for data transfers
+ */
+struct marvell_nfc_caps {
+	unsigned int max_cs_nb;
+	unsigned int max_rb_nb;
+	bool need_system_controller;
+	bool legacy_of_bindings;
+	bool is_nfcv2;
+	bool use_dma;
+};
+
+/**
+ * NAND controller structure: stores Marvell NAND controller information
+ *
+ * @controller:		Base controller structure
+ * @dev:		Parent device (used to print error messages)
+ * @regs:		NAND controller registers
+ * @ecc_clk:		ECC block clock, two times the NAND controller clock
+ * @complete:		Completion object to wait for NAND controller events
+ * @assigned_cs:	Bitmask describing already assigned CS lines
+ * @chips:		List containing all the NAND chips attached to
+ *			this NAND controller
+ * @caps:		NAND controller capabilities for each compatible string
+ * @dma_chan:		DMA channel (NFCv1 only)
+ * @dma_buf:		32-bit aligned buffer for DMA transfers (NFCv1 only)
+ */
+struct marvell_nfc {
+	struct nand_hw_control controller;
+	struct device *dev;
+	void __iomem *regs;
+	struct clk *ecc_clk;
+	struct completion complete;
+	unsigned long assigned_cs;
+	struct list_head chips;
+	struct nand_chip *selected_chip;
+	const struct marvell_nfc_caps *caps;
+
+	/* DMA (NFCv1 only) */
+	bool use_dma;
+	struct dma_chan *dma_chan;
+	u8 *dma_buf;
+};
+
+static inline struct marvell_nfc *to_marvell_nfc(struct nand_hw_control *ctrl)
+{
+	return container_of(ctrl, struct marvell_nfc, controller);
+}
+
+/**
+ * NAND controller timings expressed in NAND Controller clock cycles
+ *
+ * @tRP:		ND_nRE pulse width
+ * @tRH:		ND_nRE high duration
+ * @tWP:		ND_nWE pulse time
+ * @tWH:		ND_nWE high duration
+ * @tCS:		Enable signal setup time
+ * @tCH:		Enable signal hold time
+ * @tADL:		Address to write data delay
+ * @tAR:		ND_ALE low to ND_nRE low delay
+ * @tWHR:		ND_nWE high to ND_nRE low for status read
+ * @tRHW:		ND_nRE high duration, read to write delay
+ * @tR:			ND_nWE high to ND_nRE low for read
+ */
+struct marvell_nfc_timings {
+	/* NDTR0 fields */
+	unsigned int tRP;
+	unsigned int tRH;
+	unsigned int tWP;
+	unsigned int tWH;
+	unsigned int tCS;
+	unsigned int tCH;
+	unsigned int tADL;
+	/* NDTR1 fields */
+	unsigned int tAR;
+	unsigned int tWHR;
+	unsigned int tRHW;
+	unsigned int tR;
+};
+
+/**
+ * Derives a duration in numbers of clock cycles.
+ *
+ * @ps: Duration in pico-seconds
+ * @period_ns:  Clock period in nano-seconds
+ *
+ * Convert the duration in nano-seconds, then divide by the period and
+ * return the number of clock periods.
+ */
+#define TO_CYCLES(ps, period_ns) (DIV_ROUND_UP(ps / 1000, period_ns))
+
+/**
+ * NAND driver structure filled during the parsing of the ->exec_op() subop
+ * subset of instructions.
+ *
+ * @ndcb:		Array of values written to NDCBx registers
+ * @cle_ale_delay_ns:	Optional delay after the last CMD or ADDR cycle
+ * @rdy_timeout_ms:	Timeout for waits on Ready/Busy pin
+ * @rdy_delay_ns:	Optional delay after waiting for the RB pin
+ * @data_delay_ns:	Optional delay after the data xfer
+ * @data_instr_idx:	Index of the data instruction in the subop
+ * @data_instr:		Pointer to the data instruction in the subop
+ */
+struct marvell_nfc_op {
+	u32 ndcb[4];
+	unsigned int cle_ale_delay_ns;
+	unsigned int rdy_timeout_ms;
+	unsigned int rdy_delay_ns;
+	unsigned int data_delay_ns;
+	unsigned int data_instr_idx;
+	const struct nand_op_instr *data_instr;
+};
+
+/*
+ * Internal helper to conditionnally apply a delay (from the above structure,
+ * most of the time).
+ */
+static void cond_delay(unsigned int ns)
+{
+	if (!ns)
+		return;
+
+	if (ns < 10000)
+		ndelay(ns);
+	else
+		udelay(DIV_ROUND_UP(ns, 1000));
+}
+
+/*
+ * The controller has many flags that could generate interrupts, most of them
+ * are disabled and polling is used. For the very slow signals, using interrupts
+ * may relax the CPU charge.
+ */
+static void marvell_nfc_disable_int(struct marvell_nfc *nfc, u32 int_mask)
+{
+	u32 reg;
+
+	/* Writing 1 disables the interrupt */
+	reg = readl_relaxed(nfc->regs + NDCR);
+	writel_relaxed(reg | int_mask, nfc->regs + NDCR);
+}
+
+static void marvell_nfc_enable_int(struct marvell_nfc *nfc, u32 int_mask)
+{
+	u32 reg;
+
+	/* Writing 0 enables the interrupt */
+	reg = readl_relaxed(nfc->regs + NDCR);
+	writel_relaxed(reg & ~int_mask, nfc->regs + NDCR);
+}
+
+static void marvell_nfc_clear_int(struct marvell_nfc *nfc, u32 int_mask)
+{
+	writel_relaxed(int_mask, nfc->regs + NDSR);
+}
+
+static void marvell_nfc_force_byte_access(struct nand_chip *chip,
+					  bool force_8bit)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 ndcr;
+
+	/*
+	 * Callers of this function do not verify if the NAND is using a 16-bit
+	 * an 8-bit bus for normal operations, so we need to take care of that
+	 * here by leaving the configuration unchanged if the NAND does not have
+	 * the NAND_BUSWIDTH_16 flag set.
+	 */
+	if (!(chip->options & NAND_BUSWIDTH_16))
+		return;
+
+	ndcr = readl_relaxed(nfc->regs + NDCR);
+
+	if (force_8bit)
+		ndcr &= ~(NDCR_DWIDTH_M | NDCR_DWIDTH_C);
+	else
+		ndcr |= NDCR_DWIDTH_M | NDCR_DWIDTH_C;
+
+	writel_relaxed(ndcr, nfc->regs + NDCR);
+}
+
+static int marvell_nfc_wait_ndrun(struct nand_chip *chip)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 val;
+	int ret;
+
+	/*
+	 * The command is being processed, wait for the ND_RUN bit to be
+	 * cleared by the NFC. If not, we must clear it by hand.
+	 */
+	ret = readl_relaxed_poll_timeout(nfc->regs + NDCR, val,
+					 (val & NDCR_ND_RUN) == 0,
+					 POLL_PERIOD, POLL_TIMEOUT);
+	if (ret) {
+		dev_err(nfc->dev, "Timeout on NAND controller run mode\n");
+		writel_relaxed(readl(nfc->regs + NDCR) & ~NDCR_ND_RUN,
+			       nfc->regs + NDCR);
+		return ret;
+	}
+
+	return 0;
+}
+
+/*
+ * Any time a command has to be sent to the controller, the following sequence
+ * has to be followed:
+ * - call marvell_nfc_prepare_cmd()
+ *      -> activate the ND_RUN bit that will kind of 'start a job'
+ *      -> wait the signal indicating the NFC is waiting for a command
+ * - send the command (cmd and address cycles)
+ * - enventually send or receive the data
+ * - call marvell_nfc_end_cmd() with the corresponding flag
+ *      -> wait the flag to be triggered or cancel the job with a timeout
+ *
+ * The following helpers are here to factorize the code a bit so that
+ * specialized functions responsible for executing the actual NAND
+ * operations do not have to replicate the same code blocks.
+ */
+static int marvell_nfc_prepare_cmd(struct nand_chip *chip)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 ndcr, val;
+	int ret;
+
+	/* Poll ND_RUN and clear NDSR before issuing any command */
+	ret = marvell_nfc_wait_ndrun(chip);
+	if (ret) {
+		dev_err(nfc->dev, "Last operation did not suceed\n");
+		return ret;
+	}
+
+	ndcr = readl_relaxed(nfc->regs + NDCR);
+	writel_relaxed(readl(nfc->regs + NDSR), nfc->regs + NDSR);
+
+	/* Assert ND_RUN bit and wait the NFC to be ready */
+	writel_relaxed(ndcr | NDCR_ND_RUN, nfc->regs + NDCR);
+	ret = readl_relaxed_poll_timeout(nfc->regs + NDSR, val,
+					 val & NDSR_WRCMDREQ,
+					 POLL_PERIOD, POLL_TIMEOUT);
+	if (ret) {
+		dev_err(nfc->dev, "Timeout on WRCMDRE\n");
+		return -ETIMEDOUT;
+	}
+
+	/* Command may be written, clear WRCMDREQ status bit */
+	writel_relaxed(NDSR_WRCMDREQ, nfc->regs + NDSR);
+
+	return 0;
+}
+
+static void marvell_nfc_send_cmd(struct nand_chip *chip,
+				 struct marvell_nfc_op *nfc_op)
+{
+	struct marvell_nand_chip *marvell_nand = to_marvell_nand(chip);
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+
+	dev_dbg(nfc->dev, "\nNDCR:  0x%08x\n"
+		"NDCB0: 0x%08x\nNDCB1: 0x%08x\nNDCB2: 0x%08x\nNDCB3: 0x%08x\n",
+		(u32)readl_relaxed(nfc->regs + NDCR), nfc_op->ndcb[0],
+		nfc_op->ndcb[1], nfc_op->ndcb[2], nfc_op->ndcb[3]);
+
+	writel_relaxed(to_nand_sel(marvell_nand)->ndcb0_csel | nfc_op->ndcb[0],
+		       nfc->regs + NDCB0);
+	writel_relaxed(nfc_op->ndcb[1], nfc->regs + NDCB0);
+	writel(nfc_op->ndcb[2], nfc->regs + NDCB0);
+
+	/*
+	 * Write NDCB0 four times only if LEN_OVRD is set or if ADDR6 or ADDR7
+	 * fields are used (only available on NFCv2).
+	 */
+	if (nfc_op->ndcb[0] & NDCB0_LEN_OVRD ||
+	    NDCB0_ADDR_GET_NUM_CYC(nfc_op->ndcb[0]) >= 6) {
+		if (!WARN_ON_ONCE(!nfc->caps->is_nfcv2))
+			writel(nfc_op->ndcb[3], nfc->regs + NDCB0);
+	}
+}
+
+static int marvell_nfc_end_cmd(struct nand_chip *chip, int flag,
+			       const char *label)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 val;
+	int ret;
+
+	ret = readl_relaxed_poll_timeout(nfc->regs + NDSR, val,
+					 val & flag,
+					 POLL_PERIOD, POLL_TIMEOUT);
+
+	if (ret) {
+		dev_err(nfc->dev, "Timeout on %s (NDSR: 0x%08x)\n",
+			label, val);
+		if (nfc->dma_chan)
+			dmaengine_terminate_all(nfc->dma_chan);
+		return ret;
+	}
+
+	/*
+	 * DMA function uses this helper to poll on CMDD bits without wanting
+	 * them to be cleared.
+	 */
+	if (nfc->use_dma && (readl_relaxed(nfc->regs + NDCR) & NDCR_DMA_EN))
+		return 0;
+
+	writel_relaxed(flag, nfc->regs + NDSR);
+
+	return 0;
+}
+
+static int marvell_nfc_wait_cmdd(struct nand_chip *chip)
+{
+	struct marvell_nand_chip *marvell_nand = to_marvell_nand(chip);
+	int cs_flag = NDSR_CMDD(to_nand_sel(marvell_nand)->ndcb0_csel);
+
+	return marvell_nfc_end_cmd(chip, cs_flag, "CMDD");
+}
+
+static int marvell_nfc_wait_op(struct nand_chip *chip, unsigned int timeout_ms)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	int ret;
+
+	/* Timeout is expressed in ms */
+	if (!timeout_ms)
+		timeout_ms = IRQ_TIMEOUT;
+
+	init_completion(&nfc->complete);
+
+	marvell_nfc_enable_int(nfc, NDCR_RDYM);
+	ret = wait_for_completion_timeout(&nfc->complete,
+					  msecs_to_jiffies(timeout_ms));
+	marvell_nfc_disable_int(nfc, NDCR_RDYM);
+	marvell_nfc_clear_int(nfc, NDSR_RDY(0) | NDSR_RDY(1));
+	if (!ret) {
+		dev_err(nfc->dev, "Timeout waiting for RB signal\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+static void marvell_nfc_select_chip(struct mtd_info *mtd, int die_nr)
+{
+	struct nand_chip *chip = mtd_to_nand(mtd);
+	struct marvell_nand_chip *marvell_nand = to_marvell_nand(chip);
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 ndcr_generic;
+
+	if (chip == nfc->selected_chip && die_nr == marvell_nand->selected_die)
+		return;
+
+	if (die_nr < 0 || die_nr >= marvell_nand->nsels) {
+		nfc->selected_chip = NULL;
+		marvell_nand->selected_die = -1;
+		return;
+	}
+
+	/*
+	 * Do not change the timing registers when using the DT property
+	 * marvell,nand-keep-config; in that case ->ndtr0 and ->ndtr1 from the
+	 * marvell_nand structure are supposedly empty.
+	 */
+	writel_relaxed(marvell_nand->ndtr0, nfc->regs + NDTR0);
+	writel_relaxed(marvell_nand->ndtr1, nfc->regs + NDTR1);
+
+	/*
+	 * Reset the NDCR register to a clean state for this particular chip,
+	 * also clear ND_RUN bit.
+	 */
+	ndcr_generic = readl_relaxed(nfc->regs + NDCR) &
+		       NDCR_GENERIC_FIELDS_MASK & ~NDCR_ND_RUN;
+	writel_relaxed(ndcr_generic | marvell_nand->ndcr, nfc->regs + NDCR);
+
+	/* Also reset the interrupt status register */
+	marvell_nfc_clear_int(nfc, NDCR_ALL_INT);
+
+	nfc->selected_chip = chip;
+	marvell_nand->selected_die = die_nr;
+}
+
+static irqreturn_t marvell_nfc_isr(int irq, void *dev_id)
+{
+	struct marvell_nfc *nfc = dev_id;
+	u32 st = readl_relaxed(nfc->regs + NDSR);
+	u32 ien = (~readl_relaxed(nfc->regs + NDCR)) & NDCR_ALL_INT;
+
+	/*
+	 * RDY interrupt mask is one bit in NDCR while there are two status
+	 * bit in NDSR (RDY[cs0/cs2] and RDY[cs1/cs3]).
+	 */
+	if (st & NDSR_RDY(1))
+		st |= NDSR_RDY(0);
+
+	if (!(st & ien))
+		return IRQ_NONE;
+
+	marvell_nfc_disable_int(nfc, st & NDCR_ALL_INT);
+
+	if (!(st & (NDSR_RDDREQ | NDSR_WRDREQ | NDSR_WRCMDREQ)))
+		complete(&nfc->complete);
+
+	return IRQ_HANDLED;
+}
+
+/* HW ECC related functions */
+static void marvell_nfc_enable_hw_ecc(struct nand_chip *chip)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 ndcr = readl_relaxed(nfc->regs + NDCR);
+
+	if (!(ndcr & NDCR_ECC_EN)) {
+		writel_relaxed(ndcr | NDCR_ECC_EN, nfc->regs + NDCR);
+
+		/*
+		 * When enabling BCH, set threshold to 0 to always know the
+		 * number of corrected bitflips.
+		 */
+		if (chip->ecc.algo == NAND_ECC_BCH)
+			writel_relaxed(NDECCCTRL_BCH_EN, nfc->regs + NDECCCTRL);
+	}
+}
+
+static void marvell_nfc_disable_hw_ecc(struct nand_chip *chip)
+{
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	u32 ndcr = readl_relaxed(nfc->regs + NDCR);
+
+	if (ndcr & NDCR_ECC_EN) {
+		writel_relaxed(ndcr & ~NDCR_ECC_EN, nfc->regs + NDCR);
+		if (chip->ecc.algo == NAND_ECC_BCH)
+			writel_relaxed(0, nfc->regs + NDECCCTRL);
+	}
+}
+
+/* DMA related helpers */
+static void marvell_nfc_enable_dma(struct marvell_nfc *nfc)
+{
+	u32 reg;
+
+	reg = readl_relaxed(nfc->regs + NDCR);
+	writel_relaxed(reg | NDCR_DMA_EN, nfc->regs + NDCR);
+}
+
+static void marvell_nfc_disable_dma(struct marvell_nfc *nfc)
+{
+	u32 reg;
+
+	reg = readl_relaxed(nfc->regs + NDCR);
+	writel_relaxed(reg & ~NDCR_DMA_EN, nfc->regs + NDCR);
+}
+
+/* Read/write PIO/DMA accessors */
+static int marvell_nfc_xfer_data_dma(struct marvell_nfc *nfc,
+				     enum dma_data_direction direction,
+				     unsigned int len)
+{
+	unsigned int dma_len = min_t(int, ALIGN(len, 32), MAX_CHUNK_SIZE);
+	struct dma_async_tx_descriptor *tx;
+	struct scatterlist sg;
+	dma_cookie_t cookie;
+	int ret;
+
+	marvell_nfc_enable_dma(nfc);
+	/* Prepare the DMA transfer */
+	sg_init_one(&sg, nfc->dma_buf, dma_len);
+	dma_map_sg(nfc->dma_chan->device->dev, &sg, 1, direction);
+	tx = dmaengine_prep_slave_sg(nfc->dma_chan, &sg, 1,
+				     direction == DMA_FROM_DEVICE ?
+				     DMA_DEV_TO_MEM : DMA_MEM_TO_DEV,
+				     DMA_PREP_INTERRUPT);
+	if (!tx) {
+		dev_err(nfc->dev, "Could not prepare DMA S/G list\n");
+		return -ENXIO;
+	}
+
+	/* Do the task and wait for it to finish */
+	cookie = dmaengine_submit(tx);
+	ret = dma_submit_error(cookie);
+	if (ret)
+		return -EIO;
+
+	dma_async_issue_pending(nfc->dma_chan);
+	ret = marvell_nfc_wait_cmdd(nfc->selected_chip);
+	dma_unmap_sg(nfc->dma_chan->device->dev, &sg, 1, direction);
+	marvell_nfc_disable_dma(nfc);
+	if (ret) {
+		dev_err(nfc->dev, "Timeout waiting for DMA (status: %d)\n",
+			dmaengine_tx_status(nfc->dma_chan, cookie, NULL));
+		dmaengine_terminate_all(nfc->dma_chan);
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+static int marvell_nfc_xfer_data_in_pio(struct marvell_nfc *nfc, u8 *in,
+					unsigned int len)
+{
+	unsigned int last_len = len % FIFO_DEPTH;
+	unsigned int last_full_offset = round_down(len, FIFO_DEPTH);
+	int i;
+
+	for (i = 0; i < last_full_offset; i += FIFO_DEPTH)
+		ioread32_rep(nfc->regs + NDDB, in + i, FIFO_REP(FIFO_DEPTH));
+
+	if (last_len) {
+		u8 tmp_buf[FIFO_DEPTH];
+
+		ioread32_rep(nfc->regs + NDDB, tmp_buf, FIFO_REP(FIFO_DEPTH));
+		memcpy(in + last_full_offset, tmp_buf, last_len);
+	}
+
+	return 0;
+}
+
+static int marvell_nfc_xfer_data_out_pio(struct marvell_nfc *nfc, const u8 *out,
+					 unsigned int len)
+{
+	unsigned int last_len = len % FIFO_DEPTH;
+	unsigned int last_full_offset = round_down(len, FIFO_DEPTH);
+	int i;
+
+	for (i = 0; i < last_full_offset; i += FIFO_DEPTH)
+		iowrite32_rep(nfc->regs + NDDB, out + i, FIFO_REP(FIFO_DEPTH));
+
+	if (last_len) {
+		u8 tmp_buf[FIFO_DEPTH];
+
+		memcpy(tmp_buf, out + last_full_offset, last_len);
+		iowrite32_rep(nfc->regs + NDDB, tmp_buf, FIFO_REP(FIFO_DEPTH));
+	}
+
+	return 0;
+}
+
+static void marvell_nfc_check_empty_chunk(struct nand_chip *chip,
+					  u8 *data, int data_len,
+					  u8 *spare, int spare_len,
+					  u8 *ecc, int ecc_len,
+					  unsigned int *max_bitflips)
+{
+	struct mtd_info *mtd = nand_to_mtd(chip);
+	int bf;
+
+	/*
+	 * Blank pages (all 0xFF) that have not been written may be recognized
+	 * as bad if bitflips occur, so whenever an uncorrectable error occurs,
+	 * check if the entire page (with ECC bytes) is actually blank or not.
+	 */
+	if (!data)
+		data_len = 0;
+	if (!spare)
+		spare_len = 0;
+	if (!ecc)
+		ecc_len = 0;
+
+	bf = nand_check_erased_ecc_chunk(data, data_len, ecc, ecc_len,
+					 spare, spare_len, chip->ecc.strength);
+	if (bf < 0) {
+		mtd->ecc_stats.failed++;
+		return;
+	}
+
+	/* Update the stats and max_bitflips */
+	mtd->ecc_stats.corrected += bf;
+	*max_bitflips = max_t(unsigned int, *max_bitflips, bf);
+}
+
+/*
+ * Check a chunk is correct or not according to hardware ECC engine.
+ * mtd->ecc_stats.corrected is updated, as well as max_bitflips, however
+ * mtd->ecc_stats.failure is not, the function will instead return a non-zero
+ * value indicating that a check on the emptyness of the subpage must be
+ * performed before declaring the subpage corrupted.
+ */
+static int marvell_nfc_hw_ecc_correct(struct nand_chip *chip,
+				      unsigned int *max_bitflips)
+{
+	struct mtd_info *mtd = nand_to_mtd(chip);
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	int bf = 0;
+	u32 ndsr;
+
+	ndsr = readl_relaxed(nfc->regs + NDSR);
+
+	/* Check uncorrectable error flag */
+	if (ndsr & NDSR_UNCERR) {
+		writel_relaxed(ndsr, nfc->regs + NDSR);
+
+		/*
+		 * Do not increment ->ecc_stats.failed now, instead, return a
+		 * non-zero value to indicate that this chunk was apparently
+		 * bad, and it should be check to see if it empty or not. If
+		 * the chunk (with ECC bytes) is not declared empty, the calling
+		 * function must increment the failure count.
+		 */
+		return -EBADMSG;
+	}
+
+	/* Check correctable error flag */
+	if (ndsr & NDSR_CORERR) {
+		writel_relaxed(ndsr, nfc->regs + NDSR);
+
+		if (chip->ecc.algo == NAND_ECC_BCH)
+			bf = NDSR_ERRCNT(ndsr);
+		else
+			bf = 1;
+	}
+
+	/* Update the stats and max_bitflips */
+	mtd->ecc_stats.corrected += bf;
+	*max_bitflips = max_t(unsigned int, *max_bitflips, bf);
+
+	return 0;
+}
+
+/* Hamming read helpers */
+static int marvell_nfc_hw_ecc_hmg_do_read_page(struct nand_chip *chip,
+					       u8 *data_buf, u8 *oob_buf,
+					       bool raw, int page)
+{
+	struct marvell_nand_chip *marvell_nand = to_marvell_nand(chip);
+	struct marvell_nfc *nfc = to_marvell_nfc(chip->controller);
+	const struct marvell_hw_ecc_layout *lt = to_marvell_nand(chip)->layout;
+	struct marvell_nfc_op nfc_op = {
+		.ndcb[0] = NDCB0_CMD_TYPE(TYPE_READ) |
+			   NDCB0_ADDR_CYC(marvell_nand->addr_cyc) |
+			   NDCB0_DBC |
+			   NDCB0_CMD1(NAND_CMD_READ0) |
+			   NDCB0_CMD2(NAND_CMD_READSTART),
+		.ndcb[1] = NDCB1_ADDRS_PAGE(page),
+		.ndcb[2] = NDCB2_ADDR5_PAGE(page),
+	};
+	unsigned int oob_bytes = lt->spare_bytes + (raw ? lt->ecc_bytes : 0);
+	int ret;
+
+	/* NFCv2 needs more information about the operation being executed */
+	if (nfc->caps->is_nfcv2)
+		nfc_op.ndcb[0] |= NDCB0_CMD_XTYPE(XTYPE_MONOLITHIC_RW);
+
+	ret = marvell_nfc_prepare_cmd(chip);
+	if (ret)
+		return ret;
+
+	marvell_nfc_send_cmd(chip, &nfc_op);
+	ret = marvell_nfc_end_cmd(chip, NDSR_RDDREQ,
+				  "RDDREQ while draining FIFO (data/oob)");
+	if (ret)
+		return ret;
+
+	/*
+	 * Read the page then the OOB area. Unlike what is shown in current
+	 * documentation, spare bytes are protected by the ECC engine, and must
+	 * be at the beginning of the OOB area or running this driver on legacy
+	 * systems will prevent the discovery of the BBM/BBT.
+	 */
+	if (nfc->use_dma) {
+		marvell_nfc_xfer_data_dma(nfc, DMA_FROM_DEVICE,
+					  lt->data_bytes + oob_bytes);
+		memcpy(data_buf, nfc->dma_buf, lt->data_bytes);
+		memcpy(oob_buf, nfc->dma_buf + lt->data_bytes, oob_bytes);
+	} else {
+		marvell_nfc_xfer_data_in_pio(nfc, data_buf, lt->data_bytes);
+		marvell_nfc_xfer_data_in_pio(nfc, oob_buf, oob_bytes);
+	}
+
+	ret = marvell_nfc_wait_cmdd(chip);
+
+	return ret;
+}
+
+static int marvell_nfc_hw_ecc_hmg_read_page_raw(struct mtd_info *mtd,
+						struct nand_chip *chip, u8 *buf,
+						int oob_required, int page)
+{
+	return marvell_nfc_hw_ecc_hmg_do_r