Contributors: 29
Author Tokens Token Proportion Commits Commit Proportion
Sandeep Paulraj 1409 33.36% 1 1.28%
Brian Niebuhr 1200 28.41% 29 37.18%
Murali Karicheri 536 12.69% 5 6.41%
Sekhar Nori 258 6.11% 6 7.69%
Matt Porter 257 6.08% 1 1.28%
Franklin S Cooper Jr 130 3.08% 2 2.56%
Fabien Parent 110 2.60% 2 2.56%
Frode Isaksen 90 2.13% 4 5.13%
Peter Ujfalusi 80 1.89% 1 1.28%
Jingoo Han 34 0.80% 3 3.85%
Linus Walleij 16 0.38% 1 1.28%
Andrzej Hajda 15 0.36% 1 1.28%
David Lechner 13 0.31% 1 1.28%
Stephen Warren 12 0.28% 1 1.28%
Arvind Yadav 11 0.26% 2 2.56%
Michele Dionisio 8 0.19% 1 1.28%
Axel Lin 8 0.19% 2 2.56%
Luis de Bethencourt 6 0.14% 1 1.28%
Grant C. Likely 6 0.14% 1 1.28%
Kees Cook 5 0.12% 1 1.28%
Michael Williamson 3 0.07% 2 2.56%
Prakash Manjunathappa 3 0.07% 2 2.56%
Grygorii Strashko 3 0.07% 2 2.56%
Bartosz Golaszewski 2 0.05% 1 1.28%
Wolfram Sang 2 0.05% 1 1.28%
Sachin Kamat 2 0.05% 1 1.28%
Thomas Gleixner 2 0.05% 1 1.28%
Tejun Heo 2 0.05% 1 1.28%
Linus Torvalds 1 0.02% 1 1.28%
Total 4224 78 


// SPDX-License-Identifier: GPL-2.0-or-later
/*
 * Copyright (C) 2009 Texas Instruments.
 * Copyright (C) 2010 EF Johnson Technologies
 */

#include <linux/interrupt.h>
#include <linux/io.h>
#include <linux/gpio/consumer.h>
#include <linux/module.h>
#include <linux/delay.h>
#include <linux/platform_device.h>
#include <linux/err.h>
#include <linux/clk.h>
#include <linux/dmaengine.h>
#include <linux/dma-mapping.h>
#include <linux/of.h>
#include <linux/of_device.h>
#include <linux/spi/spi.h>
#include <linux/spi/spi_bitbang.h>
#include <linux/slab.h>

#include <linux/platform_data/spi-davinci.h>

#define CS_DEFAULT	0xFF

#define SPIFMT_PHASE_MASK	BIT(16)
#define SPIFMT_POLARITY_MASK	BIT(17)
#define SPIFMT_DISTIMER_MASK	BIT(18)
#define SPIFMT_SHIFTDIR_MASK	BIT(20)
#define SPIFMT_WAITENA_MASK	BIT(21)
#define SPIFMT_PARITYENA_MASK	BIT(22)
#define SPIFMT_ODD_PARITY_MASK	BIT(23)
#define SPIFMT_WDELAY_MASK	0x3f000000u
#define SPIFMT_WDELAY_SHIFT	24
#define SPIFMT_PRESCALE_SHIFT	8

/* SPIPC0 */
#define SPIPC0_DIFUN_MASK	BIT(11)		/* MISO */
#define SPIPC0_DOFUN_MASK	BIT(10)		/* MOSI */
#define SPIPC0_CLKFUN_MASK	BIT(9)		/* CLK */
#define SPIPC0_SPIENA_MASK	BIT(8)		/* nREADY */

#define SPIINT_MASKALL		0x0101035F
#define SPIINT_MASKINT		0x0000015F
#define SPI_INTLVL_1		0x000001FF
#define SPI_INTLVL_0		0x00000000

/* SPIDAT1 (upper 16 bit defines) */
#define SPIDAT1_CSHOLD_MASK	BIT(12)
#define SPIDAT1_WDEL		BIT(10)

/* SPIGCR1 */
#define SPIGCR1_CLKMOD_MASK	BIT(1)
#define SPIGCR1_MASTER_MASK     BIT(0)
#define SPIGCR1_POWERDOWN_MASK	BIT(8)
#define SPIGCR1_LOOPBACK_MASK	BIT(16)
#define SPIGCR1_SPIENA_MASK	BIT(24)

/* SPIBUF */
#define SPIBUF_TXFULL_MASK	BIT(29)
#define SPIBUF_RXEMPTY_MASK	BIT(31)

/* SPIDELAY */
#define SPIDELAY_C2TDELAY_SHIFT 24
#define SPIDELAY_C2TDELAY_MASK  (0xFF << SPIDELAY_C2TDELAY_SHIFT)
#define SPIDELAY_T2CDELAY_SHIFT 16
#define SPIDELAY_T2CDELAY_MASK  (0xFF << SPIDELAY_T2CDELAY_SHIFT)
#define SPIDELAY_T2EDELAY_SHIFT 8
#define SPIDELAY_T2EDELAY_MASK  (0xFF << SPIDELAY_T2EDELAY_SHIFT)
#define SPIDELAY_C2EDELAY_SHIFT 0
#define SPIDELAY_C2EDELAY_MASK  0xFF

/* Error Masks */
#define SPIFLG_DLEN_ERR_MASK		BIT(0)
#define SPIFLG_TIMEOUT_MASK		BIT(1)
#define SPIFLG_PARERR_MASK		BIT(2)
#define SPIFLG_DESYNC_MASK		BIT(3)
#define SPIFLG_BITERR_MASK		BIT(4)
#define SPIFLG_OVRRUN_MASK		BIT(6)
#define SPIFLG_BUF_INIT_ACTIVE_MASK	BIT(24)
#define SPIFLG_ERROR_MASK		(SPIFLG_DLEN_ERR_MASK \
				| SPIFLG_TIMEOUT_MASK | SPIFLG_PARERR_MASK \
				| SPIFLG_DESYNC_MASK | SPIFLG_BITERR_MASK \
				| SPIFLG_OVRRUN_MASK)

#define SPIINT_DMA_REQ_EN	BIT(16)

/* SPI Controller registers */
#define SPIGCR0		0x00
#define SPIGCR1		0x04
#define SPIINT		0x08
#define SPILVL		0x0c
#define SPIFLG		0x10
#define SPIPC0		0x14
#define SPIDAT1		0x3c
#define SPIBUF		0x40
#define SPIDELAY	0x48
#define SPIDEF		0x4c
#define SPIFMT0		0x50

#define DMA_MIN_BYTES	16

/* SPI Controller driver's private data. */
struct davinci_spi {
	struct spi_bitbang	bitbang;
	struct clk		*clk;

	u8			version;
	resource_size_t		pbase;
	void __iomem		*base;
	u32			irq;
	struct completion	done;

	const void		*tx;
	void			*rx;
	int			rcount;
	int			wcount;

	struct dma_chan		*dma_rx;
	struct dma_chan		*dma_tx;

	struct davinci_spi_platform_data pdata;

	void			(*get_rx)(u32 rx_data, struct davinci_spi *);
	u32			(*get_tx)(struct davinci_spi *);

	u8			*bytes_per_word;

	u8			prescaler_limit;
};

static struct davinci_spi_config davinci_spi_default_cfg;

static void davinci_spi_rx_buf_u8(u32 data, struct davinci_spi *dspi)
{
	if (dspi->rx) {
		u8 *rx = dspi->rx;
		*rx++ = (u8)data;
		dspi->rx = rx;
	}
}

static void davinci_spi_rx_buf_u16(u32 data, struct davinci_spi *dspi)
{
	if (dspi->rx) {
		u16 *rx = dspi->rx;
		*rx++ = (u16)data;
		dspi->rx = rx;
	}
}

static u32 davinci_spi_tx_buf_u8(struct davinci_spi *dspi)
{
	u32 data = 0;

	if (dspi->tx) {
		const u8 *tx = dspi->tx;

		data = *tx++;
		dspi->tx = tx;
	}
	return data;
}

static u32 davinci_spi_tx_buf_u16(struct davinci_spi *dspi)
{
	u32 data = 0;

	if (dspi->tx) {
		const u16 *tx = dspi->tx;

		data = *tx++;
		dspi->tx = tx;
	}
	return data;
}

static inline void set_io_bits(void __iomem *addr, u32 bits)
{
	u32 v = ioread32(addr);

	v |= bits;
	iowrite32(v, addr);
}

static inline void clear_io_bits(void __iomem *addr, u32 bits)
{
	u32 v = ioread32(addr);

	v &= ~bits;
	iowrite32(v, addr);
}

/*
 * Interface to control the chip select signal
 */
static void davinci_spi_chipselect(struct spi_device *spi, int value)
{
	struct davinci_spi *dspi;
	struct davinci_spi_config *spicfg = spi->controller_data;
	u8 chip_sel = spi->chip_select;
	u16 spidat1 = CS_DEFAULT;

	dspi = spi_master_get_devdata(spi->master);

	/* program delay transfers if tx_delay is non zero */
	if (spicfg && spicfg->wdelay)
		spidat1 |= SPIDAT1_WDEL;

	/*
	 * Board specific chip select logic decides the polarity and cs
	 * line for the controller
	 */
	if (spi->cs_gpiod) {
		/*
		 * FIXME: is this code ever executed? This host does not
		 * set SPI_MASTER_GPIO_SS so this chipselect callback should
		 * not get called from the SPI core when we are using
		 * GPIOs for chip select.
		 */
		if (value == BITBANG_CS_ACTIVE)
			gpiod_set_value(spi->cs_gpiod, 1);
		else
			gpiod_set_value(spi->cs_gpiod, 0);
	} else {
		if (value == BITBANG_CS_ACTIVE) {
			if (!(spi->mode & SPI_CS_WORD))
				spidat1 |= SPIDAT1_CSHOLD_MASK;
			spidat1 &= ~(0x1 << chip_sel);
		}
	}

	iowrite16(spidat1, dspi->base + SPIDAT1 + 2);
}

/**
 * davinci_spi_get_prescale - Calculates the correct prescale value
 * @maxspeed_hz: the maximum rate the SPI clock can run at
 *
 * This function calculates the prescale value that generates a clock rate
 * less than or equal to the specified maximum.
 *
 * Returns: calculated prescale value for easy programming into SPI registers
 * or negative error number if valid prescalar cannot be updated.
 */
static inline int davinci_spi_get_prescale(struct davinci_spi *dspi,
							u32 max_speed_hz)
{
	int ret;

	/* Subtract 1 to match what will be programmed into SPI register. */
	ret = DIV_ROUND_UP(clk_get_rate(dspi->clk), max_speed_hz) - 1;

	if (ret < dspi->prescaler_limit || ret > 255)
		return -EINVAL;

	return ret;
}

/**
 * davinci_spi_setup_transfer - This functions will determine transfer method
 * @spi: spi device on which data transfer to be done
 * @t: spi transfer in which transfer info is filled
 *
 * This function determines data transfer method (8/16/32 bit transfer).
 * It will also set the SPI Clock Control register according to
 * SPI slave device freq.
 */
static int davinci_spi_setup_transfer(struct spi_device *spi,
		struct spi_transfer *t)
{

	struct davinci_spi *dspi;
	struct davinci_spi_config *spicfg;
	u8 bits_per_word = 0;
	u32 hz = 0, spifmt = 0;
	int prescale;

	dspi = spi_master_get_devdata(spi->master);
	spicfg = spi->controller_data;
	if (!spicfg)
		spicfg = &davinci_spi_default_cfg;

	if (t) {
		bits_per_word = t->bits_per_word;
		hz = t->speed_hz;
	}

	/* if bits_per_word is not set then set it default */
	if (!bits_per_word)
		bits_per_word = spi->bits_per_word;

	/*
	 * Assign function pointer to appropriate transfer method
	 * 8bit, 16bit or 32bit transfer
	 */
	if (bits_per_word <= 8) {
		dspi->get_rx = davinci_spi_rx_buf_u8;
		dspi->get_tx = davinci_spi_tx_buf_u8;
		dspi->bytes_per_word[spi->chip_select] = 1;
	} else {
		dspi->get_rx = davinci_spi_rx_buf_u16;
		dspi->get_tx = davinci_spi_tx_buf_u16;
		dspi->bytes_per_word[spi->chip_select] = 2;
	}

	if (!hz)
		hz = spi->max_speed_hz;

	/* Set up SPIFMTn register, unique to this chipselect. */

	prescale = davinci_spi_get_prescale(dspi, hz);
	if (prescale < 0)
		return prescale;

	spifmt = (prescale << SPIFMT_PRESCALE_SHIFT) | (bits_per_word & 0x1f);

	if (spi->mode & SPI_LSB_FIRST)
		spifmt |= SPIFMT_SHIFTDIR_MASK;

	if (spi->mode & SPI_CPOL)
		spifmt |= SPIFMT_POLARITY_MASK;

	if (!(spi->mode & SPI_CPHA))
		spifmt |= SPIFMT_PHASE_MASK;

	/*
	* Assume wdelay is used only on SPI peripherals that has this field
	* in SPIFMTn register and when it's configured from board file or DT.
	*/
	if (spicfg->wdelay)
		spifmt |= ((spicfg->wdelay << SPIFMT_WDELAY_SHIFT)
				& SPIFMT_WDELAY_MASK);

	/*
	 * Version 1 hardware supports two basic SPI modes:
	 *  - Standard SPI mode uses 4 pins, with chipselect
	 *  - 3 pin SPI is a 4 pin variant without CS (SPI_NO_CS)
	 *	(distinct from SPI_3WIRE, with just one data wire;
	 *	or similar variants without MOSI or without MISO)
	 *
	 * Version 2 hardware supports an optional handshaking signal,
	 * so it can support two more modes:
	 *  - 5 pin SPI variant is standard SPI plus SPI_READY
	 *  - 4 pin with enable is (SPI_READY | SPI_NO_CS)
	 */

	if (dspi->version == SPI_VERSION_2) {

		u32 delay = 0;

		if (spicfg->odd_parity)
			spifmt |= SPIFMT_ODD_PARITY_MASK;

		if (spicfg->parity_enable)
			spifmt |= SPIFMT_PARITYENA_MASK;

		if (spicfg->timer_disable) {
			spifmt |= SPIFMT_DISTIMER_MASK;
		} else {
			delay |= (spicfg->c2tdelay << SPIDELAY_C2TDELAY_SHIFT)
						& SPIDELAY_C2TDELAY_MASK;
			delay |= (spicfg->t2cdelay << SPIDELAY_T2CDELAY_SHIFT)
						& SPIDELAY_T2CDELAY_MASK;
		}

		if (spi->mode & SPI_READY) {
			spifmt |= SPIFMT_WAITENA_MASK;
			delay |= (spicfg->t2edelay << SPIDELAY_T2EDELAY_SHIFT)
						& SPIDELAY_T2EDELAY_MASK;
			delay |= (spicfg->c2edelay << SPIDELAY_C2EDELAY_SHIFT)
						& SPIDELAY_C2EDELAY_MASK;
		}

		iowrite32(delay, dspi->base + SPIDELAY);
	}

	iowrite32(spifmt, dspi->base + SPIFMT0);

	return 0;
}

static int davinci_spi_of_setup(struct spi_device *spi)
{
	struct davinci_spi_config *spicfg = spi->controller_data;
	struct device_node *np = spi->dev.of_node;
	struct davinci_spi *dspi = spi_master_get_devdata(spi->master);
	u32 prop;

	if (spicfg == NULL && np) {
		spicfg = kzalloc(sizeof(*spicfg), GFP_KERNEL);
		if (!spicfg)
			return -ENOMEM;
		*spicfg = davinci_spi_default_cfg;
		/* override with dt configured values */
		if (!of_property_read_u32(np, "ti,spi-wdelay", &prop))
			spicfg->wdelay = (u8)prop;
		spi->controller_data = spicfg;

		if (dspi->dma_rx && dspi->dma_tx)
			spicfg->io_type = SPI_IO_TYPE_DMA;
	}

	return 0;
}

/**
 * davinci_spi_setup - This functions will set default transfer method
 * @spi: spi device on which data transfer to be done
 *
 * This functions sets the default transfer method.
 */
static int davinci_spi_setup(struct spi_device *spi)
{
	struct davinci_spi *dspi;
	struct device_node *np = spi->dev.of_node;
	bool internal_cs = true;

	dspi = spi_master_get_devdata(spi->master);

	if (!(spi->mode & SPI_NO_CS)) {
		if (np && spi->cs_gpiod)
			internal_cs = false;

		if (internal_cs)
			set_io_bits(dspi->base + SPIPC0, 1 << spi->chip_select);
	}

	if (spi->mode & SPI_READY)
		set_io_bits(dspi->base + SPIPC0, SPIPC0_SPIENA_MASK);

	if (spi->mode & SPI_LOOP)
		set_io_bits(dspi->base + SPIGCR1, SPIGCR1_LOOPBACK_MASK);
	else
		clear_io_bits(dspi->base + SPIGCR1, SPIGCR1_LOOPBACK_MASK);

	return davinci_spi_of_setup(spi);
}

static void davinci_spi_cleanup(struct spi_device *spi)
{
	struct davinci_spi_config *spicfg = spi->controller_data;

	spi->controller_data = NULL;
	if (spi->dev.of_node)
		kfree(spicfg);
}

static bool davinci_spi_can_dma(struct spi_master *master,
				struct spi_device *spi,
				struct spi_transfer *xfer)
{
	struct davinci_spi_config *spicfg = spi->controller_data;
	bool can_dma = false;

	if (spicfg)
		can_dma = (spicfg->io_type == SPI_IO_TYPE_DMA) &&
			(xfer->len >= DMA_MIN_BYTES) &&
			!is_vmalloc_addr(xfer->rx_buf) &&
			!is_vmalloc_addr(xfer->tx_buf);

	return can_dma;
}

static int davinci_spi_check_error(struct davinci_spi *dspi, int int_status)
{
	struct device *sdev = dspi->bitbang.master->dev.parent;

	if (int_status & SPIFLG_TIMEOUT_MASK) {
		dev_err(sdev, "SPI Time-out Error\n");
		return -ETIMEDOUT;
	}
	if (int_status & SPIFLG_DESYNC_MASK) {
		dev_err(sdev, "SPI Desynchronization Error\n");
		return -EIO;
	}
	if (int_status & SPIFLG_BITERR_MASK) {
		dev_err(sdev, "SPI Bit error\n");
		return -EIO;
	}

	if (dspi->version == SPI_VERSION_2) {
		if (int_status & SPIFLG_DLEN_ERR_MASK) {
			dev_err(sdev, "SPI Data Length Error\n");
			return -EIO;
		}
		if (int_status & SPIFLG_PARERR_MASK) {
			dev_err(sdev, "SPI Parity Error\n");
			return -EIO;
		}
		if (int_status & SPIFLG_OVRRUN_MASK) {
			dev_err(sdev, "SPI Data Overrun error\n");
			return -EIO;
		}
		if (int_status & SPIFLG_BUF_INIT_ACTIVE_MASK) {
			dev_err(sdev, "SPI Buffer Init Active\n");
			return -EBUSY;
		}
	}

	return 0;
}

/**
 * davinci_spi_process_events - check for and handle any SPI controller events
 * @dspi: the controller data
 *
 * This function will check the SPIFLG register and handle any events that are
 * detected there
 */
static int davinci_spi_process_events(struct davinci_spi *dspi)
{
	u32 buf, status, errors = 0, spidat1;

	buf = ioread32(dspi->base + SPIBUF);

	if (dspi->rcount > 0 && !(buf & SPIBUF_RXEMPTY_MASK)) {
		dspi->get_rx(buf & 0xFFFF, dspi);
		dspi->rcount--;
	}

	status = ioread32(dspi->base + SPIFLG);

	if (unlikely(status & SPIFLG_ERROR_MASK)) {
		errors = status & SPIFLG_ERROR_MASK;
		goto out;
	}

	if (dspi->wcount > 0 && !(buf & SPIBUF_TXFULL_MASK)) {
		spidat1 = ioread32(dspi->base + SPIDAT1);
		dspi->wcount--;
		spidat1 &= ~0xFFFF;
		spidat1 |= 0xFFFF & dspi->get_tx(dspi);
		iowrite32(spidat1, dspi->base + SPIDAT1);
	}

out:
	return errors;
}

static void davinci_spi_dma_rx_callback(void *data)
{
	struct davinci_spi *dspi = (struct davinci_spi *)data;

	dspi->rcount = 0;

	if (!dspi->wcount && !dspi->rcount)
		complete(&dspi->done);
}

static void davinci_spi_dma_tx_callback(void *data)
{
	struct davinci_spi *dspi = (struct davinci_spi *)data;

	dspi->wcount = 0;

	if (!dspi->wcount && !dspi->rcount)
		complete(&dspi->done);
}

/**
 * davinci_spi_bufs - functions which will handle transfer data
 * @spi: spi device on which data transfer to be done
 * @t: spi transfer in which transfer info is filled
 *
 * This function will put data to be transferred into data register
 * of SPI controller and then wait until the completion will be marked
 * by the IRQ Handler.
 */
static int davinci_spi_bufs(struct spi_device *spi, struct spi_transfer *t)
{
	struct davinci_spi *dspi;
	int data_type, ret = -ENOMEM;
	u32 tx_data, spidat1;
	u32 errors = 0;
	struct davinci_spi_config *spicfg;
	struct davinci_spi_platform_data *pdata;
	unsigned uninitialized_var(rx_buf_count);

	dspi = spi_master_get_devdata(spi->master);
	pdata = &dspi->pdata;
	spicfg = (struct davinci_spi_config *)spi->controller_data;
	if (!spicfg)
		spicfg = &davinci_spi_default_cfg;

	/* convert len to words based on bits_per_word */
	data_type = dspi->bytes_per_word[spi->chip_select];

	dspi->tx = t->tx_buf;
	dspi->rx = t->rx_buf;
	dspi->wcount = t->len / data_type;
	dspi->rcount = dspi->wcount;

	spidat1 = ioread32(dspi->base + SPIDAT1);

	clear_io_bits(dspi->base + SPIGCR1, SPIGCR1_POWERDOWN_MASK);
	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_SPIENA_MASK);

	reinit_completion(&dspi->done);

	if (!davinci_spi_can_dma(spi->master, spi, t)) {
		if (spicfg->io_type != SPI_IO_TYPE_POLL)
			set_io_bits(dspi->base + SPIINT, SPIINT_MASKINT);
		/* start the transfer */
		dspi->wcount--;
		tx_data = dspi->get_tx(dspi);
		spidat1 &= 0xFFFF0000;
		spidat1 |= tx_data & 0xFFFF;
		iowrite32(spidat1, dspi->base + SPIDAT1);
	} else {
		struct dma_slave_config dma_rx_conf = {
			.direction = DMA_DEV_TO_MEM,
			.src_addr = (unsigned long)dspi->pbase + SPIBUF,
			.src_addr_width = data_type,
			.src_maxburst = 1,
		};
		struct dma_slave_config dma_tx_conf = {
			.direction = DMA_MEM_TO_DEV,
			.dst_addr = (unsigned long)dspi->pbase + SPIDAT1,
			.dst_addr_width = data_type,
			.dst_maxburst = 1,
		};
		struct dma_async_tx_descriptor *rxdesc;
		struct dma_async_tx_descriptor *txdesc;

		dmaengine_slave_config(dspi->dma_rx, &dma_rx_conf);
		dmaengine_slave_config(dspi->dma_tx, &dma_tx_conf);

		rxdesc = dmaengine_prep_slave_sg(dspi->dma_rx,
				t->rx_sg.sgl, t->rx_sg.nents, DMA_DEV_TO_MEM,
				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
		if (!rxdesc)
			goto err_desc;

		if (!t->tx_buf) {
			/* To avoid errors when doing rx-only transfers with
			 * many SG entries (> 20), use the rx buffer as the
			 * dummy tx buffer so that dma reloads are done at the
			 * same time for rx and tx.
			 */
			t->tx_sg.sgl = t->rx_sg.sgl;
			t->tx_sg.nents = t->rx_sg.nents;
		}

		txdesc = dmaengine_prep_slave_sg(dspi->dma_tx,
				t->tx_sg.sgl, t->tx_sg.nents, DMA_MEM_TO_DEV,
				DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
		if (!txdesc)
			goto err_desc;

		rxdesc->callback = davinci_spi_dma_rx_callback;
		rxdesc->callback_param = (void *)dspi;
		txdesc->callback = davinci_spi_dma_tx_callback;
		txdesc->callback_param = (void *)dspi;

		if (pdata->cshold_bug)
			iowrite16(spidat1 >> 16, dspi->base + SPIDAT1 + 2);

		dmaengine_submit(rxdesc);
		dmaengine_submit(txdesc);

		dma_async_issue_pending(dspi->dma_rx);
		dma_async_issue_pending(dspi->dma_tx);

		set_io_bits(dspi->base + SPIINT, SPIINT_DMA_REQ_EN);
	}

	/* Wait for the transfer to complete */
	if (spicfg->io_type != SPI_IO_TYPE_POLL) {
		if (wait_for_completion_timeout(&dspi->done, HZ) == 0)
			errors = SPIFLG_TIMEOUT_MASK;
	} else {
		while (dspi->rcount > 0 || dspi->wcount > 0) {
			errors = davinci_spi_process_events(dspi);
			if (errors)
				break;
			cpu_relax();
		}
	}

	clear_io_bits(dspi->base + SPIINT, SPIINT_MASKALL);
	if (davinci_spi_can_dma(spi->master, spi, t))
		clear_io_bits(dspi->base + SPIINT, SPIINT_DMA_REQ_EN);

	clear_io_bits(dspi->base + SPIGCR1, SPIGCR1_SPIENA_MASK);
	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_POWERDOWN_MASK);

	/*
	 * Check for bit error, desync error,parity error,timeout error and
	 * receive overflow errors
	 */
	if (errors) {
		ret = davinci_spi_check_error(dspi, errors);
		WARN(!ret, "%s: error reported but no error found!\n",
							dev_name(&spi->dev));
		return ret;
	}

	if (dspi->rcount != 0 || dspi->wcount != 0) {
		dev_err(&spi->dev, "SPI data transfer error\n");
		return -EIO;
	}

	return t->len;

err_desc:
	return ret;
}

/**
 * dummy_thread_fn - dummy thread function
 * @irq: IRQ number for this SPI Master
 * @context_data: structure for SPI Master controller davinci_spi
 *
 * This is to satisfy the request_threaded_irq() API so that the irq
 * handler is called in interrupt context.
 */
static irqreturn_t dummy_thread_fn(s32 irq, void *data)
{
	return IRQ_HANDLED;
}

/**
 * davinci_spi_irq - Interrupt handler for SPI Master Controller
 * @irq: IRQ number for this SPI Master
 * @context_data: structure for SPI Master controller davinci_spi
 *
 * ISR will determine that interrupt arrives either for READ or WRITE command.
 * According to command it will do the appropriate action. It will check
 * transfer length and if it is not zero then dispatch transfer command again.
 * If transfer length is zero then it will indicate the COMPLETION so that
 * davinci_spi_bufs function can go ahead.
 */
static irqreturn_t davinci_spi_irq(s32 irq, void *data)
{
	struct davinci_spi *dspi = data;
	int status;

	status = davinci_spi_process_events(dspi);
	if (unlikely(status != 0))
		clear_io_bits(dspi->base + SPIINT, SPIINT_MASKINT);

	if ((!dspi->rcount && !dspi->wcount) || status)
		complete(&dspi->done);

	return IRQ_HANDLED;
}

static int davinci_spi_request_dma(struct davinci_spi *dspi)
{
	struct device *sdev = dspi->bitbang.master->dev.parent;

	dspi->dma_rx = dma_request_chan(sdev, "rx");
	if (IS_ERR(dspi->dma_rx))
		return PTR_ERR(dspi->dma_rx);

	dspi->dma_tx = dma_request_chan(sdev, "tx");
	if (IS_ERR(dspi->dma_tx)) {
		dma_release_channel(dspi->dma_rx);
		return PTR_ERR(dspi->dma_tx);
	}

	return 0;
}

#if defined(CONFIG_OF)

/* OF SPI data structure */
struct davinci_spi_of_data {
	u8	version;
	u8	prescaler_limit;
};

static const struct davinci_spi_of_data dm6441_spi_data = {
	.version = SPI_VERSION_1,
	.prescaler_limit = 2,
};

static const struct davinci_spi_of_data da830_spi_data = {
	.version = SPI_VERSION_2,
	.prescaler_limit = 2,
};

static const struct davinci_spi_of_data keystone_spi_data = {
	.version = SPI_VERSION_1,
	.prescaler_limit = 0,
};

static const struct of_device_id davinci_spi_of_match[] = {
	{
		.compatible = "ti,dm6441-spi",
		.data = &dm6441_spi_data,
	},
	{
		.compatible = "ti,da830-spi",
		.data = &da830_spi_data,
	},
	{
		.compatible = "ti,keystone-spi",
		.data = &keystone_spi_data,
	},
	{ },
};
MODULE_DEVICE_TABLE(of, davinci_spi_of_match);

/**
 * spi_davinci_get_pdata - Get platform data from DTS binding
 * @pdev: ptr to platform data
 * @dspi: ptr to driver data
 *
 * Parses and populates pdata in dspi from device tree bindings.
 *
 * NOTE: Not all platform data params are supported currently.
 */
static int spi_davinci_get_pdata(struct platform_device *pdev,
			struct davinci_spi *dspi)
{
	struct device_node *node = pdev->dev.of_node;
	struct davinci_spi_of_data *spi_data;
	struct davinci_spi_platform_data *pdata;
	unsigned int num_cs, intr_line = 0;
	const struct of_device_id *match;

	pdata = &dspi->pdata;

	match = of_match_device(davinci_spi_of_match, &pdev->dev);
	if (!match)
		return -ENODEV;

	spi_data = (struct davinci_spi_of_data *)match->data;

	pdata->version = spi_data->version;
	pdata->prescaler_limit = spi_data->prescaler_limit;
	/*
	 * default num_cs is 1 and all chipsel are internal to the chip
	 * indicated by chip_sel being NULL or cs_gpios being NULL or
	 * set to -ENOENT. num-cs includes internal as well as gpios.
	 * indicated by chip_sel being NULL. GPIO based CS is not
	 * supported yet in DT bindings.
	 */
	num_cs = 1;
	of_property_read_u32(node, "num-cs", &num_cs);
	pdata->num_chipselect = num_cs;
	of_property_read_u32(node, "ti,davinci-spi-intr-line", &intr_line);
	pdata->intr_line = intr_line;
	return 0;
}
#else
static int spi_davinci_get_pdata(struct platform_device *pdev,
			struct davinci_spi *dspi)
{
	return -ENODEV;
}
#endif

/**
 * davinci_spi_probe - probe function for SPI Master Controller
 * @pdev: platform_device structure which contains plateform specific data
 *
 * According to Linux Device Model this function will be invoked by Linux
 * with platform_device struct which contains the device specific info.
 * This function will map the SPI controller's memory, register IRQ,
 * Reset SPI controller and setting its registers to default value.
 * It will invoke spi_bitbang_start to create work queue so that client driver
 * can register transfer method to work queue.
 */
static int davinci_spi_probe(struct platform_device *pdev)
{
	struct spi_master *master;
	struct davinci_spi *dspi;
	struct davinci_spi_platform_data *pdata;
	struct resource *r;
	int ret = 0;
	u32 spipc0;

	master = spi_alloc_master(&pdev->dev, sizeof(struct davinci_spi));
	if (master == NULL) {
		ret = -ENOMEM;
		goto err;
	}

	platform_set_drvdata(pdev, master);

	dspi = spi_master_get_devdata(master);

	if (dev_get_platdata(&pdev->dev)) {
		pdata = dev_get_platdata(&pdev->dev);
		dspi->pdata = *pdata;
	} else {
		/* update dspi pdata with that from the DT */
		ret = spi_davinci_get_pdata(pdev, dspi);
		if (ret < 0)
			goto free_master;
	}

	/* pdata in dspi is now updated and point pdata to that */
	pdata = &dspi->pdata;

	dspi->bytes_per_word = devm_kcalloc(&pdev->dev,
					    pdata->num_chipselect,
					    sizeof(*dspi->bytes_per_word),
					    GFP_KERNEL);
	if (dspi->bytes_per_word == NULL) {
		ret = -ENOMEM;
		goto free_master;
	}

	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
	if (r == NULL) {
		ret = -ENOENT;
		goto free_master;
	}

	dspi->pbase = r->start;

	dspi->base = devm_ioremap_resource(&pdev->dev, r);
	if (IS_ERR(dspi->base)) {
		ret = PTR_ERR(dspi->base);
		goto free_master;
	}

	init_completion(&dspi->done);

	ret = platform_get_irq(pdev, 0);
	if (ret == 0)
		ret = -EINVAL;
	if (ret < 0)
		goto free_master;
	dspi->irq = ret;

	ret = devm_request_threaded_irq(&pdev->dev, dspi->irq, davinci_spi_irq,
				dummy_thread_fn, 0, dev_name(&pdev->dev), dspi);
	if (ret)
		goto free_master;

	dspi->bitbang.master = master;

	dspi->clk = devm_clk_get(&pdev->dev, NULL);
	if (IS_ERR(dspi->clk)) {
		ret = -ENODEV;
		goto free_master;
	}
	ret = clk_prepare_enable(dspi->clk);
	if (ret)
		goto free_master;

	master->use_gpio_descriptors = true;
	master->dev.of_node = pdev->dev.of_node;
	master->bus_num = pdev->id;
	master->num_chipselect = pdata->num_chipselect;
	master->bits_per_word_mask = SPI_BPW_RANGE_MASK(2, 16);
	master->flags = SPI_MASTER_MUST_RX;
	master->setup = davinci_spi_setup;
	master->cleanup = davinci_spi_cleanup;
	master->can_dma = davinci_spi_can_dma;

	dspi->bitbang.chipselect = davinci_spi_chipselect;
	dspi->bitbang.setup_transfer = davinci_spi_setup_transfer;
	dspi->prescaler_limit = pdata->prescaler_limit;
	dspi->version = pdata->version;

	dspi->bitbang.flags = SPI_NO_CS | SPI_LSB_FIRST | SPI_LOOP | SPI_CS_WORD;
	if (dspi->version == SPI_VERSION_2)
		dspi->bitbang.flags |= SPI_READY;

	dspi->bitbang.txrx_bufs = davinci_spi_bufs;

	ret = davinci_spi_request_dma(dspi);
	if (ret == -EPROBE_DEFER) {
		goto free_clk;
	} else if (ret) {
		dev_info(&pdev->dev, "DMA is not supported (%d)\n", ret);
		dspi->dma_rx = NULL;
		dspi->dma_tx = NULL;
	}

	dspi->get_rx = davinci_spi_rx_buf_u8;
	dspi->get_tx = davinci_spi_tx_buf_u8;

	/* Reset In/OUT SPI module */
	iowrite32(0, dspi->base + SPIGCR0);
	udelay(100);
	iowrite32(1, dspi->base + SPIGCR0);

	/* Set up SPIPC0.  CS and ENA init is done in davinci_spi_setup */
	spipc0 = SPIPC0_DIFUN_MASK | SPIPC0_DOFUN_MASK | SPIPC0_CLKFUN_MASK;
	iowrite32(spipc0, dspi->base + SPIPC0);

	if (pdata->intr_line)
		iowrite32(SPI_INTLVL_1, dspi->base + SPILVL);
	else
		iowrite32(SPI_INTLVL_0, dspi->base + SPILVL);

	iowrite32(CS_DEFAULT, dspi->base + SPIDEF);

	/* master mode default */
	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_CLKMOD_MASK);
	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_MASTER_MASK);
	set_io_bits(dspi->base + SPIGCR1, SPIGCR1_POWERDOWN_MASK);

	ret = spi_bitbang_start(&dspi->bitbang);
	if (ret)
		goto free_dma;

	dev_info(&pdev->dev, "Controller at 0x%p\n", dspi->base);

	return ret;

free_dma:
	if (dspi->dma_rx) {
		dma_release_channel(dspi->dma_rx);
		dma_release_channel(dspi->dma_tx);
	}
free_clk:
	clk_disable_unprepare(dspi->clk);
free_master:
	spi_master_put(master);
err:
	return ret;
}

/**
 * davinci_spi_remove - remove function for SPI Master Controller
 * @pdev: platform_device structure which contains plateform specific data
 *
 * This function will do the reverse action of davinci_spi_probe function
 * It will free the IRQ and SPI controller's memory region.
 * It will also call spi_bitbang_stop to destroy the work queue which was
 * created by spi_bitbang_start.
 */
static int davinci_spi_remove(struct platform_device *pdev)
{
	struct davinci_spi *dspi;
	struct spi_master *master;

	master = platform_get_drvdata(pdev);
	dspi = spi_master_get_devdata(master);

	spi_bitbang_stop(&dspi->bitbang);

	clk_disable_unprepare(dspi->clk);
	spi_master_put(master);

	if (dspi->dma_rx) {
		dma_release_channel(dspi->dma_rx);
		dma_release_channel(dspi->dma_tx);
	}

	return 0;
}

static struct platform_driver davinci_spi_driver = {
	.driver = {
		.name = "spi_davinci",
		.of_match_table = of_match_ptr(davinci_spi_of_match),
	},
	.probe = davinci_spi_probe,
	.remove = davinci_spi_remove,
};
module_platform_driver(davinci_spi_driver);

MODULE_DESCRIPTION("TI DaVinci SPI Master Controller Driver");
MODULE_LICENSE("GPL");