Merge branch 'spi-5.2' into spi-next

28 files changed, 2582 insertions, 550 deletions

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index f761655e2a36..0fba8f400c59 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -426,6 +426,12 @@ config SPI_MT65XX
 	  say Y or M here.If you are not sure, say N.
 	  SPI drivers for Mediatek MT65XX and MT81XX series ARM SoCs.
 
+config SPI_MT7621
+	tristate "MediaTek MT7621 SPI Controller"
+	depends on RALINK || COMPILE_TEST
+	help
+	  This selects a driver for the MediaTek MT7621 SPI Controller.
+
 config SPI_NPCM_PSPI
 	tristate "Nuvoton NPCM PSPI Controller"
 	depends on ARCH_NPCM || COMPILE_TEST
@@ -842,9 +848,17 @@ config SPI_XTENSA_XTFPGA
 	  16 bit words in SPI mode 0, automatically asserting CS on transfer
 	  start and deasserting on end.
 
+config SPI_ZYNQ_QSPI
+	tristate "Xilinx Zynq QSPI controller"
+	depends on ARCH_ZYNQ || COMPILE_TEST
+	help
+	  This enables support for the Zynq Quad SPI controller
+	  in master mode.
+	  This controller only supports SPI memory interface.
+
 config SPI_ZYNQMP_GQSPI
 	tristate "Xilinx ZynqMP GQSPI controller"
-	depends on SPI_MASTER && HAS_DMA
+	depends on (SPI_MASTER && HAS_DMA) || COMPILE_TEST
 	help
 	  Enables Xilinx GQSPI controller driver for Zynq UltraScale+ MPSoC.
 
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index d8fc03c9faa2..f2f78d03dc28 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -60,6 +60,7 @@ obj-$(CONFIG_SPI_MPC512x_PSC)		+= spi-mpc512x-psc.o
 obj-$(CONFIG_SPI_MPC52xx_PSC)		+= spi-mpc52xx-psc.o
 obj-$(CONFIG_SPI_MPC52xx)		+= spi-mpc52xx.o
 obj-$(CONFIG_SPI_MT65XX)                += spi-mt65xx.o
+obj-$(CONFIG_SPI_MT7621)		+= spi-mt7621.o
 obj-$(CONFIG_SPI_MXIC)			+= spi-mxic.o
 obj-$(CONFIG_SPI_MXS)			+= spi-mxs.o
 obj-$(CONFIG_SPI_NPCM_PSPI)		+= spi-npcm-pspi.o
@@ -118,6 +119,7 @@ obj-$(CONFIG_SPI_XCOMM)		+= spi-xcomm.o
 obj-$(CONFIG_SPI_XILINX)		+= spi-xilinx.o
 obj-$(CONFIG_SPI_XLP)			+= spi-xlp.o
 obj-$(CONFIG_SPI_XTENSA_XTFPGA)		+= spi-xtensa-xtfpga.o
+obj-$(CONFIG_SPI_ZYNQ_QSPI)		+= spi-zynq-qspi.o
 obj-$(CONFIG_SPI_ZYNQMP_GQSPI)		+= spi-zynqmp-gqspi.o
 
 # SPI slave protocol handlers
diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c
index b3173ebddade..9f24d5f0b431 100644
--- a/drivers/spi/atmel-quadspi.c
+++ b/drivers/spi/atmel-quadspi.c
@@ -366,7 +366,7 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 	return err;
 }
 
-const char *atmel_qspi_get_name(struct spi_mem *spimem)
+static const char *atmel_qspi_get_name(struct spi_mem *spimem)
 {
 	return dev_name(spimem->spi->dev.parent);
 }
diff --git a/drivers/spi/spi-at91-usart.c b/drivers/spi/spi-at91-usart.c
index a694d702e574..f763e14bdf12 100644
--- a/drivers/spi/spi-at91-usart.c
+++ b/drivers/spi/spi-at91-usart.c
@@ -178,12 +178,6 @@ static int at91_usart_spi_setup(struct spi_device *spi)
 	struct at91_usart_spi *aus = spi_master_get_devdata(spi->controller);
 	u32 *ausd = spi->controller_state;
 	unsigned int mr = at91_usart_spi_readl(aus, MR);
-	u8 bits = spi->bits_per_word;
-
-	if (bits != 8) {
-		dev_dbg(&spi->dev, "Only 8 bits per word are supported\n");
-		return -EINVAL;
-	}
 
 	if (spi->mode & SPI_CPOL)
 		mr |= US_MR_CPOL;
@@ -212,7 +206,7 @@ static int at91_usart_spi_setup(struct spi_device *spi)
 
 	dev_dbg(&spi->dev,
 		"setup: bpw %u mode 0x%x -> mr %d %08x\n",
-		bits, spi->mode, spi->chip_select, mr);
+		spi->bits_per_word, spi->mode, spi->chip_select, mr);
 
 	return 0;
 }
diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c
index 35aebdfd3b4e..8aa22713c483 100644
--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -335,20 +335,6 @@ static int bcm2835_spi_transfer_one_irq(struct spi_master *master,
 	return 1;
 }
 
-/*
- * DMA support
- *
- * this implementation has currently a few issues in so far as it does
- * not work arrount limitations of the HW.
- *
- * the main one being that DMA transfers are limited to 16 bit
- * (so 0 to 65535 bytes) by the SPI HW due to BCM2835_SPI_DLEN
- *
- * there may be a few more border-cases we may need to address as well
- * but unfortunately this would mean splitting up the scatter-gather
- * list making it slightly unpractical...
- */
-
 /**
  * bcm2835_spi_transfer_prologue() - transfer first few bytes without DMA
  * @master: SPI master
@@ -630,19 +616,6 @@ static bool bcm2835_spi_can_dma(struct spi_master *master,
 	if (tfr->len < BCM2835_SPI_DMA_MIN_LENGTH)
 		return false;
 
-	/* BCM2835_SPI_DLEN has defined a max transfer size as
-	 * 16 bit, so max is 65535
-	 * we can revisit this by using an alternative transfer
-	 * method - ideally this would get done without any more
-	 * interaction...
-	 */
-	if (tfr->len > 65535) {
-		dev_warn_once(&spi->dev,
-			      "transfer size of %d too big for dma-transfer\n",
-			      tfr->len);
-		return false;
-	}
-
 	/* return OK */
 	return true;
 }
@@ -707,7 +680,6 @@ static void bcm2835_dma_init(struct spi_master *master, struct device *dev)
 
 	/* all went well, so set can_dma */
 	master->can_dma = bcm2835_spi_can_dma;
-	master->max_dma_len = 65535; /* limitation by BCM2835_SPI_DLEN */
 	/* need to do TX AND RX DMA, so we need dummy buffers */
 	master->flags = SPI_MASTER_MUST_RX | SPI_MASTER_MUST_TX;
 
@@ -844,6 +816,17 @@ static int bcm2835_spi_prepare_message(struct spi_master *master,
 	struct spi_device *spi = msg->spi;
 	struct bcm2835_spi *bs = spi_master_get_devdata(master);
 	u32 cs = bcm2835_rd(bs, BCM2835_SPI_CS);
+	int ret;
+
+	/*
+	 * DMA transfers are limited to 16 bit (0 to 65535 bytes) by the SPI HW
+	 * due to DLEN. Split up transfers (32-bit FIFO aligned) if the limit is
+	 * exceeded.
+	 */
+	ret = spi_split_transfers_maxsize(master, msg, 65532,
+					  GFP_KERNEL | GFP_DMA);
+	if (ret)
+		return ret;
 
 	cs &= ~(BCM2835_SPI_CS_CPOL | BCM2835_SPI_CS_CPHA);
 
diff --git a/drivers/spi/spi-bcm2835aux.c b/drivers/spi/spi-bcm2835aux.c
index f7e054848ca5..bbf87adb3ff8 100644
--- a/drivers/spi/spi-bcm2835aux.c
+++ b/drivers/spi/spi-bcm2835aux.c
@@ -21,6 +21,7 @@
 
 #include <linux/clk.h>
 #include <linux/completion.h>
+#include <linux/debugfs.h>
 #include <linux/delay.h>
 #include <linux/err.h>
 #include <linux/interrupt.h>
@@ -36,6 +37,12 @@
 #include <linux/spi/spi.h>
 #include <linux/spinlock.h>
 
+/* define polling limits */
+static unsigned int polling_limit_us = 30;
+module_param(polling_limit_us, uint, 0664);
+MODULE_PARM_DESC(polling_limit_us,
+		 "time in us to run a transfer in polling mode - if zero no polling is used\n");
+
 /*
  * spi register defines
  *
@@ -88,10 +95,6 @@
 #define BCM2835_AUX_SPI_STAT_BUSY	0x00000040
 #define BCM2835_AUX_SPI_STAT_BITCOUNT	0x0000003F
 
-/* timeout values */
-#define BCM2835_AUX_SPI_POLLING_LIMIT_US	30
-#define BCM2835_AUX_SPI_POLLING_JIFFIES		2
-
 struct bcm2835aux_spi {
 	void __iomem *regs;
 	struct clk *clk;
@@ -102,8 +105,53 @@ struct bcm2835aux_spi {
 	int tx_len;
 	int rx_len;
 	int pending;
+
+	u64 count_transfer_polling;
+	u64 count_transfer_irq;
+	u64 count_transfer_irq_after_poll;
+
+	struct dentry *debugfs_dir;
 };
 
+#if defined(CONFIG_DEBUG_FS)
+static void bcm2835aux_debugfs_create(struct bcm2835aux_spi *bs,
+				      const char *dname)
+{
+	char name[64];
+	struct dentry *dir;
+
+	/* get full name */
+	snprintf(name, sizeof(name), "spi-bcm2835aux-%s", dname);
+
+	/* the base directory */
+	dir = debugfs_create_dir(name, NULL);
+	bs->debugfs_dir = dir;
+
+	/* the counters */
+	debugfs_create_u64("count_transfer_polling", 0444, dir,
+			   &bs->count_transfer_polling);
+	debugfs_create_u64("count_transfer_irq", 0444, dir,
+			   &bs->count_transfer_irq);
+	debugfs_create_u64("count_transfer_irq_after_poll", 0444, dir,
+			   &bs->count_transfer_irq_after_poll);
+}
+
+static void bcm2835aux_debugfs_remove(struct bcm2835aux_spi *bs)
+{
+	debugfs_remove_recursive(bs->debugfs_dir);
+	bs->debugfs_dir = NULL;
+}
+#else
+static void bcm2835aux_debugfs_create(struct bcm2835aux_spi *bs,
+				      const char *dname)
+{
+}
+
+static void bcm2835aux_debugfs_remove(struct bcm2835aux_spi *bs)
+{
+}
+#endif /* CONFIG_DEBUG_FS */
+
 static inline u32 bcm2835aux_rd(struct bcm2835aux_spi *bs, unsigned reg)
 {
 	return readl(bs->regs + reg);
@@ -123,9 +171,6 @@ static inline void bcm2835aux_rd_fifo(struct bcm2835aux_spi *bs)
 	data = bcm2835aux_rd(bs, BCM2835_AUX_SPI_IO);
 	if (bs->rx_buf) {
 		switch (count) {
-		case 4:
-			*bs->rx_buf++ = (data >> 24) & 0xff;
-			/* fallthrough */
 		case 3:
 			*bs->rx_buf++ = (data >> 16) & 0xff;
 			/* fallthrough */
@@ -178,24 +223,14 @@ static void bcm2835aux_spi_reset_hw(struct bcm2835aux_spi *bs)
 		      BCM2835_AUX_SPI_CNTL0_CLEARFIFO);
 }
 
-static irqreturn_t bcm2835aux_spi_interrupt(int irq, void *dev_id)
+static void bcm2835aux_spi_transfer_helper(struct bcm2835aux_spi *bs)
 {
-	struct spi_master *master = dev_id;
-	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
-	irqreturn_t ret = IRQ_NONE;
-
-	/* IRQ may be shared, so return if our interrupts are disabled */
-	if (!(bcm2835aux_rd(bs, BCM2835_AUX_SPI_CNTL1) &
-	      (BCM2835_AUX_SPI_CNTL1_TXEMPTY | BCM2835_AUX_SPI_CNTL1_IDLE)))
-		return ret;
+	u32 stat = bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT);
 
 	/* check if we have data to read */
-	while (bs->rx_len &&
-	       (!(bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT) &
-		  BCM2835_AUX_SPI_STAT_RX_EMPTY))) {
+	for (; bs->rx_len && (stat & BCM2835_AUX_SPI_STAT_RX_LVL);
+	     stat = bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT))
 		bcm2835aux_rd_fifo(bs);
-		ret = IRQ_HANDLED;
-	}
 
 	/* check if we have data to write */
 	while (bs->tx_len &&
@@ -203,16 +238,21 @@ static irqreturn_t bcm2835aux_spi_interrupt(int irq, void *dev_id)
 	       (!(bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT) &
 		  BCM2835_AUX_SPI_STAT_TX_FULL))) {
 		bcm2835aux_wr_fifo(bs);
-		ret = IRQ_HANDLED;
 	}
+}
 
-	/* and check if we have reached "done" */
-	while (bs->rx_len &&
-	       (!(bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT) &
-		  BCM2835_AUX_SPI_STAT_BUSY))) {
-		bcm2835aux_rd_fifo(bs);
-		ret = IRQ_HANDLED;
-	}
+static irqreturn_t bcm2835aux_spi_interrupt(int irq, void *dev_id)
+{
+	struct spi_master *master = dev_id;
+	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
+
+	/* IRQ may be shared, so return if our interrupts are disabled */
+	if (!(bcm2835aux_rd(bs, BCM2835_AUX_SPI_CNTL1) &
+	      (BCM2835_AUX_SPI_CNTL1_TXEMPTY | BCM2835_AUX_SPI_CNTL1_IDLE)))
+		return IRQ_NONE;
+
+	/* do common fifo handling */
+	bcm2835aux_spi_transfer_helper(bs);
 
 	if (!bs->tx_len) {
 		/* disable tx fifo empty interrupt */
@@ -226,8 +266,7 @@ static irqreturn_t bcm2835aux_spi_interrupt(int irq, void *dev_id)
 		complete(&master->xfer_completion);
 	}
 
-	/* and return */
-	return ret;
+	return IRQ_HANDLED;
 }
 
 static int __bcm2835aux_spi_transfer_one_irq(struct spi_master *master,
@@ -251,6 +290,9 @@ static int bcm2835aux_spi_transfer_one_irq(struct spi_master *master,
 {
 	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
 
+	/* update statistics */
+	bs->count_transfer_irq++;
+
 	/* fill in registers and fifos before enabling interrupts */
 	bcm2835aux_wr(bs, BCM2835_AUX_SPI_CNTL1, bs->cntl[1]);
 	bcm2835aux_wr(bs, BCM2835_AUX_SPI_CNTL0, bs->cntl[0]);
@@ -273,35 +315,22 @@ static int bcm2835aux_spi_transfer_one_poll(struct spi_master *master,
 {
 	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
 	unsigned long timeout;
-	u32 stat;
+
+	/* update statistics */
+	bs->count_transfer_polling++;
 
 	/* configure spi */
 	bcm2835aux_wr(bs, BCM2835_AUX_SPI_CNTL1, bs->cntl[1]);
 	bcm2835aux_wr(bs, BCM2835_AUX_SPI_CNTL0, bs->cntl[0]);
 
-	/* set the timeout */
-	timeout = jiffies + BCM2835_AUX_SPI_POLLING_JIFFIES;
+	/* set the timeout to at least 2 jiffies */
+	timeout = jiffies + 2 + HZ * polling_limit_us / 1000000;
 
 	/* loop until finished the transfer */
 	while (bs->rx_len) {
-		/* read status */
-		stat = bcm2835aux_rd(bs, BCM2835_AUX_SPI_STAT);
 
-		/* fill in tx fifo with remaining data */
-		if ((bs->tx_len) && (!(stat & BCM2835_AUX_SPI_STAT_TX_FULL))) {
-			bcm2835aux_wr_fifo(bs);
-			continue;
-		}
-
-		/* read data from fifo for both cases */
-		if (!(stat & BCM2835_AUX_SPI_STAT_RX_EMPTY)) {
-			bcm2835aux_rd_fifo(bs);
-			continue;
-		}
-		if (!(stat & BCM2835_AUX_SPI_STAT_BUSY)) {
-			bcm2835aux_rd_fifo(bs);
-			continue;
-		}
+		/* do common fifo handling */
+		bcm2835aux_spi_transfer_helper(bs);
 
 		/* there is still data pending to read check the timeout */
 		if (bs->rx_len && time_after(jiffies, timeout)) {
@@ -310,6 +339,7 @@ static int bcm2835aux_spi_transfer_one_poll(struct spi_master *master,
 					    jiffies - timeout,
 					    bs->tx_len, bs->rx_len);
 			/* forward to interrupt handler */
+			bs->count_transfer_irq_after_poll++;
 			return __bcm2835aux_spi_transfer_one_irq(master,
 							       spi, tfr);
 		}
@@ -324,8 +354,8 @@ static int bcm2835aux_spi_transfer_one(struct spi_master *master,
 				       struct spi_transfer *tfr)
 {
 	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
-	unsigned long spi_hz, clk_hz, speed;
-	unsigned long spi_used_hz;
+	unsigned long spi_hz, clk_hz, speed, spi_used_hz;
+	unsigned long hz_per_byte, byte_limit;
 
 	/* calculate the registers to handle
 	 *
@@ -369,14 +399,15 @@ static int bcm2835aux_spi_transfer_one(struct spi_master *master,
 	 * of Hz per byte per polling limit.  E.g., we can transfer 1 byte in
 	 * 30 µs per 300,000 Hz of bus clock.
 	 */
-#define HZ_PER_BYTE ((9 * 1000000) / BCM2835_AUX_SPI_POLLING_LIMIT_US)
+	hz_per_byte = polling_limit_us ? (9 * 1000000) / polling_limit_us : 0;
+	byte_limit = hz_per_byte ? spi_used_hz / hz_per_byte : 1;
+
 	/* run in polling mode for short transfers */
-	if (tfr->len < spi_used_hz / HZ_PER_BYTE)
+	if (tfr->len < byte_limit)
 		return bcm2835aux_spi_transfer_one_poll(master, spi, tfr);
 
 	/* run in interrupt mode for all others */
 	return bcm2835aux_spi_transfer_one_irq(master, spi, tfr);
-#undef HZ_PER_BYTE
 }
 
 static int bcm2835aux_spi_prepare_message(struct spi_master *master,
@@ -421,6 +452,50 @@ static void bcm2835aux_spi_handle_err(struct spi_master *master,
 	bcm2835aux_spi_reset_hw(bs);
 }
 
+static int bcm2835aux_spi_setup(struct spi_device *spi)
+{
+	int ret;
+
+	/* sanity check for native cs */
+	if (spi->mode & SPI_NO_CS)
+		return 0;
+	if (gpio_is_valid(spi->cs_gpio)) {
+		/* with gpio-cs set the GPIO to the correct level
+		 * and as output (in case the dt has the gpio not configured
+		 * as output but native cs)
+		 */
+		ret = gpio_direction_output(spi->cs_gpio,
+					    (spi->mode & SPI_CS_HIGH) ? 0 : 1);
+		if (ret)
+			dev_err(&spi->dev,
+				"could not set gpio %i as output: %i\n",
+				spi->cs_gpio, ret);
+
+		return ret;
+	}
+
+	/* for dt-backwards compatibility: only support native on CS0
+	 * known things not supported with broken native CS:
+	 * * multiple chip-selects: cs0-cs2 are all
+	 *     simultaniously asserted whenever there is a transfer
+	 *     this even includes SPI_NO_CS
+	 * * SPI_CS_HIGH: cs are always asserted low
+	 * * cs_change: cs is deasserted after each spi_transfer
+	 * * cs_delay_usec: cs is always deasserted one SCK cycle
+	 *     after the last transfer
+	 * probably more...
+	 */
+	dev_warn(&spi->dev,
+		 "Native CS is not supported - please configure cs-gpio in device-tree\n");
+
+	if (spi->chip_select == 0)
+		return 0;
+
+	dev_warn(&spi->dev, "Native CS is not working for cs > 0\n");
+
+	return -EINVAL;
+}
+
 static int bcm2835aux_spi_probe(struct platform_device *pdev)
 {
 	struct spi_master *master;
@@ -438,7 +513,19 @@ static int bcm2835aux_spi_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, master);
 	master->mode_bits = (SPI_CPOL | SPI_CS_HIGH | SPI_NO_CS);
 	master->bits_per_word_mask = SPI_BPW_MASK(8);
-	master->num_chipselect = -1;
+	/* even though the driver never officially supported native CS
+	 * allow a single native CS for legacy DT support purposes when
+	 * no cs-gpio is configured.
+	 * Known limitations for native cs are:
+	 * * multiple chip-selects: cs0-cs2 are all simultaniously asserted
+	 *     whenever there is a transfer -  this even includes SPI_NO_CS
+	 * * SPI_CS_HIGH: is ignores - cs are always asserted low
+	 * * cs_change: cs is deasserted after each spi_transfer
+	 * * cs_delay_usec: cs is always deasserted one SCK cycle after
+	 *     a spi_transfer
+	 */
+	master->num_chipselect = 1;
+	master->setup = bcm2835aux_spi_setup;
 	master->transfer_one = bcm2835aux_spi_transfer_one;
 	master->handle_err = bcm2835aux_spi_handle_err;
 	master->prepare_message = bcm2835aux_spi_prepare_message;
@@ -502,6 +589,8 @@ static int bcm2835aux_spi_probe(struct platform_device *pdev)
 		goto out_clk_disable;
 	}
 
+	bcm2835aux_debugfs_create(bs, dev_name(&pdev->dev));
+
 	return 0;
 
 out_clk_disable:
@@ -516,6 +605,8 @@ static int bcm2835aux_spi_remove(struct platform_device *pdev)
 	struct spi_master *master = platform_get_drvdata(pdev);
 	struct bcm2835aux_spi *bs = spi_master_get_devdata(master);
 
+	bcm2835aux_debugfs_remove(bs);
+
 	bcm2835aux_spi_reset_hw(bs);
 
 	/* disable the HW block by releasing the clock */
diff --git a/drivers/spi/spi-bitbang.c b/drivers/spi/spi-bitbang.c
index dd9a8c54a693..4243e53f9f7b 100644
--- a/drivers/spi/spi-bitbang.c
+++ b/drivers/spi/spi-bitbang.c
@@ -335,6 +335,42 @@ static void spi_bitbang_set_cs(struct spi_device *spi, bool enable)
 
 /*----------------------------------------------------------------------*/
 
+int spi_bitbang_init(struct spi_bitbang *bitbang)
+{
+	struct spi_master *master = bitbang->master;
+
+	if (!master || !bitbang->chipselect)
+		return -EINVAL;
+
+	mutex_init(&bitbang->lock);
+
+	if (!master->mode_bits)
+		master->mode_bits = SPI_CPOL | SPI_CPHA | bitbang->flags;
+
+	if (master->transfer || master->transfer_one_message)
+		return -EINVAL;
+
+	master->prepare_transfer_hardware = spi_bitbang_prepare_hardware;
+	master->unprepare_transfer_hardware = spi_bitbang_unprepare_hardware;
+	master->transfer_one = spi_bitbang_transfer_one;
+	master->set_cs = spi_bitbang_set_cs;
+
+	if (!bitbang->txrx_bufs) {
+		bitbang->use_dma = 0;
+		bitbang->txrx_bufs = spi_bitbang_bufs;
+		if (!master->setup) {
+			if (!bitbang->setup_transfer)
+				bitbang->setup_transfer =
+					 spi_bitbang_setup_transfer;
+			master->setup = spi_bitbang_setup;
+			master->cleanup = spi_bitbang_cleanup;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(spi_bitbang_init);
+
 /**
  * spi_bitbang_start - start up a polled/bitbanging SPI master driver
  * @bitbang: driver handle
@@ -368,33 +404,9 @@ int spi_bitbang_start(struct spi_bitbang *bitbang)
 	struct spi_master *master = bitbang->master;
 	int ret;
 
-	if (!master || !bitbang->chipselect)
-		return -EINVAL;
-
-	mutex_init(&bitbang->lock);
-
-	if (!master->mode_bits)
-		master->mode_bits = SPI_CPOL | SPI_CPHA | bitbang->flags;
-
-	if (master->transfer || master->transfer_one_message)
-		return -EINVAL;
-
-	master->prepare_transfer_hardware = spi_bitbang_prepare_hardware;
-	master->unprepare_transfer_hardware = spi_bitbang_unprepare_hardware;
-	master->transfer_one = spi_bitbang_transfer_one;
-	master->set_cs = spi_bitbang_set_cs;
-
-	if (!bitbang->txrx_bufs) {
-		bitbang->use_dma = 0;
-		bitbang->txrx_bufs = spi_bitbang_bufs;
-		if (!master->setup) {
-			if (!bitbang->setup_transfer)
-				bitbang->setup_transfer =
-					 spi_bitbang_setup_transfer;
-			master->setup = spi_bitbang_setup;
-			master->cleanup = spi_bitbang_cleanup;
-		}
-	}
+	ret = spi_bitbang_init(bitbang);
+	if (ret)
+		return ret;
 
 	/* driver may get busy before register() returns, especially
 	 * if someone registered boardinfo for devices
diff --git a/drivers/spi/spi-dw-mmio.c b/drivers/spi/spi-dw-mmio.c
index 4bd59a93d988..de952b17bc10 100644
--- a/drivers/spi/spi-dw-mmio.c
+++ b/drivers/spi/spi-dw-mmio.c
@@ -30,6 +30,7 @@
 struct dw_spi_mmio {
 	struct dw_spi  dws;
 	struct clk     *clk;
+	struct clk     *pclk;
 	void           *priv;
 };
 
@@ -172,6 +173,14 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
 	if (ret)
 		return ret;
 
+	/* Optional clock needed to access the registers */
+	dwsmmio->pclk = devm_clk_get_optional(&pdev->dev, "pclk");
+	if (IS_ERR(dwsmmio->pclk))
+		return PTR_ERR(dwsmmio->pclk);
+	ret = clk_prepare_enable(dwsmmio->pclk);
+	if (ret)
+		goto out_clk;
+
 	dws->bus_num = pdev->id;
 
 	dws->max_freq = clk_get_rate(dwsmmio->clk);
@@ -199,6 +208,8 @@ static int dw_spi_mmio_probe(struct platform_device *pdev)
 	return 0;
 
 out:
+	clk_disable_unprepare(dwsmmio->pclk);
+out_clk:
 	clk_disable_unprepare(dwsmmio->clk);
 	return ret;
 }
@@ -208,6 +219,7 @@ static int dw_spi_mmio_remove(struct platform_device *pdev)
 	struct dw_spi_mmio *dwsmmio = platform_get_drvdata(pdev);
 
 	dw_spi_remove_host(&dwsmmio->dws);
+	clk_disable_unprepare(dwsmmio->pclk);
 	clk_disable_unprepare(dwsmmio->clk);
 
 	return 0;
diff --git a/drivers/spi/spi-ep93xx.c b/drivers/spi/spi-ep93xx.c
index 79fc3940245a..47e39251bad9 100644
--- a/drivers/spi/spi-ep93xx.c
+++ b/drivers/spi/spi-ep93xx.c
@@ -28,7 +28,6 @@
 #include <linux/platform_device.h>
 #include <linux/sched.h>
 #include <linux/scatterlist.h>
-#include <linux/gpio.h>
 #include <linux/spi/spi.h>
 
 #include <linux/platform_data/dma-ep93xx.h>
@@ -676,6 +675,7 @@ static int ep93xx_spi_probe(struct platform_device *pdev)
 	if (!master)
 		return -ENOMEM;
 
+	master->use_gpio_descriptors = true;
 	master->prepare_transfer_hardware = ep93xx_spi_prepare_hardware;
 	master->unprepare_transfer_hardware = ep93xx_spi_unprepare_hardware;
 	master->prepare_message = ep93xx_spi_prepare_message;
@@ -683,31 +683,11 @@ static int ep93xx_spi_probe(struct platform_device *pdev)
 	master->bus_num = pdev->id;
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH;
 	master->bits_per_word_mask = SPI_BPW_RANGE_MASK(4, 16);
-
-	master->num_chipselect = info->num_chipselect;
-	master->cs_gpios = devm_kcalloc(&master->dev,
-					master->num_chipselect, sizeof(int),
-					GFP_KERNEL);
-	if (!master->cs_gpios) {
-		error = -ENOMEM;
-		goto fail_release_master;
-	}
-
-	for (i = 0; i < master->num_chipselect; i++) {
-		master->cs_gpios[i] = info->chipselect[i];
-
-		if (!gpio_is_valid(master->cs_gpios[i]))
-			continue;
-
-		error = devm_gpio_request_one(&pdev->dev, master->cs_gpios[i],
-					      GPIOF_OUT_INIT_HIGH,
-					      "ep93xx-spi");
-		if (error) {
-			dev_err(&pdev->dev, "could not request cs gpio %d\n",
-				master->cs_gpios[i]);
-			goto fail_release_master;
-		}
-	}
+	/*
+	 * The SPI core will count the number of GPIO descriptors to figure
+	 * out the number of chip selects available on the platform.
+	 */
+	master->num_chipselect = 0;
 
 	platform_set_drvdata(pdev, master);
 
diff --git a/drivers/spi/spi-fsl-lib.h b/drivers/spi/spi-fsl-lib.h
index f303f306b38e..483734bc1b1e 100644
--- a/drivers/spi/spi-fsl-lib.h
+++ b/drivers/spi/spi-fsl-lib.h
@@ -95,8 +95,10 @@ static inline u32 mpc8xxx_spi_read_reg(__be32 __iomem *reg)
 
 struct mpc8xxx_spi_probe_info {
 	struct fsl_spi_platform_data pdata;
+	int ngpios;
 	int *gpios;
 	bool *alow_flags;
+	__be32 __iomem *immr_spi_cs;
 };
 
 extern u32 mpc8xxx_spi_tx_buf_u8(struct mpc8xxx_spi *mpc8xxx_spi);
diff --git a/drivers/spi/spi-fsl-lpspi.c b/drivers/spi/spi-fsl-lpspi.c
index 9e117c4635d8..d08e9324140e 100644
--- a/drivers/spi/spi-fsl-lpspi.c
+++ b/drivers/spi/spi-fsl-lpspi.c
@@ -8,7 +8,10 @@
 #include <linux/clk.h>
 #include <linux/completion.h>
 #include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
 #include <linux/err.h>
+#include <linux/gpio.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/irq.h>
@@ -16,7 +19,12 @@
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_device.h>
+#include <linux/of_gpio.h>
+#include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
+#include <linux/platform_data/dma-imx.h>
+#include <linux/platform_data/spi-imx.h>
+#include <linux/pm_runtime.h>
 #include <linux/slab.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_bitbang.h>
@@ -24,6 +32,11 @@
 
 #define DRIVER_NAME "fsl_lpspi"
 
+#define FSL_LPSPI_RPM_TIMEOUT 50 /* 50ms */
+
+/* The maximum bytes that edma can transfer once.*/
+#define FSL_LPSPI_MAX_EDMA_BYTES  ((1 << 15) - 1)
+
 /* i.MX7ULP LPSPI registers */
 #define IMX7ULP_VERID	0x0
 #define IMX7ULP_PARAM	0x4
@@ -57,6 +70,8 @@
 #define IER_FCIE	BIT(9)
 #define IER_RDIE	BIT(1)
 #define IER_TDIE	BIT(0)
+#define DER_RDDE	BIT(1)
+#define DER_TDDE	BIT(0)
 #define CFGR1_PCSCFG	BIT(27)
 #define CFGR1_PINCFG	(BIT(24)|BIT(25))
 #define CFGR1_PCSPOL	BIT(8)
@@ -84,8 +99,11 @@ struct lpspi_config {
 struct fsl_lpspi_data {
 	struct device *dev;
 	void __iomem *base;
-	struct clk *clk;
+	unsigned long base_phys;
+	struct clk *clk_ipg;
+	struct clk *clk_per;
 	bool is_slave;
+	bool is_first_byte;
 
 	void *rx_buf;
 	const void *tx_buf;
@@ -101,6 +119,13 @@ struct fsl_lpspi_data {
 	struct completion xfer_done;
 
 	bool slave_aborted;
+
+	/* DMA */
+	bool usedma;
+	struct completion dma_rx_completion;
+	struct completion dma_tx_completion;
+
+	int chipselect[0];
 };
 
 static const struct of_device_id fsl_lpspi_dt_ids[] = {
@@ -147,12 +172,48 @@ static void fsl_lpspi_intctrl(struct fsl_lpspi_data *fsl_lpspi,
 	writel(enable, fsl_lpspi->base + IMX7ULP_IER);
 }
 
+static int fsl_lpspi_bytes_per_word(const int bpw)
+{
+	return DIV_ROUND_UP(bpw, BITS_PER_BYTE);
+}
+
+static bool fsl_lpspi_can_dma(struct spi_controller *controller,
+			      struct spi_device *spi,
+			      struct spi_transfer *transfer)
+{
+	unsigned int bytes_per_word;
+
+	if (!controller->dma_rx)
+		return false;
+
+	bytes_per_word = fsl_lpspi_bytes_per_word(transfer->bits_per_word);
+
+	switch (bytes_per_word)
+	{
+		case 1:
+		case 2:
+		case 4:
+			break;
+		default:
+			return false;
+	}
+
+	return true;
+}
+
 static int lpspi_prepare_xfer_hardware(struct spi_controller *controller)
 {
 	struct fsl_lpspi_data *fsl_lpspi =
 				spi_controller_get_devdata(controller);
+	int ret;
+
+	ret = pm_runtime_get_sync(fsl_lpspi->dev);
+	if (ret < 0) {
+		dev_err(fsl_lpspi->dev, "failed to enable clock\n");
+		return ret;
+	}
 
-	return clk_prepare_enable(fsl_lpspi->clk);
+	return 0;
 }
 
 static int lpspi_unprepare_xfer_hardware(struct spi_controller *controller)
@@ -160,7 +221,22 @@ static int lpspi_unprepare_xfer_hardware(struct spi_controller *controller)
 	struct fsl_lpspi_data *fsl_lpspi =
 				spi_controller_get_devdata(controller);
 
-	clk_disable_unprepare(fsl_lpspi->clk);
+	pm_runtime_mark_last_busy(fsl_lpspi->dev);
+	pm_runtime_put_autosuspend(fsl_lpspi->dev);
+
+	return 0;
+}
+
+static int fsl_lpspi_prepare_message(struct spi_controller *controller,
+				     struct spi_message