Merge remote-tracking branch 'spi/for-5.12' into spi-next

28 files changed, 2244 insertions, 2423 deletions

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index aadaea052f51..5aab7c6cc439 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -203,7 +203,7 @@ config SPI_CADENCE
 
 config SPI_CADENCE_QUADSPI
 	tristate "Cadence Quad SPI controller"
-	depends on OF && (ARM || ARM64 || COMPILE_TEST)
+	depends on OF && (ARM || ARM64 || X86 || COMPILE_TEST)
 	help
 	  Enable support for the Cadence Quad SPI Flash controller.
 
@@ -292,13 +292,6 @@ config SPI_DLN2
 	 This driver can also be built as a module.  If so, the module
 	 will be called spi-dln2.
 
-config SPI_EFM32
-	tristate "EFM32 SPI controller"
-	depends on OF && ARM && (ARCH_EFM32 || COMPILE_TEST)
-	select SPI_BITBANG
-	help
-	  Driver for the spi controller found on Energy Micro's EFM32 SoCs.
-
 config SPI_EP93XX
 	tristate "Cirrus Logic EP93xx SPI controller"
 	depends on ARCH_EP93XX || COMPILE_TEST
@@ -650,7 +643,7 @@ config SPI_RPCIF
 	tristate "Renesas RPC-IF SPI driver"
 	depends on RENESAS_RPCIF
 	help
-	  SPI driver for Renesas R-Car Gen3 RPC-IF.
+	  SPI driver for Renesas R-Car Gen3 or RZ/G2 RPC-IF.
 
 config SPI_RSPI
 	tristate "Renesas RSPI/QSPI controller"
@@ -751,13 +744,6 @@ config SPI_SIFIVE
 	help
 	  This exposes the SPI controller IP from SiFive.
 
-config SPI_SIRF
-	tristate "CSR SiRFprimaII SPI controller"
-	depends on SIRF_DMA
-	select SPI_BITBANG
-	help
-	  SPI driver for CSR SiRFprimaII SoCs
-
 config SPI_SLAVE_MT27XX
 	tristate "MediaTek SPI slave device"
 	depends on ARCH_MEDIATEK || COMPILE_TEST
@@ -843,6 +829,15 @@ config SPI_MXS
 	help
 	  SPI driver for Freescale MXS devices.
 
+config SPI_TEGRA210_QUAD
+	tristate "NVIDIA Tegra QSPI Controller"
+	depends on ARCH_TEGRA || COMPILE_TEST
+	depends on RESET_CONTROLLER
+	help
+	  QSPI driver for NVIDIA Tegra QSPI Controller interface. This
+	  controller is different from the SPI controller and is available
+	  on Tegra SoCs starting from Tegra210.
+
 config SPI_TEGRA114
 	tristate "NVIDIA Tegra114 SPI Controller"
 	depends on (ARCH_TEGRA && TEGRA20_APB_DMA) || COMPILE_TEST
@@ -885,12 +880,6 @@ config SPI_TOPCLIFF_PCH
 	  This driver also supports the ML7213/ML7223/ML7831, a companion chip
 	  for the Atom E6xx series and compatible with the Intel EG20T PCH.
 
-config SPI_TXX9
-	tristate "Toshiba TXx9 SPI controller"
-	depends on GPIOLIB && (CPU_TX49XX || COMPILE_TEST)
-	help
-	  SPI driver for Toshiba TXx9 MIPS SoCs
-
 config SPI_UNIPHIER
 	tristate "Socionext UniPhier SPI Controller"
 	depends on (ARCH_UNIPHIER || COMPILE_TEST) && OF
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 6fea5821662e..0f06fc0813c6 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -42,7 +42,6 @@ spi-dw-$(CONFIG_SPI_DW_DMA)		+= spi-dw-dma.o
 obj-$(CONFIG_SPI_DW_BT1)		+= spi-dw-bt1.o
 obj-$(CONFIG_SPI_DW_MMIO)		+= spi-dw-mmio.o
 obj-$(CONFIG_SPI_DW_PCI)		+= spi-dw-pci.o
-obj-$(CONFIG_SPI_EFM32)			+= spi-efm32.o
 obj-$(CONFIG_SPI_EP93XX)		+= spi-ep93xx.o
 obj-$(CONFIG_SPI_FALCON)		+= spi-falcon.o
 obj-$(CONFIG_SPI_FSI)			+= spi-fsi.o
@@ -94,6 +93,7 @@ obj-$(CONFIG_SPI_QCOM_QSPI)		+= spi-qcom-qspi.o
 obj-$(CONFIG_SPI_QUP)			+= spi-qup.o
 obj-$(CONFIG_SPI_ROCKCHIP)		+= spi-rockchip.o
 obj-$(CONFIG_SPI_RB4XX)			+= spi-rb4xx.o
+obj-$(CONFIG_MACH_REALTEK_RTL)		+= spi-realtek-rtl.o
 obj-$(CONFIG_SPI_RPCIF)			+= spi-rpc-if.o
 obj-$(CONFIG_SPI_RSPI)			+= spi-rspi.o
 obj-$(CONFIG_SPI_S3C24XX)		+= spi-s3c24xx-hw.o
@@ -105,7 +105,6 @@ obj-$(CONFIG_SPI_SH_HSPI)		+= spi-sh-hspi.o
 obj-$(CONFIG_SPI_SH_MSIOF)		+= spi-sh-msiof.o
 obj-$(CONFIG_SPI_SH_SCI)		+= spi-sh-sci.o
 obj-$(CONFIG_SPI_SIFIVE)		+= spi-sifive.o
-obj-$(CONFIG_SPI_SIRF)		+= spi-sirf.o
 obj-$(CONFIG_SPI_SLAVE_MT27XX)          += spi-slave-mt27xx.o
 obj-$(CONFIG_SPI_SPRD)			+= spi-sprd.o
 obj-$(CONFIG_SPI_SPRD_ADI)		+= spi-sprd-adi.o
@@ -115,6 +114,7 @@ obj-$(CONFIG_SPI_ST_SSC4)		+= spi-st-ssc4.o
 obj-$(CONFIG_SPI_SUN4I)			+= spi-sun4i.o
 obj-$(CONFIG_SPI_SUN6I)			+= spi-sun6i.o
 obj-$(CONFIG_SPI_SYNQUACER)		+= spi-synquacer.o
+obj-$(CONFIG_SPI_TEGRA210_QUAD)		+= spi-tegra210-quad.o
 obj-$(CONFIG_SPI_TEGRA114)		+= spi-tegra114.o
 obj-$(CONFIG_SPI_TEGRA20_SFLASH)	+= spi-tegra20-sflash.o
 obj-$(CONFIG_SPI_TEGRA20_SLINK)		+= spi-tegra20-slink.o
@@ -122,7 +122,6 @@ obj-$(CONFIG_SPI_TLE62X0)		+= spi-tle62x0.o
 spi-thunderx-objs			:= spi-cavium.o spi-cavium-thunderx.o
 obj-$(CONFIG_SPI_THUNDERX)		+= spi-thunderx.o
 obj-$(CONFIG_SPI_TOPCLIFF_PCH)		+= spi-topcliff-pch.o
-obj-$(CONFIG_SPI_TXX9)			+= spi-txx9.o
 obj-$(CONFIG_SPI_UNIPHIER)		+= spi-uniphier.o
 obj-$(CONFIG_SPI_XCOMM)		+= spi-xcomm.o
 obj-$(CONFIG_SPI_XILINX)		+= spi-xilinx.o
diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c
index 10bc5390ab91..95d4fa32c299 100644
--- a/drivers/spi/atmel-quadspi.c
+++ b/drivers/spi/atmel-quadspi.c
@@ -657,6 +657,7 @@ static int __maybe_unused atmel_qspi_suspend(struct device *dev)
 	struct spi_controller *ctrl = dev_get_drvdata(dev);
 	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
 
+	atmel_qspi_write(QSPI_CR_QSPIDIS, aq, QSPI_CR);
 	clk_disable_unprepare(aq->qspick);
 	clk_disable_unprepare(aq->pclk);
 
diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c
index 948396b382d7..f429436082af 100644
--- a/drivers/spi/spi-atmel.c
+++ b/drivers/spi/spi-atmel.c
@@ -1590,7 +1590,7 @@ static int atmel_spi_probe(struct platform_device *pdev)
 		if (ret == 0) {
 			as->use_dma = true;
 		} else if (ret == -EPROBE_DEFER) {
-			return ret;
+			goto out_unmap_regs;
 		}
 	} else if (as->caps.has_pdc_support) {
 		as->use_pdc = true;
diff --git a/drivers/spi/spi-au1550.c b/drivers/spi/spi-au1550.c
index dfb7196f4caf..4b59a1b1bf7e 100644
--- a/drivers/spi/spi-au1550.c
+++ b/drivers/spi/spi-au1550.c
@@ -26,7 +26,7 @@
 
 #include <asm/mach-au1x00/au1550_spi.h>
 
-static unsigned usedma = 1;
+static unsigned int usedma = 1;
 module_param(usedma, uint, 0644);
 
 /*
@@ -43,9 +43,9 @@ struct au1550_spi {
 	volatile psc_spi_t __iomem *regs;
 	int irq;
 
-	unsigned len;
-	unsigned tx_count;
-	unsigned rx_count;
+	unsigned int len;
+	unsigned int tx_count;
+	unsigned int rx_count;
 	const u8 *tx;
 	u8 *rx;
 
@@ -56,14 +56,14 @@ struct au1550_spi {
 
 	struct completion master_done;
 
-	unsigned usedma;
+	unsigned int usedma;
 	u32 dma_tx_id;
 	u32 dma_rx_id;
 	u32 dma_tx_ch;
 	u32 dma_rx_ch;
 
 	u8 *dma_rx_tmpbuf;
-	unsigned dma_rx_tmpbuf_size;
+	unsigned int dma_rx_tmpbuf_size;
 	u32 dma_rx_tmpbuf_addr;
 
 	struct spi_master *master;
@@ -74,8 +74,7 @@ struct au1550_spi {
 
 
 /* we use an 8-bit memory device for dma transfers to/from spi fifo */
-static dbdev_tab_t au1550_spi_mem_dbdev =
-{
+static dbdev_tab_t au1550_spi_mem_dbdev = {
 	.dev_id			= DBDMA_MEM_CHAN,
 	.dev_flags		= DEV_FLAGS_ANYUSE|DEV_FLAGS_SYNC,
 	.dev_tsize		= 0,
@@ -99,7 +98,7 @@ static void au1550_spi_bits_handlers_set(struct au1550_spi *hw, int bpw);
  *    BRG valid range is 4..63
  *    DIV valid range is 0..3
  */
-static u32 au1550_spi_baudcfg(struct au1550_spi *hw, unsigned speed_hz)
+static u32 au1550_spi_baudcfg(struct au1550_spi *hw, unsigned int speed_hz)
 {
 	u32 mainclk_hz = hw->pdata->mainclk_hz;
 	u32 div, brg;
@@ -161,7 +160,7 @@ static void au1550_spi_reset_fifos(struct au1550_spi *hw)
 static void au1550_spi_chipsel(struct spi_device *spi, int value)
 {
 	struct au1550_spi *hw = spi_master_get_devdata(spi->master);
-	unsigned cspol = spi->mode & SPI_CS_HIGH ? 1 : 0;
+	unsigned int cspol = spi->mode & SPI_CS_HIGH ? 1 : 0;
 	u32 cfg, stat;
 
 	switch (value) {
@@ -221,7 +220,7 @@ static void au1550_spi_chipsel(struct spi_device *spi, int value)
 static int au1550_spi_setupxfer(struct spi_device *spi, struct spi_transfer *t)
 {
 	struct au1550_spi *hw = spi_master_get_devdata(spi->master);
-	unsigned bpw, hz;
+	unsigned int bpw, hz;
 	u32 cfg, stat;
 
 	if (t) {
@@ -276,7 +275,7 @@ static int au1550_spi_setupxfer(struct spi_device *spi, struct spi_transfer *t)
  * spi master done event irq is not generated unless rx fifo is empty (emptied)
  * so we need rx tmp buffer to use for rx dma if user does not provide one
  */
-static int au1550_spi_dma_rxtmp_alloc(struct au1550_spi *hw, unsigned size)
+static int au1550_spi_dma_rxtmp_alloc(struct au1550_spi *hw, unsigned int size)
 {
 	hw->dma_rx_tmpbuf = kmalloc(size, GFP_KERNEL);
 	if (!hw->dma_rx_tmpbuf)
@@ -399,10 +398,10 @@ static int au1550_spi_dma_txrxb(struct spi_device *spi, struct spi_transfer *t)
 			DMA_FROM_DEVICE);
 	}
 	/* unmap buffers if mapped above */
-	if (t->rx_buf && t->rx_dma == 0 )
+	if (t->rx_buf && t->rx_dma == 0)
 		dma_unmap_single(hw->dev, dma_rx_addr, t->len,
 			DMA_FROM_DEVICE);
-	if (t->tx_buf && t->tx_dma == 0 )
+	if (t->tx_buf && t->tx_dma == 0)
 		dma_unmap_single(hw->dev, dma_tx_addr, t->len,
 			DMA_TO_DEVICE);
 
@@ -447,8 +446,8 @@ static irqreturn_t au1550_spi_dma_irq_callback(struct au1550_spi *hw)
 				"dma transfer: receive FIFO overflow!\n");
 		else
 			dev_err(hw->dev,
-				"dma transfer: unexpected SPI error "
-				"(event=0x%x stat=0x%x)!\n", evnt, stat);
+				"dma transfer: unexpected SPI error (event=0x%x stat=0x%x)!\n",
+				evnt, stat);
 
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
@@ -493,12 +492,12 @@ static void au1550_spi_tx_word_##size(struct au1550_spi *hw)		\
 	wmb(); /* drain writebuffer */					\
 }
 
-AU1550_SPI_RX_WORD(8,0xff)
-AU1550_SPI_RX_WORD(16,0xffff)
-AU1550_SPI_RX_WORD(32,0xffffff)
-AU1550_SPI_TX_WORD(8,0xff)
-AU1550_SPI_TX_WORD(16,0xffff)
-AU1550_SPI_TX_WORD(32,0xffffff)
+AU1550_SPI_RX_WORD(8, 0xff)
+AU1550_SPI_RX_WORD(16, 0xffff)
+AU1550_SPI_RX_WORD(32, 0xffffff)
+AU1550_SPI_TX_WORD(8, 0xff)
+AU1550_SPI_TX_WORD(16, 0xffff)
+AU1550_SPI_TX_WORD(32, 0xffffff)
 
 static int au1550_spi_pio_txrxb(struct spi_device *spi, struct spi_transfer *t)
 {
@@ -567,8 +566,8 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 		au1550_spi_mask_ack_all(hw);
 		au1550_spi_reset_fifos(hw);
 		dev_err(hw->dev,
-			"pio transfer: unexpected SPI error "
-			"(event=0x%x stat=0x%x)!\n", evnt, stat);
+			"pio transfer: unexpected SPI error (event=0x%x stat=0x%x)!\n",
+			evnt, stat);
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
 	}
@@ -636,12 +635,14 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 static int au1550_spi_txrx_bufs(struct spi_device *spi, struct spi_transfer *t)
 {
 	struct au1550_spi *hw = spi_master_get_devdata(spi->master);
+
 	return hw->txrx_bufs(spi, t);
 }
 
 static irqreturn_t au1550_spi_irq(int irq, void *dev)
 {
 	struct au1550_spi *hw = dev;
+
 	return hw->irq_callback(hw);
 }
 
@@ -872,6 +873,7 @@ static int au1550_spi_probe(struct platform_device *pdev)
 	{
 		int min_div = (2 << 0) * (2 * (4 + 1));
 		int max_div = (2 << 3) * (2 * (63 + 1));
+
 		master->max_speed_hz = hw->pdata->mainclk_hz / min_div;
 		master->min_speed_hz =
 				hw->pdata->mainclk_hz / (max_div + 1) + 1;
@@ -972,8 +974,7 @@ static int __init au1550_spi_init(void)
 	if (usedma) {
 		ddma_memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev);
 		if (!ddma_memid)
-			printk(KERN_ERR "au1550-spi: cannot add memory"
-					"dbdma device\n");
+			printk(KERN_ERR "au1550-spi: cannot add memory dbdma device\n");
 	}
 	return platform_driver_register(&au1550_spi_drv);
 }
diff --git a/drivers/spi/spi-bcm-qspi.c b/drivers/spi/spi-bcm-qspi.c
index c028446c7460..707fe3a5d8ef 100644
--- a/drivers/spi/spi-bcm-qspi.c
+++ b/drivers/spi/spi-bcm-qspi.c
@@ -881,7 +881,7 @@ static int bcm_qspi_bspi_exec_mem_op(struct spi_device *spi,
 	 * when using flex mode we need to send
 	 * the upper address byte to bspi
 	 */
-	if (bcm_qspi_bspi_ver_three(qspi) == false) {
+	if (!bcm_qspi_bspi_ver_three(qspi)) {
 		addr = from & 0xff000000;
 		bcm_qspi_write(qspi, BSPI,
 			       BSPI_BSPI_FLASH_UPPER_ADDR_BYTE, addr);
diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c
index 197485f2c2b2..8965fe61c8b4 100644
--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -386,7 +386,7 @@ static irqreturn_t bcm2835_spi_interrupt(int irq, void *dev_id)
 		/* Transfer complete - reset SPI HW */
 		bcm2835_spi_reset_hw(bs);
 		/* wake up the framework */
-		complete(&bs->ctlr->xfer_completion);
+		spi_finalize_current_transfer(bs->ctlr);
 	}
 
 	return IRQ_HANDLED;
@@ -608,7 +608,7 @@ static void bcm2835_spi_dma_rx_done(void *data)
 	bcm2835_spi_reset_hw(bs);
 
 	/* and mark as completed */;
-	complete(&ctlr->xfer_completion);
+	spi_finalize_current_transfer(ctlr);
 }
 
 /**
@@ -640,7 +640,7 @@ static void bcm2835_spi_dma_tx_done(void *data)
 
 	bcm2835_spi_undo_prologue(bs);
 	bcm2835_spi_reset_hw(bs);
-	complete(&ctlr->xfer_completion);
+	spi_finalize_current_transfer(ctlr);
 }
 
 /**
@@ -1307,6 +1307,8 @@ static int bcm2835_spi_probe(struct platform_device *pdev)
 		return dev_err_probe(&pdev->dev, PTR_ERR(bs->clk),
 				     "could not get clk\n");
 
+	ctlr->max_speed_hz = clk_get_rate(bs->clk) / 2;
+
 	bs->irq = platform_get_irq(pdev, 0);
 	if (bs->irq <= 0)
 		return bs->irq ? bs->irq : -ENODEV;
diff --git a/drivers/spi/spi-bcm2835aux.c b/drivers/spi/spi-bcm2835aux.c
index 1a26865c42f8..75589ac6e95f 100644
--- a/drivers/spi/spi-bcm2835aux.c
+++ b/drivers/spi/spi-bcm2835aux.c
@@ -254,7 +254,7 @@ static irqreturn_t bcm2835aux_spi_interrupt(int irq, void *dev_id)
 	/* and if rx_len is 0 then disable interrupts and wake up completion */
 	if (!bs->rx_len) {
 		bcm2835aux_wr(bs, BCM2835_AUX_SPI_CNTL1, bs->cntl[1]);
-		complete(&master->xfer_completion);
+		spi_finalize_current_transfer(master);
 	}
 
 	return IRQ_HANDLED;
diff --git a/drivers/spi/spi-cadence-quadspi.c b/drivers/spi/spi-cadence-quadspi.c
index ba7d40c2922f..442cc7c53a47 100644
--- a/drivers/spi/spi-cadence-quadspi.c
+++ b/drivers/spi/spi-cadence-quadspi.c
@@ -52,6 +52,7 @@ struct cqspi_flash_pdata {
 	u8		inst_width;
 	u8		addr_width;
 	u8		data_width;
+	bool		dtr;
 	u8		cs;
 };
 
@@ -75,6 +76,7 @@ struct cqspi_st {
 	bool			is_decoded_cs;
 	u32			fifo_depth;
 	u32			fifo_width;
+	u32			num_chipselect;
 	bool			rclk_en;
 	u32			trigger_address;
 	u32			wr_delay;
@@ -111,6 +113,8 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_CONFIG_CHIPSELECT_LSB		10
 #define CQSPI_REG_CONFIG_DMA_MASK		BIT(15)
 #define CQSPI_REG_CONFIG_BAUD_LSB		19
+#define CQSPI_REG_CONFIG_DTR_PROTO		BIT(24)
+#define CQSPI_REG_CONFIG_DUAL_OPCODE		BIT(30)
 #define CQSPI_REG_CONFIG_IDLE_LSB		31
 #define CQSPI_REG_CONFIG_CHIPSELECT_MASK	0xF
 #define CQSPI_REG_CONFIG_BAUD_MASK		0xF
@@ -173,6 +177,9 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_SDRAMLEVEL_RD_MASK		0xFFFF
 #define CQSPI_REG_SDRAMLEVEL_WR_MASK		0xFFFF
 
+#define CQSPI_REG_WR_COMPLETION_CTRL		0x38
+#define CQSPI_REG_WR_DISABLE_AUTO_POLL		BIT(14)
+
 #define CQSPI_REG_IRQSTATUS			0x40
 #define CQSPI_REG_IRQMASK			0x44
 
@@ -188,6 +195,7 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_CMDCTRL			0x90
 #define CQSPI_REG_CMDCTRL_EXECUTE_MASK		BIT(0)
 #define CQSPI_REG_CMDCTRL_INPROGRESS_MASK	BIT(1)
+#define CQSPI_REG_CMDCTRL_DUMMY_LSB		7
 #define CQSPI_REG_CMDCTRL_WR_BYTES_LSB		12
 #define CQSPI_REG_CMDCTRL_WR_EN_LSB		15
 #define CQSPI_REG_CMDCTRL_ADD_BYTES_LSB		16
@@ -198,6 +206,7 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_CMDCTRL_WR_BYTES_MASK		0x7
 #define CQSPI_REG_CMDCTRL_ADD_BYTES_MASK	0x3
 #define CQSPI_REG_CMDCTRL_RD_BYTES_MASK		0x7
+#define CQSPI_REG_CMDCTRL_DUMMY_MASK		0x1F
 
 #define CQSPI_REG_INDIRECTWR			0x70
 #define CQSPI_REG_INDIRECTWR_START_MASK		BIT(0)
@@ -214,6 +223,14 @@ struct cqspi_driver_platdata {
 #define CQSPI_REG_CMDWRITEDATALOWER		0xA8
 #define CQSPI_REG_CMDWRITEDATAUPPER		0xAC
 
+#define CQSPI_REG_POLLING_STATUS		0xB0
+#define CQSPI_REG_POLLING_STATUS_DUMMY_LSB	16
+
+#define CQSPI_REG_OP_EXT_LOWER			0xE0
+#define CQSPI_REG_OP_EXT_READ_LSB		24
+#define CQSPI_REG_OP_EXT_WRITE_LSB		16
+#define CQSPI_REG_OP_EXT_STIG_LSB		0
+
 /* Interrupt status bits */
 #define CQSPI_REG_IRQ_MODE_ERR			BIT(0)
 #define CQSPI_REG_IRQ_UNDERFLOW			BIT(1)
@@ -288,6 +305,80 @@ static unsigned int cqspi_calc_rdreg(struct cqspi_flash_pdata *f_pdata)
 	return rdreg;
 }
 
+static unsigned int cqspi_calc_dummy(const struct spi_mem_op *op, bool dtr)
+{
+	unsigned int dummy_clk;
+
+	dummy_clk = op->dummy.nbytes * (8 / op->dummy.buswidth);
+	if (dtr)
+		dummy_clk /= 2;
+
+	return dummy_clk;
+}
+
+static int cqspi_set_protocol(struct cqspi_flash_pdata *f_pdata,
+			      const struct spi_mem_op *op)
+{
+	f_pdata->inst_width = CQSPI_INST_TYPE_SINGLE;
+	f_pdata->addr_width = CQSPI_INST_TYPE_SINGLE;
+	f_pdata->data_width = CQSPI_INST_TYPE_SINGLE;
+	f_pdata->dtr = op->data.dtr && op->cmd.dtr && op->addr.dtr;
+
+	switch (op->data.buswidth) {
+	case 0:
+		break;
+	case 1:
+		f_pdata->data_width = CQSPI_INST_TYPE_SINGLE;
+		break;
+	case 2:
+		f_pdata->data_width = CQSPI_INST_TYPE_DUAL;
+		break;
+	case 4:
+		f_pdata->data_width = CQSPI_INST_TYPE_QUAD;
+		break;
+	case 8:
+		f_pdata->data_width = CQSPI_INST_TYPE_OCTAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* Right now we only support 8-8-8 DTR mode. */
+	if (f_pdata->dtr) {
+		switch (op->cmd.buswidth) {
+		case 0:
+			break;
+		case 8:
+			f_pdata->inst_width = CQSPI_INST_TYPE_OCTAL;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		switch (op->addr.buswidth) {
+		case 0:
+			break;
+		case 8:
+			f_pdata->addr_width = CQSPI_INST_TYPE_OCTAL;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		switch (op->data.buswidth) {
+		case 0:
+			break;
+		case 8:
+			f_pdata->data_width = CQSPI_INST_TYPE_OCTAL;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
 static int cqspi_wait_idle(struct cqspi_st *cqspi)
 {
 	const unsigned int poll_idle_retry = 3;
@@ -345,19 +436,85 @@ static int cqspi_exec_flash_cmd(struct cqspi_st *cqspi, unsigned int reg)
 	return cqspi_wait_idle(cqspi);
 }
 
+static int cqspi_setup_opcode_ext(struct cqspi_flash_pdata *f_pdata,
+				  const struct spi_mem_op *op,
+				  unsigned int shift)
+{
+	struct cqspi_st *cqspi = f_pdata->cqspi;
+	void __iomem *reg_base = cqspi->iobase;
+	unsigned int reg;
+	u8 ext;
+
+	if (op->cmd.nbytes != 2)
+		return -EINVAL;
+
+	/* Opcode extension is the LSB. */
+	ext = op->cmd.opcode & 0xff;
+
+	reg = readl(reg_base + CQSPI_REG_OP_EXT_LOWER);
+	reg &= ~(0xff << shift);
+	reg |= ext << shift;
+	writel(reg, reg_base + CQSPI_REG_OP_EXT_LOWER);
+
+	return 0;
+}
+
+static int cqspi_enable_dtr(struct cqspi_flash_pdata *f_pdata,
+			    const struct spi_mem_op *op, unsigned int shift,
+			    bool enable)
+{
+	struct cqspi_st *cqspi = f_pdata->cqspi;
+	void __iomem *reg_base = cqspi->iobase;
+	unsigned int reg;
+	int ret;
+
+	reg = readl(reg_base + CQSPI_REG_CONFIG);
+
+	/*
+	 * We enable dual byte opcode here. The callers have to set up the
+	 * extension opcode based on which type of operation it is.
+	 */
+	if (enable) {
+		reg |= CQSPI_REG_CONFIG_DTR_PROTO;
+		reg |= CQSPI_REG_CONFIG_DUAL_OPCODE;
+
+		/* Set up command opcode extension. */
+		ret = cqspi_setup_opcode_ext(f_pdata, op, shift);
+		if (ret)
+			return ret;
+	} else {
+		reg &= ~CQSPI_REG_CONFIG_DTR_PROTO;
+		reg &= ~CQSPI_REG_CONFIG_DUAL_OPCODE;
+	}
+
+	writel(reg, reg_base + CQSPI_REG_CONFIG);
+
+	return cqspi_wait_idle(cqspi);
+}
+
 static int cqspi_command_read(struct cqspi_flash_pdata *f_pdata,
 			      const struct spi_mem_op *op)
 {
 	struct cqspi_st *cqspi = f_pdata->cqspi;
 	void __iomem *reg_base = cqspi->iobase;
 	u8 *rxbuf = op->data.buf.in;
-	u8 opcode = op->cmd.opcode;
+	u8 opcode;
 	size_t n_rx = op->data.nbytes;
 	unsigned int rdreg;
 	unsigned int reg;
+	unsigned int dummy_clk;
 	size_t read_len;
 	int status;
 
+	status = cqspi_set_protocol(f_pdata, op);
+	if (status)
+		return status;
+
+	status = cqspi_enable_dtr(f_pdata, op, CQSPI_REG_OP_EXT_STIG_LSB,
+				  f_pdata->dtr);
+	if (status)
+		return status;
+
 	if (!n_rx || n_rx > CQSPI_STIG_DATA_LEN_MAX || !rxbuf) {
 		dev_err(&cqspi->pdev->dev,
 			"Invalid input argument, len %zu rxbuf 0x%p\n",
@@ -365,11 +522,24 @@ static int cqspi_command_read(struct cqspi_flash_pdata *f_pdata,
 		return -EINVAL;
 	}
 
+	if (f_pdata->dtr)
+		opcode = op->cmd.opcode >> 8;
+	else
+		opcode = op->cmd.opcode;
+
 	reg = opcode << CQSPI_REG_CMDCTRL_OPCODE_LSB;
 
 	rdreg = cqspi_calc_rdreg(f_pdata);
 	writel(rdreg, reg_base + CQSPI_REG_RD_INSTR);
 
+	dummy_clk = cqspi_calc_dummy(op, f_pdata->dtr);
+	if (dummy_clk > CQSPI_DUMMY_CLKS_MAX)
+		return -EOPNOTSUPP;
+
+	if (dummy_clk)
+		reg |= (dummy_clk & CQSPI_REG_CMDCTRL_DUMMY_MASK)
+		     << CQSPI_REG_CMDCTRL_DUMMY_LSB;
+
 	reg |= (0x1 << CQSPI_REG_CMDCTRL_RD_EN_LSB);
 
 	/* 0 means 1 byte. */
@@ -401,12 +571,22 @@ static int cqspi_command_write(struct cqspi_flash_pdata *f_pdata,
 {
 	struct cqspi_st *cqspi = f_pdata->cqspi;
 	void __iomem *reg_base = cqspi->iobase;
-	const u8 opcode = op->cmd.opcode;
+	u8 opcode;
 	const u8 *txbuf = op->data.buf.out;
 	size_t n_tx = op->data.nbytes;
 	unsigned int reg;
 	unsigned int data;
 	size_t write_len;
+	int ret;
+
+	ret = cqspi_set_protocol(f_pdata, op);
+	if (ret)
+		return ret;
+
+	ret = cqspi_enable_dtr(f_pdata, op, CQSPI_REG_OP_EXT_STIG_LSB,
+			       f_pdata->dtr);
+	if (ret)
+		return ret;
 
 	if (n_tx > CQSPI_STIG_DATA_LEN_MAX || (n_tx && !txbuf)) {
 		dev_err(&cqspi->pdev->dev,
@@ -415,6 +595,14 @@ static int cqspi_command_write(struct cqspi_flash_pdata *f_pdata,
 		return -EINVAL;
 	}
 
+	reg = cqspi_calc_rdreg(f_pdata);
+	writel(reg, reg_base + CQSPI_REG_RD_INSTR);
+
+	if (f_pdata->dtr)
+		opcode = op->cmd.opcode >> 8;
+	else
+		opcode = op->cmd.opcode;
+
 	reg = opcode << CQSPI_REG_CMDCTRL_OPCODE_LSB;
 
 	if (op->addr.nbytes) {
@@ -454,14 +642,27 @@ static int cqspi_read_setup(struct cqspi_flash_pdata *f_pdata,
 	void __iomem *reg_base = cqspi->iobase;
 	unsigned int dummy_clk = 0;
 	unsigned int reg;
+	int ret;
+	u8 opcode;
+
+	ret = cqspi_enable_dtr(f_pdata, op, CQSPI_REG_OP_EXT_READ_LSB,
+			       f_pdata->dtr);
+	if (ret)
+		return ret;
 
-	reg = op->cmd.opcode << CQSPI_REG_RD_INSTR_OPCODE_LSB;
+	if (f_pdata->dtr)
+		opcode = op->cmd.opcode >> 8;
+	else
+		opcode = op->cmd.opcode;
+
+	reg = opcode << CQSPI_REG_RD_INSTR_OPCODE_LSB;
 	reg |= cqspi_calc_rdreg(f_pdata);
 
 	/* Setup dummy clock cycles */
-	dummy_clk = op->dummy.nbytes * 8;
+	dummy_clk = cqspi_calc_dummy(op, f_pdata->dtr);
+
 	if (dummy_clk > CQSPI_DUMMY_CLKS_MAX)
-		dummy_clk = CQSPI_DUMMY_CLKS_MAX;
+		return -EOPNOTSUPP;
 
 	if (dummy_clk)
 		reg |= (dummy_clk & CQSPI_REG_RD_INSTR_DUMMY_MASK)
@@ -573,15 +774,43 @@ static int cqspi_write_setup(struct cqspi_flash_pdata *f_pdata,
 			     const struct spi_mem_op *op)
 {
 	unsigned int reg;
+	int ret;
 	struct cqspi_st *cqspi = f_pdata->cqspi;
 	void __iomem *reg_base = cqspi->iobase;
+	u8 opcode;
+
+	ret = cqspi_enable_dtr(f_pdata, op, CQSPI_REG_OP_EXT_WRITE_LSB,
+			       f_pdata->dtr);
+	if (ret)
+		return ret;
+
+	if (f_pdata->dtr)
+		opcode = op->cmd.opcode >> 8;
+	else
+		opcode = op->cmd.opcode;
 
 	/* Set opcode. */
-	reg = op->cmd.opcode << CQSPI_REG_WR_INSTR_OPCODE_LSB;
+	reg = opcode << CQSPI_REG_WR_INSTR_OPCODE_LSB;
+	reg |= f_pdata->data_width << CQSPI_REG_WR_INSTR_TYPE_DATA_LSB;
+	reg |= f_pdata->addr_width << CQSPI_REG_WR_INSTR_TYPE_ADDR_LSB;
 	writel(reg, reg_base + CQSPI_REG_WR_INSTR);
 	reg = cqspi_calc_rdreg(f_pdata);
 	writel(reg, reg_base + CQSPI_REG_RD_INSTR);
 
+	if (f_pdata->dtr) {
+		/*
+		 * Some flashes like the cypress Semper flash expect a 4-byte
+		 * dummy address with the Read SR command in DTR mode, but this
+		 * controller does not support sending address with the Read SR
+		 * command. So, disable write completion polling on the
+		 * controller's side. spi-nor will take care of polling the
+		 * status register.
+		 */
+		reg = readl(reg_base + CQSPI_REG_WR_COMPLETION_CTRL);
+		reg |= CQSPI_REG_WR_DISABLE_AUTO_POLL;
+		writel(reg, reg_base + CQSPI_REG_WR_COMPLETION_CTRL);
+	}
+
 	reg = readl(reg_base + CQSPI_REG_SIZE);
 	reg &= ~CQSPI_REG_SIZE_ADDRESS_MASK;
 	reg |= (op->addr.nbytes - 1);
@@ -835,35 +1064,6 @@ static void cqspi_configure(struct cqspi_flash_pdata *f_pdata,
 		cqspi_controller_enable(cqspi, 1);
 }
 
-static int cqspi_set_protocol(struct cqspi_flash_pdata *f_pdata,
-			      const struct spi_mem_op *op)
-{
-	f_pdata->inst_width = CQSPI_INST_TYPE_SINGLE;
-	f_pdata->addr_width = CQSPI_INST_TYPE_SINGLE;
-	f_pdata->data_width = CQSPI_INST_TYPE_SINGLE;
-
-	if (op->data.dir == SPI_MEM_DATA_IN) {
-		switch (op->data.buswidth) {
-		case 1:
-			f_pdata->data_width = CQSPI_INST_TYPE_SINGLE;
-			break;
-		case 2:
-			f_pdata->data_width = CQSPI_INST_TYPE_DUAL;
-			break;
-		case 4:
-			f_pdata->data_width = CQSPI_INST_TYPE_QUAD;
-			break;
-		case 8:
-			f_pdata->data_width = CQSPI_INST_TYPE_OCTAL;
-			break;
-		default:
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
 static ssize_t cqspi_write(struct cqspi_flash_pdata *f_pdata,
 			   const struct spi_mem_op *op)
 {
@@ -881,7 +1081,16 @@ static ssize_t cqspi_write(struct cqspi_flash_pdata *f_pdata,
 	if (ret)
 		return ret;
 
-	if (cqspi->use_direct_mode && ((to + len) <= cqspi->ahb_size)) {
+	/*
+	 * Some flashes like the Cypress Semper flash expect a dummy 4-byte
+	 * address (all 0s) with the read status register command in DTR mode.
+	 * But this controller does not support sending dummy address bytes to
+	 * the flash when it is polling the write completion register in DTR
+	 * mode. So, we can not use direct mode when in DTR mode for writing
+	 * data.
+	 */
+	if (!f_pdata->dtr && cqspi->use_direct_mode &&
+	    ((to + len) <= cqspi->ahb_size)) {
 		memcpy_toio(cqspi->ahb_base + to, buf, len);
 		return cqspi_wait_idle(cqspi);
 	}
@@ -942,7 +1151,7 @@ static int cqspi_direct_read_execute(struct cqspi_flash_pdata *f_pdata,
 
 	dma_async_issue_pending(cqspi->rx_chan);
 	if (!wait_for_completion_timeout(&cqspi->rx_dma_complete,
-					 msecs_to_j