Merge tag 'spi-for-linus' of git://git.secretlab.ca/git/linux-2.6

Pull SPI changes for v3.4 from Grant Likely:
 "Mostly a bunch of new drivers and driver bug fixes; but this also
  includes a few patches that create a core message queue infrastructure
  for the spi subsystem instead of making each driver open code it."

* tag 'spi-for-linus' of git://git.secretlab.ca/git/linux-2.6: (34 commits)
  spi/fsl-espi: Make sure pm is within 2..32
  spi/fsl-espi: make the clock computation easier to read
  spi: sh-hspi: modify write/read method
  spi: sh-hspi: control spi clock more correctly
  spi: sh-hspi: convert to using core message queue
  spi: s3c64xx: Fix build
  spi: s3c64xx: remove unnecessary callback msg->complete
  spi: remove redundant variable assignment
  spi: release lock on error path in spi_pump_messages()
  spi: Compatibility with direction which is used in samsung DMA operation
  spi-topcliff-pch: add recovery processing in case wait-event timeout
  spi-topcliff-pch: supports a spi mode setup and bit order setup by IO control
  spi-topcliff-pch: Fix issue for transmitting over 4KByte
  spi-topcliff-pch: Modify pci-bus number dynamically to get DMA device info
  spi/imx: simplify error handling to free gpios
  spi: Convert to DEFINE_PCI_DEVICE_TABLE
  spi: add Broadcom BCM63xx SPI controller driver
  SPI: add CSR SiRFprimaII SPI controller driver
  spi-topcliff-pch: fix -Wuninitialized warning
  spi: Mark spi_register_board_info() __devinit
  ...

22 files changed, 2890 insertions, 422 deletions

diff --git a/Documentation/devicetree/bindings/spi/omap-spi.txt b/Documentation/devicetree/bindings/spi/omap-spi.txt
new file mode 100644
index 000000000000..81df374adbb9
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/omap-spi.txt
@@ -0,0 +1,20 @@
+OMAP2+ McSPI device
+
+Required properties:
+- compatible :
+  - "ti,omap2-spi" for OMAP2 & OMAP3.
+  - "ti,omap4-spi" for OMAP4+.
+- ti,spi-num-cs : Number of chipselect supported  by the instance.
+- ti,hwmods: Name of the hwmod associated to the McSPI
+
+
+Example:
+
+mcspi1: mcspi@1 {
+    #address-cells = <1>;
+    #size-cells = <0>;
+    compatible = "ti,omap4-mcspi";
+    ti,hwmods = "mcspi1";
+    ti,spi-num-cs = <4>;
+};
+
diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary
index 4884cb33845d..7312ec14dd89 100644
--- a/Documentation/spi/spi-summary
+++ b/Documentation/spi/spi-summary
@@ -1,7 +1,7 @@
 Overview of Linux kernel SPI support
 ====================================
 
-21-May-2007
+02-Feb-2012
 
 What is SPI?
 ------------
@@ -483,9 +483,9 @@ also initialize its own internal state.  (See below about bus numbering
 and those methods.)
 
 After you initialize the spi_master, then use spi_register_master() to
-publish it to the rest of the system.  At that time, device nodes for
-the controller and any predeclared spi devices will be made available,
-and the driver model core will take care of binding them to drivers.
+publish it to the rest of the system. At that time, device nodes for the
+controller and any predeclared spi devices will be made available, and
+the driver model core will take care of binding them to drivers.
 
 If you need to remove your SPI controller driver, spi_unregister_master()
 will reverse the effect of spi_register_master().
@@ -521,21 +521,53 @@ SPI MASTER METHODS
 		** When you code setup(), ASSUME that the controller
 		** is actively processing transfers for another device.
 
-    master->transfer(struct spi_device *spi, struct spi_message *message)
-    	This must not sleep.  Its responsibility is arrange that the
-	transfer happens and its complete() callback is issued.  The two
-	will normally happen later, after other transfers complete, and
-	if the controller is idle it will need to be kickstarted.
-
     master->cleanup(struct spi_device *spi)
 	Your controller driver may use spi_device.controller_state to hold
 	state it dynamically associates with that device.  If you do that,
 	be sure to provide the cleanup() method to free that state.
 
+    master->prepare_transfer_hardware(struct spi_master *master)
+	This will be called by the queue mechanism to signal to the driver
+	that a message is coming in soon, so the subsystem requests the
+	driver to prepare the transfer hardware by issuing this call.
+	This may sleep.
+
+    master->unprepare_transfer_hardware(struct spi_master *master)
+	This will be called by the queue mechanism to signal to the driver
+	that there are no more messages pending in the queue and it may
+	relax the hardware (e.g. by power management calls). This may sleep.
+
+    master->transfer_one_message(struct spi_master *master,
+				 struct spi_message *mesg)
+	The subsystem calls the driver to transfer a single message while
+	queuing transfers that arrive in the meantime. When the driver is
+	finished with this message, it must call
+	spi_finalize_current_message() so the subsystem can issue the next
+	transfer. This may sleep.
+
+    DEPRECATED METHODS
+
+    master->transfer(struct spi_device *spi, struct spi_message *message)
+	This must not sleep. Its responsibility is arrange that the
+	transfer happens and its complete() callback is issued. The two
+	will normally happen later, after other transfers complete, and
+	if the controller is idle it will need to be kickstarted. This
+	method is not used on queued controllers and must be NULL if
+	transfer_one_message() and (un)prepare_transfer_hardware() are
+	implemented.
+
 
 SPI MESSAGE QUEUE
 
-The bulk of the driver will be managing the I/O queue fed by transfer().
+If you are happy with the standard queueing mechanism provided by the
+SPI subsystem, just implement the queued methods specified above. Using
+the message queue has the upside of centralizing a lot of code and
+providing pure process-context execution of methods. The message queue
+can also be elevated to realtime priority on high-priority SPI traffic.
+
+Unless the queueing mechanism in the SPI subsystem is selected, the bulk
+of the driver will be managing the I/O queue fed by the now deprecated
+function transfer().
 
 That queue could be purely conceptual.  For example, a driver used only
 for low-frequency sensor access might be fine using synchronous PIO.
@@ -561,4 +593,6 @@ Stephen Street
 Mark Underwood
 Andrew Victor
 Vitaly Wool
-
+Grant Likely
+Mark Brown
+Linus Walleij
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 8293658e7cf9..0b06e360628a 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -94,6 +94,12 @@ config SPI_AU1550
 	  If you say yes to this option, support will be included for the
 	  PSC SPI controller found on Au1550, Au1200 and Au1300 series.
 
+config SPI_BCM63XX
+	tristate "Broadcom BCM63xx SPI controller"
+	depends on BCM63XX
+	help
+          Enable support for the SPI controller on the Broadcom BCM63xx SoCs.
+
 config SPI_BITBANG
 	tristate "Utilities for Bitbanging SPI masters"
 	help
@@ -126,7 +132,7 @@ config SPI_COLDFIRE_QSPI
 
 config SPI_DAVINCI
 	tristate "Texas Instruments DaVinci/DA8x/OMAP-L/AM1x SoC SPI controller"
-	depends on SPI_MASTER && ARCH_DAVINCI
+	depends on ARCH_DAVINCI
 	select SPI_BITBANG
 	help
 	  SPI master controller for DaVinci/DA8x/OMAP-L/AM1x SPI modules.
@@ -188,7 +194,7 @@ config SPI_MPC52xx_PSC
 
 config SPI_MPC512x_PSC
 	tristate "Freescale MPC512x PSC SPI controller"
-	depends on SPI_MASTER && PPC_MPC512x
+	depends on PPC_MPC512x
 	help
 	  This enables using the Freescale MPC5121 Programmable Serial
 	  Controller in SPI master mode.
@@ -238,7 +244,7 @@ config SPI_OMAP24XX
 
 config SPI_OMAP_100K
 	tristate "OMAP SPI 100K"
-	depends on SPI_MASTER && (ARCH_OMAP850 || ARCH_OMAP730)
+	depends on ARCH_OMAP850 || ARCH_OMAP730
 	help
 	  OMAP SPI 100K master controller for omap7xx boards.
 
@@ -262,7 +268,7 @@ config SPI_PL022
 
 config SPI_PPC4xx
 	tristate "PPC4xx SPI Controller"
-	depends on PPC32 && 4xx && SPI_MASTER
+	depends on PPC32 && 4xx
 	select SPI_BITBANG
 	help
 	  This selects a driver for the PPC4xx SPI Controller.
@@ -279,6 +285,12 @@ config SPI_PXA2XX
 config SPI_PXA2XX_PCI
 	def_bool SPI_PXA2XX && X86_32 && PCI
 
+config SPI_RSPI
+	tristate "Renesas RSPI controller"
+	depends on SUPERH
+	help
+	  SPI driver for Renesas RSPI blocks.
+
 config SPI_S3C24XX
 	tristate "Samsung S3C24XX series SPI"
 	depends on ARCH_S3C2410 && EXPERIMENTAL
@@ -324,9 +336,22 @@ config SPI_SH_SCI
 	help
 	  SPI driver for SuperH SCI blocks.
 
+config SPI_SH_HSPI
+	tristate "SuperH HSPI controller"
+	depends on ARCH_SHMOBILE
+	help
+	  SPI driver for SuperH HSPI blocks.
+
+config SPI_SIRF
+	tristate "CSR SiRFprimaII SPI controller"
+	depends on ARCH_PRIMA2
+	select SPI_BITBANG
+	help
+	  SPI driver for CSR SiRFprimaII SoCs
+
 config SPI_STMP3XXX
 	tristate "Freescale STMP37xx/378x SPI/SSP controller"
-	depends on ARCH_STMP3XXX && SPI_MASTER
+	depends on ARCH_STMP3XXX
 	help
 	  SPI driver for Freescale STMP37xx/378x SoC SSP interface
 
@@ -384,7 +409,6 @@ config SPI_NUC900
 
 config SPI_DESIGNWARE
 	tristate "DesignWare SPI controller core support"
-	depends on SPI_MASTER
 	help
 	  general driver for SPI controller core from DesignWare
 
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 61c3261c388c..a1d48e0ba3dc 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_SPI_ALTERA)		+= spi-altera.o
 obj-$(CONFIG_SPI_ATMEL)			+= spi-atmel.o
 obj-$(CONFIG_SPI_ATH79)			+= spi-ath79.o
 obj-$(CONFIG_SPI_AU1550)		+= spi-au1550.o
+obj-$(CONFIG_SPI_BCM63XX)		+= spi-bcm63xx.o
 obj-$(CONFIG_SPI_BFIN)			+= spi-bfin5xx.o
 obj-$(CONFIG_SPI_BFIN_SPORT)		+= spi-bfin-sport.o
 obj-$(CONFIG_SPI_BITBANG)		+= spi-bitbang.o
@@ -44,13 +45,16 @@ obj-$(CONFIG_SPI_PL022)			+= spi-pl022.o
 obj-$(CONFIG_SPI_PPC4xx)		+= spi-ppc4xx.o
 obj-$(CONFIG_SPI_PXA2XX)		+= spi-pxa2xx.o
 obj-$(CONFIG_SPI_PXA2XX_PCI)		+= spi-pxa2xx-pci.o
+obj-$(CONFIG_SPI_RSPI)			+= spi-rspi.o
 obj-$(CONFIG_SPI_S3C24XX)		+= spi-s3c24xx-hw.o
 spi-s3c24xx-hw-y			:= spi-s3c24xx.o
 spi-s3c24xx-hw-$(CONFIG_SPI_S3C24XX_FIQ) += spi-s3c24xx-fiq.o
 obj-$(CONFIG_SPI_S3C64XX)		+= spi-s3c64xx.o
 obj-$(CONFIG_SPI_SH)			+= spi-sh.o
+obj-$(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-$(CONFIG_SPI_SIRF)		+= spi-sirf.o
 obj-$(CONFIG_SPI_STMP3XXX)		+= spi-stmp.o
 obj-$(CONFIG_SPI_TEGRA)			+= spi-tegra.o
 obj-$(CONFIG_SPI_TI_SSP)		+= spi-ti-ssp.o
diff --git a/drivers/spi/spi-bcm63xx.c b/drivers/spi/spi-bcm63xx.c
new file mode 100644
index 000000000000..f01b2648452e
--- /dev/null
+++ b/drivers/spi/spi-bcm63xx.c
@@ -0,0 +1,486 @@
+/*
+ * Broadcom BCM63xx SPI controller support
+ *
+ * Copyright (C) 2009-2011 Florian Fainelli <florian@openwrt.org>
+ * Copyright (C) 2010 Tanguy Bouzeloc <tanguy.bouzeloc@efixo.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the
+ * Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/spi/spi.h>
+#include <linux/completion.h>
+#include <linux/err.h>
+
+#include <bcm63xx_dev_spi.h>
+
+#define PFX		KBUILD_MODNAME
+#define DRV_VER		"0.1.2"
+
+struct bcm63xx_spi {
+	spinlock_t		lock;
+	int			stopping;
+	struct completion	done;
+
+	void __iomem		*regs;
+	int			irq;
+
+	/* Platform data */
+	u32			speed_hz;
+	unsigned		fifo_size;
+
+	/* Data buffers */
+	const unsigned char	*tx_ptr;
+	unsigned char		*rx_ptr;
+
+	/* data iomem */
+	u8 __iomem		*tx_io;
+	const u8 __iomem	*rx_io;
+
+	int			remaining_bytes;
+
+	struct clk		*clk;
+	struct platform_device	*pdev;
+};
+
+static inline u8 bcm_spi_readb(struct bcm63xx_spi *bs,
+				unsigned int offset)
+{
+	return bcm_readb(bs->regs + bcm63xx_spireg(offset));
+}
+
+static inline u16 bcm_spi_readw(struct bcm63xx_spi *bs,
+				unsigned int offset)
+{
+	return bcm_readw(bs->regs + bcm63xx_spireg(offset));
+}
+
+static inline void bcm_spi_writeb(struct bcm63xx_spi *bs,
+				  u8 value, unsigned int offset)
+{
+	bcm_writeb(value, bs->regs + bcm63xx_spireg(offset));
+}
+
+static inline void bcm_spi_writew(struct bcm63xx_spi *bs,
+				  u16 value, unsigned int offset)
+{
+	bcm_writew(value, bs->regs + bcm63xx_spireg(offset));
+}
+
+static const unsigned bcm63xx_spi_freq_table[SPI_CLK_MASK][2] = {
+	{ 20000000, SPI_CLK_20MHZ },
+	{ 12500000, SPI_CLK_12_50MHZ },
+	{  6250000, SPI_CLK_6_250MHZ },
+	{  3125000, SPI_CLK_3_125MHZ },
+	{  1563000, SPI_CLK_1_563MHZ },
+	{   781000, SPI_CLK_0_781MHZ },
+	{   391000, SPI_CLK_0_391MHZ }
+};
+
+static int bcm63xx_spi_setup_transfer(struct spi_device *spi,
+				      struct spi_transfer *t)
+{
+	struct bcm63xx_spi *bs = spi_master_get_devdata(spi->master);
+	u8 bits_per_word;
+	u8 clk_cfg, reg;
+	u32 hz;
+	int i;
+
+	bits_per_word = (t) ? t->bits_per_word : spi->bits_per_word;
+	hz = (t) ? t->speed_hz : spi->max_speed_hz;
+	if (bits_per_word != 8) {
+		dev_err(&spi->dev, "%s, unsupported bits_per_word=%d\n",
+			__func__, bits_per_word);
+		return -EINVAL;
+	}
+
+	if (spi->chip_select > spi->master->num_chipselect) {
+		dev_err(&spi->dev, "%s, unsupported slave %d\n",
+			__func__, spi->chip_select);
+		return -EINVAL;
+	}
+
+	/* Find the closest clock configuration */
+	for (i = 0; i < SPI_CLK_MASK; i++) {
+		if (hz <= bcm63xx_spi_freq_table[i][0]) {
+			clk_cfg = bcm63xx_spi_freq_table[i][1];
+			break;
+		}
+	}
+
+	/* No matching configuration found, default to lowest */
+	if (i == SPI_CLK_MASK)
+		clk_cfg = SPI_CLK_0_391MHZ;
+
+	/* clear existing clock configuration bits of the register */
+	reg = bcm_spi_readb(bs, SPI_CLK_CFG);
+	reg &= ~SPI_CLK_MASK;
+	reg |= clk_cfg;
+
+	bcm_spi_writeb(bs, reg, SPI_CLK_CFG);
+	dev_dbg(&spi->dev, "Setting clock register to %02x (hz %d)\n",
+		clk_cfg, hz);
+
+	return 0;
+}
+
+/* the spi->mode bits understood by this driver: */
+#define MODEBITS (SPI_CPOL | SPI_CPHA)
+
+static int bcm63xx_spi_setup(struct spi_device *spi)
+{
+	struct bcm63xx_spi *bs;
+	int ret;
+
+	bs = spi_master_get_devdata(spi->master);
+
+	if (bs->stopping)
+		return -ESHUTDOWN;
+
+	if (!spi->bits_per_word)
+		spi->bits_per_word = 8;
+
+	if (spi->mode & ~MODEBITS) {
+		dev_err(&spi->dev, "%s, unsupported mode bits %x\n",
+			__func__, spi->mode & ~MODEBITS);
+		return -EINVAL;
+	}
+
+	ret = bcm63xx_spi_setup_transfer(spi, NULL);
+	if (ret < 0) {
+		dev_err(&spi->dev, "setup: unsupported mode bits %x\n",
+			spi->mode & ~MODEBITS);
+		return ret;
+	}
+
+	dev_dbg(&spi->dev, "%s, mode %d, %u bits/w, %u nsec/bit\n",
+		__func__, spi->mode & MODEBITS, spi->bits_per_word, 0);
+
+	return 0;
+}
+
+/* Fill the TX FIFO with as many bytes as possible */
+static void bcm63xx_spi_fill_tx_fifo(struct bcm63xx_spi *bs)
+{
+	u8 size;
+
+	/* Fill the Tx FIFO with as many bytes as possible */
+	size = bs->remaining_bytes < bs->fifo_size ? bs->remaining_bytes :
+		bs->fifo_size;
+	memcpy_toio(bs->tx_io, bs->tx_ptr, size);
+	bs->remaining_bytes -= size;
+}
+
+static int bcm63xx_txrx_bufs(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct bcm63xx_spi *bs = spi_master_get_devdata(spi->master);
+	u16 msg_ctl;
+	u16 cmd;
+
+	dev_dbg(&spi->dev, "txrx: tx %p, rx %p, len %d\n",
+		t->tx_buf, t->rx_buf, t->len);
+
+	/* Transmitter is inhibited */
+	bs->tx_ptr = t->tx_buf;
+	bs->rx_ptr = t->rx_buf;
+	init_completion(&bs->done);
+
+	if (t->tx_buf) {
+		bs->remaining_bytes = t->len;
+		bcm63xx_spi_fill_tx_fifo(bs);
+	}
+
+	/* Enable the command done interrupt which
+	 * we use to determine completion of a command */
+	bcm_spi_writeb(bs, SPI_INTR_CMD_DONE, SPI_INT_MASK);
+
+	/* Fill in the Message control register */
+	msg_ctl = (t->len << SPI_BYTE_CNT_SHIFT);
+
+	if (t->rx_buf && t->tx_buf)
+		msg_ctl |= (SPI_FD_RW << SPI_MSG_TYPE_SHIFT);
+	else if (t->rx_buf)
+		msg_ctl |= (SPI_HD_R << SPI_MSG_TYPE_SHIFT);
+	else if (t->tx_buf)
+		msg_ctl |= (SPI_HD_W << SPI_MSG_TYPE_SHIFT);
+
+	bcm_spi_writew(bs, msg_ctl, SPI_MSG_CTL);
+
+	/* Issue the transfer */
+	cmd = SPI_CMD_START_IMMEDIATE;
+	cmd |= (0 << SPI_CMD_PREPEND_BYTE_CNT_SHIFT);
+	cmd |= (spi->chip_select << SPI_CMD_DEVICE_ID_SHIFT);
+	bcm_spi_writew(bs, cmd, SPI_CMD);
+	wait_for_completion(&bs->done);
+
+	/* Disable the CMD_DONE interrupt */
+	bcm_spi_writeb(bs, 0, SPI_INT_MASK);
+
+	return t->len - bs->remaining_bytes;
+}
+
+static int bcm63xx_transfer(struct spi_device *spi, struct spi_message *m)
+{
+	struct bcm63xx_spi *bs = spi_master_get_devdata(spi->master);
+	struct spi_transfer *t;
+	int ret = 0;
+
+	if (unlikely(list_empty(&m->transfers)))
+		return -EINVAL;
+
+	if (bs->stopping)
+		return -ESHUTDOWN;
+
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		ret += bcm63xx_txrx_bufs(spi, t);
+	}
+
+	m->complete(m->context);
+
+	return ret;
+}
+
+/* This driver supports single master mode only. Hence
+ * CMD_DONE is the only interrupt we care about
+ */
+static irqreturn_t bcm63xx_spi_interrupt(int irq, void *dev_id)
+{
+	struct spi_master *master = (struct spi_master *)dev_id;
+	struct bcm63xx_spi *bs = spi_master_get_devdata(master);
+	u8 intr;
+	u16 cmd;
+
+	/* Read interupts and clear them immediately */
+	intr = bcm_spi_readb(bs, SPI_INT_STATUS);
+	bcm_spi_writeb(bs, SPI_INTR_CLEAR_ALL, SPI_INT_STATUS);
+	bcm_spi_writeb(bs, 0, SPI_INT_MASK);
+
+	/* A tansfer completed */
+	if (intr & SPI_INTR_CMD_DONE) {
+		u8 rx_tail;
+
+		rx_tail = bcm_spi_readb(bs, SPI_RX_TAIL);
+
+		/* Read out all the data */
+		if (rx_tail)
+			memcpy_fromio(bs->rx_ptr, bs->rx_io, rx_tail);
+
+		/* See if there is more data to send */
+		if (bs->remaining_bytes > 0) {
+			bcm63xx_spi_fill_tx_fifo(bs);
+
+			/* Start the transfer */
+			bcm_spi_writew(bs, SPI_HD_W << SPI_MSG_TYPE_SHIFT,
+				       SPI_MSG_CTL);
+			cmd = bcm_spi_readw(bs, SPI_CMD);
+			cmd |= SPI_CMD_START_IMMEDIATE;
+			cmd |= (0 << SPI_CMD_PREPEND_BYTE_CNT_SHIFT);
+			bcm_spi_writeb(bs, SPI_INTR_CMD_DONE, SPI_INT_MASK);
+			bcm_spi_writew(bs, cmd, SPI_CMD);
+		} else {
+			complete(&bs->done);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+
+static int __devinit bcm63xx_spi_probe(struct platform_device *pdev)
+{
+	struct resource *r;
+	struct device *dev = &pdev->dev;
+	struct bcm63xx_spi_pdata *pdata = pdev->dev.platform_data;
+	int irq;
+	struct spi_master *master;
+	struct clk *clk;
+	struct bcm63xx_spi *bs;
+	int ret;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!r) {
+		dev_err(dev, "no iomem\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(dev, "no irq\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	clk = clk_get(dev, "spi");
+	if (IS_ERR(clk)) {
+		dev_err(dev, "no clock for device\n");
+		ret = PTR_ERR(clk);
+		goto out;
+	}
+
+	master = spi_alloc_master(dev, sizeof(*bs));
+	if (!master) {
+		dev_err(dev, "out of memory\n");
+		ret = -ENOMEM;
+		goto out_clk;
+	}
+
+	bs = spi_master_get_devdata(master);
+	init_completion(&bs->done);
+
+	platform_set_drvdata(pdev, master);
+	bs->pdev = pdev;
+
+	if (!devm_request_mem_region(&pdev->dev, r->start,
+					resource_size(r), PFX)) {
+		dev_err(dev, "iomem request failed\n");
+		ret = -ENXIO;
+		goto out_err;
+	}
+
+	bs->regs = devm_ioremap_nocache(&pdev->dev, r->start,
+							resource_size(r));
+	if (!bs->regs) {
+		dev_err(dev, "unable to ioremap regs\n");
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	bs->irq = irq;
+	bs->clk = clk;
+	bs->fifo_size = pdata->fifo_size;
+
+	ret = devm_request_irq(&pdev->dev, irq, bcm63xx_spi_interrupt, 0,
+							pdev->name, master);
+	if (ret) {
+		dev_err(dev, "unable to request irq\n");
+		goto out_err;
+	}
+
+	master->bus_num = pdata->bus_num;
+	master->num_chipselect = pdata->num_chipselect;
+	master->setup = bcm63xx_spi_setup;
+	master->transfer = bcm63xx_transfer;
+	bs->speed_hz = pdata->speed_hz;
+	bs->stopping = 0;
+	bs->tx_io = (u8 *)(bs->regs + bcm63xx_spireg(SPI_MSG_DATA));
+	bs->rx_io = (const u8 *)(bs->regs + bcm63xx_spireg(SPI_RX_DATA));
+	spin_lock_init(&bs->lock);
+
+	/* Initialize hardware */
+	clk_enable(bs->clk);
+	bcm_spi_writeb(bs, SPI_INTR_CLEAR_ALL, SPI_INT_STATUS);
+
+	/* register and we are done */
+	ret = spi_register_master(master);
+	if (ret) {
+		dev_err(dev, "spi register failed\n");
+		goto out_clk_disable;
+	}
+
+	dev_info(dev, "at 0x%08x (irq %d, FIFOs size %d) v%s\n",
+		 r->start, irq, bs->fifo_size, DRV_VER);
+
+	return 0;
+
+out_clk_disable:
+	clk_disable(clk);
+out_err:
+	platform_set_drvdata(pdev, NULL);
+	spi_master_put(master);
+out_clk:
+	clk_put(clk);
+out:
+	return ret;
+}
+
+static int __devexit bcm63xx_spi_remove(struct platform_device *pdev)
+{
+	struct spi_master *master = platform_get_drvdata(pdev);
+	struct bcm63xx_spi *bs = spi_master_get_devdata(master);
+
+	/* reset spi block */
+	bcm_spi_writeb(bs, 0, SPI_INT_MASK);
+	spin_lock(&bs->lock);
+	bs->stopping = 1;
+
+	/* HW shutdown */
+	clk_disable(bs->clk);
+	clk_put(bs->clk);
+
+	spin_unlock(&bs->lock);
+	platform_set_drvdata(pdev, 0);
+	spi_unregister_master(master);
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int bcm63xx_spi_suspend(struct device *dev)
+{
+	struct spi_master *master =
+			platform_get_drvdata(to_platform_device(dev));
+	struct bcm63xx_spi *bs = spi_master_get_devdata(master);
+
+	clk_disable(bs->clk);
+
+	return 0;
+}
+
+static int bcm63xx_spi_resume(struct device *dev)
+{
+	struct spi_master *master =
+			platform_get_drvdata(to_platform_device(dev));
+	struct bcm63xx_spi *bs = spi_master_get_devdata(master);
+
+	clk_enable(bs->clk);
+
+	return 0;
+}
+
+static const struct dev_pm_ops bcm63xx_spi_pm_ops = {
+	.suspend	= bcm63xx_spi_suspend,
+	.resume		= bcm63xx_spi_resume,
+};
+
+#define BCM63XX_SPI_PM_OPS	(&bcm63xx_spi_pm_ops)
+#else
+#define BCM63XX_SPI_PM_OPS	NULL
+#endif
+
+static struct platform_driver bcm63xx_spi_driver = {
+	.driver = {
+		.name	= "bcm63xx-spi",
+		.owner	= THIS_MODULE,
+		.pm	= BCM63XX_SPI_PM_OPS,
+	},
+	.probe		= bcm63xx_spi_probe,
+	.remove		= __devexit_p(bcm63xx_spi_remove),
+};
+
+module_platform_driver(bcm63xx_spi_driver);
+
+MODULE_ALIAS("platform:bcm63xx_spi");
+MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
+MODULE_AUTHOR("Tanguy Bouzeloc <tanguy.bouzeloc@efixo.com>");
+MODULE_DESCRIPTION("Broadcom BCM63xx SPI Controller driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi-dw-pci.c b/drivers/spi/spi-dw-pci.c
index f64250ea1611..14f7cc9523f0 100644
--- a/drivers/spi/spi-dw-pci.c
+++ b/drivers/spi/spi-dw-pci.c
@@ -149,7 +149,7 @@ static int spi_resume(struct pci_dev *pdev)
 #define spi_resume	NULL
 #endif
 
-static const struct pci_device_id pci_ids[] __devinitdata = {
+static DEFINE_PCI_DEVICE_TABLE(pci_ids) = {
 	/* Intel MID platform SPI controller 0 */
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x0800) },
 	{},
diff --git a/drivers/spi/spi-fsl-espi.c b/drivers/spi/spi-fsl-espi.c
index d770f03705c3..7523a2429d09 100644
--- a/drivers/spi/spi-fsl-espi.c
+++ b/drivers/spi/spi-fsl-espi.c
@@ -180,18 +180,20 @@ static int fsl_espi_setup_transfer(struct spi_device *spi,
 
 	if ((mpc8xxx_spi->spibrg / hz) > 64) {
 		cs->hw_mode |= CSMODE_DIV16;
-		pm = (mpc8xxx_spi->spibrg - 1) / (hz * 64) + 1;
+		pm = DIV_ROUND_UP(mpc8xxx_spi->spibrg, hz * 16 * 4);
 
-		WARN_ONCE(pm > 16, "%s: Requested speed is too low: %d Hz. "
+		WARN_ONCE(pm > 33, "%s: Requested speed is too low: %d Hz. "
 			  "Will use %d Hz instead.\n", dev_name(&spi->dev),
-			  hz, mpc8xxx_spi->spibrg / 1024);
-		if (pm > 16)
-			pm = 16;
+				hz, mpc8xxx_spi->spibrg / (4 * 16 * (32 + 1)));
+		if (pm > 33)
+			pm = 33;
 	} else {
-		pm = (mpc8xxx_spi->spibrg - 1) / (hz * 4) + 1;
+		pm = DIV_ROUND_UP(mpc8xxx_spi->spibrg, hz * 4);
 	}
 	if (pm)
 		pm--;
+	if (pm < 2)
+		pm = 2;
 
 	cs->hw_mode |= CSMODE_PM(pm);
 
diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c
index c6e697f5e007..31054e3de4c1 100644
--- a/drivers/spi/spi-imx.c
+++ b/drivers/spi/spi-imx.c
@@ -793,13 +793,8 @@ static int __devinit spi_imx_probe(struct platform_device *pdev)
 
 		ret = gpio_request(spi_imx->chipselect[i], DRIVER_NAME);
 		if (ret) {
-			while (i > 0) {
-				i--;
-				if (spi_imx->chipselect[i] >= 0)
-					gpio_free(spi_imx->chipselect[i]);
-			}
 			dev_err(&pdev->dev, "can't get cs gpios\n");
-			goto out_master_put;
+			goto out_gpio_free;
 		}
 	}
 
@@ -881,10 +876,10 @@ out_iounmap:
 out_release_mem:
 	release_mem_region(res->start, resource_size(res));
 out_gpio_free:
-	for (i = 0; i < master->num_chipselect; i++)
+	while (--i >= 0) {
 		if (spi_imx->chipselect[i] >= 0)
 			gpio_free(spi_imx->chipselect[i]);
-out_master_put:
+	}
 	spi_master_put(master);
 	kfree(master);
 	platform_set_drvdata(pdev, NULL);
diff --git a/drivers/spi/spi-nuc900.c b/drivers/spi/spi-nuc900.c
index 182e9c873822..dae8be229c5d 100644
--- a/drivers/spi/spi-nuc900.c
+++ b/drivers/spi/spi-nuc900.c
@@ -360,8 +360,6 @@ static int __devinit nuc900_spi_probe(struct platform_device *pdev)
 	}
 
 	hw = spi_master_get_devdata(master);
-	memset(hw, 0, sizeof(struct nuc900_spi));
-
 	hw->master = spi_master_get(master);
 	hw->pdata  = pdev->dev.platform_data;
 	hw->dev = &pdev->dev;
diff --git a/drivers/spi/spi-omap2-mcspi.c b/drivers/spi/spi-omap2-mcspi.c
index 0b0dfb71c640..bb9274c2526d 100644
--- a/drivers/spi/spi-omap2-mcspi.c
+++ b/drivers/spi/spi-omap2-mcspi.c
@@ -34,6 +34,8 @@
 #include <linux/io.h>
 #include <linux/slab.h>
 #include <linux/pm_runtime.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <linux/spi/spi.h>
 
@@ -1079,15 +1081,39 @@ static int omap_mcspi_runtime_resume(struct device *dev)
 	return 0;
 }
 
+static struct omap2_mcspi_platform_config omap2_pdata = {
+	.regs_offset = 0,
+};
+
+static struct omap2_mcspi_platform_config omap4_pdata = {
+	.regs_offset = OMAP4_MCSPI_REG_OFFSET,
+};
+
+static const struct of_device_id omap_mcspi_of_match[] = {
+	{
+		.compatible = "ti,omap2-mcspi",
+		.data = &omap2_pdata,
+	},
+	{
+		.compatible = "ti,omap4-mcspi",
+		.data = &omap4_pdata,
+	},
+	{ },
+};
+MODULE_DEVICE_TABLE(of, omap_mcspi_of_match);
 
 static int __init omap2_mcspi_probe(struct platform_device *pdev)
 {
 	struct spi_master	*master;
-	struct omap2_mcspi_platform_config *pdata = pdev->dev.platform_data;
+	struct omap2_mcspi_platform_config *pdata;
 	struct omap2_mcspi	*mcspi;
 	struct resource		*r;
 	int			status = 0, i;
 	char			wq_name[20];
+	u32			regs_offset = 0;
+	static int		bus_num = 1;
+	struct device_node	*node = pdev->dev.of_node;
+	const struct of_device_id *match;
 
 	master = spi_alloc_master(&pdev->dev, sizeof *mcspi);
 	if (master == NULL) {
@@ -1098,13 +1124,26 @@ static int __init omap2_mcspi_probe(struct platform_device *pdev)
 	/* the spi->mode bits understood by this driver: */
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH;
 
-	if (pdev->id != -1)
-		master->bus_num = pdev->id;
-
 	master->setup = omap2_mcspi_setup;
 	master->transfer = omap2_mcspi_transfer;
 	master->cleanup = omap2_mcspi_cleanup;
-	master->num_chipselect = pdata->num_cs;
+	master->dev.of_node = node;
+
+	match = of_match_device(omap_mcspi_of_match, &pdev->dev);
+	if (match) {
+		u32 num_cs = 1; /* default number of chipselect */
+		pdata = match->data;
+
+		of_property_read_u32(node, "ti,spi-num-cs", &num_cs);
+		master->num_chipselect = num_cs;
+		master->bus_num = bus_num++;
+	} else {
+		pdata = pdev->dev.platform_data;
+		master->num_chipselect = pdata->num_cs;
+		if (pdev->id != -1)
+			master->bus_num = pdev->id;
+	}
+	regs_offset = pdata->regs_offset;
 
 	dev_set_drvdata(&pdev->dev, master);
 
@@ -1124,8 +1163,8 @@ static int __init omap2_mcspi_probe(struct platform_device *pdev)
 		goto free_master;
 	}
 
-	r->start += pdata->regs_offset;
-	r->end += pdata->regs_offset;
+	r->start += regs_offset;
+	r->end += regs_offset;
 	mcspi->phys = r->start;
 	if (!request_mem_region(r->start, resource_size(r),
 				dev_name(&pdev->dev))) {
@@ -1285,7 +1324,8 @@ static struct platform_driver omap2_mcspi_driver = {
 	.driver = {
 		.name =		"omap2_mcspi",
 		.owner =	THIS_MODULE,
-		.pm =		&omap2_mcspi_pm_ops
+		.pm =		&omap2_mcspi_pm_ops,
+		.of_match_table = omap_mcspi_of_match,
 	},
 	.remove =	__exit_p(omap2_mcspi_remove),
 };
diff --git a/drivers/spi/spi-pl022.c b/drivers/spi/spi-pl022.c
index f37ad2271ad5..dc8485d1e883 100644
--- a/drivers/spi/spi-pl022.c
+++ b/drivers/spi/spi-pl022.c
@@ -29,7 +29,6 @@
 #include <linux/errno.h>
 #include <linux/interrupt.h>
 #include <linux/spi/spi.h>
-#include <linux/workqueue.h>
 #include <linux/delay.h>
 #include <linux/clk.h>
 #include <linux/err.h>
@@ -330,12 +329,13 @@ struct vendor_data {
  * @clk: outgoing clock "SPICLK" for the SPI bus
  * @master: SPI framework hookup
  * @master_info: controller-specific data from machine setup
- * @workqueue: a workqueue on which any spi_message request is queued
- * @pump_messages: work struct for scheduling work to the workqueue
+ * @kworker: thread struct for message pump
+ * @kworker_task: pointer to task for message pump kworker thread
+ * @pump_messages: work struct for scheduling work to the message pump
  * @queue_lock: spinlock to syncronise access to message queue
  * @queue: message queue
- * @busy: workqueue is busy
- * @running: workqueue is running
+ * @busy: message pump is busy
+ * @running: message pump is running
  * @pump_transfers: Tasklet used in Interrupt Transfer mode
  * @cur_msg: Pointer to current spi_message being processed
  * @cur_transfer: Pointer to current spi_transfer
@@ -365,14 +365,7 @@ struct pl022 {
 	struct clk			*clk;
 	struct spi_master		*master;
 	struct pl022_ssp_controller	*master_info;
-	/* Driver message queue */
-	struct workqueue_struct		*workqueue;
-	struct work_struct		pump_messages;
-	spinlock_t			queue_lock;
-	struct list_head		queue;
-	bool				busy;
-	bool				running;
-	/* Message transfer pump */
+	/* Message per-transfer pump */
 	struct tasklet_struct		pump_transfers;
 	struct spi_message		*cur_msg;
 	struct spi_transfer		*cur_transfer;
@@ -394,6 +387,7 @@ struct pl022 {
 	struct sg_table			sgt_rx;
 	struct sg_table			sgt_tx;
 	char				*dummypage;
+	bool				dma_running;
 #endif
 };
 
@@ -448,8 +442,6 @@ static void null_cs_control(u32 command)
 static void giveback(struct pl022 *pl022)
 {
 	struct spi_transfer *last_transfer;
-	unsigned long flags;
-	struct spi_message *msg;
 	pl022->next_msg_cs_active = false;
 
 	last_transfer = list_entry(pl022->cur_msg->transfers.prev,
@@ -477,15 +469,8 @@ static void giveback(struct pl022 *pl022)
 		 * sent the current message could be unloaded, which
 		 * could invalidate the cs_control() callback...
 		 */
-
 		/* get a pointer to the next message, if any */
-		spin_lock_irqsave(&pl022->queue_lock, flags);
-		if (list_empty(&pl022->queue))
-			next_msg = NULL;
-		else
-			next_msg = list_entry(pl022->queue.next,
-					struct spi_message, queue);
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
+		next_msg = spi_get_next_queued_message(pl022->master);
 
 		/*
 		 * see if the next and current messages point
@@ -497,19 +482,13 @@ static void giveback(struct pl022 *pl022)
 			pl022->cur_chip->cs_control(SSP_CHIP_DESELECT);
 		else
 			pl022->next_msg_cs_active = true;
+
 	}
 
-	spin_lock_irqsave(&pl022->queue_lock, flags);
-	msg = pl022->cur_msg;
 	pl022->cur_msg = NULL;
 	pl022->cur_transfer = NULL;
 	pl022->cur_chip = NULL;
-	queue_work(pl022->workqueue, &pl022->pump_messages);
-	spin_unlock_irqrestore(&pl022->queue_lock, flags);
-
-	msg->state = NULL;
-	if (msg->complete)
-		msg->complete(msg->context);
+	spi_finalize_current_message(pl022->master);
 }
 
 /**
@@ -1063,6 +1042,7 @@ static int configure_dma(struct pl022 *pl022)
 	dmaengine_submit(txdesc);
 	dma_async_issue_pending(rxchan);
 	dma_async_issue_pending(txchan);
+	pl022->dma_running = true;
 
 	return 0;
 
@@ -1141,11 +1121,12 @@ static void terminate_dma(struct pl022 *pl022)
 	dmaengine_terminate_all(rxchan);
 	dmaengine_terminate_all(txchan);
 	unmap_free_dma_scatter(pl022);
+	pl022->dma_running = false;
 }
 
 static void pl022_dma_remove(struct pl022 *pl022)
 {
-	if (pl022->busy)
+	if (pl022->dma_running)
 		terminate_dma(pl022);
 	if (pl022->dma_tx_channel)
 		dma_release_channel(pl022->dma_tx_channel);
@@ -1493,73 +1474,20 @@ out:
 	return;
 }
 
-/**
- * pump_messages - Workqueue function which processes spi message queue
- * @data: pointer to private data of SSP driver
- *
- * This function checks if there is any spi message in the queue that
- * needs processing and delegate control to appropriate function
- * do_polling_transfer()/do_interrupt_dma_transfer()
- * based on the kind of the transfer
- *
- */
-static void pump_messages(struct work_struct *work)
+static int pl022_transfer_one_message(struct spi_master *master,
+				      struct spi_message *msg)
 {
-	struct pl022 *pl022 =
-		container_of(work, struct pl022, pump_messages);
-	unsigned long flags;
-	bool was_busy = false;
-
-	/* Lock queue and check for queue work */
-	spin_lock_irqsave(&pl022->queue_lock, flags);
-	if (list_empty(&pl022->queue) || !pl022->running) {
-		if (pl022->busy) {
-			/* nothing more to do - disable spi/ssp and power off */
-			writew((readw(SSP_CR1(pl022->virtbase)) &
-				(~SSP_CR1_MASK_SSE)), SSP_CR1(pl022->virtbase));
-
-			if (pl022->master_info->autosuspend_delay > 0) {
-				pm_runtime_mark_last_busy(&pl022->adev->dev);
-				pm_runtime_put_autosuspend(&pl022->adev->dev);
-			} else {
-				pm_runtime_put(&pl022->adev->dev);
-			}
-		}
-		pl022->busy = false;
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
-		return;
-	}
-
-	/* Make sure we are not already running a message */
-	if (pl022->cur_msg) {
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
-		return;
-	}
-	/* Extract head of queue */
-	pl022->cur_msg =
-	    list_entry(pl022->queue.next, struct spi_message, queue);
-
-	list_del_init(&pl022->cur_msg->queue);
-	if (pl022->busy)
-		was_busy = true;
-	else
-		pl022->busy = true;
-	spin_unlock_irqrestore(&pl022->queue_lock, flags);
+	struct pl022 *pl022 = spi_master_get_devdata(master);
 
 	/* Initial message state */
-	pl022->cur_msg->state = STATE_START;
-	pl022->cur_transfer = list_entry(pl022->cur_msg->transfers.next,
-					    struct spi_transfer, transfer_list);
+	pl022->cur_msg = msg;
+	msg->state = STATE_START;
+
+	pl022->cur_transfer = list_entry(msg->transfers.next,
+					 struct spi_transfer, transfer_list);
 
 	/* Setup the SPI using the per chip configuration */
-	pl022->cur_chip = spi_get_ctldata(pl022->cur_msg->spi);
-	if (!was_busy)
-		/*
-		 * We enable the core voltage and clocks here, then the clocks
-		 * and core will be disabled when this workqueue is run again
-		 * and there is no more work to be done.
-		 */
-		pm_runtime_get_sync(&pl022->adev->dev);
+	pl022->cur_chip = spi_get_ctldata(msg->spi);
 
 	restore_state(pl022);
 	flush(pl022);
@@ -1568,95 +1496,37 @@ static void pump_messages(struct work_struct *work)
 		do_polling_transfer(pl022);
 	else
 		do_interrupt_dma_transfer(pl022);
-}
-
-static int __init init_queue(struct pl022 *pl022)
-{
-	INIT_LIST_HEAD(&pl022->queue);
-	spin_lock_init(&pl022->queue_lock);
-
-	pl022->running = false;
-	pl022->busy = false;
-
-	tasklet_init(&pl022->pump_transfers, pump_transfers,
-			(unsigned long)pl022);
-
-	INIT_WORK(&pl022->pump_messages, pump_messages);
-	pl022->workqueue = create_singlethread_workqueue(
-					dev_name(pl022->master->dev.parent));
-	if (pl022->workqueue == NULL)
-		return -EBUSY;
 
 	return 0;
 }
 
-static int start_queue(struct pl022 *pl022)
+static int pl022_prepare_transfer_hardware(struct spi_master *master)
 {
-	unsigned long flags;
-
-	spin_lock_irqsave(&pl022->queue_lock, flags);
-
-	if (pl022->running || pl022->busy) {
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
-		return -EBUSY;
-	}
-
-	pl022->running = true;
-	pl022->cur_msg = NULL;
-	pl022->cur_transfer = NULL;
-	pl022->cur_chip = NULL;
-	pl022->next_msg_cs_active = false;
-	spin_unlock_irqrestore(&pl022->queue_lock, flags);
-
-	queue_work(pl022->workqueue, &pl022->pump_messages);
+	struct pl022 *pl022 = spi_master_get_devdata(master);
 
+	/*
+	 * Just make sure we have all we need to run the transfer by syncing
+	 * with the runtime PM framework.
+	 */
+	pm_runtime_get_sync(&pl022->adev->dev);
 	return 0;
 }
 
-static int stop_queue(struct pl022 *pl022)
+static int pl022_unprepare_transfer_hardware(struct spi_master *master)
 {
-	unsigned long flags;
-	unsigned limit = 500;
-	int status = 0;
+	struct pl022 *pl022 = spi_master_get_devdata(master);
 
-	spin_lock_irqsave(&pl022->queue_lock, flags);
+	/* nothing more to do - disable spi/ssp and power off */
+	writew((readw(SSP_CR1(pl022->virtbase)) &
+		(~SSP_CR1_MASK_SSE)), SSP_CR1(pl022->virtbase));
 
-	/* This is a bit lame, but is optimized for the common execution path.
-	 * A wait_queue on the pl022->busy could be used, but then the common
-	 * execution path (pump_messages) would be required to call wake_up or
-	 * friends on every SPI message. Do this instead */
-	while ((!list_empty(&pl022->queue) || pl022->busy) && limit--) {
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
-		msleep(10);
-		spin_lock_irqsave(&pl022->queue_lock, flags);
+	if (pl022->master_info->autosuspend_delay > 0) {
+		pm_runtime_mark_last_busy(&pl022->adev->dev);
+		pm_runtime_put_autosuspend(&pl022->adev->dev);
+	} else {
+		pm_runtime_put(&pl022->adev->dev);
 	}
 
-	if (!list_empty(&pl022->queue) || pl022->busy)
-		status = -EBUSY;
-	else
-		pl022->running = false;
-
-	spin_unlock_irqrestore(&pl022->queue_lock, flags);
-
-	return status;
-}
-
-static int destroy_queue(struct pl022 *pl022)
-{
-	int status;
-
-	status = stop_queue(pl022);
-	/* we are unloading the module or failing to load (only two calls
-	 * to this routine), and neither call can handle a return value.
-	 * However, destroy_workqueue calls flush_workqueue, and that will
-	 * block until all work is done.  If the reason that stop_queue
-	 * timed out is that the work will never finish, then it does no
-	 * good to call destroy_workqueue, so return anyway. */
-	if (status != 0)
-		return status;
-
-	destroy_workqueue(pl022->workqueue);
-
 	return 0;
 }
 
@@ -1776,38 +1646,6 @@ static int verify_controller_parameters(struct pl022 *pl022,
 	return 0;
 }
 
-/**
- * pl022_transfer - transfer function registered to SPI master framework
- * @spi: spi device which is requesting transfer
- * @msg: spi message which is to handled is queued to driver queue
- *
- * This function is registered to the SPI framework for this SPI master
- * controller. It will queue the spi_message in the queue of driver if
- * the queue is not stopped and return.
- */
-static int pl022_transfer(struct spi_device *spi, struct spi_message *msg)
-{
-	struct pl022 *pl022 = spi_master_get_devdata(spi->master);
-	unsigned long flags;
-
-	spin_lock_irqsave(&pl022->queue_lock, flags);
-
-	if (!pl022->running) {
-		spin_unlock_irqrestore(&pl022->queue_lock, flags);
-		return -ESHUTDOWN;
-	}
-	msg->actual_length = 0;
-	msg->status = -EINPROGRESS;
-	msg->state = STATE_START;
-
-	list_add_tail(&msg->queue, &pl022->queue);
-	if (pl022->running && !pl022->busy)
-		queue_work(pl022->workqueue, &pl022->pump_messages);
-
-	spin_unlock_irqrestore(&pl022->queue_lock, flags);
-	return 0;
-}
-
 static inline u32 spi_rate(u32 rate, u16 cpsdvsr, u16 scr)
 {
 	return rate / (cpsdvsr * (1 + scr));
@@ -2170,7 +2008,10 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id)
 	master->num_chipselect = platform_info->num_chipselect;
 	master->cleanup = pl022_cleanup;
 	master->setup = pl022_setup;
-	master->transfer = pl022_transfer;
+	master->prepare_transfer_hardware = pl022_prepare_transfer_hardware;
+	master->transfer_one_message = pl022_transfer_one_message;
+	master->unprepare_transfer_hardware = pl022_unprepare_transfer_hardware;
+	master->rt = platform_info->rt;
 
 	/*
 	 * Supports mode 0-3, loopback, and active low CS. Transfers are
@@ -2214,6 +2055,10 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id)
 		goto err_no_clk_en;
 	}
 
+	/* Initialize transfer pump */
+	tasklet_init(&pl022->pump_transfers, pump_transfers,
+		     (unsigned long)pl022);
+
 	/* Disable SSP */
 	writew((readw(SSP_CR1(pl022->virtbase)) & (~SSP_CR1_MASK_SSE)),
 	       SSP_CR1(pl022->virtbase));
@@ -2233,17 +2078,6 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id)
 			platform_info->enable_dma = 0;
 	}
 
-	/* Initialize and start queue */
-	status = init_queue(pl022);
-	if (status != 0) {
-		dev_err(&adev->dev, "probe - problem initializing queue\n");
-		goto err_init_queue;
-	}
-	status = start_queue(pl022);
-	if (status != 0) {
-		dev_err(&adev->dev, "probe - problem starting queue\n");
-		goto err_start_queue;
-	}
 	/* Register with the SPI framework */
 	amba_set_drvdata(adev, pl022);
 	status = spi_register_master(master);
@@ -2269,9 +2103,6 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id)
 	return 0;
 
  err_spi_register:
- err_start_queue:
- err_init_queue:
-	destroy_queue(pl022);
 	if (platform_info->enable_dma)
 		pl022_dma_remove(pl022);
 
@@ -2307,9 +2138,6 @@ pl022_remove(struct amba_device *adev)
 	 */
 	pm_runtime_get_noresume(&adev->dev);
 
-	/* Remove the queue */
-	if (destroy_queue(pl022) != 0)
-		dev_err(&adev->dev, "queue remove failed\n");
 	load_ssp_default_config(pl022);
 	if (pl022->master_info->enable_dma)
 		pl022_dma_remove(pl022);
@@ -2331,12 +2159,12 @@ pl022_remove(struct amba_device *adev)
 static int pl022_suspend(struct device *dev)
 {
 	struct pl022 *pl022 = dev_get_drvdata(dev);
-	int status = 0;
+	int ret;
 
-	status = stop_queue(pl022);
-	if (status) {
-		dev_warn(dev, "suspend cannot stop queue\n");
-		return status;
+	ret = spi_master_suspend(pl022->master);
+	if (ret) {
+		dev_warn(dev, "cannot suspend master\n");
+		return ret;
 	}
 
 	dev_dbg(dev, "suspended\n");
@@ -2346,16 +2174,16 @@ static int pl022_suspend(struct device *dev)
 static int pl022_resume(struct device *dev)
 {
 	struct pl022 *pl022 = dev_get_drvdata(dev);
-	int status = 0;
+	int ret;
 
 	/* Start the queue running */
-	status = start_queue(pl022);
-	if (status)
-		dev_err(dev, "problem starting queue (%d)\n", status);
+	ret = spi_master_resume(pl022->master);
+	if (ret)
+		dev_err(dev, "problem starting queue (%d)\n", ret);
 	else
 		dev_dbg(dev, "resumed\n");
 
-	return status;
+	return ret;
 }
 #endif	/* CONFIG_PM */
 
diff --git a/drivers/spi/spi-pxa2xx-pci.c b/drivers/spi/spi-pxa2xx-pci.c
index 8caa07d58e69..3fb44afe27b4 100644
--- a/drivers/spi/spi-pxa2xx-pci.c
+++ b/drivers/spi/spi-pxa2xx-pci.c
@@ -151,7 +151,7 @@ static void __devexit ce4100_spi_remove(struct pci_dev *dev)
 	kfree(spi_info);
 }
 
-static struct pci_device_id ce4100_spi_devices[] __devinitdata = {
+static DEFINE_PCI_DEVICE_TABLE(ce4100_spi_devices) = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2e6a) },
 	{ },
 };
diff --git a/drivers/spi/spi-rspi.c b/drivers/spi/spi-rspi.c
new file mode 100644
index 000000000000..354f170eab95
--- /dev/null
+++ b/drivers/spi/spi-rspi.c
@@ -0,0 +1,521 @@
+/*
+ * SH RSPI driver
+ *
+ * Copyright (C) 2012  Renesas Solutions Corp.
+ *
+ * Based on spi-sh.c:
+ * Copyright (C) 2011 Renesas Solutions Corp.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/list.h>
+#include <linux/workqueue.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/spi/spi.h>
+
+#define RSPI_SPCR		0x00
+#define RSPI_SSLP		0x01
+#define RSPI_SPPCR		0x02
+#define RSPI_SPSR		0x03
+#define RSPI_SPDR		0x04
+#define RSPI_SPSCR		0x08
+#define RSPI_SPSSR		0x09
+#define RSPI_SPBR		0x0a
+#define RSPI_SPDCR		0x0b
+#define RSPI_SPCKD		0x0c
+#define RSPI_SSLND		0x0d
+#define RSPI_SPND		0x0e
+#define RSPI_SPCR2		0x0f
+#define RSPI_SPCMD0		0x10
+#define RSPI_SPCMD1		0x12
+#define RSPI_SPCMD2		0x14
+#define RSPI_SPCMD3		0x16
+#define RSPI_SPCMD4		0x18
+#define RSPI_SPCMD5		0x1a
+#define RSPI_SPCMD6		0x1c
+#define RSPI_SPCMD7		0x1e
+
+/* SPCR */
+#define SPCR_SPRIE		0x80
+#define SPCR_SPE		0x40
+#define SPCR_SPTIE		0x20
+#define SPCR_SPEIE		0x10
+#define SPCR_MSTR		0x08
+#define SPCR_MODFEN		0x04
+#define SPCR_TXMD		0x02
+#define SPCR_SPMS		0x01
+
+/* SSLP */
+#define SSLP_SSL1P		0x02
+#define SSLP_SSL0P		0x01
+
+/* SPPCR */
+#define SPPCR_MOIFE		0x20
+#define SPPCR_MOIFV		0x10
+#define SPPCR_SPOM		0x04
+#define SPPCR_SPLP2		0x02
+#define SPPCR_SPLP		0x01
+
+/* SPSR */
+#define SPSR_SPRF		0x80
+#define SPSR_SPTEF		0x20
+#define SPSR_PERF		0x08
+#define SPSR_MODF		0x04
+#define SPSR_IDLNF		0x02
+#define SPSR_OVRF		0x01
+
+/* SPSCR */
+#define SPSCR_SPSLN_MASK	0x07
+
+/* SPSSR */
+#define SPSSR_SPECM_MASK	0x70
+#define SPSSR_SPCP_MASK		0x07
+
+/* SPDCR */
+#define SPDCR_SPLW		0x20
+#define SPDCR_SPRDTD		0x10
+#define SPDCR_SLSEL1		0x08
+#define SPDCR_SLSEL0		0x04
+#define SPDCR_SLSEL_MASK	0x0c
+#define SPDCR_SPFC1		0x02
+#define SPDCR_SPFC0		0x01
+
+/* SPCKD */
+#define SPCKD_SCKDL_MASK	0x07
+
+/* SSLND */
+#define SSLND_SLNDL_MASK	0x07
+
+/* SPND */
+#define SPND_SPNDL_MASK		0x07
+
+/* SPCR2 */
+#define SPCR2_PTE		0x08
+#define SPCR2_SPIE		0x04
+#define SPCR2_SPOE		0x02
+#define SPCR2_SPPE		0x01
+
+/* SPCMDn */
+#define SPCMD_SCKDEN		0x8000
+#define SPCMD_SLNDEN		0x4000
+#define SPCMD_SPNDEN		0x2000
+#define SPCMD_LSBF		0x1000
+#define SPCMD_SPB_MASK		0x0f00
+#define SPCMD_SPB_8_TO_16(bit)	(((bit - 1) << 8) & SPCMD_SPB_MASK)
+#define SPCMD_SPB_20BIT		0x0000
+#define SPCMD_SPB_24BIT		0x0100
+#define SPCMD_SPB_32BIT		0x0200
+#define SPCMD_SSLKP		0x0080
+#define SPCMD_SSLA_MASK		0x0030
+#define SPCMD_BRDV_MASK		0x000c
+#define SPCMD_CPOL		0x0002
+#define SPCMD_CPHA		0x0001
+
+struct rspi_data {
+	void __iomem *addr;
+	u32 max_speed_hz;
+	struct spi_master *master;
+	struct list_head queue;
+	struct work_struct ws;
+	wait_queue_head_t wait;
+	spinlock_t lock;
+	struct clk *clk;
+	unsigned char spsr;
+};
+
+static void rspi_write8(struct rspi_data *rspi, u8 data, u16 offset)
+{
+	iowrite8(data, rspi->addr + offset);
+}
+
+static void rspi_write16(struct rspi_data *rspi, u16 data, u16 offset)
+{
+	iowrite16(data, rspi->addr + offset);
+}
+
+static u8 rspi_read8(struct rspi_data *rspi, u16 offset)
+{
+	return ioread8(rspi->addr + offset);
+}
+
+static u16 rspi_read16(struct rspi_data *rspi, u16 offset)
+{
+	return ioread16(rspi->addr + offset);
+}
+
+static unsigned char rspi_calc_spbr(struct rspi_data *rspi)
+{
+	int tmp;
+	unsigned char spbr;
+
+	tmp = clk_get_rate(rspi->clk) / (2 * rspi->max_speed_hz) - 1;
+	spbr = clamp(tmp, 0, 255);
+
+	return spbr;
+}
+
+static void rspi_enable_irq(struct rspi_data *rspi, u8 enable)
+{
+	rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) | enable, RSPI_SPCR);
+}
+
+static void rspi_disable_irq(struct rspi_data *rspi, u8 disable)
+{
+	rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) & ~disable, RSPI_SPCR);
+}
+
+static int rspi_wait_for_interrupt(struct rspi_data *rspi, u8 wait_mask,
+				   u8 enable_bit)
+{
+	int ret;
+
+	rspi->spsr = rspi_read8(rspi, RSPI_SPSR);
+	rspi_enable_irq(rspi, enable_bit);
+	ret = wait_event_timeout(rspi->wait, rspi->spsr & wait_mask, HZ);
+	if (ret == 0 && !(rspi->spsr & wait_mask))
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+static void rspi_assert_ssl(struct rspi_data *rspi)
+{
+	rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) | SPCR_SPE, RSPI_SPCR);
+}
+
+static void rspi_negate_ssl(struct rspi_data *rspi)
+{
+	rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) & ~SPCR_SPE, RSPI_SPCR);
+}
+
+static int rspi_set_config_register(struct rspi_data *rspi, int access_size)
+{
+	/* Sets output mode(CMOS) and MOSI signal(from previous transfer) */
+	rspi_write8(rspi, 0x00, RSPI_SPPCR);
+
+	/* Sets transfer bit rate */
+	rspi_write8(rspi, rspi_calc_spbr(rspi), RSPI_SPBR);
+
+	/* Sets number of frames to be used: 1 frame */
+	rspi_write8(rspi, 0x00, RSPI_SPDCR);
+
+	/* Sets RSPCK, SSL, next-access delay value */
+	rspi_write8(rspi, 0x00, RSPI_SPCKD);
+	rspi_write8(rspi, 0x00, RSPI_SSLND);
+	rspi_write8(rspi, 0x00, RSPI_SPND);
+
+	/* Sets parity, interrupt mask */
+	rspi_write8(rspi, 0x00, RSPI_SPCR2);
+
+	/* Sets SPCMD */
+	rspi_write16(rspi, SPCMD_SPB_8_TO_16(access_size) | SPCMD_SSLKP,
+		     RSPI_SPCMD0);
+
+	/* Sets RSPI mode */
+	rspi_write8(rspi, SPCR_MSTR, RSPI_SPCR);
+
+	return 0;
+}
+
+static int rspi_send_pio(struct rspi_data *rspi, struct spi_message *mesg,
+			 struct spi_transfer *t)
+{
+	int remain = t->len;
+	u8 *data;
+
+	data = (u8 *)t->tx_buf;
+	while (remain > 0) {
+		rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) | SPCR_TXMD,
+			    RSPI_SPCR);
+
+		if (rspi_wait_for_interrupt(rspi, SPSR_SPTEF, SPCR_SPTIE) < 0) {
+			dev_err(&rspi->master->dev,
+				"%s: tx empty timeout\n", __func__);
+			return -ETIMEDOUT;
+		}
+
+		rspi_write16(rspi, *data, RSPI_SPDR);
+		data++;
+		remain--;
+	}
+
+	/* Waiting for the last transmition */
+	rspi_wait_for_interrupt(rspi, SPSR_SPTEF, SPCR_SPTIE);
+
+	return 0;
+}
+
+static int rspi_receive_pio(struct rspi_data *rspi, struct spi_message *mesg,
+			    struct spi_transfer *t)
+{
+	int remain = t->len;
+	u8 *data;
+	unsigned char spsr;
+
+	spsr = rspi_read8(rspi, RSPI_SPSR);
+	if (spsr & SPSR_SPRF)
+		rspi_read16(rspi, RSPI_SPDR);	/* dummy read */
+	if (spsr & SPSR_OVRF)
+		rspi_write8(rspi, rspi_read8(rspi, RSPI_SPSR) & ~SPSR_OVRF,
+			    RSPI_SPCR);
+
+	data = (u8 *)t->rx_buf;
+	while (remain > 0) {
+		rspi_write8(rspi, rspi_read8(rspi, RSPI_SPCR) & ~SPCR_TXMD,
+			    RSPI_SPCR);
+
+		if (rspi_wait_for_interrupt(rspi, SPSR_SPTEF, SPCR_SPTIE) < 0) {
+			dev_err(&rspi->master->dev,
+				"%s: tx empty timeout\n", __func__);
+			return -ETIMEDOUT;
+		}
+		/* dummy write for generate clock */
+		rspi_write16(rspi, 0x00, RSPI_SPDR);
+
+		if (rspi_wait_for_interrupt(rspi, SPSR_SPRF, SPCR_SPRIE) < 0) {
+			dev_err(&rspi->master->dev,
+				"%s: receive timeout\n", __func__);
+			return -ETIMEDOUT;
+		}
+		/* SPDR allows 16 or 32-bit access only */
+		*data = (u8)rspi_read16(rspi, RSPI_SPDR);
+
+		data++;
+		remain--;
+	}
+
+	return 0;
+}
+
+static void rspi_work(struct work_struct *work)
+{
+	struct rspi_data *rspi = container_of(work, struct rspi_data, ws);
+	struct spi_message *mesg;
+	struct spi_transfer *t;
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&rspi->lock, flags);
+	while (!list_empty(&rspi->queue)) {
+		mesg = list_entry(rspi->queue.next, struct spi_message, queue);
+		list_del_init(&mesg->queue);
+		spin_unlock_irqrestore(&rspi->lock, flags);
+
+		rspi_assert_ssl(rspi);
+
+		list_for_each_entry(t, &mesg->transfers, transfer_list) {
+			if (t->tx_buf) {
+				ret = rspi_send_pio(rspi, mesg, t);
+				if (ret < 0)
+					goto error;
+			}
+			if (t->rx_buf) {
+				ret = rspi_receive_pio(rspi, mesg, t);
+				if (ret < 0)
+					goto error;
+			}
+			mesg->actual_length += t->len;
+		}
+		rspi_negate_ssl(rspi);
+
+		mesg->status = 0;
+		mesg->complete(mesg->context);
+
+		spin_lock_irqsave(&rspi->lock, flags);
+	}
+
+	return;
+
+error:
+	mesg->status = ret;
+	mesg->complete(mesg->context);
+}
+
+static int rspi_setup(struct spi_device *spi)
+{
+	struct rspi_data *rspi = spi_master_get_devdata(spi->master);
+
+	if (!spi->bits_per_word)
+		spi->bits_per_word = 8;
+	rspi->max_speed_hz = spi->max_speed_hz;
+
+	rspi_set_config_register(rspi, 8);
+
+	return 0;
+}
+
+static int rspi_transfer(struct spi_device *spi, struct spi_message *mesg)
+{
+	struct rspi_data *rspi = spi_master_get_devdata(spi->master);
+	unsigned long flags;
+
+	mesg->actual_length = 0;
+	mesg->status = -EINPROGRESS;
+
+	spin_lock_irqsave(&rspi->lock, flags);
+	list_add_tail(&mesg->queue, &rspi->queue);
+	schedule_work(&rspi->ws);
+	spin_unlock_irqrestore(&rspi->lock, flags);
+
+	return 0;
+}
+
+static void rspi_cleanup(struct spi_device *spi)
+{
+}
+
+static irqreturn_t rspi_irq(int irq, void *_sr)
+{
+	struct rspi_data *rspi = (struct rspi_data *)_sr;
+	unsigned long spsr;
+	irqreturn_t ret = IRQ_NONE;
+	unsigned char disable_irq = 0;
+
+	rspi->spsr = spsr = rspi_read8(rspi, RSPI_SPSR);
+	if (spsr & SPSR_SPRF)
+		disable_irq |= SPCR_SPRIE;
+	if (spsr & SPSR_SPTEF)
+		disable_irq |= SPCR_SPTIE;
+
+	if (disable_irq) {
+		ret = IRQ_HANDLED;
+		rspi_disable_irq(rspi, disable_irq);
+		wake_up(&rspi->wait);
+	}
+
+	return ret;
+}
+
+static int __devexit rspi_remove(struct platform_device *pdev)
+{
+	struct rspi_data *rspi = dev_get_drvdata(&pdev->dev);
+
+	spi_unregister_master(rspi->master);
+	free_irq(platform_get_irq(pdev, 0), rspi);
+	clk_put(rspi->clk);
+	iounmap(rspi->addr);
+	spi_master_put(rspi->master);
+
+	return 0;
+}
+
+static int __devinit rspi_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct spi_master *master;
+	struct rspi_data *rspi;
+	int ret, irq;
+	char clk_name[16];
+
+	/* get base addr */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (unlikely(res == NULL)) {
+		dev_err(&pdev->dev, "invalid resource\n");
+		return -EINVAL;
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "platform_get_irq error\n");
+		return -ENODEV;
+	}
+
+	master = spi_alloc_master(&pdev->dev, sizeof(struct rspi_data));
+	if (master == NULL) {
+		dev_err(&pdev->dev, "spi_alloc_master error.\n");
+		return -ENOMEM;
+	}
+
+	rspi = spi_master_get_devdata(master);
+	dev_set_drvdata(&pdev->dev, rspi);
+
+	rspi->master = master;
+	rspi->addr = ioremap(res->start, resource_size(res));
+	if (rspi->addr == NULL) {
+		dev_err(&pdev->dev, "ioremap error.\n");
+		ret = -ENOMEM;
+		goto error1;
+	}
+
+	snprintf(clk_name, sizeof(clk_name), "rspi%d", pdev->id);
+	rspi->clk = clk_get(&pdev->dev, clk_name);
+	if (IS_ERR(rspi->clk)) {
+		dev_err(&pdev->dev, "cannot get clock\n");
+		ret = PTR_ERR(rspi->clk);
+		goto error2;
+	}
+	clk_enable(rspi->clk);
+
+	INIT_LIST_HEAD(&rspi->queue);
+	spin_lock_init(&rspi->lock);
+	INIT_WORK(&rspi->ws, rspi_work);
+	init_waitqueue_head(&rspi->wait);
+
+	master->num_chipselect = 2;
+	master->bus_num = pdev->id;
+	master->setup = rspi_setup;
+	master->transfer = rspi_transfer;
+	master->cleanup = rspi_cleanup;
+
+	ret = request_irq(irq, rspi_irq, 0, dev_name(&pdev->dev), rspi);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "request_irq error\n");
+		goto error3;
+	}
+
+	ret = spi_register_master(master);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "spi_register_master error.\n");
+		goto error4;
+	}
+
+	dev_info(&pdev->dev, "probed\n");
+
+	return 0;
+
+error4:
+	free_irq(irq, rspi);
+error3:
+	clk_put(rspi->clk);
+error2:
+	iounmap(rspi->addr);
+error1:
+	spi_master_put(master);
+
+	return ret;
+}
+
+static struct platform_driver rspi_driver = {
+	.probe =	rspi_probe,
+	.remove =	__devexit_p(rspi_remove),
+	.driver		= {
+		.name = "rspi",
+		.owner	= THIS_MODULE,
+	},
+};
+module_platform_driver(rspi_driver);
+
+MODULE_DESCRIPTION("Renesas RSPI bus driver");
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Yoshihiro Shimoda");
+MODULE_ALIAS("platform:rspi");
diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
index dcf7e1006426..972a94c58be3 100644
--- a/drivers/spi/spi-s3c64xx.c
+++ b/drivers/spi/spi-s3c64xx.c
@@ -20,10 +20,12 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/workqueue.h>
+#include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/clk.h>
 #include <linux/dma-mapping.h>
 #include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 #include <linux/spi/spi.h>
 
 #include <mach/dma.h>
@@ -126,8 +128,6 @@
 
 #define msecs_to_loops(t) (loops_per_jiffy / 1000 * HZ * t)
 
-#define SUSPND    (1<<0)
-#define SPIBUSY   (1<<1)
 #define RXBUSY    (1<<2)
 #define TXBUSY    (1<<3)
 
@@ -142,10 +142,8 @@ struct s3c64xx_spi_dma_data {
  * @clk: Pointer to the spi clock.
  * @src_clk: Pointer to the clock used to generate SPI signals.
  * @master: Pointer to the SPI Protocol master.
- * @workqueue: Work queue for the SPI xfer requests.
  * @cntrlr_info: Platform specific data for the controller this driver manages.
  * @tgl_spi: Pointer to the last CS left untoggled by the cs_change hint.
- * @work: Work
  * @queue: To log SPI xfer requests.
  * @lock: Controller specific lock.
  * @state: Set of FLAGS to indicate status.
@@ -153,6 +151,7 @@ struct s3c64xx_spi_dma_data {
  * @tx_dmach: Controller's DMA channel for Tx.
  * @sfr_start: BUS address of SPI controller regs.
  * @regs: Pointer to ioremap'ed controller registers.
+ * @irq: interrupt
  * @xfer_completion: To indicate completion of xfer task.
  * @cur_mode: Stores the active configuration of the controller.
  * @cur_bpw: Stores the active bits per word settings.
@@ -164,10 +163,8 @@ struct s3c64xx_spi_driver_data {
 	struct clk                      *src_clk;
 	struct platform_device          *pdev;
 	struct spi_master               *master;
-	struct workqueue_struct	        *workqueue;
 	struct s3c64xx_spi_info  *cntrlr_info;
 	struct spi_device               *tgl_spi;
-	struct work_struct              work;
 	struct list_head                queue;
 	spinlock_t                      lock;
 	unsigned long                   sfr_start;
@@ -239,7 +236,7 @@ static void s3c64xx_spi_dmacb(void *data)
 	struct s3c64xx_spi_dma_data *dma = data;
 	unsigned long flags;
 
-	if (dma->direction == DMA_FROM_DEVICE)
+	if (dma->direction == DMA_DEV_TO_MEM)
 		sdd = container_of(data,
 			struct s3c64xx_spi_driver_data, rx_dma);
 	else
@@ -248,7 +245,7 @@ static void s3c64xx_spi_dmacb(void *data)
 
 	spin_lock_irqsave(&sdd->lock, flags);
 
-	if (dma->direction == DMA_FROM_DEVICE) {
+	if (dma->direction == DMA_DEV_TO_MEM) {
 		sdd->state &= ~RXBUSY;
 		if (!(sdd->state & TXBUSY))
 			complete(&sdd->xfer_completion);
@@ -267,7 +264,7 @@ static void prepare_dma(struct s3c64xx_spi_dma_data *dma,
 	struct s3c64xx_spi_driver_data *sdd;
 	struct samsung_dma_prep_info info;
 
-	if (dma->direction == DMA_FROM_DEVICE)
+	if (dma->direction == DMA_DEV_TO_MEM)
 		sdd = container_of((void *)dma,
 			struct s3c64xx_spi_driver_data, rx_dma);
 	else
@@ -634,9 +631,10 @@ static void s3c64xx_spi_unmap_mssg(struct s3c64xx_spi_driver_data *sdd,
 	}
 }
 
-static void handle_msg(struct s3c64xx_spi_driver_data *sdd,
-					struct spi_message *msg)
+static int s3c64xx_spi_transfer_one_message(struct spi_master *master,
+					    struct spi_message *msg)
 {
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
 	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
 	struct spi_device *spi = msg->spi;
 	struct s3c64xx_spi_csinfo *cs = spi->controller_data;
@@ -766,73 +764,33 @@ out:
 
 	msg->status = status;
 
-	if (msg->complete)
-		msg->complete(msg->context);
+	spi_finalize_current_message(master);
+
+	return 0;
 }
 
-static void s3c64xx_spi_work(struct work_struct *work)
+static int s3c64xx_spi_prepare_transfer(struct spi_master *spi)
 {
-	struct s3c64xx_spi_driver_data *sdd = container_of(work,
-					struct s3c64xx_spi_driver_data, work);
-	unsigned long flags;
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(spi);
 
 	/* Acquire DMA channels */
 	while (!acquire_dma(sdd))
 		msleep(10);
 
-	spin_lock_irqsave(&sdd->lock, flags);
-
-	while (!list_empty(&sdd->queue)
-				&& !(sdd->state & SUSPND)) {
-
-		struct spi_message *msg;
-
-		msg = container_of(sdd->queue.next, struct spi_message, queue);
-
-		list_del_init(&msg->queue);
-
-		/* Set Xfer busy flag */
-		sdd->state |= SPIBUSY;
-
-		spin_unlock_irqrestore(&sdd->lock, flags);
-
-		handle_msg(sdd, msg);
+	pm_runtime_get_sync(&sdd->pdev->dev);
 
-		spin_lock_irqsave(&sdd->lock, flags);
-
-		sdd->state &= ~SPIBUSY;
-	}
+	return 0;
+}
 
-	spin_unlock_irqrestore(&sdd->lock, flags);
+static int s3c64xx_spi_unprepare_transfer(struct spi_master *spi)
+{
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(spi);
 
 	/* Free DMA channels */
 	sdd->ops->release(sdd->rx_dma.ch, &s3c64xx_spi_dma_client);
 	sdd->ops->release(sdd->tx_dma.ch, &s3c64xx_spi_dma_client);
-}
 
-static int s3c64xx_spi_transfer(struct spi_device *spi,
-						struct spi_message *msg)
-{
-	struct s3c64xx_spi_driver_data *sdd;
-	unsigned long flags;
-
-	sdd = spi_master_get_devdata(spi->master);
-
-	spin_lock_irqsave(&sdd->lock, flags);
-
-	if (sdd->state & SUSPND) {
-		spin_unlock_irqrestore(&sdd->lock, flags);
-		return -ESHUTDOWN;
-	}
-
-	msg->status = -EINPROGRESS;
-	msg->actual_length = 0;
-
-	list_add_tail(&msg->queue, &sdd->queue);
-
-	queue_work(sdd->workqueue, &sdd->work);
-
-	spin_unlock_irqrestore(&sdd->lock, flags);
+	pm_runtime_put(&sdd->pdev->dev);
 
 	return 0;
 }
@@ -872,13 +830,6 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 		}
 	}
 
-	if (sdd->state & SUSPND) {
-		spin_unlock_irqrestore(&sdd->lock, flags);
-		dev_err(&spi->dev,
-			"setup: SPI-%d not active!\n", spi->master->bus_num);
-		return -ESHUTDOWN;
-	}
-
 	spin_unlock_irqrestore(&sdd->lock, flags);
 
 	if (spi->bits_per_word != 8
@@ -890,6 +841,8 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 		goto setup_exit;
 	}
 
+	pm_runtime_get_sync(&sdd->pdev->dev);
+
 	/* Check if we can provide the requested rate */
 	if (!sci->clk_from_cmu) {
 		u32 psr, speed;
@@ -922,6 +875,8 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 			err = -EINVAL;
 	}
 
+	pm_runtime_put(&sdd->pdev->dev);
+
 setup_exit:
 
 	/* setup() returns with device de-selected */
@@ -930,6 +885,33 @@ setup_exit:
 	return err;
 }
 
+static irqreturn_t s3c64xx_spi_irq(int irq, void *data)
+{
+	struct s3c64xx_spi_driver_data *sdd = data;
+	struct spi_master *spi = sdd->master;
+	unsigned int val;
+
+	val = readl(sdd->regs + S3C64XX_SPI_PENDING_CLR);
+
+	val &= S3C64XX_SPI_PND_RX_OVERRUN_CLR |
+		S3C64XX_SPI_PND_RX_UNDERRUN_CLR |
+		S3C64XX_SPI_PND_TX_OVERRUN_CLR |
+		S3C64XX_SPI_PND_TX_UNDERRUN_CLR;
+
+	writel(val, sdd->regs + S3C64XX_SPI_PENDING_CLR);
+
+	if (val & S3C64XX_SPI_PND_RX_OVERRUN_CLR)
+		dev_err(&spi->dev, "RX overrun\n");
+	if (val & S3C64XX_SPI_PND_RX_UNDERRUN_CLR)
+		dev_err(&spi->dev, "RX underrun\n");
+	if (val & S3C64XX_SPI_PND_TX_OVERRUN_CLR)
+		dev_err(&spi->dev, "TX overrun\n");
+	if (val & S3C64XX_SPI_PND_TX_UNDERRUN_CLR)
+		dev_err(&spi->dev, "TX underrun\n");
+
+	return IRQ_HANDLED;
+}
+
 static void s3c64xx_spi_hwinit(struct s3c64xx_spi_driver_data *sdd, int channel)
 {
 	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
@@ -970,7 +952,7 @@ static int __init s3c64xx_spi_probe(struct platform_device *pdev)
 	struct s3c64xx_spi_driver_data *sdd;
 	struct s3c64xx_spi_info *sci;
 	struct spi_master *master;
-	int ret;
+	int ret, irq;
 	char clk_name[16];
 
 	if (pdev->id < 0) {
@@ -1006,6 +988,12 @@ static int __init s3c64xx_spi_probe(struct platform_device *pdev)
 		return -ENXIO;
 	}
 
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_warn(&pdev->dev, "Failed to get IRQ: %d\n", irq);
+		return irq;
+	}
+
 	master = spi_alloc_master(&pdev->dev,
 				sizeof(struct s3c64xx_spi_driver_data));
 	if (master == NULL) {
@@ -1021,15 +1009,17 @@ static int __init s3c64xx_spi_probe(struct platform_device *pdev)
 	sdd->pdev = pdev;
 	sdd->sfr_start = mem_res->start;
 	sdd->tx_dma.dmach = dmatx_res->start;
-	sdd->tx_dma.direction = DMA_TO_DEVICE;
+	sdd->tx_dma.direction = DMA_MEM_TO_DEV;
 	sdd->rx_dma.dmach = dmarx_res->start;
-	sdd->rx_dma.direction = DMA_FROM_DEVICE;
+	sdd->rx_dma.direction = DMA_DEV_TO_MEM;
 
 	sdd->cur_bpw = 8;
 
 	master->bus_num = pdev->id;
 	master->setup = s3c64xx_spi_setup;
-	master->transfer = s3c64xx_spi_transfer;
+	master->prepare_transfer_hardware = s3c64xx_spi_prepare_transfer;
+	master->transfer_one_message = s3c64xx_spi_transfer_one_message;
+	master->unprepare_transfer_hardware = s3c64xx_spi_unprepare_transfer;
 	master->num_chipselect = sci->num_cs;
 	master->dma_alignment = 8;
 	/* the spi->mode bits understood by this driver: */
@@ -1084,22 +1074,24 @@ static int __init s3c64xx_spi_probe(struct platform_device *pdev)
 		goto err6;
 	}
 
-	sdd->workqueue = create_singlethread_workqueue(
-						dev_name(master->dev.parent));
-	if (sdd->workqueue == NULL) {
-		dev_err(&pdev->dev, "Unable to create workqueue\n");
-		ret = -ENOMEM;
-		goto err7;
-	}
-
 	/* Setup Deufult Mode */
 	s3c64xx_spi_hwinit(sdd, pdev->id);
 
 	spin_lock_init(&sdd->lock);
 	init_completion(&sdd->xfer_completion);
-	INIT_WORK(&sdd->work, s3c64xx_spi_work);
 	INIT_LIST_HEAD(&sdd->queue);
 
+	ret = request_irq(irq, s3c64xx_spi_irq, 0, "spi-s3c64xx", sdd);
+	if (ret != 0) {
+		dev_err(&pdev->dev, "Failed to request IRQ %d: %d\n",
+			irq, ret);
+		goto err7;
+	}
+
+	writel(S3C64XX_SPI_INT_RX_OVERRUN_EN | S3C64XX_SPI_INT_RX_UNDERRUN_EN |
+	       S3C64XX_SPI_INT_TX_OVERRUN_EN | S3C64XX_SPI_INT_TX_UNDERRUN_EN,
+	       sdd->regs + S3C64XX_SPI_INT_EN);
+
 	if (spi_register_master(master)) {
 		dev_err(&pdev->dev, "cannot register SPI master\n");
 		ret = -EBUSY;
@@ -1113,10 +1105,12 @@ static int __init s3c64xx_spi_probe(struct platform_device *pdev)
 					mem_res->end, mem_res->start,
 					sdd->rx_dma.dmach, sdd->tx_dma.dmach);
 
+	pm_runtime_enable(&pdev->dev);
+
 	return 0;
 
 err8:
-	destroy_workqueue(sdd->workqueue);
+	free_irq(irq, sdd);
 err7:
 	clk_disable(sdd->src_clk);
 err6:
@@ -1142,18 +1136,14 @@ static int s3c64xx_spi_remove(struct platform_device *pdev)
 	struct spi_master *master = spi_master_get(platform_get_drvdata(pdev));
 	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
 	struct resource	*mem_res;
-	unsigned long flags;
-
-	spin_lock_irqsave(&sdd->lock, flags);
-	sdd->state |= SUSPND;
-	spin_unlock_irqrestore(&sdd->lock, flags);
 
-	while (sdd->state & SPIBUSY)
-		msleep(10);
+	pm_runtime_disable(&pdev->dev);
 
 	spi_unregister_master(master);
 
-	destroy_workqueue(sdd->workqueue);
+	writel(0, sdd->regs + S3C64XX_SPI_INT_EN);
+
+	free_irq(platform_get_irq(pdev, 0), sdd);
 
 	clk_disable(sdd->src_clk);
 	clk_put(sdd->src_clk);
@@ -1174,18 +1164,12 @@ static int s3c64xx_spi_remove(struct platform_device *pdev)
 }
 
 #ifdef CONFIG_PM
-static int s3c64xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
+static int s3c64xx_spi_suspend(struct device *dev)
 {
-	struct spi_master *master = spi_master_get(platform_get_drvdata(pdev));
+	struct spi_master *master = spi_master_get(dev_get_drvdata(dev));
 	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
-	unsigned long flags;
-
-	spin_lock_irqsave(&sdd->lock, flags);
-	sdd->state |= SUSPND;
-	spin_unlock_irqrestore(&sdd->lock, flags);
 
-	while (sdd->state & SPIBUSY)
-		msleep(10);
+	spi_master_suspend(master);
 
 	/* Disable the clock */
 	clk_disable(sdd->src_clk);
@@ -1196,12 +1180,12 @@ static int s3c64xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
 	return 0;
 }
 
-static int s3c64xx_spi_resume(struct platform_device *pdev)
+static int s3c64xx_spi_resume(struct device *dev)
 {
-	struct spi_master *master = spi_master_get(platform_get_drvdata(pdev));
+	struct platform_device *pdev = to_platform_device(dev);
+	struct spi_master *master = spi_master_get(dev_get_drvdata(dev));
 	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
 	struct s3c64xx_spi_info *sci = sdd->cntrlr_info;
-	unsigned long flags;
 
 	sci->cfg_gpio(pdev);
 
@@ -1211,25 +1195,49 @@ static int s3c64xx_spi_resume(struct platform_device *pdev)
 
 	s3c64xx_spi_hwinit(sdd, pdev->id);
 
-	spin_lock_irqsave(&sdd->lock, flags);
-	sdd->state &= ~SUSPND;
-	spin_unlock_irqrestore(&sdd->lock, flags);
+	spi_master_resume(master);
 
 	return 0;
 }
-#else
-#define s3c64xx_spi_suspend	NULL
-#define s3c64xx_spi_resume	NULL
 #endif /* CONFIG_PM */
 
+#ifdef CONFIG_PM_RUNTIME
+static int s3c64xx_spi_runtime_suspend(struct device *dev)
+{
+	struct spi_master *master = spi_master_get(dev_get_drvdata(dev));
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
+
+	clk_disable(sdd->clk);
+	clk_disable(sdd->src_clk);
+
+	return 0;
+}
+
+static int s3c64xx_spi_runtime_resume(struct device *dev)
+{
+	struct spi_master *master = spi_master_get(dev_get_drvdata(dev));
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
+
+	clk_enable(sdd->src_clk);
+	clk_enable(sdd->clk);
+
+	return 0;
+}
+#endif /* CONFIG_PM_RUNTIME */
+
+static const struct dev_pm_ops s3c64xx_spi_pm = {
+	SET_SYSTEM_SLEEP_PM_OPS(s3c64xx_spi_suspend, s3c64xx_spi_resume)
+	SET_RUNTIME_PM_OPS(s3c64xx_spi_runtime_suspend,
+			   s3c64xx_spi_runtime_resume, NULL)
+};
+
 static struct platform_driver s3c64xx_spi_driver = {
 	.driver = {
 		.name	= "s3c64xx-spi",
 		.owner = THIS_MODULE,
+		.pm = &s3c64xx_spi_pm,
 	},
 	.remove = s3c64xx_spi_remove,
-	.suspend = s3c64xx_spi_suspend,
-	.resume = s3c64xx_spi_resume,
 };
 MODULE_ALIAS("platform:s3c64xx-spi");
 
diff --git a/drivers/spi/spi-sh-hspi.c b/drivers/spi/spi-sh-hspi.c
new file mode 100644
index 000000000000..934138c7b3d3
--- /dev/null
+++ b/drivers/spi/spi-sh-hspi.c
@@ -0,0 +1,331 @@
+/*
+ * SuperH HSPI bus driver
+ *
+ * Copyright (C) 2011  Kuninori Morimoto
+ *
+ * Based on spi-sh.c:
+ * Based on pxa2xx_spi.c:
+ * Copyright (C) 2011 Renesas Solutions Corp.
+ * Copyright (C) 2005 Stephen Street / StreetFire Sound Labs
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ *
+ */
+
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/timer.h>
+#include <linux/delay.h>
+#include <linux/list.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/io.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/sh_hspi.h>
+
+#define SPCR	0x00
+#define SPSR	0x04
+#define SPSCR	0x08
+#define SPTBR	0x0C
+#define SPRBR	0x10
+#define SPCR2	0x14
+
+/* SPSR */
+#define RXFL	(1 << 2)
+
+#define hspi2info(h)	(h->dev->platform_data)
+
+struct hspi_priv {
+	void __iomem *addr;
+	struct spi_master *master;
+	struct device *dev;
+	struct clk *clk;
+};
+
+/*
+ *		basic function
+ */
+static void hspi_write(struct hspi_priv *hspi, int reg, u32 val)
+{
+	iowrite32(val, hspi->addr + reg);
+}
+
+static u32 hspi_read(struct hspi_priv *hspi, int reg)
+{
+	return ioread32(hspi->addr + reg);
+}
+
+/*
+ *		transfer function
+ */
+static int hspi_status_check_timeout(struct hspi_priv *hspi, u32 mask, u32 val)
+{
+	int t = 256;
+
+	while (t--) {
+		if ((mask & hspi_read(hspi, SPSR)) == val)
+			return 0;
+
+		msleep(20);
+	}
+
+	dev_err(hspi->dev, "timeout\n");
+	return -ETIMEDOUT;
+}
+
+/*
+ *		spi master function
+ */
+static int hspi_prepare_transfer(struct spi_master *master)
+{
+	struct hspi_priv *hspi = spi_master_get_devdata(master);
+
+	pm_runtime_get_sync(hspi->dev);
+	return 0;
+}
+
+static int hspi_unprepare_transfer(struct spi_master *master)
+{
+	struct hspi_priv *hspi = spi_master_get_devdata(master);
+
+	pm_runtime_put_sync(hspi->dev);
+	return 0;
+}
+
+static void hspi_hw_setup(struct hspi_priv *hspi,
+			  struct spi_message *msg,
+			  struct spi_transfer *t)
+{
+	struct spi_device *spi = msg->spi;
+	struct device *dev = hspi->dev;
+	u32 target_rate;
+	u32 spcr, idiv_clk;
+	u32 rate, best_rate, min, tmp;
+
+	target_rate = t ? t->speed_hz : 0;
+	if (!target_rate)
+		target_rate = spi->max_speed_hz;
+
+	/*
+	 * find best IDIV/CLKCx settings
+	 */
+	min = ~0;
+	best_rate = 0;
+	spcr = 0;
+	for (idiv_clk = 0x00; idiv_clk <= 0x3F; idiv_clk++) {
+		rate = clk_get_rate(hspi->clk);
+
+		/* IDIV calculation */
+		if (idiv_clk & (1 << 5))
+			rate /= 128;
+		else
+			rate /= 16;
+
+		/* CLKCx calculation */
+		rate /= (((idiv_clk & 0x1F) + 1) * 2) ;
+
+		/* save best settings */
+		tmp = abs(target_rate - rate);
+		if (tmp < min) {
+			min = tmp;
+			spcr = idiv_clk;
+			best_rate = rate;
+		}
+	}
+
+	if (spi->mode & SPI_CPHA)
+		spcr |= 1 << 7;
+	if (spi->mode & SPI_CPOL)
+		spcr |= 1 << 6;
+
+	dev_dbg(dev, "speed %d/%d\n", target_rate, best_rate);
+
+	hspi_write(hspi, SPCR, spcr);
+	hspi_write(hspi, SPSR, 0x0);
+	hspi_write(hspi, SPSCR, 0x1);	/* master mode */
+}
+
+static int hspi_transfer_one_message(struct spi_master *master,
+				     struct spi_message *msg)
+{
+	struct hspi_priv *hspi = spi_master_get_devdata(master);
+	struct spi_transfer *t;
+	u32 tx;
+	u32 rx;
+	int ret, i;
+
+	dev_dbg(hspi->dev, "%s\n", __func__);
+
+	ret = 0;
+	list_for_each_entry(t, &msg->transfers, transfer_list) {
+		hspi_hw_setup(hspi, msg, t);
+
+		for (i = 0; i < t->len; i++) {
+
+			/* wait remains */
+			ret = hspi_status_check_timeout(hspi, 0x1, 0);
+			if (ret < 0)
+				break;
+
+			tx = 0;
+			if (t->tx_buf)
+				tx = (u32)((u8 *)t->tx_buf)[i];
+
+			hspi_write(hspi, SPTBR, tx);
+
+			/* wait recive */
+			ret = hspi_status_check_timeout(hspi, 0x4, 0x4);
+			if (ret < 0)
+				break;
+
+			rx = hspi_read(hspi, SPRBR);
+			if (t->rx_buf)
+				((u8 *)t->rx_buf)[i] = (u8)rx;
+
+		}
+
+		msg->actual_length += t->len;
+	}
+
+	msg->status = ret;
+	spi_finalize_current_message(master);
+
+	return ret;
+}
+
+static int hspi_setup(struct spi_device *spi)
+{
+	struct hspi_priv *hspi = spi_master_get_devdata(spi->master);
+	struct device *dev = hspi->dev;
+
+	if (8 != spi->bits_per_word) {
+		dev_err(dev, "bits_per_word should be 8\n");
+		return -EIO;
+	}
+
+	dev_dbg(dev, "%s setup\n", spi->modalias);
+
+	return 0;
+}
+
+static void hspi_cleanup(struct spi_device *spi)
+{
+	struct hspi_priv *hspi = spi_master_get_devdata(spi->master);
+	struct device *dev = hspi->dev;
+
+	dev_dbg(dev, "%s cleanup\n", spi->modalias);
+}
+
+static int __devinit hspi_probe(struct platform_device *pdev)
+{
+	struct resource *res;
+	struct spi_master *master;
+	struct hspi_priv *hspi;
+	struct clk *clk;
+	int ret;
+
+	/* get base addr */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res) {
+		dev_err(&pdev->dev, "invalid resource\n");
+		return -EINVAL;
+	}
+
+	master = spi_alloc_master(&pdev->dev, sizeof(*hspi));
+	if (!master) {
+		dev_err(&pdev->dev, "spi_alloc_master error.\n");
+		return -ENOMEM;
+	}
+
+	clk = clk_get(NULL, "shyway_clk");
+	if (!clk) {
+		dev_err(&pdev->dev, "shyway_clk is required\n");
+		ret = -EINVAL;
+		goto error0;
+	}
+
+	hspi = spi_master_get_devdata(master);
+	dev_set_drvdata(&pdev->dev, hspi);
+
+	/* init hspi */
+	hspi->master	= master;
+	hspi->dev	= &pdev->dev;
+	hspi->clk	= clk;
+	hspi->addr	= devm_ioremap(hspi->dev,
+				       res->start, resource_size(res));
+	if (!hspi->addr) {
+		dev_err(&pdev->dev, "ioremap error.\n");
+		ret = -ENOMEM;
+		goto error1;
+	}
+
+	master->num_chipselect	= 1;
+	master->bus_num		= pdev->id;
+	master->setup		= hspi_setup;
+	master->cleanup		= hspi_cleanup;
+	master->mode_bits	= SPI_CPOL | SPI_CPHA;
+	master->prepare_transfer_hardware	= hspi_prepare_transfer;
+	master->transfer_one_message		= hspi_transfer_one_message;
+	master->unprepare_transfer_hardware	= hspi_unprepare_transfer;
+	ret = spi_register_master(master);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "spi_register_master error.\n");
+		goto error2;
+	}
+
+	pm_runtime_enable(&pdev->dev);
+
+	dev_info(&pdev->dev, "probed\n");
+
+	return 0;
+
+ error2:
+	devm_iounmap(hspi->dev, hspi->addr);
+ error1:
+	clk_put(clk);
+ error0:
+	spi_master_put(master);
+
+	return ret;
+}
+
+static int __devexit hspi_remove(struct platform_device *pdev)
+{
+	struct hspi_priv *hspi = dev_get_drvdata(&pdev->dev);
+
+	pm_runtime_disable(&pdev->dev);
+
+	clk_put(hspi->clk);
+	spi_unregister_master(hspi->master);
+	devm_iounmap(hspi->dev, hspi->addr);
+
+	return 0;
+}
+
+static struct platform_driver hspi_driver = {
+	.probe = hspi_probe,
+	.remove = __devexit_p(hspi_remove),
+	.driver = {
+		.name = "sh-hspi",
+		.owner = THIS_MODULE,
+	},
+};
+module_platform_driver(hspi_driver);
+
+MODULE_DESCRIPTION("SuperH HSPI bus driver");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>");
+MODULE_ALIAS("platform:sh_spi");
diff --git a/drivers/spi/spi-sh.c b/drivers/spi/spi-sh.c
index 70c8af9f7ccc..79442c31bcd9 100644
--- a/drivers/spi/spi-sh.c
+++ b/drivers/spi/spi-sh.c
@@ -92,17 +92,26 @@ struct spi_sh_data {
 	unsigned long cr1;
 	wait_queue_head_t wait;
 	spinlock_t lock;
+	int width;
 };
 
 static void spi_sh_write(struct spi_sh_data *ss, unsigned long data,
 			     unsigned long offset)
 {
-	writel(data, ss->addr + offset);
+	if (ss->width == 8)
+		iowrite8(data, ss->addr + (offset >> 2));
+	else if (ss->width == 32)
+		iowrite32(data, ss->addr + offset);
 }
 
 static unsigned long spi_sh_read(struct spi_sh_data *ss, unsigned long offset)
 {
-	return readl(ss->addr + offset);
+	if (ss->width == 8)
+		return ioread8(ss->addr + (offset >> 2));
+	else if (ss->width == 32)
+		return ioread32(ss->addr + offset);
+	else
+		return 0;
 }
 
 static void spi_sh_set_bit(struct spi_sh_data *ss, unsigned long val,
@@ -464,6 +473,18 @@ static int __devinit spi_sh_probe(struct platform_device *pdev)
 	ss = spi_master_get_devdata(master);
 	dev_set_drvdata(&pdev->dev, ss);
 
+	switch (res->flags & IORESOURCE_MEM_TYPE_MASK) {
+	case IORESOURCE_MEM_8BIT:
+		ss->width = 8;
+		break;
+	case IORESOURCE_MEM_32BIT:
+		ss->width = 32;
+		break;
+	default:
+		dev_err(&pdev->dev, "No support width\n");
+		ret = -ENODEV;
+		goto error1;
+	}
 	ss->irq = irq;
 	ss->master = master;
 	ss->addr = ioremap(res->start, resource_size(res));
diff --git a/drivers/spi/spi-sirf.c b/drivers/spi/spi-sirf.c
new file mode 100644
index 000000000000..52fe495bb32a
--- /dev/null
+++ b/drivers/spi/spi-sirf.c
@@ -0,0 +1,687 @@
+/*
+ * SPI bus driver for CSR SiRFprimaII
+ *
+ * Copyright (c) 2011 Cambridge Silicon Radio Limited, a CSR plc group company.
+ *
+ * Licensed under GPLv2 or later.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/clk.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/bitops.h>
+#include <linux/err.h>
+#include <linux/platform_device.h>
+#include <linux/of_gpio.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/spi_bitbang.h>
+#include <linux/pinctrl/pinmux.h>
+
+#define DRIVER_NAME "sirfsoc_spi"
+
+#define SIRFSOC_SPI_CTRL		0x0000
+#define SIRFSOC_SPI_CMD			0x0004
+#define SIRFSOC_SPI_TX_RX_EN		0x0008
+#define SIRFSOC_SPI_INT_EN		0x000C
+#define SIRFSOC_SPI_INT_STATUS		0x0010
+#define SIRFSOC_SPI_TX_DMA_IO_CTRL	0x0100
+#define SIRFSOC_SPI_TX_DMA_IO_LEN	0x0104
+#define SIRFSOC_SPI_TXFIFO_CTRL		0x0108
+#define SIRFSOC_SPI_TXFIFO_LEVEL_CHK	0x010C
+#define SIRFSOC_SPI_TXFIFO_OP		0x0110
+#define SIRFSOC_SPI_TXFIFO_STATUS	0x0114
+#define SIRFSOC_SPI_TXFIFO_DATA		0x0118
+#define SIRFSOC_SPI_RX_DMA_IO_CTRL	0x0120
+#define SIRFSOC_SPI_RX_DMA_IO_LEN	0x0124
+#define SIRFSOC_SPI_RXFIFO_CTRL		0x0128
+#define SIRFSOC_SPI_RXFIFO_LEVEL_CHK	0x012C
+#define SIRFSOC_SPI_RXFIFO_OP		0x0130
+#define SIRFSOC_SPI_RXFIFO_STATUS	0x0134
+#define SIRFSOC_SPI_RXFIFO_DATA		0x0138
+#define SIRFSOC_SPI_DUMMY_DELAY_CTL	0x0144
+
+/* SPI CTRL register defines */
+#define SIRFSOC_SPI_SLV_MODE		BIT(16)
+#define SIRFSOC_SPI_CMD_MODE		BIT(17)
+#define SIRFSOC_SPI_CS_IO_OUT		BIT(18)
+#define SIRFSOC_SPI_CS_IO_MODE		BIT(19)
+#define SIRFSOC_SPI_CLK_IDLE_STAT	BIT(20)
+#define SIRFSOC_SPI_CS_IDLE_STAT	BIT(21)
+#define SIRFSOC_SPI_TRAN_MSB		BIT(22)
+#define SIRFSOC_SPI_DRV_POS_EDGE	BIT(23)
+#define SIRFSOC_SPI_CS_HOLD_TIME	BIT(24)
+#define SIRFSOC_SPI_CLK_SAMPLE_MODE	BIT(25)
+#define SIRFSOC_SPI_TRAN_DAT_FORMAT_8	(0 << 26)
+#define SIRFSOC_SPI_TRAN_DAT_FORMAT_12	(1 << 26)
+#define SIRFSOC_SPI_TRAN_DAT_FORMAT_16	(2 << 26)
+#define SIRFSOC_SPI_TRAN_DAT_FORMAT_32	(3 << 26)
+#define SIRFSOC_SPI_CMD_BYTE_NUM(x)		((x & 3) << 28)
+#define SIRFSOC_SPI_ENA_AUTO_CLR		BIT(30)
+#define SIRFSOC_SPI_MUL_DAT_MODE		BIT(31)
+
+/* Interrupt Enable */
+#define SIRFSOC_SPI_RX_DONE_INT_EN		BIT(0)
+#define SIRFSOC_SPI_TX_DONE_INT_EN		BIT(1)
+#define SIRFSOC_SPI_RX_OFLOW_INT_EN		BIT(2)
+#define SIRFSOC_SPI_TX_UFLOW_INT_EN		BIT(3)
+#define SIRFSOC_SPI_RX_IO_DMA_INT_EN	BIT(4)
+#define SIRFSOC_SPI_TX_IO_DMA_INT_EN	BIT(5)
+#define SIRFSOC_SPI_RXFIFO_FULL_INT_EN	BIT(6)
+#define SIRFSOC_SPI_TXFIFO_EMPTY_INT_EN	BIT(7)
+#define SIRFSOC_SPI_RXFIFO_THD_INT_EN	BIT(8)
+#define SIRFSOC_SPI_TXFIFO_THD_INT_EN	BIT(9)
+#define SIRFSOC_SPI_FRM_END_INT_EN	BIT(10)
+
+#define SIRFSOC_SPI_INT_MASK_ALL		0x1FFF
+
+/* Interrupt status */
+#define SIRFSOC_SPI_RX_DONE		BIT(0)
+#define SIRFSOC_SPI_TX_DONE		BIT(1)
+#define SIRFSOC_SPI_RX_OFLOW		BIT(2)
+#define SIRFSOC_SPI_TX_UFLOW		BIT(3)
+#define SIRFSOC_SPI_RX_FIFO_FULL	BIT(6)
+#define SIRFSOC_SPI_TXFIFO_EMPTY	BIT(7)
+#define SIRFSOC_SPI_RXFIFO_THD_REACH	BIT(8)
+#define SIRFSOC_SPI_TXFIFO_THD_REACH	BIT(9)
+#define SIRFSOC_SPI_FRM_END		BIT(10)
+
+/* TX RX enable */
+#define SIRFSOC_SPI_RX_EN		BIT(0)
+#define SIRFSOC_SPI_TX_EN		BIT(1)
+#define SIRFSOC_SPI_CMD_TX_EN		BIT(2)
+
+#define SIRFSOC_SPI_IO_MODE_SEL		BIT(0)
+#define SIRFSOC_SPI_RX_DMA_FLUSH	BIT(2)
+
+/* FIFO OPs */
+#define SIRFSOC_SPI_FIFO_RESET		BIT(0)
+#define SIRFSOC_SPI_FIFO_START		BIT(1)
+
+/* FIFO CTRL */
+#define SIRFSOC_SPI_FIFO_WIDTH_BYTE	(0 << 0)
+#define SIRFSOC_SPI_FIFO_WIDTH_WORD	(1 << 0)
+#define SIRFSOC_SPI_FIFO_WIDTH_DWORD	(2 << 0)
+
+/* FIFO Status */
+#define	SIRFSOC_SPI_FIFO_LEVEL_MASK	0xFF
+#define SIRFSOC_SPI_FIFO_FULL		BIT(8)
+#define SIRFSOC_SPI_FIFO_EMPTY		BIT(9)
+
+/* 256 bytes rx/tx FIFO */
+#define SIRFSOC_SPI_FIFO_SIZE		256
+#define SIRFSOC_SPI_DAT_FRM_LEN_MAX	(64 * 1024)
+
+#define SIRFSOC_SPI_FIFO_SC(x)		((x) & 0x3F)
+#define SIRFSOC_SPI_FIFO_LC(x)		(((x) & 0x3F) << 10)
+#define SIRFSOC_SPI_FIFO_HC(x)		(((x) & 0x3F) << 20)
+#define SIRFSOC_SPI_FIFO_THD(x)		(((x) & 0xFF) << 2)
+
+struct sirfsoc_spi {
+	struct spi_bitbang bitbang;
+	struct completion done;
+
+	void __iomem *base;
+	u32 ctrl_freq;  /* SPI controller clock speed */
+	struct clk *clk;
+	struct pinmux *pmx;
+
+	/* rx & tx bufs from the spi_transfer */
+	const void *tx;
+	void *rx;
+
+	/* place received word into rx buffer */
+	void (*rx_word) (struct sirfsoc_spi *);
+	/* get word from tx buffer for sending */
+	void (*tx_word) (struct sirfsoc_spi *);
+
+	/* number of words left to be tranmitted/received */
+	unsigned int left_tx_cnt;
+	unsigned int left_rx_cnt;
+
+	/* tasklet to push tx msg into FIFO */
+	struct tasklet_struct tasklet_tx;
+
+	int chipselect[0];
+};
+
+static void spi_sirfsoc_rx_word_u8(struct sirfsoc_spi *sspi)
+{
+	u32 data;
+	u8 *rx = sspi->rx;
+
+	data = readl(sspi->base + SIRFSOC_SPI_RXFIFO_DATA);
+
+	if (rx) {
+		*rx++ = (u8) data;
+		sspi->rx = rx;
+	}
+
+	sspi->left_rx_cnt--;
+}
+
+static void spi_sirfsoc_tx_word_u8(struct sirfsoc_spi *sspi)
+{
+	u32 data = 0;
+	const u8 *tx = sspi->tx;
+
+	if (tx) {
+		data = *tx++;
+		sspi->tx = tx;
+	}
+
+	writel(data, sspi->base + SIRFSOC_SPI_TXFIFO_DATA);
+	sspi->left_tx_cnt--;
+}
+
+static void spi_sirfsoc_rx_word_u16(struct sirfsoc_spi *sspi)
+{
+	u32 data;
+	u16 *rx = sspi->rx;
+
+	data = readl(sspi->base + SIRFSOC_SPI_RXFIFO_DATA);
+
+	if (rx) {
+		*rx++ = (u16) data;
+		sspi->rx = rx;
+	}
+
+	sspi->left_rx_cnt--;
+}
+
+static void spi_sirfsoc_tx_word_u16(struct sirfsoc_spi *sspi)
+{
+	u32 data = 0;
+	const u16 *tx = sspi->tx;
+
+	if (tx) {
+		data = *tx++;
+		sspi->tx = tx;
+	}
+
+	writel(data, sspi->base + SIRFSOC_SPI_TXFIFO_DATA);
+	sspi->left_tx_cnt--;
+}
+
+static void spi_sirfsoc_rx_word_u32(struct sirfsoc_spi *sspi)
+{
+	u32 data;
+	u32 *rx = sspi->rx;
+
+	data = readl(sspi->base + SIRFSOC_SPI_RXFIFO_DATA);
+
+	if (rx) {
+		*rx++ = (u32) data;
+		sspi->rx = rx;
+	}
+
+	sspi->left_rx_cnt--;
+
+}
+
+static void spi_sirfsoc_tx_word_u32(struct sirfsoc_spi *sspi)
+{
+	u32 data = 0;
+	const u32 *tx = sspi->tx;
+
+	if (tx) {
+		data = *tx++;
+		sspi->tx = tx;
+	}
+
+	writel(data, sspi->base + SIRFSOC_SPI_TXFIFO_DATA);
+	sspi->left_tx_cnt--;
+}
+
+static void spi_sirfsoc_tasklet_tx(unsigned long arg)
+{
+	struct sirfsoc_spi *sspi = (struct sirfsoc_spi *)arg;
+
+	/* Fill Tx FIFO while there are left words to be transmitted */
+	while (!((readl(sspi->base + SIRFSOC_SPI_TXFIFO_STATUS) &
+			SIRFSOC_SPI_FIFO_FULL)) &&
+			sspi->left_tx_cnt)
+		sspi->tx_word(sspi);
+}
+
+static irqreturn_t spi_sirfsoc_irq(int irq, void *dev_id)
+{
+	struct sirfsoc_spi *sspi = dev_id;
+	u32 spi_stat = readl(sspi->base + SIRFSOC_SPI_INT_STATUS);
+
+	writel(spi_stat, sspi->base + SIRFSOC_SPI_INT_STATUS);
+
+	/* Error Conditions */
+	if (spi_stat & SIRFSOC_SPI_RX_OFLOW ||
+			spi_stat & SIRFSOC_SPI_TX_UFLOW) {
+		complete(&sspi->done);
+		writel(0x0, sspi->base + SIRFSOC_SPI_INT_EN);
+	}
+
+	if (spi_stat & SIRFSOC_SPI_FRM_END) {
+		while (!((readl(sspi->base + SIRFSOC_SPI_RXFIFO_STATUS)
+				& SIRFSOC_SPI_FIFO_EMPTY)) &&
+				sspi->left_rx_cnt)
+			sspi->rx_word(sspi);
+
+		/* Received all words */
+		if ((sspi->left_rx_cnt == 0) && (sspi->left_tx_cnt == 0)) {
+			complete(&sspi->done);
+			writel(0x0, sspi->base + SIRFSOC_SPI_INT_EN);
+		}
+	}
+
+	if (spi_stat & SIRFSOC_SPI_RXFIFO_THD_REACH ||
+		spi_stat & SIRFSOC_SPI_TXFIFO_THD_REACH ||
+		spi_stat & SIRFSOC_SPI_RX_FIFO_FULL ||
+		spi_stat & SIRFSOC_SPI_TXFIFO_EMPTY)
+		tasklet_schedule(&sspi->tasklet_tx);
+
+	return IRQ_HANDLED;
+}
+
+static int spi_sirfsoc_transfer(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct sirfsoc_spi *sspi;
+	int timeout = t->len * 10;
+	sspi = spi_master_get_devdata(spi->master);
+
+	sspi->tx = t->tx_buf;
+	sspi->rx = t->rx_buf;
+	sspi->left_tx_cnt = sspi->left_rx_cnt = t->len;
+	INIT_COMPLETION(sspi->done);
+
+	writel(SIRFSOC_SPI_INT_MASK_ALL, sspi->base + SIRFSOC_SPI_INT_STATUS);
+
+	if (t->len == 1) {
+		writel(readl(sspi->base + SIRFSOC_SPI_CTRL) |
+			SIRFSOC_SPI_ENA_AUTO_CLR,
+			sspi->base + SIRFSOC_SPI_CTRL);
+		writel(0, sspi->base + SIRFSOC_SPI_TX_DMA_IO_LEN);
+		writel(0, sspi->base + SIRFSOC_SPI_RX_DMA_IO_LEN);
+	} else if ((t->len > 1) && (t->len < SIRFSOC_SPI_DAT_FRM_LEN_MAX)) {
+		writel(readl(sspi->base + SIRFSOC_SPI_CTRL) |
+				SIRFSOC_SPI_MUL_DAT_MODE |
+				SIRFSOC_SPI_ENA_AUTO_CLR,
+			sspi->base + SIRFSOC_SPI_CTRL);
+		writel(t->len - 1, sspi->base + SIRFSOC_SPI_TX_DMA_IO_LEN);
+		writel(t->len - 1, sspi->base + SIRFSOC_SPI_RX_DMA_IO_LEN);
+	} else {
+		writel(readl(sspi->base + SIRFSOC_SPI_CTRL),
+			sspi->base + SIRFSOC_SPI_CTRL);
+		writel(0, sspi->base + SIRFSOC_SPI_TX_DMA_IO_LEN);
+		writel(0, sspi->base + SIRFSOC_SPI_RX_DMA_IO_LEN);
+	}
+
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+
+	/* Send the first word to trigger the whole tx/rx process */
+	sspi->tx_word(sspi);
+
+	writel(SIRFSOC_SPI_RX_OFLOW_INT_EN | SIRFSOC_SPI_TX_UFLOW_INT_EN |
+		SIRFSOC_SPI_RXFIFO_THD_INT_EN | SIRFSOC_SPI_TXFIFO_THD_INT_EN |
+		SIRFSOC_SPI_FRM_END_INT_EN | SIRFSOC_SPI_RXFIFO_FULL_INT_EN |
+		SIRFSOC_SPI_TXFIFO_EMPTY_INT_EN, sspi->base + SIRFSOC_SPI_INT_EN);
+	writel(SIRFSOC_SPI_RX_EN | SIRFSOC_SPI_TX_EN, sspi->base + SIRFSOC_SPI_TX_RX_EN);
+
+	if (wait_for_completion_timeout(&sspi->done, timeout) == 0)
+		dev_err(&spi->dev, "transfer timeout\n");
+
+	/* TX, RX FIFO stop */
+	writel(0, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(0, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+	writel(0, sspi->base + SIRFSOC_SPI_TX_RX_EN);
+	writel(0, sspi->base + SIRFSOC_SPI_INT_EN);
+
+	return t->len - sspi->left_rx_cnt;
+}
+
+static void spi_sirfsoc_chipselect(struct spi_device *spi, int value)
+{
+	struct sirfsoc_spi *sspi = spi_master_get_devdata(spi->master);
+
+	if (sspi->chipselect[spi->chip_select] == 0) {
+		u32 regval = readl(sspi->base + SIRFSOC_SPI_CTRL);
+		regval |= SIRFSOC_SPI_CS_IO_OUT;
+		switch (value) {
+		case BITBANG_CS_ACTIVE:
+			if (spi->mode & SPI_CS_HIGH)
+				regval |= SIRFSOC_SPI_CS_IO_OUT;
+			else
+				regval &= ~SIRFSOC_SPI_CS_IO_OUT;
+			break;
+		case BITBANG_CS_INACTIVE:
+			if (spi->mode & SPI_CS_HIGH)
+				regval &= ~SIRFSOC_SPI_CS_IO_OUT;
+			else
+				regval |= SIRFSOC_SPI_CS_IO_OUT;
+			break;
+		}
+		writel(regval, sspi->base + SIRFSOC_SPI_CTRL);
+	} else {
+		int gpio = sspi->chipselect[spi->chip_select];
+		gpio_direction_output(gpio, spi->mode & SPI_CS_HIGH ? 0 : 1);
+	}
+}
+
+static int
+spi_sirfsoc_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct sirfsoc_spi *sspi;
+	u8 bits_per_word = 0;
+	int hz = 0;
+	u32 regval;
+	u32 txfifo_ctrl, rxfifo_ctrl;
+	u32 fifo_size = SIRFSOC_SPI_FIFO_SIZE / 4;
+
+	sspi = spi_master_get_devdata(spi->master);
+
+	bits_per_word = t && t->bits_per_word ? t->bits_per_word :
+		spi->bits_per_word;
+	hz = t && t->speed_hz ? t->speed_hz : spi->max_speed_hz;
+
+	/* Enable IO mode for RX, TX */
+	writel(SIRFSOC_SPI_IO_MODE_SEL, sspi->base + SIRFSOC_SPI_TX_DMA_IO_CTRL);
+	writel(SIRFSOC_SPI_IO_MODE_SEL, sspi->base + SIRFSOC_SPI_RX_DMA_IO_CTRL);
+	regval = (sspi->ctrl_freq / (2 * hz)) - 1;
+
+	if (regval > 0xFFFF || regval < 0) {
+		dev_err(&spi->dev, "Speed %d not supported\n", hz);
+		return -EINVAL;
+	}
+
+	switch (bits_per_word) {
+	case 8:
+		regval |= SIRFSOC_SPI_TRAN_DAT_FORMAT_8;
+		sspi->rx_word = spi_sirfsoc_rx_word_u8;
+		sspi->tx_word = spi_sirfsoc_tx_word_u8;
+		txfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_BYTE;
+		rxfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_BYTE;
+		break;
+	case 12:
+	case 16:
+		regval |= (bits_per_word ==  12) ? SIRFSOC_SPI_TRAN_DAT_FORMAT_12 :
+			SIRFSOC_SPI_TRAN_DAT_FORMAT_16;
+		sspi->rx_word = spi_sirfsoc_rx_word_u16;
+		sspi->tx_word = spi_sirfsoc_tx_word_u16;
+		txfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_WORD;
+		rxfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_WORD;
+		break;
+	case 32:
+		regval |= SIRFSOC_SPI_TRAN_DAT_FORMAT_32;
+		sspi->rx_word = spi_sirfsoc_rx_word_u32;
+		sspi->tx_word = spi_sirfsoc_tx_word_u32;
+		txfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_DWORD;
+		rxfifo_ctrl = SIRFSOC_SPI_FIFO_THD(SIRFSOC_SPI_FIFO_SIZE / 2) |
+					SIRFSOC_SPI_FIFO_WIDTH_DWORD;
+		break;
+	default:
+		dev_err(&spi->dev, "Bits per word %d not supported\n",
+		       bits_per_word);
+		return -EINVAL;
+	}
+
+	if (!(spi->mode & SPI_CS_HIGH))
+		regval |= SIRFSOC_SPI_CS_IDLE_STAT;
+	if (!(spi->mode & SPI_LSB_FIRST))
+		regval |= SIRFSOC_SPI_TRAN_MSB;
+	if (spi->mode & SPI_CPOL)
+		regval |= SIRFSOC_SPI_CLK_IDLE_STAT;
+
+	/*
+	 * Data should be driven at least 1/2 cycle before the fetch edge to make
+	 * sure that data gets stable at the fetch edge.
+	 */
+	if (((spi->mode & SPI_CPOL) && (spi->mode & SPI_CPHA)) ||
+	    (!(spi->mode & SPI_CPOL) && !(spi->mode & SPI_CPHA)))
+		regval &= ~SIRFSOC_SPI_DRV_POS_EDGE;
+	else
+		regval |= SIRFSOC_SPI_DRV_POS_EDGE;
+
+	writel(SIRFSOC_SPI_FIFO_SC(fifo_size - 2) |
+			SIRFSOC_SPI_FIFO_LC(fifo_size / 2) |
+			SIRFSOC_SPI_FIFO_HC(2),
+		sspi->base + SIRFSOC_SPI_TXFIFO_LEVEL_CHK);
+	writel(SIRFSOC_SPI_FIFO_SC(2) |
+			SIRFSOC_SPI_FIFO_LC(fifo_size / 2) |
+			SIRFSOC_SPI_FIFO_HC(fifo_size - 2),
+		sspi->base + SIRFSOC_SPI_RXFIFO_LEVEL_CHK);
+	writel(txfifo_ctrl, sspi->base + SIRFSOC_SPI_TXFIFO_CTRL);
+	writel(rxfifo_ctrl, sspi->base + SIRFSOC_SPI_RXFIFO_CTRL);
+
+	writel(regval, sspi->base + SIRFSOC_SPI_CTRL);
+	return 0;
+}
+
+static int spi_sirfsoc_setup(struct spi_device *spi)
+{
+	struct sirfsoc_spi *sspi;
+
+	if (!spi->max_speed_hz)
+		return -EINVAL;
+
+	sspi = spi_master_get_devdata(spi->master);
+
+	if (!spi->bits_per_word)
+		spi->bits_per_word = 8;
+
+	return spi_sirfsoc_setup_transfer(spi, NULL);
+}
+
+static int __devinit spi_sirfsoc_probe(struct platform_device *pdev)
+{
+	struct sirfsoc_spi *sspi;
+	struct spi_master *master;
+	struct resource *mem_res;
+	int num_cs, cs_gpio, irq;
+	int i;
+	int ret;
+
+	ret = of_property_read_u32(pdev->dev.of_node,
+			"sirf,spi-num-chipselects", &num_cs);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "Unable to get chip select number\n");
+		goto err_cs;
+	}
+
+	master = spi_alloc_master(&pdev->dev, sizeof(*sspi) + sizeof(int) * num_cs);
+	if (!master) {
+		dev_err(&pdev->dev, "Unable to allocate SPI master\n");
+		return -ENOMEM;
+	}
+	platform_set_drvdata(pdev, master);
+	sspi = spi_master_get_devdata(master);
+
+	mem_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!mem_res) {
+		dev_err(&pdev->dev, "Unable to get IO resource\n");
+		ret = -ENODEV;
+		goto free_master;
+	}
+	master->num_chipselect = num_cs;
+
+	for (i = 0; i < master->num_chipselect; i++) {
+		cs_gpio = of_get_named_gpio(pdev->dev.of_node, "cs-gpios", i);
+		if (cs_gpio < 0) {
+			dev_err(&pdev->dev, "can't get cs gpio from DT\n");
+			ret = -ENODEV;
+			goto free_master;
+		}
+
+		sspi->chipselect[i] = cs_gpio;
+		if (cs_gpio == 0)
+			continue; /* use cs from spi controller */
+
+		ret = gpio_request(cs_gpio, DRIVER_NAME);
+		if (ret) {
+			while (i > 0) {
+				i--;
+				if (sspi->chipselect[i] > 0)
+					gpio_free(sspi->chipselect[i]);
+			}
+			dev_err(&pdev->dev, "fail to request cs gpios\n");
+			goto free_master;
+		}
+	}
+
+	sspi->base = devm_request_and_ioremap(&pdev->dev, mem_res);
+	if (!sspi->base) {
+		dev_err(&pdev->dev, "IO remap failed!\n");
+		ret = -ENOMEM;
+		goto free_master;
+	}
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		ret = -ENXIO;
+		goto free_master;
+	}
+	ret = devm_request_irq(&pdev->dev, irq, spi_sirfsoc_irq, 0,
+				DRIVER_NAME, sspi);
+	if (ret)
+		goto free_master;
+
+	sspi->bitbang.master = spi_master_get(master);
+	sspi->bitbang.chipselect = spi_sirfsoc_chipselect;
+	sspi->bitbang.setup_transfer = spi_sirfsoc_setup_transfer;
+	sspi->bitbang.txrx_bufs = spi_sirfsoc_transfer;
+	sspi->bitbang.master->setup = spi_sirfsoc_setup;
+	master->bus_num = pdev->id;
+	sspi->bitbang.master->dev.of_node = pdev->dev.of_node;
+
+	sspi->pmx = pinmux_get(&pdev->dev, NULL);
+	ret = IS_ERR(sspi->pmx);
+	if (ret)
+		goto free_master;
+
+	pinmux_enable(sspi->pmx);
+
+	sspi->clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(sspi->clk)) {
+		ret = -EINVAL;
+		goto free_pmx;
+	}
+	clk_enable(sspi->clk);
+	sspi->ctrl_freq = clk_get_rate(sspi->clk);
+
+	init_completion(&sspi->done);
+
+	tasklet_init(&sspi->tasklet_tx, spi_sirfsoc_tasklet_tx,
+		     (unsigned long)sspi);
+
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+	/* We are not using dummy delay between command and data */
+	writel(0, sspi->base + SIRFSOC_SPI_DUMMY_DELAY_CTL);
+
+	ret = spi_bitbang_start(&sspi->bitbang);
+	if (ret)
+		goto free_clk;
+
+	dev_info(&pdev->dev, "registerred, bus number = %d\n", master->bus_num);
+
+	return 0;
+
+free_clk:
+	clk_disable(sspi->clk);
+	clk_put(sspi->clk);
+free_pmx:
+	pinmux_disable(sspi->pmx);
+	pinmux_put(sspi->pmx);
+free_master:
+	spi_master_put(master);
+err_cs:
+	return ret;
+}
+
+static int  __devexit spi_sirfsoc_remove(struct platform_device *pdev)
+{
+	struct spi_master *master;
+	struct sirfsoc_spi *sspi;
+	int i;
+
+	master = platform_get_drvdata(pdev);
+	sspi = spi_master_get_devdata(master);
+
+	spi_bitbang_stop(&sspi->bitbang);
+	for (i = 0; i < master->num_chipselect; i++) {
+		if (sspi->chipselect[i] > 0)
+			gpio_free(sspi->chipselect[i]);
+	}
+	clk_disable(sspi->clk);
+	clk_put(sspi->clk);
+	pinmux_disable(sspi->pmx);
+	pinmux_put(sspi->pmx);
+	spi_master_put(master);
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int spi_sirfsoc_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct spi_master *master = platform_get_drvdata(pdev);
+	struct sirfsoc_spi *sspi = spi_master_get_devdata(master);
+
+	clk_disable(sspi->clk);
+	return 0;
+}
+
+static int spi_sirfsoc_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct spi_master *master = platform_get_drvdata(pdev);
+	struct sirfsoc_spi *sspi = spi_master_get_devdata(master);
+
+	clk_enable(sspi->clk);
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_RESET, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_RXFIFO_OP);
+	writel(SIRFSOC_SPI_FIFO_START, sspi->base + SIRFSOC_SPI_TXFIFO_OP);
+
+	return 0;
+}
+
+static const struct dev_pm_ops spi_sirfsoc_pm_ops = {
+	.suspend = spi_sirfsoc_suspend,
+	.resume = spi_sirfsoc_resume,
+};
+#endif
+
+static const struct of_device_id spi_sirfsoc_of_match[] = {
+	{ .compatible = "sirf,prima2-spi", },
+	{}
+};
+MODULE_DEVICE_TABLE(of, sirfsoc_spi_of_match);
+
+static struct platform_driver spi_sirfsoc_driver = {
+	.driver = {
+		.name = DRIVER_NAME,
+		.owner = THIS_MODULE,
+#ifdef CONFIG_PM
+		.pm     = &spi_sirfsoc_pm_ops,
+#endif
+		.of_match_table = spi_sirfsoc_of_match,
+	},
+	.probe = spi_sirfsoc_probe,
+	.remove = __devexit_p(spi_sirfsoc_remove),
+};
+module_platform_driver(spi_sirfsoc_driver);
+
+MODULE_DESCRIPTION("SiRF SoC SPI master driver");
+MODULE_AUTHOR("Zhiwu Song <Zhiwu.Song@csr.com>, "
+		"Barry Song <Baohua.Song@csr.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/spi/spi-topcliff-pch.c b/drivers/spi/spi-topcliff-pch.c
index 10182eb50068..5c6fa5ed3366 100644
--- a/drivers/spi/spi-topcliff-pch.c
+++ b/drivers/spi/spi-topcliff-pch.c
@@ -196,6 +196,7 @@ struct pch_spi_data {
 	struct pch_spi_dma_ctrl dma;
 	int use_dma;
 	u8 irq_reg_sts;
+	int save_total_len;
 };
 
 /**
@@ -216,7 +217,7 @@ struct pch_pd_dev_save {
 	struct pch_spi_board_data *board_dat;
 };
 
-static struct pci_device_id pch_spi_pcidev_id[] = {
+static DEFINE_PCI_DEVICE_TABLE(pch_spi_pcidev_id) = {
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_GE_SPI),    1, },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7213_SPI), 2, },
 	{ PCI_VDEVICE(ROHM, PCI_DEVICE_ID_ML7223_SPI), 1, },
@@ -318,22 +319,23 @@ static void pch_spi_handler_sub(struct pch_spi_data *data, u32 reg_spsr_val,
 		data->tx_index = tx_index;
 		data->rx_index = rx_index;
 
-	}
-
-	/* if transfer complete interrupt */
-	if (reg_spsr_val & SPSR_FI_BIT) {
-		if ((tx_index == bpw_len) && (rx_index == tx_index)) {
-			/* disable interrupts */
-			pch_spi_setclr_reg(data->master, PCH_SPCR, 0, PCH_ALL);
-
-			/* transfer is completed;
-			   inform pch_spi_process_messages */
-			data->transfer_complete = true;
-			data->transfer_active = false;
-			wake_up(&data->wait);
-		} else {
-			dev_err(&data->master->dev,
-				"%s : Transfer is not completed", __func__);
+		/* if transfer complete interrupt */
+		if (reg_spsr_val & SPSR_FI_BIT) {
+			if ((tx_index == bpw_len) && (rx_index == tx_index)) {
+				/* disable interrupts */
+				pch_spi_setclr_reg(data->master, PCH_SPCR, 0,
+						   PCH_ALL);
+
+				/* transfer is completed;
+				   inform pch_spi_process_messages */
+				data->transfer_complete = true;
+				data->transfer_active = false;
+				wake_up(&data->wait);
+			} else {
+				dev_err(&data->master->dev,
+					"%s : Transfer is not completed",
+					__func__);
+			}
 		}
 	}
 }
@@ -822,11 +824,13 @@ static void pch_spi_copy_rx_data_for_dma(struct pch_spi_data *data, int bpw)
 		rx_dma_buf = data->dma.rx_buf_virt;
 		for (j = 0; j < data->bpw_len; j++)
 			*rx_buf++ = *rx_dma_buf++ & 0xFF;
+		data->cur_trans->rx_buf = rx_buf;
 	} else {
 		rx_sbuf = data->cur_trans->rx_buf;
 		rx_dma_sbuf = data->dma.rx_buf_virt;
 		for (j = 0; j < data->bpw_len; j++)
 			*rx_sbuf++ = *rx_dma_sbuf++;
+		data->cur_trans->rx_buf = rx_sbuf;
 	}
 }
 
@@ -852,6 +856,9 @@ static int pch_spi_start_transfer(struct pch_spi_data *data)
 	rtn = wait_event_interruptible_timeout(data->wait,
 					       data->transfer_complete,
 					       msecs_to_jiffies(2 * HZ));
+	if (!rtn)
+		dev_err(&data->master->dev,
+			"%s wait-event timeout\n", __func__);
 
 	dma_sync_sg_for_cpu(&data->master->dev, dma->sg_rx_p, dma->nent,
 			    DMA_FROM_DEVICE);
@@ -923,7 +930,8 @@ static void pch_spi_request_dma(struct pch_spi_data *data, int bpw)
 	dma_cap_set(DMA_SLAVE, mask);
 
 	/* Get DMA's dev information */
-	dma_dev = pci_get_bus_and_slot(2, PCI_DEVFN(12, 0));
+	dma_dev = pci_get_bus_and_slot(data->board_dat->pdev->bus->number,
+				       PCI_DEVFN(12, 0));
 
 	/* Set Tx DMA */
 	param = &dma->param_tx;
@@ -987,6 +995,7 @@ static void pch_spi_handle_dma(struct pch_spi_data *data, int *bpw)
 	int i;
 	int size;
 	int rem;
+	int head;
 	unsigned long flags;
 	struct pch_spi_dma_ctrl *dma;
 
@@ -1015,6 +1024,11 @@ static void pch_spi_handle_dma(struct pch_spi_data *data, int *bpw)
 	}
 	data->bpw_len = data->cur_trans->len / (*bpw / 8);
 
+	if (data->bpw_len > PCH_BUF_SIZE) {
+		data->bpw_len = PCH_BUF_SIZE;
+		data->cur_trans->len -= PCH_BUF_SIZE;
+	}
+
 	/* copy Tx Data */
 	if (data->cur_trans->tx_buf != NULL) {
 		if (*bpw == 8) {
@@ -1029,10 +1043,17 @@ static void pch_spi_handle_dma(struct pch_spi_data *data, int *bpw)
 				*tx_dma_sbuf++ = *tx_sbuf++;
 		}
 	}
+
+	/* Calculate Rx parameter for DMA transmitting */
 	if (data->bpw_len > PCH_DMA_TRANS_SIZE) {
-		num = data->bpw_len / PCH_DMA_TRANS_SIZE + 1;
+		if (data->bpw_len % PCH_DMA_TRANS_SIZE) {
+			num = data->bpw_len / PCH_DMA_TRANS_SIZE + 1;
+			rem = data->bpw_len % PCH_DMA_TRANS_SIZE;
+		} else {
+			num = data->bpw_len / PCH_DMA_TRANS_SIZE;
+			rem = PCH_DMA_TRANS_SIZE;
+		}
 		size = PCH_DMA_TRANS_SIZE;
-		rem = data->bpw_len % PCH_DMA_TRANS_SIZE;
 	} else {
 		num = 1;
 		size = data->bpw_len;
@@ -1092,15 +1113,23 @@ static void pch_spi_handle_dma(struct pch_spi_data *data, int *bpw)
 	dma->nent = num;
 	dma->desc_rx = desc_rx;
 
-	/* TX */
-	if (data->bpw_len > PCH_DMA_TRANS_SIZE) {
-		num = data->bpw_len / PCH_DMA_TRANS_SIZE;
+	/* Calculate Tx parameter for DMA transmitting */
+	if (data->bpw_len > PCH_MAX_FIFO_DEPTH) {
+		head = PCH_MAX_FIFO_DEPTH - PCH_DMA_TRANS_SIZE;
+		if (data->bpw_len % PCH_DMA_TRANS_SIZE > 4) {
+			num = data->bpw_len / PCH_DMA_TRANS_SIZE + 1;
+			rem = data->bpw_len % PCH_DMA_TRANS_SIZE - head;
+		} else {
+			num = data->bpw_len / PCH_DMA_TRANS_SIZE;
+			rem = data->bpw_len % PCH_DMA_TRANS_SIZE +
+			      PCH_DMA_TRANS_SIZE - head;
+		}
 		size = PCH_DMA_TRANS_SIZE;
-		rem = 16;
 	} else {
 		num = 1;
 		size = data->bpw_len;
 		rem = data->bpw_len;
+		head = 0;
 	}
 
 	dma->sg_tx_p = kzalloc(sizeof(struct scatterlist)*num, GFP_ATOMIC);
@@ -1110,11 +1139,17 @@ static void pch_spi_handle_dma(struct pch_spi_data *data, int *bpw)
 	for (i = 0; i < num; i++, sg++) {
 		if (i == 0) {
 			sg->offset = 0;
+			sg_set_page(sg, virt_to_page(dma->tx_buf_virt), size + head,
+				    sg->offset);
+			sg_dma_len(sg) = size + head;
+		} else if (i == (num - 1)) {
+			sg->offset = head + size * i;
+			sg->offset = sg->offset * (*bpw / 8);
 			sg_set_page(sg, virt_to_page(dma->tx_buf_virt), rem,
 				    sg->offset);
 			sg_dma_len(sg) = rem;
 		} else {
-			sg->offset = rem + size * (i - 1);
+			sg->offset = head + size * i;
 			sg->offset = sg->offset * (*bpw / 8);
 			sg_set_page(sg, virt_to_page(dma->tx_buf_virt), size,
 				    sg->offset);
@@ -1202,6 +1237,7 @@ static void pch_spi_process_messages(struct work_struct *pwork)
 				    data->current_msg->spi->bits_per_word);
 	pch_spi_writereg(data->master, PCH_SSNXCR, SSN_NO_CONTROL);
 	do {
+		int cnt;
 		/* If we are already processing a message get the next
 		transfer structure from the message otherwise retrieve
 		the 1st transfer request from the message. */
@@ -1221,11 +1257,28 @@ static void pch_spi_process_messages(struct work_struct *pwork)
 		}
 		spin_unlock(&data->lock);
 
+		if (!data->cur_trans->len)
+			goto out;
+		cnt = (data->cur_trans->len - 1) / PCH_BUF_SIZE + 1;
+		data->save_total_len = data->cur_trans->len;
 		if (data->use_dma) {
-			pch_spi_handle_dma(data, &bpw);
-			if (!pch_spi_start_transfer(data))
-				goto out;
-			pch_spi_copy_rx_data_for_dma(data, bpw);
+			int i;
+			char *save_rx_buf = data->cur_trans->rx_buf;
+			for (i = 0; i < cnt; i ++) {
+				pch_spi_handle_dma(data, &bpw);
+				if (!pch_spi_start_transfer(data)) {
+					data->transfer_complete = true;
+					data->current_msg->status = -EIO;
+					data->current_msg->complete
+						   (data->current_msg->context);
+					data->bcurrent_msg_processing = false;
+					data->current_msg = NULL;
+					data->cur_trans = NULL;
+					goto out;
+				}
+				pch_spi_copy_rx_data_for_dma(data, bpw);
+			}
+			data->cur_trans->rx_buf = save_rx_buf;
 		} else {
 			pch_spi_set_tx(data, &bpw);
 			pch_spi_set_ir(data);
@@ -1236,6 +1289,7 @@ static void pch_spi_process_messages(struct work_struct *pwork)
 			data->pkt_tx_buff = NULL;
 		}
 		/* increment message count */
+		data->cur_trans->len = data->save_total_len;
 		data->current_msg->actual_length += data->cur_trans->len;
 
 		dev_dbg(&data->master->dev,
@@ -1388,6 +1442,7 @@ static int __devinit pch_spi_pd_probe(struct platform_device *plat_dev)
 	master->num_chipselect = PCH_MAX_CS;
 	master->setup = pch_spi_setup;
 	master->transfer = pch_spi_transfer;
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_LSB_FIRST;
 
 	data->board_dat = board_dat;
 	data->plat_dev = plat_dev;
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index b2ccdea30cb9..3d8f662e4fe9 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -30,6 +30,9 @@
 #include <linux/of_spi.h>
 #include <linux/pm_runtime.h>
 #include <linux/export.h>
+#include <linux/sched.h>
+#include <linux/delay.h>
+#include <linux/kthread.h>
 
 static void spidev_release(struct device *dev)
 {
@@ -481,7 +484,7 @@ static void spi_match_master_to_boardinfo(struct spi_master *master,
  * The board info passed can safely be __initdata ... but be careful of
  * any embedded pointers (platform_data, etc), they're copied as-is.
  */
-int __init
+int __devinit
 spi_register_board_info(struct spi_board_info const *info, unsigned n)
 {
 	struct boardinfo *bi;
@@ -507,6 +510,294 @@ spi_register_board_info(struct spi_board_info const *info, unsigned n)
 
 /*-------------------------------------------------------------------------*/
 
+/**
+ * spi_pump_messages - kthread work function which processes spi message queue
+ * @work: pointer to kthread work struct contained in the master struct
+ *
+ * This function checks if there is any spi message in the queue that
+ * needs processing and if so call out to the driver to initialize hardware
+ * and transfer each message.
+ *
+ */
+static void spi_pump_messages(struct kthread_work *work)
+{
+	struct spi_master *master =
+		container_of(work, struct spi_master, pump_messages);
+	unsigned long flags;
+	bool was_busy = false;
+	int ret;
+
+	/* Lock queue and check for queue work */
+	spin_lock_irqsave(&master->queue_lock, flags);
+	if (list_empty(&master->queue) || !master->running) {
+		if (master->busy) {
+			ret = master->unprepare_transfer_hardware(master);
+			if (ret) {
+				spin_unlock_irqrestore(&master->queue_lock, flags);
+				dev_err(&master->dev,
+					"failed to unprepare transfer hardware\n");
+				return;
+			}
+		}
+		master->busy = false;
+		spin_unlock_irqrestore(&master->queue_lock, flags);
+		return;
+	}
+
+	/* Make sure we are not already running a message */
+	if (master->cur_msg) {
+		spin_unlock_irqrestore(&master->queue_lock, flags);
+		return;
+	}
+	/* Extract head of queue */
+	master->cur_msg =
+	    list_entry(master->queue.next, struct spi_message, queue);
+
+	list_del_init(&master->cur_msg->queue);
+	if (master->busy)
+		was_busy = true;
+	else
+		master->busy = true;
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+
+	if (!was_busy) {
+		ret = master->prepare_transfer_hardware(master);
+		if (ret) {
+			dev_err(&master->dev,
+				"failed to prepare transfer hardware\n");
+			return;
+		}
+	}
+
+	ret = master->transfer_one_message(master, master->cur_msg);
+	if (ret) {
+		dev_err(&master->dev,
+			"failed to transfer one message from queue\n");
+		return;
+	}
+}
+
+static int spi_init_queue(struct spi_master *master)
+{
+	struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 };
+
+	INIT_LIST_HEAD(&master->queue);
+	spin_lock_init(&master->queue_lock);
+
+	master->running = false;
+	master->busy = false;
+
+	init_kthread_worker(&master->kworker);
+	master->kworker_task = kthread_run(kthread_worker_fn,
+					   &master->kworker,
+					   dev_name(&master->dev));
+	if (IS_ERR(master->kworker_task)) {
+		dev_err(&master->dev, "failed to create message pump task\n");
+		return -ENOMEM;
+	}
+	init_kthread_work(&master->pump_messages, spi_pump_messages);
+
+	/*
+	 * Master config will indicate if this controller should run the
+	 * message pump with high (realtime) priority to reduce the transfer
+	 * latency on the bus by minimising the delay between a transfer
+	 * request and the scheduling of the message pump thread. Without this
+	 * setting the message pump thread will remain at default priority.
+	 */
+	if (master->rt) {
+		dev_info(&master->dev,
+			"will run message pump with realtime priority\n");
+		sched_setscheduler(master->kworker_task, SCHED_FIFO, &param);
+	}
+
+	return 0;
+}
+
+/**
+ * spi_get_next_queued_message() - called by driver to check for queued
+ * messages
+ * @master: the master to check for queued messages
+ *
+ * If there are more messages in the queue, the next message is returned from
+ * this call.
+ */
+struct spi_message *spi_get_next_queued_message(struct spi_master *master)
+{
+	struct spi_message *next;
+	unsigned long flags;
+
+	/* get a pointer to the next message, if any */
+	spin_lock_irqsave(&master->queue_lock, flags);
+	if (list_empty(&master->queue))
+		next = NULL;
+	else
+		next = list_entry(master->queue.next,
+				  struct spi_message, queue);
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+
+	return next;
+}
+EXPORT_SYMBOL_GPL(spi_get_next_queued_message);
+
+/**
+ * spi_finalize_current_message() - the current message is complete
+ * @master: the master to return the message to
+ *
+ * Called by the driver to notify the core that the message in the front of the
+ * queue is complete and can be removed from the queue.
+ */
+void spi_finalize_current_message(struct spi_master *master)
+{
+	struct spi_message *mesg;
+	unsigned long flags;
+
+	spin_lock_irqsave(&master->queue_lock, flags);
+	mesg = master->cur_msg;
+	master->cur_msg = NULL;
+
+	queue_kthread_work(&master->kworker, &master->pump_messages);
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+
+	mesg->state = NULL;
+	if (mesg->complete)
+		mesg->complete(mesg->context);
+}
+EXPORT_SYMBOL_GPL(spi_finalize_current_message);
+
+static int spi_start_queue(struct spi_master *master)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&master->queue_lock, flags);
+
+	if (master->running || master->busy) {
+		spin_unlock_irqrestore(&master->queue_lock, flags);
+		return -EBUSY;
+	}
+
+	master->running = true;
+	master->cur_msg = NULL;
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+
+	queue_kthread_work(&master->kworker, &master->pump_messages);
+
+	return 0;
+}
+
+static int spi_stop_queue(struct spi_master *master)
+{
+	unsigned long flags;
+	unsigned limit = 500;
+	int ret = 0;
+
+	spin_lock_irqsave(&master->queue_lock, flags);
+
+	/*
+	 * This is a bit lame, but is optimized for the common execution path.
+	 * A wait_queue on the master->busy could be used, but then the common
+	 * execution path (pump_messages) would be required to call wake_up or
+	 * friends on every SPI message. Do this instead.
+	 */
+	while ((!list_empty(&master->queue) || master->busy) && limit--) {
+		spin_unlock_irqrestore(&master->queue_lock, flags);
+		msleep(10);
+		spin_lock_irqsave(&master->queue_lock, flags);
+	}
+
+	if (!list_empty(&master->queue) || master->busy)
+		ret = -EBUSY;
+	else
+		master->running = false;
+
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+
+	if (ret) {
+		dev_warn(&master->dev,
+			 "could not stop message queue\n");
+		return ret;
+	}
+	return ret;
+}
+
+static int spi_destroy_queue(struct spi_master *master)
+{
+	int ret;
+
+	ret = spi_stop_queue(master);
+
+	/*
+	 * flush_kthread_worker will block until all work is done.
+	 * If the reason that stop_queue timed out is that the work will never
+	 * finish, then it does no good to call flush/stop thread, so
+	 * return anyway.
+	 */
+	if (ret) {
+		dev_err(&master->dev, "problem destroying queue\n");
+		return ret;
+	}
+
+	flush_kthread_worker(&master->kworker);
+	kthread_stop(master->kworker_task);
+
+	return 0;
+}
+
+/**
+ * spi_queued_transfer - transfer function for queued transfers
+ * @spi: spi device which is requesting transfer
+ * @msg: spi message which is to handled is queued to driver queue
+ */
+static int spi_queued_transfer(struct spi_device *spi, struct spi_message *msg)
+{
+	struct spi_master *master = spi->master;
+	unsigned long flags;
+
+	spin_lock_irqsave(&master->queue_lock, flags);
+
+	if (!master->running) {
+		spin_unlock_irqrestore(&master->queue_lock, flags);
+		return -ESHUTDOWN;
+	}
+	msg->actual_length = 0;
+	msg->status = -EINPROGRESS;
+
+	list_add_tail(&msg->queue, &master->queue);
+	if (master->running && !master->busy)
+		queue_kthread_work(&master->kworker, &master->pump_messages);
+
+	spin_unlock_irqrestore(&master->queue_lock, flags);
+	return 0;
+}
+
+static int spi_master_initialize_queue(struct spi_master *master)
+{
+	int ret;
+
+	master->queued = true;
+	master->transfer = spi_queued_transfer;
+
+	/* Initialize and start queue */
+	ret = spi_init_queue(master);
+	if (ret) {
+		dev_err(&master->dev, "problem initializing queue\n");
+		goto err_init_queue;
+	}
+	ret = spi_start_queue(master);
+	if (ret) {
+		dev_err(&master->dev, "problem starting queue\n");
+		goto err_start_queue;
+	}
+
+	return 0;
+
+err_start_queue:
+err_init_queue:
+	spi_destroy_queue(master);
+	return ret;
+}
+
+/*-------------------------------------------------------------------------*/
+
 static void spi_master_release(struct device *dev)
 {
 	struct spi_master *master;
@@ -522,6 +813,7 @@ static struct class spi_master_class = {
 };
 
 
+
 /**
  * spi_alloc_master - allocate SPI master controller
  * @dev: the controller, possibly using the platform_bus
@@ -539,7 +831,8 @@ static struct class spi_master_class = {
  *
  * The caller is responsible for assigning the bus number and initializing
  * the master's methods before calling spi_register_master(); and (after errors
- * adding the device) calling spi_master_put() to prevent a memory leak.
+ * adding the device) calling spi_master_put() and kfree() to prevent a memory
+ * leak.
  */
 struct spi_master *spi_alloc_master(struct device *dev, unsigned size)
 {
@@ -621,14 +914,23 @@ int spi_register_master(struct spi_master *master)
 	dev_dbg(dev, "registered master %s%s\n", dev_name(&master->dev),
 			dynamic ? " (dynamic)" : "");
 
+	/* If we're using a queued driver, start the queue */
+	if (master->transfer)
+		dev_info(dev, "master is unqueued, this is deprecated\n");
+	else {
+		status = spi_master_initialize_queue(master);
+		if (status) {
+			device_unregister(&master->dev);
+			goto done;
+		}
+	}
+
 	mutex_lock(&board_lock);
 	list_add_tail(&master->list, &spi_master_list);
 	list_for_each_entry(bi, &board_list, list)
 		spi_match_master_to_boardinfo(master, &bi->board_info);
 	mutex_unlock(&board_lock);
 
-	status = 0;
-
 	/* Register devices from the device tree */
 	of_register_spi_devices(master);
 done:
@@ -636,7 +938,6 @@ done:
 }
 EXPORT_SYMBOL_GPL(spi_register_master);
 
-
 static int __unregister(struct device *dev, void *null)
 {
 	spi_unregister_device(to_spi_device(dev));
@@ -657,6 +958,11 @@ void spi_unregister_master(struct spi_master *master)
 {
 	int dummy;
 
+	if (master->queued) {
+		if (spi_destroy_queue(master))
+			dev_err(&master->dev, "queue remove failed\n");
+	}
+
 	mutex_lock(&board_lock);
 	list_del(&master->list);
 	mutex_unlock(&board_lock);
@@ -666,6 +972,37 @@ void spi_unregister_master(struct spi_master *master)
 }
 EXPORT_SYMBOL_GPL(spi_unregister_master);
 
+int spi_master_suspend(struct spi_master *master)
+{
+	int ret;
+
+	/* Basically no-ops for non-queued masters */
+	if (!master->queued)
+		return 0;
+
+	ret = spi_stop_queue(master);
+	if (ret)
+		dev_err(&master->dev, "queue stop failed\n");
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(spi_master_suspend);
+
+int spi_master_resume(struct spi_master *master)
+{
+	int ret;
+
+	if (!master->queued)
+		return 0;
+
+	ret = spi_start_queue(master);
+	if (ret)
+		dev_err(&master->dev, "queue restart failed\n");
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(spi_master_resume);
+
 static int __spi_master_match(struct device *dev, void *data)
 {
 	struct spi_master *m;
diff --git a/include/linux/amba/pl022.h b/include/linux/amba/pl022.h
index 572f637299c9..3672f40f3455 100644
--- a/include/linux/amba/pl022.h
+++ b/include/linux/amba/pl022.h
@@ -241,6 +241,8 @@ struct dma_chan;
  * @autosuspend_delay: delay in ms following transfer completion before the
  *     runtime power management system suspends the device. A setting of 0
  *     indicates no delay and the device will be suspended immediately.
+ * @rt: indicates the controller should run the message pump with realtime
+ *     priority to minimise the transfer latency on the bus.
  */
 struct pl022_ssp_controller {
 	u16 bus_id;
@@ -250,6 +252,7 @@ struct pl022_ssp_controller {
 	void *dma_rx_param;
 	void *dma_tx_param;
 	int autosuspend_delay;
+	bool rt;
 };
 
 /**
diff --git a/include/linux/spi/sh_hspi.h b/include/linux/spi/sh_hspi.h
new file mode 100644
index 000000000000..a1121f872ac1
--- /dev/null
+++ b/include/linux/spi/sh_hspi.h
@@ -0,0 +1,23 @@
+/*
+ * Copyright (C) 2011 Kuninori Morimoto
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#ifndef SH_HSPI_H
+#define SH_HSPI_H
+
+struct sh_hspi_info {
+};
+
+#endif
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 176fce9cc6b1..98679b061b63 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -22,6 +22,7 @@
 #include <linux/device.h>
 #include <linux/mod_devicetable.h>
 #include <linux/slab.h>
+#include <linux/kthread.h>
 
 /*
  * INTERFACES between SPI master-side drivers and SPI infrastructure.
@@ -235,6 +236,27 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv)
  *	the device whose settings are being modified.
  * @transfer: adds a message to the controller's transfer queue.
  * @cleanup: frees controller-specific state
+ * @queued: whether this master is providing an internal message queue
+ * @kworker: thread struct for message pump
+ * @kworker_task: pointer to task for message pump kworker thread
+ * @pump_messages: work struct for scheduling work to the message pump
+ * @queue_lock: spinlock to syncronise access to message queue
+ * @queue: message queue
+ * @cur_msg: the currently in-flight message
+ * @busy: message pump is busy
+ * @running: message pump is running
+ * @rt: whether this queue is set to run as a realtime task
+ * @prepare_transfer_hardware: a message will soon arrive from the queue
+ *	so the subsystem requests the driver to prepare the transfer hardware
+ *	by issuing this call
+ * @transfer_one_message: the subsystem calls the driver to transfer a single
+ *	message while queuing transfers that arrive in the meantime. When the
+ *	driver is finished with this message, it must call
+ *	spi_finalize_current_message() so the subsystem can issue the next
+ *	transfer
+ * @prepare_transfer_hardware: there are currently no more messages on the
+ *	queue so the subsystem notifies the driver that it may relax the
+ *	hardware by issuing this call
  *
  * Each SPI master controller can communicate with one or more @spi_device
  * children.  These make a small bus, sharing MOSI, MISO and SCK signals
@@ -318,6 +340,28 @@ struct spi_master {
 
 	/* called on release() to free memory provided by spi_master */
 	void			(*cleanup)(struct spi_device *spi);
+
+	/*
+	 * These hooks are for drivers that want to use the generic
+	 * master transfer queueing mechanism. If these are used, the
+	 * transfer() function above must NOT be specified by the driver.
+	 * Over time we expect SPI drivers to be phased over to this API.
+	 */
+	bool				queued;
+	struct kthread_worker		kworker;
+	struct task_struct		*kworker_task;
+	struct kthread_work		pump_messages;
+	spinlock_t			queue_lock;
+	struct list_head		queue;
+	struct spi_message		*cur_msg;
+	bool				busy;
+	bool				running;
+	bool				rt;
+
+	int (*prepare_transfer_hardware)(struct spi_master *master);
+	int (*transfer_one_message)(struct spi_master *master,
+				    struct spi_message *mesg);
+	int (*unprepare_transfer_hardware)(struct spi_master *master);
 };
 
 static inline void *spi_master_get_devdata(struct spi_master *master)
@@ -343,6 +387,13 @@ static inline void spi_master_put(struct spi_master *master)
 		put_device(&master->dev);
 }
 
+/* PM calls that need to be issued by the driver */
+extern int spi_master_suspend(struct spi_master *master);
+extern int spi_master_resume(struct spi_master *master);
+
+/* Calls the driver make to interact with the message queue */
+extern struct spi_message *spi_get_next_queued_message(struct spi_master *master);
+extern void spi_finalize_current_message(struct spi_master *master);
 
 /* the spi driver core manages memory for the spi_master classdev */
 extern struct spi_master *
@@ -549,7 +600,7 @@ static inline struct spi_message *spi_message_alloc(unsigned ntrans, gfp_t flags
 			+ ntrans * sizeof(struct spi_transfer),
 			flags);
 	if (m) {
-		int i;
+		unsigned i;
 		struct spi_transfer *t = (struct spi_transfer *)(m + 1);
 
 		INIT_LIST_HEAD(&m->transfers);