Merge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/spi-2.6

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/spi-2.6:
  [PATCH] SPI: spi_bitbang: clocking fixes
  [PATCH] spi: Update to PXA2xx SPI Driver
  [PATCH] SPI: busnum == 0 needs to work
  [PATCH] SPI: devices can require LSB-first encodings
  [PATCH] SPI: Renamed bitbang_transfer_setup to spi_bitbang_setup_transfer and export it
  [PATCH] SPI: Add David as the SPI subsystem maintainer
  [PATCH] SPI: spi bounce buffer has a minimum length
  [PATCH] SPI: spi whitespace fixes
  [PATCH] SPI: add PXA2xx SSP SPI Driver
  [PATCH] SPI: per-transfer overrides for wordsize and clocking

11 files changed, 1941 insertions, 41 deletions

diff --git a/Documentation/spi/pxa2xx b/Documentation/spi/pxa2xx
new file mode 100644
index 000000000000..9c45f3df2e18
--- /dev/null
+++ b/Documentation/spi/pxa2xx
@@ -0,0 +1,234 @@
+PXA2xx SPI on SSP driver HOWTO
+===================================================
+This a mini howto on the pxa2xx_spi driver.  The driver turns a PXA2xx
+synchronous serial port into a SPI master controller
+(see Documentation/spi/spi_summary). The driver has the following features
+
+- Support for any PXA2xx SSP
+- SSP PIO and SSP DMA data transfers.
+- External and Internal (SSPFRM) chip selects.
+- Per slave device (chip) configuration.
+- Full suspend, freeze, resume support.
+
+The driver is built around a "spi_message" fifo serviced by workqueue and a
+tasklet. The workqueue, "pump_messages", drives message fifo and the tasklet
+(pump_transfer) is responsible for queuing SPI transactions and setting up and
+launching the dma/interrupt driven transfers.
+
+Declaring PXA2xx Master Controllers
+-----------------------------------
+Typically a SPI master is defined in the arch/.../mach-*/board-*.c as a
+"platform device".  The master configuration is passed to the driver via a table
+found in include/asm-arm/arch-pxa/pxa2xx_spi.h:
+
+struct pxa2xx_spi_master {
+	enum pxa_ssp_type ssp_type;
+	u32 clock_enable;
+	u16 num_chipselect;
+	u8 enable_dma;
+};
+
+The "pxa2xx_spi_master.ssp_type" field must have a value between 1 and 3 and
+informs the driver which features a particular SSP supports.
+
+The "pxa2xx_spi_master.clock_enable" field is used to enable/disable the
+corresponding SSP peripheral block in the "Clock Enable Register (CKEN"). See
+the "PXA2xx Developer Manual" section "Clocks and Power Management".
+
+The "pxa2xx_spi_master.num_chipselect" field is used to determine the number of
+slave device (chips) attached to this SPI master.
+
+The "pxa2xx_spi_master.enable_dma" field informs the driver that SSP DMA should
+be used.  This caused the driver to acquire two DMA channels: rx_channel and
+tx_channel.  The rx_channel has a higher DMA service priority the tx_channel.
+See the "PXA2xx Developer Manual" section "DMA Controller".
+
+NSSP MASTER SAMPLE
+------------------
+Below is a sample configuration using the PXA255 NSSP.
+
+static struct resource pxa_spi_nssp_resources[] = {
+	[0] = {
+		.start	= __PREG(SSCR0_P(2)), /* Start address of NSSP */
+		.end	= __PREG(SSCR0_P(2)) + 0x2c, /* Range of registers */
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_NSSP, /* NSSP IRQ */
+		.end	= IRQ_NSSP,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct pxa2xx_spi_master pxa_nssp_master_info = {
+	.ssp_type = PXA25x_NSSP, /* Type of SSP */
+	.clock_enable = CKEN9_NSSP, /* NSSP Peripheral clock */
+	.num_chipselect = 1, /* Matches the number of chips attached to NSSP */
+	.enable_dma = 1, /* Enables NSSP DMA */
+};
+
+static struct platform_device pxa_spi_nssp = {
+	.name = "pxa2xx-spi", /* MUST BE THIS VALUE, so device match driver */
+	.id = 2, /* Bus number, MUST MATCH SSP number 1..n */
+	.resource = pxa_spi_nssp_resources,
+	.num_resources = ARRAY_SIZE(pxa_spi_nssp_resources),
+	.dev = {
+		.platform_data = &pxa_nssp_master_info, /* Passed to driver */
+	},
+};
+
+static struct platform_device *devices[] __initdata = {
+	&pxa_spi_nssp,
+};
+
+static void __init board_init(void)
+{
+	(void)platform_add_device(devices, ARRAY_SIZE(devices));
+}
+
+Declaring Slave Devices
+-----------------------
+Typically each SPI slave (chip) is defined in the arch/.../mach-*/board-*.c
+using the "spi_board_info" structure found in "linux/spi/spi.h". See
+"Documentation/spi/spi_summary" for additional information.
+
+Each slave device attached to the PXA must provide slave specific configuration
+information via the structure "pxa2xx_spi_chip" found in
+"include/asm-arm/arch-pxa/pxa2xx_spi.h".  The pxa2xx_spi master controller driver
+will uses the configuration whenever the driver communicates with the slave
+device.
+
+struct pxa2xx_spi_chip {
+	u8 tx_threshold;
+	u8 rx_threshold;
+	u8 dma_burst_size;
+	u32 timeout_microsecs;
+	u8 enable_loopback;
+	void (*cs_control)(u32 command);
+};
+
+The "pxa2xx_spi_chip.tx_threshold" and "pxa2xx_spi_chip.rx_threshold" fields are
+used to configure the SSP hardware fifo.  These fields are critical to the
+performance of pxa2xx_spi driver and misconfiguration will result in rx
+fifo overruns (especially in PIO mode transfers). Good default values are
+
+	.tx_threshold = 12,
+	.rx_threshold = 4,
+
+The "pxa2xx_spi_chip.dma_burst_size" field is used to configure PXA2xx DMA
+engine and is related the "spi_device.bits_per_word" field.  Read and understand
+the PXA2xx "Developer Manual" sections on the DMA controller and SSP Controllers
+to determine the correct value. An SSP configured for byte-wide transfers would
+use a value of 8.
+
+The "pxa2xx_spi_chip.timeout_microsecs" fields is used to efficiently handle
+trailing bytes in the SSP receiver fifo.  The correct value for this field is
+dependent on the SPI bus speed ("spi_board_info.max_speed_hz") and the specific
+slave device.  Please note the the PXA2xx SSP 1 does not support trailing byte
+timeouts and must busy-wait any trailing bytes.
+
+The "pxa2xx_spi_chip.enable_loopback" field is used to place the SSP porting
+into internal loopback mode.  In this mode the SSP controller internally
+connects the SSPTX pin the the SSPRX pin.  This is useful for initial setup
+testing.
+
+The "pxa2xx_spi_chip.cs_control" field is used to point to a board specific
+function for asserting/deasserting a slave device chip select.  If the field is
+NULL, the pxa2xx_spi master controller driver assumes that the SSP port is
+configured to use SSPFRM instead.
+
+NSSP SALVE SAMPLE
+-----------------
+The pxa2xx_spi_chip structure is passed to the pxa2xx_spi driver in the
+"spi_board_info.controller_data" field. Below is a sample configuration using
+the PXA255 NSSP.
+
+/* Chip Select control for the CS8415A SPI slave device */
+static void cs8415a_cs_control(u32 command)
+{
+	if (command & PXA2XX_CS_ASSERT)
+		GPCR(2) = GPIO_bit(2);
+	else
+		GPSR(2) = GPIO_bit(2);
+}
+
+/* Chip Select control for the CS8405A SPI slave device */
+static void cs8405a_cs_control(u32 command)
+{
+	if (command & PXA2XX_CS_ASSERT)
+		GPCR(3) = GPIO_bit(3);
+	else
+		GPSR(3) = GPIO_bit(3);
+}
+
+static struct pxa2xx_spi_chip cs8415a_chip_info = {
+	.tx_threshold = 12, /* SSP hardward FIFO threshold */
+	.rx_threshold = 4, /* SSP hardward FIFO threshold */
+	.dma_burst_size = 8, /* Byte wide transfers used so 8 byte bursts */
+	.timeout_microsecs = 64, /* Wait at least 64usec to handle trailing */
+	.cs_control = cs8415a_cs_control, /* Use external chip select */
+};
+
+static struct pxa2xx_spi_chip cs8405a_chip_info = {
+	.tx_threshold = 12, /* SSP hardward FIFO threshold */
+	.rx_threshold = 4, /* SSP hardward FIFO threshold */
+	.dma_burst_size = 8, /* Byte wide transfers used so 8 byte bursts */
+	.timeout_microsecs = 64, /* Wait at least 64usec to handle trailing */
+	.cs_control = cs8405a_cs_control, /* Use external chip select */
+};
+
+static struct spi_board_info streetracer_spi_board_info[] __initdata = {
+	{
+		.modalias = "cs8415a", /* Name of spi_driver for this device */
+		.max_speed_hz = 3686400, /* Run SSP as fast a possbile */
+		.bus_num = 2, /* Framework bus number */
+		.chip_select = 0, /* Framework chip select */
+		.platform_data = NULL; /* No spi_driver specific config */
+		.controller_data = &cs8415a_chip_info, /* Master chip config */
+		.irq = STREETRACER_APCI_IRQ, /* Slave device interrupt */
+	},
+	{
+		.modalias = "cs8405a", /* Name of spi_driver for this device */
+		.max_speed_hz = 3686400, /* Run SSP as fast a possbile */
+		.bus_num = 2, /* Framework bus number */
+		.chip_select = 1, /* Framework chip select */
+		.controller_data = &cs8405a_chip_info, /* Master chip config */
+		.irq = STREETRACER_APCI_IRQ, /* Slave device interrupt */
+	},
+};
+
+static void __init streetracer_init(void)
+{
+	spi_register_board_info(streetracer_spi_board_info,
+				ARRAY_SIZE(streetracer_spi_board_info));
+}
+
+
+DMA and PIO I/O Support
+-----------------------
+The pxa2xx_spi driver support both DMA and interrupt driven PIO message
+transfers.  The driver defaults to PIO mode and DMA transfers must enabled by
+setting the "enable_dma" flag in the "pxa2xx_spi_master" structure and and
+ensuring that the "pxa2xx_spi_chip.dma_burst_size" field is non-zero.  The DMA
+mode support both coherent and stream based DMA mappings.
+
+The following logic is used to determine the type of I/O to be used on
+a per "spi_transfer" basis:
+
+if !enable_dma or dma_burst_size == 0 then
+	always use PIO transfers
+
+if spi_message.is_dma_mapped and rx_dma_buf != 0 and tx_dma_buf != 0 then
+	use coherent DMA mode
+
+if rx_buf and tx_buf are aligned on 8 byte boundary then
+	use streaming DMA mode
+
+otherwise
+	use PIO transfer
+
+THANKS TO
+---------
+
+David Brownell and others for mentoring the development of this driver.
+
diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary
index a5ffba33a351..068732d32276 100644
--- a/Documentation/spi/spi-summary
+++ b/Documentation/spi/spi-summary
@@ -414,7 +414,33 @@ to get the driver-private data allocated for that device.
 The driver will initialize the fields of that spi_master, including the
 bus number (maybe the same as the platform device ID) and three methods
 used to interact with the SPI core and SPI protocol drivers.  It will
-also initialize its own internal state.
+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.
+
+If you need to remove your SPI controller driver, spi_unregister_master()
+will reverse the effect of spi_register_master().
+
+
+BUS NUMBERING
+
+Bus numbering is important, since that's how Linux identifies a given
+SPI bus (shared SCK, MOSI, MISO).  Valid bus numbers start at zero.  On
+SOC systems, the bus numbers should match the numbers defined by the chip
+manufacturer.  For example, hardware controller SPI2 would be bus number 2,
+and spi_board_info for devices connected to it would use that number.
+
+If you don't have such hardware-assigned bus number, and for some reason
+you can't just assign them, then provide a negative bus number.  That will
+then be replaced by a dynamically assigned number. You'd then need to treat
+this as a non-static configuration (see above).
+
+
+SPI MASTER METHODS
 
     master->setup(struct spi_device *spi)
 	This sets up the device clock rate, SPI mode, and word sizes.
@@ -431,6 +457,9 @@ also initialize its own internal state.
 	state it dynamically associates with that device.  If you do that,
 	be sure to provide the cleanup() method to free that state.
 
+
+SPI MESSAGE QUEUE
+
 The bulk of the driver will be managing the I/O queue fed by transfer().
 
 That queue could be purely conceptual.  For example, a driver used only
@@ -440,6 +469,9 @@ But the queue will probably be very real, using message->queue, PIO,
 often DMA (especially if the root filesystem is in SPI flash), and
 execution contexts like IRQ handlers, tasklets, or workqueues (such
 as keventd).  Your driver can be as fancy, or as simple, as you need.
+Such a transfer() method would normally just add the message to a
+queue, and then start some asynchronous transfer engine (unless it's
+already running).
 
 
 THANKS TO
diff --git a/MAINTAINERS b/MAINTAINERS
index 246153dee7f1..753584cf4e7e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2518,6 +2518,12 @@ M:	perex@suse.cz
 L:	alsa-devel@alsa-project.org
 S:	Maintained
 
+SPI SUBSYSTEM
+P:	David Brownell
+M:	dbrownell@users.sourceforge.net
+L:	spi-devel-general@lists.sourceforge.net
+S:	Maintained
+
 TPM DEVICE DRIVER
 P:	Kylene Hall
 M:	kjhall@us.ibm.com
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 7a75faeb0526..9ce1d01469b1 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -75,6 +75,14 @@ config SPI_BUTTERFLY
 	  inexpensive battery powered microcontroller evaluation board.
 	  This same cable can be used to flash new firmware.
 
+config SPI_PXA2XX
+	tristate "PXA2xx SSP SPI master"
+	depends on SPI_MASTER && ARCH_PXA && EXPERIMENTAL
+	help
+	  This enables using a PXA2xx SSP port as a SPI master controller.
+	  The driver can be configured to use any SSP port and additional
+	  documentation can be found a Documentation/spi/pxa2xx.
+
 #
 # Add new SPI master controllers in alphabetical order above this line
 #
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index c2c87e845abf..1bca5f95de25 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_SPI_MASTER)		+= spi.o
 # SPI master controller drivers (bus)
 obj-$(CONFIG_SPI_BITBANG)		+= spi_bitbang.o
 obj-$(CONFIG_SPI_BUTTERFLY)		+= spi_butterfly.o
+obj-$(CONFIG_SPI_PXA2XX)		+= pxa2xx_spi.o
 # 	... add above this line ...
 
 # SPI protocol drivers (device/link on bus)
diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
new file mode 100644
index 000000000000..596bf820b70c
--- /dev/null
+++ b/drivers/spi/pxa2xx_spi.c
@@ -0,0 +1,1467 @@
+/*
+ * 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; 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/ioport.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/spi/spi.h>
+#include <linux/workqueue.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/hardware.h>
+#include <asm/delay.h>
+#include <asm/dma.h>
+
+#include <asm/arch/hardware.h>
+#include <asm/arch/pxa-regs.h>
+#include <asm/arch/pxa2xx_spi.h>
+
+MODULE_AUTHOR("Stephen Street");
+MODULE_DESCRIPTION("PXA2xx SSP SPI Contoller");
+MODULE_LICENSE("GPL");
+
+#define MAX_BUSES 3
+
+#define DMA_INT_MASK (DCSR_ENDINTR | DCSR_STARTINTR | DCSR_BUSERR)
+#define RESET_DMA_CHANNEL (DCSR_NODESC | DMA_INT_MASK)
+#define IS_DMA_ALIGNED(x) (((u32)(x)&0x07)==0)
+
+#define DEFINE_SSP_REG(reg, off) \
+static inline u32 read_##reg(void *p) { return __raw_readl(p + (off)); } \
+static inline void write_##reg(u32 v, void *p) { __raw_writel(v, p + (off)); }
+
+DEFINE_SSP_REG(SSCR0, 0x00)
+DEFINE_SSP_REG(SSCR1, 0x04)
+DEFINE_SSP_REG(SSSR, 0x08)
+DEFINE_SSP_REG(SSITR, 0x0c)
+DEFINE_SSP_REG(SSDR, 0x10)
+DEFINE_SSP_REG(SSTO, 0x28)
+DEFINE_SSP_REG(SSPSP, 0x2c)
+
+#define START_STATE ((void*)0)
+#define RUNNING_STATE ((void*)1)
+#define DONE_STATE ((void*)2)
+#define ERROR_STATE ((void*)-1)
+
+#define QUEUE_RUNNING 0
+#define QUEUE_STOPPED 1
+
+struct driver_data {
+	/* Driver model hookup */
+	struct platform_device *pdev;
+
+	/* SPI framework hookup */
+	enum pxa_ssp_type ssp_type;
+	struct spi_master *master;
+
+	/* PXA hookup */
+	struct pxa2xx_spi_master *master_info;
+
+	/* DMA setup stuff */
+	int rx_channel;
+	int tx_channel;
+	u32 *null_dma_buf;
+
+	/* SSP register addresses */
+	void *ioaddr;
+	u32 ssdr_physical;
+
+	/* SSP masks*/
+	u32 dma_cr1;
+	u32 int_cr1;
+	u32 clear_sr;
+	u32 mask_sr;
+
+	/* Driver message queue */
+	struct workqueue_struct	*workqueue;
+	struct work_struct pump_messages;
+	spinlock_t lock;
+	struct list_head queue;
+	int busy;
+	int run;
+
+	/* Message Transfer pump */
+	struct tasklet_struct pump_transfers;
+
+	/* Current message transfer state info */
+	struct spi_message* cur_msg;
+	struct spi_transfer* cur_transfer;
+	struct chip_data *cur_chip;
+	size_t len;
+	void *tx;
+	void *tx_end;
+	void *rx;
+	void *rx_end;
+	int dma_mapped;
+	dma_addr_t rx_dma;
+	dma_addr_t tx_dma;
+	size_t rx_map_len;
+	size_t tx_map_len;
+	u8 n_bytes;
+	u32 dma_width;
+	int cs_change;
+	void (*write)(struct driver_data *drv_data);
+	void (*read)(struct driver_data *drv_data);
+	irqreturn_t (*transfer_handler)(struct driver_data *drv_data);
+	void (*cs_control)(u32 command);
+};
+
+struct chip_data {
+	u32 cr0;
+	u32 cr1;
+	u32 to;
+	u32 psp;
+	u32 timeout;
+	u8 n_bytes;
+	u32 dma_width;
+	u32 dma_burst_size;
+	u32 threshold;
+	u32 dma_threshold;
+	u8 enable_dma;
+	u8 bits_per_word;
+	u32 speed_hz;
+	void (*write)(struct driver_data *drv_data);
+	void (*read)(struct driver_data *drv_data);
+	void (*cs_control)(u32 command);
+};
+
+static void pump_messages(void *data);
+
+static int flush(struct driver_data *drv_data)
+{
+	unsigned long limit = loops_per_jiffy << 1;
+
+	void *reg = drv_data->ioaddr;
+
+	do {
+		while (read_SSSR(reg) & SSSR_RNE) {
+			read_SSDR(reg);
+		}
+	} while ((read_SSSR(reg) & SSSR_BSY) && limit--);
+	write_SSSR(SSSR_ROR, reg);
+
+	return limit;
+}
+
+static void restore_state(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	/* Clear status and disable clock */
+	write_SSSR(drv_data->clear_sr, reg);
+	write_SSCR0(drv_data->cur_chip->cr0 & ~SSCR0_SSE, reg);
+
+	/* Load the registers */
+	write_SSCR1(drv_data->cur_chip->cr1, reg);
+	write_SSCR0(drv_data->cur_chip->cr0, reg);
+	if (drv_data->ssp_type != PXA25x_SSP) {
+		write_SSTO(0, reg);
+		write_SSPSP(drv_data->cur_chip->psp, reg);
+	}
+}
+
+static void null_cs_control(u32 command)
+{
+}
+
+static void null_writer(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+	u8 n_bytes = drv_data->n_bytes;
+
+	while ((read_SSSR(reg) & SSSR_TNF)
+			&& (drv_data->tx < drv_data->tx_end)) {
+		write_SSDR(0, reg);
+		drv_data->tx += n_bytes;
+	}
+}
+
+static void null_reader(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+	u8 n_bytes = drv_data->n_bytes;
+
+	while ((read_SSSR(reg) & SSSR_RNE)
+			&& (drv_data->rx < drv_data->rx_end)) {
+		read_SSDR(reg);
+		drv_data->rx += n_bytes;
+	}
+}
+
+static void u8_writer(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_TNF)
+			&& (drv_data->tx < drv_data->tx_end)) {
+		write_SSDR(*(u8 *)(drv_data->tx), reg);
+		++drv_data->tx;
+	}
+}
+
+static void u8_reader(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_RNE)
+			&& (drv_data->rx < drv_data->rx_end)) {
+		*(u8 *)(drv_data->rx) = read_SSDR(reg);
+		++drv_data->rx;
+	}
+}
+
+static void u16_writer(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_TNF)
+			&& (drv_data->tx < drv_data->tx_end)) {
+		write_SSDR(*(u16 *)(drv_data->tx), reg);
+		drv_data->tx += 2;
+	}
+}
+
+static void u16_reader(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_RNE)
+			&& (drv_data->rx < drv_data->rx_end)) {
+		*(u16 *)(drv_data->rx) = read_SSDR(reg);
+		drv_data->rx += 2;
+	}
+}
+static void u32_writer(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_TNF)
+			&& (drv_data->tx < drv_data->tx_end)) {
+		write_SSDR(*(u32 *)(drv_data->tx), reg);
+		drv_data->tx += 4;
+	}
+}
+
+static void u32_reader(struct driver_data *drv_data)
+{
+	void *reg = drv_data->ioaddr;
+
+	while ((read_SSSR(reg) & SSSR_RNE)
+			&& (drv_data->rx < drv_data->rx_end)) {
+		*(u32 *)(drv_data->rx) = read_SSDR(reg);
+		drv_data->rx += 4;
+	}
+}
+
+static void *next_transfer(struct driver_data *drv_data)
+{
+	struct spi_message *msg = drv_data->cur_msg;
+	struct spi_transfer *trans = drv_data->cur_transfer;
+
+	/* Move to next transfer */
+	if (trans->transfer_list.next != &msg->transfers) {
+		drv_data->cur_transfer =
+			list_entry(trans->transfer_list.next,
+					struct spi_transfer,
+					transfer_list);
+		return RUNNING_STATE;
+	} else
+		return DONE_STATE;
+}
+
+static int map_dma_buffers(struct driver_data *drv_data)
+{
+	struct spi_message *msg = drv_data->cur_msg;
+	struct device *dev = &msg->spi->dev;
+
+	if (!drv_data->cur_chip->enable_dma)
+		return 0;
+
+	if (msg->is_dma_mapped)
+		return  drv_data->rx_dma && drv_data->tx_dma;
+
+	if (!IS_DMA_ALIGNED(drv_data->rx) || !IS_DMA_ALIGNED(drv_data->tx))
+		return 0;
+
+	/* Modify setup if rx buffer is null */
+	if (drv_data->rx == NULL) {
+		*drv_data->null_dma_buf = 0;
+		drv_data->rx = drv_data->null_dma_buf;
+		drv_data->rx_map_len = 4;
+	} else
+		drv_data->rx_map_len = drv_data->len;
+
+
+	/* Modify setup if tx buffer is null */
+	if (drv_data->tx == NULL) {
+		*drv_data->null_dma_buf = 0;
+		drv_data->tx = drv_data->null_dma_buf;
+		drv_data->tx_map_len = 4;
+	} else
+		drv_data->tx_map_len = drv_data->len;
+
+	/* Stream map the rx buffer */
+	drv_data->rx_dma = dma_map_single(dev, drv_data->rx,
+						drv_data->rx_map_len,
+						DMA_FROM_DEVICE);
+	if (dma_mapping_error(drv_data->rx_dma))
+		return 0;
+
+	/* Stream map the tx buffer */
+	drv_data->tx_dma = dma_map_single(dev, drv_data->tx,
+						drv_data->tx_map_len,
+						DMA_TO_DEVICE);
+
+	if (dma_mapping_error(drv_data->tx_dma)) {
+		dma_unmap_single(dev, drv_data->rx_dma,
+					drv_data->rx_map_len, DMA_FROM_DEVICE);
+		return 0;
+	}
+
+	return 1;
+}
+
+static void unmap_dma_buffers(struct driver_data *drv_data)
+{
+	struct device *dev;
+
+	if (!drv_data->dma_mapped)
+		return;
+
+	if (!drv_data->cur_msg->is_dma_mapped) {
+		dev = &drv_data->cur_msg->spi->dev;
+		dma_unmap_single(dev, drv_data->rx_dma,
+					drv_data->rx_map_len, DMA_FROM_DEVICE);
+		dma_unmap_single(dev, drv_data->tx_dma,
+					drv_data->tx_map_len, DMA_TO_DEVICE);
+	}
+
+	drv_data->dma_mapped = 0;
+}
+
+/* caller already set message->status; dma and pio irqs are blocked */
+static void giveback(struct spi_message *message, struct driver_data *drv_data)
+{
+	struct spi_transfer* last_transfer;
+
+	last_transfer = list_entry(message->transfers.prev,
+					struct spi_transfer,
+					transfer_list);
+
+	if (!last_transfer->cs_change)
+		drv_data->cs_control(PXA2XX_CS_DEASSERT);
+
+	message->state = NULL;
+	if (message->complete)
+		message->complete(message->context);
+
+	drv_data->cur_msg = NULL;
+	drv_data->cur_transfer = NULL;
+	drv_data->cur_chip = NULL;
+	queue_work(drv_data->workqueue, &drv_data->pump_messages);
+}
+
+static int wait_ssp_rx_stall(void *ioaddr)
+{
+	unsigned long limit = loops_per_jiffy << 1;
+
+	while ((read_SSSR(ioaddr) & SSSR_BSY) && limit--)
+		cpu_relax();
+
+	return limit;
+}
+
+static int wait_dma_channel_stop(int channel)
+{
+	unsigned long limit = loops_per_jiffy << 1;
+
+	while (!(DCSR(channel) & DCSR_STOPSTATE) && limit--)
+		cpu_relax();
+
+	return limit;
+}
+
+static void dma_handler(int channel, void *data, struct pt_regs *regs)
+{
+	struct driver_data *drv_data = data;
+	struct spi_message *msg = drv_data->cur_msg;
+	void *reg = drv_data->ioaddr;
+	u32 irq_status = DCSR(channel) & DMA_INT_MASK;
+	u32 trailing_sssr = 0;
+
+	if (irq_status & DCSR_BUSERR) {
+
+		/* Disable interrupts, clear status and reset DMA */
+		if (drv_data->ssp_type != PXA25x_SSP)
+			write_SSTO(0, reg);
+		write_SSSR(drv_data->clear_sr, reg);
+		write_SSCR1(read_SSCR1(reg) & ~drv_data->dma_cr1, reg);
+		DCSR(drv_data->rx_channel) = RESET_DMA_CHANNEL;
+		DCSR(drv_data->tx_channel) = RESET_DMA_CHANNEL;
+
+		if (flush(drv_data) == 0)
+			dev_err(&drv_data->pdev->dev,
+					"dma_handler: flush fail\n");
+
+		unmap_dma_buffers(drv_data);
+
+		if (channel == drv_data->tx_channel)
+			dev_err(&drv_data->pdev->dev,
+				"dma_handler: bad bus address on "
+				"tx channel %d, source %x target = %x\n",
+				channel, DSADR(channel), DTADR(channel));
+		else
+			dev_err(&drv_data->pdev->dev,
+				"dma_handler: bad bus address on "
+				"rx channel %d, source %x target = %x\n",
+				channel, DSADR(channel), DTADR(channel));
+
+		msg->state = ERROR_STATE;
+		tasklet_schedule(&drv_data->pump_transfers);
+	}
+
+	/* PXA255x_SSP has no timeout interrupt, wait for tailing bytes */
+	if ((drv_data->ssp_type == PXA25x_SSP)
+		&& (channel == drv_data->tx_channel)
+		&& (irq_status & DCSR_ENDINTR)) {
+
+		/* Wait for rx to stall */
+		if (wait_ssp_rx_stall(drv_data->ioaddr) == 0)
+			dev_err(&drv_data->pdev->dev,
+				"dma_handler: ssp rx stall failed\n");
+
+		/* Clear and disable interrupts on SSP and DMA channels*/
+		write_SSSR(drv_data->clear_sr, reg);
+		write_SSCR1(read_SSCR1(reg) & ~drv_data->dma_cr1, reg);
+		DCSR(drv_data->tx_channel) = RESET_DMA_CHANNEL;
+		DCSR(drv_data->rx_channel) = RESET_DMA_CHANNEL;
+		if (wait_dma_channel_stop(drv_data->rx_channel) == 0)
+			dev_err(&drv_data->pdev->dev,
+				"dma_handler: dma rx channel stop failed\n");
+
+		unmap_dma_buffers(drv_data);
+
+		/* Read trailing bytes */
+		/* Calculate number of trailing bytes, read them */
+		trailing_sssr = read_SSSR(reg);
+		if ((trailing_sssr & 0xf008) != 0xf000) {
+			drv_data->rx = drv_data->rx_end -
+					(((trailing_sssr >> 12) & 0x0f) + 1);
+			drv_data->read(drv_data);
+		}
+		msg->actual_length += drv_data->len;
+
+		/* Release chip select if requested, transfer delays are
+		 * handled in pump_transfers */
+		if (drv_data->cs_change)
+			drv_data->cs_control(PXA2XX_CS_DEASSERT);
+
+		/* Move to next transfer */
+		msg->state = next_transfer(drv_data);
+
+		/* Schedule transfer tasklet */
+		tasklet_schedule(&drv_data->pump_transfers);
+	}
+}
+
+static irqreturn_t dma_transfer(struct driver_data *drv_data)
+{
+	u32 irq_status;
+	u32 trailing_sssr = 0;
+	struct spi_message *msg = drv_data->cur_msg;
+	void *reg = drv_data->ioaddr;
+
+	irq_status = read_SSSR(reg) & drv_data->mask_sr;
+	if (irq_status & SSSR_ROR) {
+		/* Clear and disable interrupts on SSP and DMA channels*/
+		if (drv_data->ssp_type != PXA25x_SSP)
+			write_SSTO(0, reg);
+		write_SSSR(drv_data->clear_sr, reg);
+		write_SSCR1(read_SSCR1(reg) & ~drv_data->dma_cr1, reg);
+		DCSR(drv_data->tx_channel) = RESET_DMA_CHANNEL;
+		DCSR(drv_data->rx_channel) = RESET_DMA_CHANNEL;
+		unmap_dma_buffers(drv_data);
+
+		if (flush(drv_data) == 0)
+			dev_err(&drv_data->pdev->dev,
+					"dma_transfer: flush fail\n");
+
+		dev_warn(&drv_data->pdev->dev, "dma_transfer: fifo overun\n");
+
+		drv_data->cur_msg->state = ERROR_STATE;
+		tasklet_schedule(&drv_data->pump_transfers);
+
+		return IRQ_HANDLED;
+	}
+
+	/* Check for false positive timeout */
+	if ((irq_status & SSSR_TINT) && DCSR(drv_data->tx_channel) & DCSR_RUN) {
+		write_SSSR(SSSR_TINT, reg);
+		return IRQ_HANDLED;
+	}
+
+	if (irq_status & SSSR_TINT || drv_data->rx == drv_data->rx_end) {
+
+		/* Clear and disable interrupts on SSP and DMA channels*/
+		if (drv_data->ssp_type != PXA25x_SSP)
+			write_SSTO(0, reg);
+		write_SSSR(drv_data->clear_sr, reg);
+		write_SSCR1(read_SSCR1(reg) & ~drv_data->dma_cr1, reg);
+		DCSR(drv_data->tx_channel) = RESET_DMA_CHANNEL;
+		DCSR(drv_data->rx_channel) = RESET_DMA_CHANNEL;
+
+		if (wait_dma_channel_stop(drv_data->rx_channel) == 0)
+			dev_err(&drv_data->pdev->dev,
+				"dma_transfer: dma rx channel stop failed\n");
+
+		if (wait_ssp_rx_stall(drv_data->ioaddr) == 0)
+			dev_err(&drv_data->pdev->dev,
+				"dma_transfer: ssp rx stall failed\n");
+
+		unmap_dma_buffers(drv_data);
+
+		/* Calculate number of trailing bytes, read them */
+		trailing_sssr = read_SSSR(reg);
+		if ((trailing_sssr & 0xf008) != 0xf000) {
+			drv_data->rx = drv_data->rx_end -
+					(((trailing_sssr >> 12) & 0x0f) + 1);
+			drv_data->read(drv_data);
+		}
+		msg->actual_length += drv_data->len;
+
+		/* Release chip select if requested, transfer delays are
+		 * handled in pump_transfers */
+		if (drv_data->cs_change)
+			drv_data->cs_control(PXA2XX_CS_DEASSERT);
+
+		/* Move to next transfer */
+		msg->state = next_transfer(drv_data);
+
+		/* Schedule transfer tasklet */
+		tasklet_schedule(&drv_data->pump_transfers);
+
+		return IRQ_HANDLED;
+	}
+
+	/* Opps problem detected */
+	return IRQ_NONE;
+}
+
+static irqreturn_t interrupt_transfer(struct driver_data *drv_data)
+{
+	u32 irq_status;
+	struct spi_message *msg = drv_data->cur_msg;
+	void *reg = drv_data->ioaddr;
+	irqreturn_t handled = IRQ_NONE;
+	unsigned long limit = loops_per_jiffy << 1;
+
+	while ((irq_status = (read_SSSR(reg) & drv_data->mask_sr))) {
+
+		if (irq_status & SSSR_ROR) {
+
+			/* Clear and disable interrupts */
+			if (drv_data->ssp_type != PXA25x_SSP)
+				write_SSTO(0, reg);
+			write_SSSR(drv_data->clear_sr, reg);
+			write_SSCR1(read_SSCR1(reg) & ~drv_data->int_cr1, reg);
+
+			if (flush(drv_data) == 0)
+				dev_err(&drv_data->pdev->dev,
+					"interrupt_transfer: flush fail\n");
+
+			dev_warn(&drv_data->pdev->dev,
+					"interrupt_transfer: fifo overun\n");
+
+			msg->state = ERROR_STATE;
+			tasklet_schedule(&drv_data->pump_transfers);
+
+			return IRQ_HANDLED;
+		}
+
+		/* Look for false positive timeout */
+		if ((irq_status & SSSR_TINT)
+				&& (drv_data->rx < drv_data->rx_end))
+			write_SSSR(SSSR_TINT, reg);
+
+		/* Pump data */
+		drv_data->read(drv_data);
+		drv_data->write(drv_data);
+
+		if (drv_data->tx == drv_data->tx_end) {
+			/* Disable tx interrupt */
+			write_SSCR1(read_SSCR1(reg) & ~SSCR1_TIE, reg);
+
+			/* PXA25x_SSP has no timeout, read trailing bytes */
+			if (drv_data->ssp_type == PXA25x_SSP) {
+				while ((read_SSSR(reg) & SSSR_BSY) && limit--)
+					drv_data->read(drv_data);
+
+				if (limit == 0)
+					dev_err(&drv_data->pdev->dev,
+						"interrupt_transfer: "
+						"trailing byte read failed\n");
+			}
+		}
+
+		if ((irq_status & SSSR_TINT)
+				|| (drv_data->rx == drv_data->rx_end)) {
+
+			/* Clear timeout */
+			if (drv_data->ssp_type != PXA25x_SSP)
+				write_SSTO(0, reg);
+			write_SSSR(drv_data->clear_sr, reg);
+			write_SSCR1(read_SSCR1(reg) & ~drv_data->int_cr1, reg);
+
+			/* Update total byte transfered */
+			msg->actual_length += drv_data->len;
+
+			/* Release chip select if requested, transfer delays are
+			 * handled in pump_transfers */
+			if (drv_data->cs_change)
+				drv_data->cs_control(PXA2XX_CS_DEASSERT);
+
+			/* Move to next transfer */
+			msg->state = next_transfer(drv_data);
+
+			/* Schedule transfer tasklet */
+			tasklet_schedule(&drv_data->pump_transfers);
+
+			return IRQ_HANDLED;
+		}
+
+		/* We did something */
+		handled = IRQ_HANDLED;
+	}
+
+	return handled;
+}
+
+static irqreturn_t ssp_int(int irq, void *dev_id, struct pt_regs *regs)
+{
+	struct driver_data *drv_data = (struct driver_data *)dev_id;
+
+	if (!drv_data->cur_msg) {
+		dev_err(&drv_data->pdev->dev, "bad message state "
+				"in interrupt handler\n");
+		/* Never fail */
+		return IRQ_HANDLED;
+	}
+
+	return drv_data->transfer_handler(drv_data);
+}
+
+static void pump_transfers(unsigned long data)
+{
+	struct driver_data *drv_data = (struct driver_data *)data;
+	struct spi_message *message = NULL;
+	struct spi_transfer *transfer = NULL;
+	struct spi_transfer *previous = NULL;
+	struct chip_data *chip = NULL;
+	void *reg = drv_data->ioaddr;
+	u32 clk_div = 0;
+	u8 bits = 0;
+	u32 speed = 0;
+	u32 cr0;
+
+	/* Get current state information */
+	message = drv_data->cur_msg;
+	transfer = drv_data->cur_transfer;
+	chip = drv_data->cur_chip;
+
+	/* Handle for abort */
+	if (message->state == ERROR_STATE) {
+		message->status = -EIO;
+		giveback(message, drv_data);
+		return;
+	}
+
+	/* Handle end of message */
+	if (message->state == DONE_STATE) {
+		message->status = 0;
+		giveback(message, drv_data);
+		return;
+	}
+
+	/* Delay if requested at end of transfer*/
+	if (message->state == RUNNING_STATE) {
+		previous = list_entry(transfer->transfer_list.prev,
+					struct spi_transfer,
+					transfer_list);
+		if (previous->delay_usecs)
+			udelay(previous->delay_usecs);
+	}
+
+	/* Setup the transfer state based on the type of transfer */
+	if (flush(drv_data) == 0) {
+		dev_err(&drv_data->pdev->dev, "pump_transfers: flush failed\n");
+		message->status = -EIO;
+		giveback(message, drv_data);
+		return;
+	}
+	drv_data->n_bytes = chip->n_bytes;
+	drv_data->dma_width = chip->dma_width;
+	drv_data->cs_control = chip->cs_control;
+	drv_data->tx = (void *)transfer->tx_buf;
+	drv_data->tx_end = drv_data->tx + transfer->len;
+	drv_data->rx = transfer->rx_buf;
+	drv_data->rx_end = drv_data->rx + transfer->len;
+	drv_data->rx_dma = transfer->rx_dma;
+	drv_data->tx_dma = transfer->tx_dma;
+	drv_data->len = transfer->len;
+	drv_data->write = drv_data->tx ? chip->write : null_writer;
+	drv_data->read = drv_data->rx ? chip->read : null_reader;
+	drv_data->cs_change = transfer->cs_change;
+
+	/* Change speed and bit per word on a per transfer */
+	if (transfer->speed_hz || transfer->bits_per_word) {
+
+		/* Disable clock */
+		write_SSCR0(chip->cr0 & ~SSCR0_SSE, reg);
+		cr0 = chip->cr0;
+		bits = chip->bits_per_word;
+		speed = chip->speed_hz;
+
+		if (transfer->speed_hz)
+			speed = transfer->speed_hz;
+
+		if (transfer->bits_per_word)
+			bits = transfer->bits_per_word;
+
+		if (reg == SSP1_VIRT)
+			clk_div = SSP1_SerClkDiv(speed);
+		else if (reg == SSP2_VIRT)
+			clk_div = SSP2_SerClkDiv(speed);
+		else if (reg == SSP3_VIRT)
+			clk_div = SSP3_SerClkDiv(speed);
+
+		if (bits <= 8) {
+			drv_data->n_bytes = 1;
+			drv_data->dma_width = DCMD_WIDTH1;
+			drv_data->read = drv_data->read != null_reader ?
+						u8_reader : null_reader;
+			drv_data->write = drv_data->write != null_writer ?
+						u8_writer : null_writer;
+		} else if (bits <= 16) {
+			drv_data->n_bytes = 2;
+			drv_data->dma_width = DCMD_WIDTH2;
+			drv_data->read = drv_data->read != null_reader ?
+						u16_reader : null_reader;
+			drv_data->write = drv_data->write != null_writer ?
+						u16_writer : null_writer;
+		} else if (bits <= 32) {
+			drv_data->n_bytes = 4;
+			drv_data->dma_width = DCMD_WIDTH4;
+			drv_data->read = drv_data->read != null_reader ?
+						u32_reader : null_reader;
+			drv_data->write = drv_data->write != null_writer ?
+						u32_writer : null_writer;
+		}
+
+		cr0 = clk_div
+			| SSCR0_Motorola
+			| SSCR0_DataSize(bits & 0x0f)
+			| SSCR0_SSE
+			| (bits > 16 ? SSCR0_EDSS : 0);
+
+		/* Start it back up */
+		write_SSCR0(cr0, reg);
+	}
+
+	message->state = RUNNING_STATE;
+
+	/* Try to map dma buffer and do a dma transfer if successful */
+	if ((drv_data->dma_mapped = map_dma_buffers(drv_data))) {
+
+		/* Ensure we have the correct interrupt handler */
+		drv_data->transfer_handler = dma_transfer;
+
+		/* Setup rx DMA Channel */
+		DCSR(drv_data->rx_channel) = RESET_DMA_CHANNEL;
+		DSADR(drv_data->rx_channel) = drv_data->ssdr_physical;
+		DTADR(drv_data->rx_channel) = drv_data->rx_dma;
+		if (drv_data->rx == drv_data->null_dma_buf)
+			/* No target address increment */
+			DCMD(drv_data->rx_channel) = DCMD_FLOWSRC
+							| drv_data->dma_width
+							| chip->dma_burst_size
+							| drv_data->len;
+		else
+			DCMD(drv_data->rx_channel) = DCMD_INCTRGADDR
+							| DCMD_FLOWSRC
+							| drv_data->dma_width
+							| chip->dma_burst_size
+							| drv_data->len;
+
+		/* Setup tx DMA Channel */
+		DCSR(drv_data->tx_channel) = RESET_DMA_CHANNEL;
+		DSADR(drv_data->tx_channel) = drv_data->tx_dma;
+		DTADR(drv_data->tx_channel) = drv_data->ssdr_physical;
+		if (drv_data->tx == drv_data->null_dma_buf)
+			/* No source address increment */
+			DCMD(drv_data->tx_channel) = DCMD_FLOWTRG
+							| drv_data->dma_width
+							| chip->dma_burst_size
+							| drv_data->len;
+		else
+			DCMD(drv_data->tx_channel) = DCMD_INCSRCADDR
+							| DCMD_FLOWTRG
+							| drv_data->dma_width
+							| chip->dma_burst_size
+							| drv_data->len;
+
+		/* Enable dma end irqs on SSP to detect end of transfer */
+		if (drv_data->ssp_type == PXA25x_SSP)
+			DCMD(drv_data->tx_channel) |= DCMD_ENDIRQEN;
+
+		/* Fix me, need to handle cs polarity */
+		drv_data->cs_control(PXA2XX_CS_ASSERT);
+
+		/* Go baby, go */
+		write_SSSR(drv_data->clear_sr, reg);
+		DCSR(drv_data->rx_channel) |= DCSR_RUN;
+		DCSR(drv_data->tx_channel) |= DCSR_RUN;
+		if (drv_data->ssp_type != PXA25x_SSP)
+			write_SSTO(chip->timeout, reg);
+		write_SSCR1(chip->cr1
+				| chip->dma_threshold
+				| drv_data->dma_cr1,
+				reg);
+	} else {
+		/* Ensure we have the correct interrupt handler	*/
+		drv_data->transfer_handler = interrupt_transfer;
+
+		/* Fix me, need to handle cs polarity */
+		drv_data->cs_control(PXA2XX_CS_ASSERT);
+
+		/* Go baby, go */
+		write_SSSR(drv_data->clear_sr, reg);
+		if (drv_data->ssp_type != PXA25x_SSP)
+			write_SSTO(chip->timeout, reg);
+		write_SSCR1(chip->cr1
+				| chip->threshold
+				| drv_data->int_cr1,
+				reg);
+	}
+}
+
+static void pump_messages(void *data)
+{
+	struct driver_data *drv_data = data;
+	unsigned long flags;
+
+	/* Lock queue and check for queue work */
+	spin_lock_irqsave(&drv_data->lock, flags);
+	if (list_empty(&drv_data->queue) || drv_data->run == QUEUE_STOPPED) {
+		drv_data->busy = 0;
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+		return;
+	}
+
+	/* Make sure we are not already running a message */
+	if (drv_data->cur_msg) {
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+		return;
+	}
+
+	/* Extract head of queue */
+	drv_data->cur_msg = list_entry(drv_data->queue.next,
+					struct spi_message, queue);
+	list_del_init(&drv_data->cur_msg->queue);
+	drv_data->busy = 1;
+	spin_unlock_irqrestore(&drv_data->lock, flags);
+
+	/* Initial message state*/
+	drv_data->cur_msg->state = START_STATE;
+	drv_data->cur_transfer = list_entry(drv_data->cur_msg->transfers.next,
+						struct spi_transfer,
+						transfer_list);
+
+	/* Setup the SSP using the per chip configuration */
+	drv_data->cur_chip = spi_get_ctldata(drv_data->cur_msg->spi);
+	restore_state(drv_data);
+
+	/* Mark as busy and launch transfers */
+	tasklet_schedule(&drv_data->pump_transfers);
+}
+
+static int transfer(struct spi_device *spi, struct spi_message *msg)
+{
+	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
+	unsigned long flags;
+
+	spin_lock_irqsave(&drv_data->lock, flags);
+
+	if (drv_data->run == QUEUE_STOPPED) {
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+		return -ESHUTDOWN;
+	}
+
+	msg->actual_length = 0;
+	msg->status = -EINPROGRESS;
+	msg->state = START_STATE;
+
+	list_add_tail(&msg->queue, &drv_data->queue);
+
+	if (drv_data->run == QUEUE_RUNNING && !drv_data->busy)
+		queue_work(drv_data->workqueue, &drv_data->pump_messages);
+
+	spin_unlock_irqrestore(&drv_data->lock, flags);
+
+	return 0;
+}
+
+static int setup(struct spi_device *spi)
+{
+	struct pxa2xx_spi_chip *chip_info = NULL;
+	struct chip_data *chip;
+	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
+	unsigned int clk_div;
+
+	if (!spi->bits_per_word)
+		spi->bits_per_word = 8;
+
+	if (drv_data->ssp_type != PXA25x_SSP
+			&& (spi->bits_per_word < 4 || spi->bits_per_word > 32))
+		return -EINVAL;
+	else if (spi->bits_per_word < 4 || spi->bits_per_word > 16)
+		return -EINVAL;
+
+	/* Only alloc (or use chip_info) on first setup */
+	chip = spi_get_ctldata(spi);
+	if (chip == NULL) {
+		chip = kzalloc(sizeof(struct chip_data), GFP_KERNEL);
+		if (!chip)
+			return -ENOMEM;
+
+		chip->cs_control = null_cs_control;
+		chip->enable_dma = 0;
+		chip->timeout = 5;
+		chip->threshold = SSCR1_RxTresh(1) | SSCR1_TxTresh(1);
+		chip->dma_burst_size = drv_data->master_info->enable_dma ?
+					DCMD_BURST8 : 0;
+
+		chip_info = spi->controller_data;
+	}
+
+	/* chip_info isn't always needed */
+	if (chip_info) {
+		if (chip_info->cs_control)
+			chip->cs_control = chip_info->cs_control;
+
+		chip->timeout = (chip_info->timeout_microsecs * 10000) / 2712;
+
+		chip->threshold = SSCR1_RxTresh(chip_info->rx_threshold)
+					| SSCR1_TxTresh(chip_info->tx_threshold);
+
+		chip->enable_dma = chip_info->dma_burst_size != 0
+					&& drv_data->master_info->enable_dma;
+		chip->dma_threshold = 0;
+
+		if (chip->enable_dma) {
+			if (chip_info->dma_burst_size <= 8) {
+				chip->dma_threshold = SSCR1_RxTresh(8)
+							| SSCR1_TxTresh(8);
+				chip->dma_burst_size = DCMD_BURST8;
+			} else if (chip_info->dma_burst_size <= 16) {
+				chip->dma_threshold = SSCR1_RxTresh(16)
+							| SSCR1_TxTresh(16);
+				chip->dma_burst_size = DCMD_BURST16;
+			} else {
+				chip->dma_threshold = SSCR1_RxTresh(32)
+							| SSCR1_TxTresh(32);
+				chip->dma_burst_size = DCMD_BURST32;
+			}
+		}
+
+
+		if (chip_info->enable_loopback)
+			chip->cr1 = SSCR1_LBM;
+	}
+
+	if (drv_data->ioaddr == SSP1_VIRT)
+		clk_div = SSP1_SerClkDiv(spi->max_speed_hz);
+	else if (drv_data->ioaddr == SSP2_VIRT)
+		clk_div = SSP2_SerClkDiv(spi->max_speed_hz);
+	else if (drv_data->ioaddr == SSP3_VIRT)
+		clk_div = SSP3_SerClkDiv(spi->max_speed_hz);
+	else
+		return -ENODEV;
+	chip->speed_hz = spi->max_speed_hz;
+
+	chip->cr0 = clk_div
+			| SSCR0_Motorola
+			| SSCR0_DataSize(spi->bits_per_word & 0x0f)
+			| SSCR0_SSE
+			| (spi->bits_per_word > 16 ? SSCR0_EDSS : 0);
+	chip->cr1 |= (((spi->mode & SPI_CPHA) != 0) << 4)
+			| (((spi->mode & SPI_CPOL) != 0) << 3);
+
+	/* NOTE:  PXA25x_SSP _could_ use external clocking ... */
+	if (drv_data->ssp_type != PXA25x_SSP)
+		dev_dbg(&spi->dev, "%d bits/word, %d Hz, mode %d\n",
+				spi->bits_per_word,
+				(CLOCK_SPEED_HZ)
+					/ (1 + ((chip->cr0 & SSCR0_SCR) >> 8)),
+				spi->mode & 0x3);
+	else
+		dev_dbg(&spi->dev, "%d bits/word, %d Hz, mode %d\n",
+				spi->bits_per_word,
+				(CLOCK_SPEED_HZ/2)
+					/ (1 + ((chip->cr0 & SSCR0_SCR) >> 8)),
+				spi->mode & 0x3);
+
+	if (spi->bits_per_word <= 8) {
+		chip->n_bytes = 1;
+		chip->dma_width = DCMD_WIDTH1;
+		chip->read = u8_reader;
+		chip->write = u8_writer;
+	} else if (spi->bits_per_word <= 16) {
+		chip->n_bytes = 2;
+		chip->dma_width = DCMD_WIDTH2;
+		chip->read = u16_reader;
+		chip->write = u16_writer;
+	} else if (spi->bits_per_word <= 32) {
+		chip->cr0 |= SSCR0_EDSS;
+		chip->n_bytes = 4;
+		chip->dma_width = DCMD_WIDTH4;
+		chip->read = u32_reader;
+		chip->write = u32_writer;
+	} else {
+		dev_err(&spi->dev, "invalid wordsize\n");
+		kfree(chip);
+		return -ENODEV;
+	}
+	chip->bits_per_word = spi->bits_per_word;
+
+	spi_set_ctldata(spi, chip);
+
+	return 0;
+}
+
+static void cleanup(const struct spi_device *spi)
+{
+	struct chip_data *chip = spi_get_ctldata((struct spi_device *)spi);
+
+	kfree(chip);
+}
+
+static int init_queue(struct driver_data *drv_data)
+{
+	INIT_LIST_HEAD(&drv_data->queue);
+	spin_lock_init(&drv_data->lock);
+
+	drv_data->run = QUEUE_STOPPED;
+	drv_data->busy = 0;
+
+	tasklet_init(&drv_data->pump_transfers,
+			pump_transfers,	(unsigned long)drv_data);
+
+	INIT_WORK(&drv_data->pump_messages, pump_messages, drv_data);
+	drv_data->workqueue = create_singlethread_workqueue(
+					drv_data->master->cdev.dev->bus_id);
+	if (drv_data->workqueue == NULL)
+		return -EBUSY;
+
+	return 0;
+}
+
+static int start_queue(struct driver_data *drv_data)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&drv_data->lock, flags);
+
+	if (drv_data->run == QUEUE_RUNNING || drv_data->busy) {
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+		return -EBUSY;
+	}
+
+	drv_data->run = QUEUE_RUNNING;
+	drv_data->cur_msg = NULL;
+	drv_data->cur_transfer = NULL;
+	drv_data->cur_chip = NULL;
+	spin_unlock_irqrestore(&drv_data->lock, flags);
+
+	queue_work(drv_data->workqueue, &drv_data->pump_messages);
+
+	return 0;
+}
+
+static int stop_queue(struct driver_data *drv_data)
+{
+	unsigned long flags;
+	unsigned limit = 500;
+	int status = 0;
+
+	spin_lock_irqsave(&drv_data->lock, flags);
+
+	/* This is a bit lame, but is optimized for the common execution path.
+	 * A wait_queue on the drv_data->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 */
+	drv_data->run = QUEUE_STOPPED;
+	while (!list_empty(&drv_data->queue) && drv_data->busy && limit--) {
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+		msleep(10);
+		spin_lock_irqsave(&drv_data->lock, flags);
+	}
+
+	if (!list_empty(&drv_data->queue) || drv_data->busy)
+		status = -EBUSY;
+
+	spin_unlock_irqrestore(&drv_data->lock, flags);
+
+	return status;
+}
+
+static int destroy_queue(struct driver_data *drv_data)
+{
+	int status;
+
+	status = stop_queue(drv_data);
+	if (status != 0)
+		return status;
+
+	destroy_workqueue(drv_data->workqueue);
+
+	return 0;
+}
+
+static int pxa2xx_spi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct pxa2xx_spi_master *platform_info;
+	struct spi_master *master;
+	struct driver_data *drv_data = 0;
+	struct resource *memory_resource;
+	int irq;
+	int status = 0;
+
+	platform_info = dev->platform_data;
+
+	if (platform_info->ssp_type == SSP_UNDEFINED) {
+		dev_err(&pdev->dev, "undefined SSP\n");
+		return -ENODEV;
+	}
+
+	/* Allocate master with space for drv_data and null dma buffer */
+	master = spi_alloc_master(dev, sizeof(struct driver_data) + 16);
+	if (!master) {
+		dev_err(&pdev->dev, "can not alloc spi_master\n");
+		return -ENOMEM;
+	}
+	drv_data = spi_master_get_devdata(master);
+	drv_data->master = master;
+	drv_data->master_info = platform_info;
+	drv_data->pdev = pdev;
+
+	master->bus_num = pdev->id;
+	master->num_chipselect = platform_info->num_chipselect;
+	master->cleanup = cleanup;
+	master->setup = setup;
+	master->transfer = transfer;
+
+	drv_data->ssp_type = platform_info->ssp_type;
+	drv_data->null_dma_buf = (u32 *)ALIGN((u32)(drv_data +
+						sizeof(struct driver_data)), 8);
+
+	/* Setup register addresses */
+	memory_resource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!memory_resource) {
+		dev_err(&pdev->dev, "memory resources not defined\n");
+		status = -ENODEV;
+		goto out_error_master_alloc;
+	}
+
+	drv_data->ioaddr = (void *)io_p2v(memory_resource->start);
+	drv_data->ssdr_physical = memory_resource->start + 0x00000010;
+	if (platform_info->ssp_type == PXA25x_SSP) {
+		drv_data->int_cr1 = SSCR1_TIE | SSCR1_RIE;
+		drv_data->dma_cr1 = 0;
+		drv_data->clear_sr = SSSR_ROR;
+		drv_data->mask_sr = SSSR_RFS | SSSR_TFS | SSSR_ROR;
+	} else {
+		drv_data->int_cr1 = SSCR1_TIE | SSCR1_RIE | SSCR1_TINTE;
+		drv_data->dma_cr1 = SSCR1_TSRE | SSCR1_RSRE | SSCR1_TINTE;
+		drv_data->clear_sr = SSSR_ROR | SSSR_TINT;
+		drv_data->mask_sr = SSSR_TINT | SSSR_RFS | SSSR_TFS | SSSR_ROR;
+	}
+
+	/* Attach to IRQ */
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "irq resource not defined\n");
+		status = -ENODEV;
+		goto out_error_master_alloc;
+	}
+
+	status = request_irq(irq, ssp_int, SA_INTERRUPT, dev->bus_id, drv_data);
+	if (status < 0) {
+		dev_err(&pdev->dev, "can not get IRQ\n");
+		goto out_error_master_alloc;
+	}
+
+	/* Setup DMA if requested */
+	drv_data->tx_channel = -1;
+	drv_data->rx_channel = -1;
+	if (platform_info->enable_dma) {
+
+		/* Get two DMA channels	(rx and tx) */
+		drv_data->rx_channel = pxa_request_dma("pxa2xx_spi_ssp_rx",
+							DMA_PRIO_HIGH,
+							dma_handler,
+							drv_data);
+		if (drv_data->rx_channel < 0) {
+			dev_err(dev, "problem (%d) requesting rx channel\n",
+				drv_data->rx_channel);
+			status = -ENODEV;
+			goto out_error_irq_alloc;
+		}
+		drv_data->tx_channel = pxa_request_dma("pxa2xx_spi_ssp_tx",
+							DMA_PRIO_MEDIUM,
+							dma_handler,
+							drv_data);
+		if (drv_data->tx_channel < 0) {
+			dev_err(dev, "problem (%d) requesting tx channel\n",
+				drv_data->tx_channel);
+			status = -ENODEV;
+			goto out_error_dma_alloc;
+		}
+
+		if (drv_data->ioaddr == SSP1_VIRT) {
+				DRCMRRXSSDR = DRCMR_MAPVLD
+						| drv_data->rx_channel;
+				DRCMRTXSSDR = DRCMR_MAPVLD
+						| drv_data->tx_channel;
+		} else if (drv_data->ioaddr == SSP2_VIRT) {
+				DRCMRRXSS2DR = DRCMR_MAPVLD
+						| drv_data->rx_channel;
+				DRCMRTXSS2DR = DRCMR_MAPVLD
+						| drv_data->tx_channel;
+		} else if (drv_data->ioaddr == SSP3_VIRT) {
+				DRCMRRXSS3DR = DRCMR_MAPVLD
+						| drv_data->rx_channel;
+				DRCMRTXSS3DR = DRCMR_MAPVLD
+						| drv_data->tx_channel;
+		} else {
+			dev_err(dev, "bad SSP type\n");
+			goto out_error_dma_alloc;
+		}
+	}
+
+	/* Enable SOC clock */
+	pxa_set_cken(platform_info->clock_enable, 1);
+
+	/* Load default SSP configuration */
+	write_SSCR0(0, drv_data->ioaddr);
+	write_SSCR1(SSCR1_RxTresh(4) | SSCR1_TxTresh(12), drv_data->ioaddr);
+	write_SSCR0(SSCR0_SerClkDiv(2)
+			| SSCR0_Motorola
+			| SSCR0_DataSize(8),
+			drv_data->ioaddr);
+	if (drv_data->ssp_type != PXA25x_SSP)
+		write_SSTO(0, drv_data->ioaddr);
+	write_SSPSP(0, drv_data->ioaddr);
+
+	/* Initial and start queue */
+	status = init_queue(drv_data);
+	if (status != 0) {
+		dev_err(&pdev->dev, "problem initializing queue\n");
+		goto out_error_clock_enabled;
+	}
+	status = start_queue(drv_data);
+	if (status != 0) {
+		dev_err(&pdev->dev, "problem starting queue\n");
+		goto out_error_clock_enabled;
+	}
+
+	/* Register with the SPI framework */
+	platform_set_drvdata(pdev, drv_data);
+	status = spi_register_master(master);
+	if (status != 0) {
+		dev_err(&pdev->dev, "problem registering spi master\n");
+		goto out_error_queue_alloc;
+	}
+
+	return status;
+
+out_error_queue_alloc:
+	destroy_queue(drv_data);
+
+out_error_clock_enabled:
+	pxa_set_cken(platform_info->clock_enable, 0);
+
+out_error_dma_alloc:
+	if (drv_data->tx_channel != -1)
+		pxa_free_dma(drv_data->tx_channel);
+	if (drv_data->rx_channel != -1)
+		pxa_free_dma(drv_data->rx_channel);
+
+out_error_irq_alloc:
+	free_irq(irq, drv_data);
+
+out_error_master_alloc:
+	spi_master_put(master);
+	return status;
+}
+
+static int pxa2xx_spi_remove(struct platform_device *pdev)
+{
+	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	int irq;
+	int status = 0;
+
+	if (!drv_data)
+		return 0;
+
+	/* Remove the queue */
+	status = destroy_queue(drv_data);
+	if (status != 0)
+		return status;
+
+	/* Disable the SSP at the peripheral and SOC level */
+	write_SSCR0(0, drv_data->ioaddr);
+	pxa_set_cken(drv_data->master_info->clock_enable, 0);
+
+	/* Release DMA */
+	if (drv_data->master_info->enable_dma) {
+		if (drv_data->ioaddr == SSP1_VIRT) {
+			DRCMRRXSSDR = 0;
+			DRCMRTXSSDR = 0;
+		} else if (drv_data->ioaddr == SSP2_VIRT) {
+			DRCMRRXSS2DR = 0;
+			DRCMRTXSS2DR = 0;
+		} else if (drv_data->ioaddr == SSP3_VIRT) {
+			DRCMRRXSS3DR = 0;
+			DRCMRTXSS3DR = 0;
+		}
+		pxa_free_dma(drv_data->tx_channel);
+		pxa_free_dma(drv_data->rx_channel);
+	}
+
+	/* Release IRQ */
+	irq = platform_get_irq(pdev, 0);
+	if (irq >= 0)
+		free_irq(irq, drv_data);
+
+	/* Disconnect from the SPI framework */
+	spi_unregister_master(drv_data->master);
+
+	/* Prevent double remove */
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static void pxa2xx_spi_shutdown(struct platform_device *pdev)
+{
+	int status = 0;
+
+	if ((status = pxa2xx_spi_remove(pdev)) != 0)
+		dev_err(&pdev->dev, "shutdown failed with %d\n", status);
+}
+
+#ifdef CONFIG_PM
+static int suspend_devices(struct device *dev, void *pm_message)
+{
+	pm_message_t *state = pm_message;
+
+	if (dev->power.power_state.event != state->event) {
+		dev_warn(dev, "pm state does not match request\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+static int pxa2xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	int status = 0;
+
+	/* Check all childern for current power state */
+	if (device_for_each_child(&pdev->dev, &state, suspend_devices) != 0) {
+		dev_warn(&pdev->dev, "suspend aborted\n");
+		return -1;
+	}
+
+	status = stop_queue(drv_data);
+	if (status != 0)
+		return status;
+	write_SSCR0(0, drv_data->ioaddr);
+	pxa_set_cken(drv_data->master_info->clock_enable, 0);
+
+	return 0;
+}
+
+static int pxa2xx_spi_resume(struct platform_device *pdev)
+{
+	struct driver_data *drv_data = platform_get_drvdata(pdev);
+	int status = 0;
+
+	/* Enable the SSP clock */
+	pxa_set_cken(drv_data->master_info->clock_enable, 1);
+
+	/* Start the queue running */
+	status = start_queue(drv_data);
+	if (status != 0) {
+		dev_err(&pdev->dev, "problem starting queue (%d)\n", status);
+		return status;
+	}
+
+	return 0;
+}
+#else
+#define pxa2xx_spi_suspend NULL
+#define pxa2xx_spi_resume NULL
+#endif /* CONFIG_PM */
+
+static struct platform_driver driver = {
+	.driver = {
+		.name = "pxa2xx-spi",
+		.bus = &platform_bus_type,
+		.owner = THIS_MODULE,
+	},
+	.probe = pxa2xx_spi_probe,
+	.remove = __devexit_p(pxa2xx_spi_remove),
+	.shutdown = pxa2xx_spi_shutdown,
+	.suspend = pxa2xx_spi_suspend,
+	.resume = pxa2xx_spi_resume,
+};
+
+static int __init pxa2xx_spi_init(void)
+{
+	platform_driver_register(&driver);
+
+	return 0;
+}
+module_init(pxa2xx_spi_init);
+
+static void __exit pxa2xx_spi_exit(void)
+{
+	platform_driver_unregister(&driver);
+}
+module_exit(pxa2xx_spi_exit);
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 94f5e8ed83a7..7a3f733051e9 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -395,7 +395,7 @@ EXPORT_SYMBOL_GPL(spi_alloc_master);
 int __init_or_module
 spi_register_master(struct spi_master *master)
 {
-	static atomic_t		dyn_bus_id = ATOMIC_INIT(0);
+	static atomic_t		dyn_bus_id = ATOMIC_INIT((1<<16) - 1);
 	struct device		*dev = master->cdev.dev;
 	int			status = -ENODEV;
 	int			dynamic = 0;
@@ -404,7 +404,7 @@ spi_register_master(struct spi_master *master)
 		return -ENODEV;
 
 	/* convention:  dynamically assigned bus IDs count down from the max */
-	if (master->bus_num == 0) {
+	if (master->bus_num < 0) {
 		master->bus_num = atomic_dec_return(&dyn_bus_id);
 		dynamic = 1;
 	}
@@ -522,7 +522,8 @@ int spi_sync(struct spi_device *spi, struct spi_message *message)
 }
 EXPORT_SYMBOL_GPL(spi_sync);
 
-#define	SPI_BUFSIZ	(SMP_CACHE_BYTES)
+/* portable code must never pass more than 32 bytes */
+#define	SPI_BUFSIZ	max(32,SMP_CACHE_BYTES)
 
 static u8	*buf;
 
diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c
index f037e5593269..dd2f950b21a7 100644
--- a/drivers/spi/spi_bitbang.c
+++ b/drivers/spi/spi_bitbang.c
@@ -138,6 +138,45 @@ static unsigned bitbang_txrx_32(
 	return t->len - count;
 }
 
+int spi_bitbang_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct spi_bitbang_cs	*cs = spi->controller_state;
+	u8			bits_per_word;
+	u32			hz;
+
+	if (t) {
+		bits_per_word = t->bits_per_word;
+		hz = t->speed_hz;
+	} else {
+		bits_per_word = 0;
+		hz = 0;
+	}
+
+	/* spi_transfer level calls that work per-word */
+	if (!bits_per_word)
+		bits_per_word = spi->bits_per_word;
+	if (bits_per_word <= 8)
+		cs->txrx_bufs = bitbang_txrx_8;
+	else if (bits_per_word <= 16)
+		cs->txrx_bufs = bitbang_txrx_16;
+	else if (bits_per_word <= 32)
+		cs->txrx_bufs = bitbang_txrx_32;
+	else
+		return -EINVAL;
+
+	/* nsecs = (clock period)/2 */
+	if (!hz)
+		hz = spi->max_speed_hz;
+	if (hz) {
+		cs->nsecs = (1000000000/2) / hz;
+		if (cs->nsecs > (MAX_UDELAY_MS * 1000 * 1000))
+			return -EINVAL;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(spi_bitbang_setup_transfer);
+
 /**
  * spi_bitbang_setup - default setup for per-word I/O loops
  */
@@ -145,8 +184,16 @@ int spi_bitbang_setup(struct spi_device *spi)
 {
 	struct spi_bitbang_cs	*cs = spi->controller_state;
 	struct spi_bitbang	*bitbang;
+	int			retval;
 
-	if (!spi->max_speed_hz)
+	bitbang = spi_master_get_devdata(spi->master);
+
+	/* REVISIT: some systems will want to support devices using lsb-first
+	 * bit encodings on the wire.  In pure software that would be trivial,
+	 * just bitbang_txrx_le_cphaX() routines shifting the other way, and
+	 * some hardware controllers also have this support.
+	 */
+	if ((spi->mode & SPI_LSB_FIRST) != 0)
 		return -EINVAL;
 
 	if (!cs) {
@@ -155,32 +202,20 @@ int spi_bitbang_setup(struct spi_device *spi)
 			return -ENOMEM;
 		spi->controller_state = cs;
 	}
-	bitbang = spi_master_get_devdata(spi->master);
 
 	if (!spi->bits_per_word)
 		spi->bits_per_word = 8;
 
-	/* spi_transfer level calls that work per-word */
-	if (spi->bits_per_word <= 8)
-		cs->txrx_bufs = bitbang_txrx_8;
-	else if (spi->bits_per_word <= 16)
-		cs->txrx_bufs = bitbang_txrx_16;
-	else if (spi->bits_per_word <= 32)
-		cs->txrx_bufs = bitbang_txrx_32;
-	else
-		return -EINVAL;
-
 	/* per-word shift register access, in hardware or bitbanging */
 	cs->txrx_word = bitbang->txrx_word[spi->mode & (SPI_CPOL|SPI_CPHA)];
 	if (!cs->txrx_word)
 		return -EINVAL;
 
-	/* nsecs = (clock period)/2 */
-	cs->nsecs = (1000000000/2) / (spi->max_speed_hz);
-	if (cs->nsecs > MAX_UDELAY_MS * 1000)
-		return -EINVAL;
+	retval = spi_bitbang_setup_transfer(spi, NULL);
+	if (retval < 0)
+		return retval;
 
-	dev_dbg(&spi->dev, "%s, mode %d, %u bits/w, %u nsec\n",
+	dev_dbg(&spi->dev, "%s, mode %d, %u bits/w, %u nsec/bit\n",
 			__FUNCTION__, spi->mode & (SPI_CPOL | SPI_CPHA),
 			spi->bits_per_word, 2 * cs->nsecs);
 
@@ -246,6 +281,8 @@ static void bitbang_work(void *_bitbang)
 		unsigned		tmp;
 		unsigned		cs_change;
 		int			status;
+		int			(*setup_transfer)(struct spi_device *,
+						struct spi_transfer *);
 
 		m = container_of(bitbang->queue.next, struct spi_message,
 				queue);
@@ -262,6 +299,7 @@ static void bitbang_work(void *_bitbang)
 		tmp = 0;
 		cs_change = 1;
 		status = 0;
+		setup_transfer = NULL;
 
 		list_for_each_entry (t, &m->transfers, transfer_list) {
 			if (bitbang->shutdown) {
@@ -269,6 +307,20 @@ static void bitbang_work(void *_bitbang)
 				break;
 			}
 
+			/* override or restore speed and wordsize */
+			if (t->speed_hz || t->bits_per_word) {
+				setup_transfer = bitbang->setup_transfer;
+				if (!setup_transfer) {
+					status = -ENOPROTOOPT;
+					break;
+				}
+			}
+			if (setup_transfer) {
+				status = setup_transfer(spi, t);
+				if (status < 0)
+					break;
+			}
+
 			/* set up default clock polarity, and activate chip;
 			 * this implicitly updates clock and spi modes as
 			 * previously recorded for this device via setup().
@@ -325,6 +377,10 @@ static void bitbang_work(void *_bitbang)
 		m->status = status;
 		m->complete(m->context);
 
+		/* restore speed and wordsize */
+		if (setup_transfer)
+			setup_transfer(spi, NULL);
+
 		/* normally deactivate chipselect ... unless no error and
 		 * cs_change has hinted that the next message will probably
 		 * be for this chip too.
@@ -348,6 +404,7 @@ int spi_bitbang_transfer(struct spi_device *spi, struct spi_message *m)
 {
 	struct spi_bitbang	*bitbang;
 	unsigned long		flags;
+	int			status = 0;
 
 	m->actual_length = 0;
 	m->status = -EINPROGRESS;
@@ -357,11 +414,15 @@ int spi_bitbang_transfer(struct spi_device *spi, struct spi_message *m)
 		return -ESHUTDOWN;
 
 	spin_lock_irqsave(&bitbang->lock, flags);
-	list_add_tail(&m->queue, &bitbang->queue);
-	queue_work(bitbang->workqueue, &bitbang->work);
+	if (!spi->max_speed_hz)
+		status = -ENETDOWN;
+	else {
+		list_add_tail(&m->queue, &bitbang->queue);
+		queue_work(bitbang->workqueue, &bitbang->work);
+	}
 	spin_unlock_irqrestore(&bitbang->lock, flags);
 
-	return 0;
+	return status;
 }
 EXPORT_SYMBOL_GPL(spi_bitbang_transfer);
 
@@ -406,6 +467,9 @@ int spi_bitbang_start(struct spi_bitbang *bitbang)
 		bitbang->use_dma = 0;
 		bitbang->txrx_bufs = spi_bitbang_bufs;
 		if (!bitbang->master->setup) {
+			if (!bitbang->setup_transfer)
+				bitbang->setup_transfer =
+					 spi_bitbang_setup_transfer;
 			bitbang->master->setup = spi_bitbang_setup;
 			bitbang->master->cleanup = spi_bitbang_cleanup;
 		}
diff --git a/include/asm-arm/arch-pxa/pxa2xx_spi.h b/include/asm-arm/arch-pxa/pxa2xx_spi.h
new file mode 100644
index 000000000000..1e70908b816f
--- /dev/null
+++ b/include/asm-arm/arch-pxa/pxa2xx_spi.h
@@ -0,0 +1,68 @@
+/*
+ * 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; 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef PXA2XX_SPI_H_
+#define PXA2XX_SPI_H_
+
+#define PXA2XX_CS_ASSERT (0x01)
+#define PXA2XX_CS_DEASSERT (0x02)
+
+#if defined(CONFIG_PXA25x)
+#define CLOCK_SPEED_HZ 3686400
+#define SSP1_SerClkDiv(x) (((CLOCK_SPEED_HZ/2/(x+1))<<8)&0x0000ff00)
+#define SSP2_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
+#define SSP3_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
+#elif defined(CONFIG_PXA27x)
+#define CLOCK_SPEED_HZ 13000000
+#define SSP1_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
+#define SSP2_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
+#define SSP3_SerClkDiv(x) (((CLOCK_SPEED_HZ/(x+1))<<8)&0x000fff00)
+#endif
+
+#define SSP1_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(1)))))
+#define SSP2_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(2)))))
+#define SSP3_VIRT ((void *)(io_p2v(__PREG(SSCR0_P(3)))))
+
+enum pxa_ssp_type {
+	SSP_UNDEFINED = 0,
+	PXA25x_SSP,  /* pxa 210, 250, 255, 26x */
+	PXA25x_NSSP, /* pxa 255, 26x (including ASSP) */
+	PXA27x_SSP,
+};
+
+/* device.platform_data for SSP controller devices */
+struct pxa2xx_spi_master {
+	enum pxa_ssp_type ssp_type;
+	u32 clock_enable;
+	u16 num_chipselect;
+	u8 enable_dma;
+};
+
+/* spi_board_info.controller_data for SPI slave devices,
+ * copied to spi_device.platform_data ... mostly for dma tuning
+ */
+struct pxa2xx_spi_chip {
+	u8 tx_threshold;
+	u8 rx_threshold;
+	u8 dma_burst_size;
+	u32 timeout_microsecs;
+	u8 enable_loopback;
+	void (*cs_control)(u32 command);
+};
+
+#endif /*PXA2XX_SPI_H_*/
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index b05f1463a267..e928c0dcc297 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -31,18 +31,23 @@ extern struct bus_type spi_bus_type;
  * @master: SPI controller used with the device.
  * @max_speed_hz: Maximum clock rate to be used with this chip
  *	(on this board); may be changed by the device's driver.
+ *	The spi_transfer.speed_hz can override this for each transfer.
  * @chip-select: Chipselect, distinguishing chips handled by "master".
  * @mode: The spi mode defines how data is clocked out and in.
  *	This may be changed by the device's driver.
+ *	The "active low" default for chipselect mode can be overridden,
+ *	as can the "MSB first" default for each word in a transfer.
  * @bits_per_word: Data transfers involve one or more words; word sizes
- * 	like eight or 12 bits are common.  In-memory wordsizes are
+ *	like eight or 12 bits are common.  In-memory wordsizes are
  *	powers of two bytes (e.g. 20 bit samples use 32 bits).
- *	This may be changed by the device's driver.
+ *	This may be changed by the device's driver, or left at the
+ *	default (0) indicating protocol words are eight bit bytes.
+ *	The spi_transfer.bits_per_word can override this for each transfer.
  * @irq: Negative, or the number passed to request_irq() to receive
- * 	interrupts from this device.
+ *	interrupts from this device.
  * @controller_state: Controller's runtime state
  * @controller_data: Board-specific definitions for controller, such as
- * 	FIFO initialization parameters; from board_info.controller_data
+ *	FIFO initialization parameters; from board_info.controller_data
  *
  * An spi_device is used to interchange data between an SPI slave
  * (usually a discrete chip) and CPU memory.
@@ -65,6 +70,7 @@ struct spi_device {
 #define	SPI_MODE_2	(SPI_CPOL|0)
 #define	SPI_MODE_3	(SPI_CPOL|SPI_CPHA)
 #define	SPI_CS_HIGH	0x04			/* chipselect active high? */
+#define	SPI_LSB_FIRST	0x08			/* per-word bits-on-wire */
 	u8			bits_per_word;
 	int			irq;
 	void			*controller_state;
@@ -73,7 +79,6 @@ struct spi_device {
 
 	// likely need more hooks for more protocol options affecting how
 	// the controller talks to each chip, like:
-	//  - bit order (default is wordwise msb-first)
 	//  - memory packing (12 bit samples into low bits, others zeroed)
 	//  - priority
 	//  - drop chipselect after each word
@@ -143,13 +148,13 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv)
  * struct spi_master - interface to SPI master controller
  * @cdev: class interface to this driver
  * @bus_num: board-specific (and often SOC-specific) identifier for a
- * 	given SPI controller.
+ *	given SPI controller.
  * @num_chipselect: chipselects are used to distinguish individual
- * 	SPI slaves, and are numbered from zero to num_chipselects.
- * 	each slave has a chipselect signal, but it's common that not
- * 	every chipselect is connected to a slave.
+ *	SPI slaves, and are numbered from zero to num_chipselects.
+ *	each slave has a chipselect signal, but it's common that not
+ *	every chipselect is connected to a slave.
  * @setup: updates the device mode and clocking records used by a
- * 	device's SPI controller; protocol code may call this.
+ *	device's SPI controller; protocol code may call this.
  * @transfer: adds a message to the controller's transfer queue.
  * @cleanup: frees controller-specific state
  *
@@ -167,13 +172,13 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv)
 struct spi_master {
 	struct class_device	cdev;
 
-	/* other than zero (== assign one dynamically), bus_num is fully
+	/* other than negative (== assign one dynamically), bus_num is fully
 	 * board-specific.  usually that simplifies to being SOC-specific.
-	 * example:  one SOC has three SPI controllers, numbered 1..3,
+	 * example:  one SOC has three SPI controllers, numbered 0..2,
 	 * and one board's schematics might show it using SPI-2.  software
 	 * would normally use bus_num=2 for that controller.
 	 */
-	u16			bus_num;
+	s16			bus_num;
 
 	/* chipselects will be integral to many controllers; some others
 	 * might use board-specific GPIOs.
@@ -268,10 +273,14 @@ extern struct spi_master *spi_busnum_to_master(u16 busnum);
  * @tx_dma: DMA address of tx_buf, if spi_message.is_dma_mapped
  * @rx_dma: DMA address of rx_buf, if spi_message.is_dma_mapped
  * @len: size of rx and tx buffers (in bytes)
+ * @speed_hz: Select a speed other then the device default for this
+ *      transfer. If 0 the default (from spi_device) is used.
+ * @bits_per_word: select a bits_per_word other then the device default
+ *      for this transfer. If 0 the default (from spi_device) is used.
  * @cs_change: affects chipselect after this transfer completes
  * @delay_usecs: microseconds to delay after this transfer before
- * 	(optionally) changing the chipselect status, then starting
- * 	the next transfer or completing this spi_message.
+ *	(optionally) changing the chipselect status, then starting
+ *	the next transfer or completing this spi_message.
  * @transfer_list: transfers are sequenced through spi_message.transfers
  *
  * SPI transfers always write the same number of bytes as they read.
@@ -322,7 +331,9 @@ struct spi_transfer {
 	dma_addr_t	rx_dma;
 
 	unsigned	cs_change:1;
+	u8		bits_per_word;
 	u16		delay_usecs;
+	u32		speed_hz;
 
 	struct list_head transfer_list;
 };
@@ -356,7 +367,7 @@ struct spi_transfer {
  * and its transfers, ignore them until its completion callback.
  */
 struct spi_message {
-	struct list_head 	transfers;
+	struct list_head	transfers;
 
 	struct spi_device	*spi;
 
@@ -374,7 +385,7 @@ struct spi_message {
 	 */
 
 	/* completion is reported through a callback */
-	void 			(*complete)(void *context);
+	void			(*complete)(void *context);
 	void			*context;
 	unsigned		actual_length;
 	int			status;
diff --git a/include/linux/spi/spi_bitbang.h b/include/linux/spi/spi_bitbang.h
index c961fe9bf3eb..16ce178f54d7 100644
--- a/include/linux/spi/spi_bitbang.h
+++ b/include/linux/spi/spi_bitbang.h
@@ -30,6 +30,12 @@ struct spi_bitbang {
 
 	struct spi_master	*master;
 
+	/* setup_transfer() changes clock and/or wordsize to match settings
+	 * for this transfer; zeroes restore defaults from spi_device.
+	 */
+	int	(*setup_transfer)(struct spi_device *spi,
+			struct spi_transfer *t);
+
 	void	(*chipselect)(struct spi_device *spi, int is_on);
 #define	BITBANG_CS_ACTIVE	1	/* normally nCS, active low */
 #define	BITBANG_CS_INACTIVE	0
@@ -51,6 +57,8 @@ struct spi_bitbang {
 extern int spi_bitbang_setup(struct spi_device *spi);
 extern void spi_bitbang_cleanup(const struct spi_device *spi);
 extern int spi_bitbang_transfer(struct spi_device *spi, struct spi_message *m);
+extern int spi_bitbang_setup_transfer(struct spi_device *spi,
+				      struct spi_transfer *t);
 
 /* start or stop queue processing */
 extern int spi_bitbang_start(struct spi_bitbang *spi);