Merge branch 'linus' into stackprotector

Conflicts:
	arch/x86/kernel/Makefile
	include/asm-x86/pda.h

16 files changed, 1089 insertions, 292 deletions

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 66ec5d8808de..b9d0efb6803f 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -49,25 +49,26 @@ config SPI_MASTER
 	  controller and the protocol drivers for the SPI slave chips
 	  that are connected.
 
+if SPI_MASTER
+
 comment "SPI Master Controller Drivers"
-	depends on SPI_MASTER
 
 config SPI_ATMEL
 	tristate "Atmel SPI Controller"
-	depends on (ARCH_AT91 || AVR32) && SPI_MASTER
+	depends on (ARCH_AT91 || AVR32)
 	help
 	  This selects a driver for the Atmel SPI Controller, present on
 	  many AT32 (AVR32) and AT91 (ARM) chips.
 
 config SPI_BFIN
 	tristate "SPI controller driver for ADI Blackfin5xx"
-	depends on SPI_MASTER && BLACKFIN
+	depends on BLACKFIN
 	help
 	  This is the SPI controller master driver for Blackfin 5xx processor.
 
 config SPI_AU1550
 	tristate "Au1550/Au12x0 SPI Controller"
-	depends on SPI_MASTER && (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL
+	depends on (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  If you say yes to this option, support will be included for the
@@ -78,7 +79,6 @@ config SPI_AU1550
 
 config SPI_BITBANG
 	tristate "Bitbanging SPI master"
-	depends on SPI_MASTER && EXPERIMENTAL
 	help
 	  With a few GPIO pins, your system can bitbang the SPI protocol.
 	  Select this to get SPI support through I/O pins (GPIO, parallel
@@ -92,7 +92,7 @@ config SPI_BITBANG
 
 config SPI_BUTTERFLY
 	tristate "Parallel port adapter for AVR Butterfly (DEVELOPMENT)"
-	depends on SPI_MASTER && PARPORT && EXPERIMENTAL
+	depends on PARPORT
 	select SPI_BITBANG
 	help
 	  This uses a custom parallel port cable to connect to an AVR
@@ -102,14 +102,14 @@ config SPI_BUTTERFLY
 
 config SPI_IMX
 	tristate "Freescale iMX SPI controller"
-	depends on SPI_MASTER && ARCH_IMX && EXPERIMENTAL
+	depends on ARCH_IMX && EXPERIMENTAL
 	help
 	  This enables using the Freescale iMX SPI controller in master
 	  mode.
 
 config SPI_LM70_LLP
 	tristate "Parallel port adapter for LM70 eval board (DEVELOPMENT)"
-	depends on SPI_MASTER && PARPORT && EXPERIMENTAL
+	depends on PARPORT && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  This driver supports the NS LM70 LLP Evaluation Board,
@@ -118,14 +118,14 @@ config SPI_LM70_LLP
 
 config SPI_MPC52xx_PSC
 	tristate "Freescale MPC52xx PSC SPI controller"
-	depends on SPI_MASTER && PPC_MPC52xx && EXPERIMENTAL
+	depends on PPC_MPC52xx && EXPERIMENTAL
 	help
 	  This enables using the Freescale MPC52xx Programmable Serial
 	  Controller in master SPI mode.
 
 config SPI_MPC83xx
 	tristate "Freescale MPC83xx/QUICC Engine SPI controller"
-	depends on SPI_MASTER && (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL
+	depends on (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL
 	help
 	  This enables using the Freescale MPC83xx and QUICC Engine SPI
 	  controllers in master mode.
@@ -137,21 +137,27 @@ config SPI_MPC83xx
 
 config SPI_OMAP_UWIRE
 	tristate "OMAP1 MicroWire"
-	depends on SPI_MASTER && ARCH_OMAP1
+	depends on ARCH_OMAP1
 	select SPI_BITBANG
 	help
 	  This hooks up to the MicroWire controller on OMAP1 chips.
 
 config SPI_OMAP24XX
 	tristate "McSPI driver for OMAP24xx/OMAP34xx"
-	depends on SPI_MASTER && (ARCH_OMAP24XX || ARCH_OMAP34XX)
+	depends on ARCH_OMAP24XX || ARCH_OMAP34XX
 	help
 	  SPI master controller for OMAP24xx/OMAP34xx Multichannel SPI
 	  (McSPI) modules.
 
+config SPI_ORION
+	tristate "Orion SPI master (EXPERIMENTAL)"
+	depends on PLAT_ORION && EXPERIMENTAL
+	help
+	  This enables using the SPI master controller on the Orion chips.
+
 config SPI_PXA2XX
 	tristate "PXA2xx SSP SPI master"
-	depends on SPI_MASTER && ARCH_PXA && EXPERIMENTAL
+	depends on ARCH_PXA && EXPERIMENTAL
 	select PXA_SSP
 	help
 	  This enables using a PXA2xx SSP port as a SPI master controller.
@@ -160,14 +166,14 @@ config SPI_PXA2XX
 
 config SPI_S3C24XX
 	tristate "Samsung S3C24XX series SPI"
-	depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL
+	depends on ARCH_S3C2410 && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  SPI driver for Samsung S3C24XX series ARM SoCs
 
 config SPI_S3C24XX_GPIO
 	tristate "Samsung S3C24XX series SPI by GPIO"
-	depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL
+	depends on ARCH_S3C2410 && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  SPI driver for Samsung S3C24XX series ARM SoCs using
@@ -177,20 +183,20 @@ config SPI_S3C24XX_GPIO
 
 config SPI_SH_SCI
 	tristate "SuperH SCI SPI controller"
-	depends on SPI_MASTER && SUPERH
+	depends on SUPERH
 	select SPI_BITBANG
 	help
 	  SPI driver for SuperH SCI blocks.
 
 config SPI_TXX9
 	tristate "Toshiba TXx9 SPI controller"
-	depends on SPI_MASTER && GENERIC_GPIO && CPU_TX49XX
+	depends on GENERIC_GPIO && CPU_TX49XX
 	help
 	  SPI driver for Toshiba TXx9 MIPS SoCs
 
 config SPI_XILINX
 	tristate "Xilinx SPI controller"
-	depends on SPI_MASTER && XILINX_VIRTEX && EXPERIMENTAL
+	depends on XILINX_VIRTEX && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  This exposes the SPI controller IP from the Xilinx EDK.
@@ -207,11 +213,10 @@ config SPI_XILINX
 # being probably the most widely used ones.
 #
 comment "SPI Protocol Masters"
-	depends on SPI_MASTER
 
 config SPI_AT25
 	tristate "SPI EEPROMs from most vendors"
-	depends on SPI_MASTER && SYSFS
+	depends on SYSFS
 	help
 	  Enable this driver to get read/write support to most SPI EEPROMs,
 	  after you configure the board init code to know about each eeprom
@@ -222,7 +227,7 @@ config SPI_AT25
 
 config SPI_SPIDEV
 	tristate "User mode SPI device driver support"
-	depends on SPI_MASTER && EXPERIMENTAL
+	depends on EXPERIMENTAL
 	help
 	  This supports user mode SPI protocol drivers.
 
@@ -231,7 +236,7 @@ config SPI_SPIDEV
 
 config SPI_TLE62X0
 	tristate "Infineon TLE62X0 (for power switching)"
-	depends on SPI_MASTER && SYSFS
+	depends on SYSFS
 	help
 	  SPI driver for Infineon TLE62X0 series line driver chips,
 	  such as the TLE6220, TLE6230 and TLE6240.  This provides a
@@ -242,6 +247,8 @@ config SPI_TLE62X0
 # Add new SPI protocol masters in alphabetical order above this line
 #
 
+endif # SPI_MASTER
+
 # (slave support would go here)
 
 endif # SPI
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 7fca043ce723..ccf18de34e1e 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -21,6 +21,7 @@ obj-$(CONFIG_SPI_LM70_LLP)		+= spi_lm70llp.o
 obj-$(CONFIG_SPI_PXA2XX)		+= pxa2xx_spi.o
 obj-$(CONFIG_SPI_OMAP_UWIRE)		+= omap_uwire.o
 obj-$(CONFIG_SPI_OMAP24XX)		+= omap2_mcspi.o
+obj-$(CONFIG_SPI_ORION)			+= orion_spi.o
 obj-$(CONFIG_SPI_MPC52xx_PSC)		+= mpc52xx_psc_spi.o
 obj-$(CONFIG_SPI_MPC83xx)		+= spi_mpc83xx.o
 obj-$(CONFIG_SPI_S3C24XX_GPIO)		+= spi_s3c24xx_gpio.o
diff --git a/drivers/spi/atmel_spi.c b/drivers/spi/atmel_spi.c
index e81d59d78910..02f9320f3efc 100644
--- a/drivers/spi/atmel_spi.c
+++ b/drivers/spi/atmel_spi.c
@@ -20,9 +20,9 @@
 #include <linux/spi/spi.h>
 
 #include <asm/io.h>
-#include <asm/arch/board.h>
-#include <asm/arch/gpio.h>
-#include <asm/arch/cpu.h>
+#include <mach/board.h>
+#include <mach/gpio.h>
+#include <mach/cpu.h>
 
 #include "atmel_spi.h"
 
@@ -184,7 +184,8 @@ static void atmel_spi_next_xfer(struct spi_master *master,
 {
 	struct atmel_spi	*as = spi_master_get_devdata(master);
 	struct spi_transfer	*xfer;
-	u32			len, remaining, total;
+	u32			len, remaining;
+	u32			ieval;
 	dma_addr_t		tx_dma, rx_dma;
 
 	if (!as->current_transfer)
@@ -197,6 +198,8 @@ static void atmel_spi_next_xfer(struct spi_master *master,
 		xfer = NULL;
 
 	if (xfer) {
+		spi_writel(as, PTCR, SPI_BIT(RXTDIS) | SPI_BIT(TXTDIS));
+
 		len = xfer->len;
 		atmel_spi_next_xfer_data(master, xfer, &tx_dma, &rx_dma, &len);
 		remaining = xfer->len - len;
@@ -234,6 +237,8 @@ static void atmel_spi_next_xfer(struct spi_master *master,
 	as->next_transfer = xfer;
 
 	if (xfer) {
+		u32	total;
+
 		total = len;
 		atmel_spi_next_xfer_data(master, xfer, &tx_dma, &rx_dma, &len);
 		as->next_remaining_bytes = total - len;
@@ -250,9 +255,11 @@ static void atmel_spi_next_xfer(struct spi_master *master,
 			"  next xfer %p: len %u tx %p/%08x rx %p/%08x\n",
 			xfer, xfer->len, xfer->tx_buf, xfer->tx_dma,
 			xfer->rx_buf, xfer->rx_dma);
+		ieval = SPI_BIT(ENDRX) | SPI_BIT(OVRES);
 	} else {
 		spi_writel(as, RNCR, 0);
 		spi_writel(as, TNCR, 0);
+		ieval = SPI_BIT(RXBUFF) | SPI_BIT(ENDRX) | SPI_BIT(OVRES);
 	}
 
 	/* REVISIT: We're waiting for ENDRX before we start the next
@@ -265,7 +272,7 @@ static void atmel_spi_next_xfer(struct spi_master *master,
 	 *
 	 * It should be doable, though. Just not now...
 	 */
-	spi_writel(as, IER, SPI_BIT(ENDRX) | SPI_BIT(OVRES));
+	spi_writel(as, IER, ieval);
 	spi_writel(as, PTCR, SPI_BIT(TXTEN) | SPI_BIT(RXTEN));
 }
 
@@ -313,14 +320,14 @@ atmel_spi_dma_map_xfer(struct atmel_spi *as, struct spi_transfer *xfer)
 		xfer->tx_dma = dma_map_single(dev,
 				(void *) xfer->tx_buf, xfer->len,
 				DMA_TO_DEVICE);
-		if (dma_mapping_error(xfer->tx_dma))
+		if (dma_mapping_error(dev, xfer->tx_dma))
 			return -ENOMEM;
 	}
 	if (xfer->rx_buf) {
 		xfer->rx_dma = dma_map_single(dev,
 				xfer->rx_buf, xfer->len,
 				DMA_FROM_DEVICE);
-		if (dma_mapping_error(xfer->rx_dma)) {
+		if (dma_mapping_error(dev, xfer->rx_dma)) {
 			if (xfer->tx_buf)
 				dma_unmap_single(dev,
 						xfer->tx_dma, xfer->len,
@@ -396,7 +403,7 @@ atmel_spi_interrupt(int irq, void *dev_id)
 
 		ret = IRQ_HANDLED;
 
-		spi_writel(as, IDR, (SPI_BIT(ENDTX) | SPI_BIT(ENDRX)
+		spi_writel(as, IDR, (SPI_BIT(RXBUFF) | SPI_BIT(ENDRX)
 				     | SPI_BIT(OVRES)));
 
 		/*
@@ -418,7 +425,7 @@ atmel_spi_interrupt(int irq, void *dev_id)
 		if (xfer->delay_usecs)
 			udelay(xfer->delay_usecs);
 
-		dev_warn(master->dev.parent, "fifo overrun (%u/%u remaining)\n",
+		dev_warn(master->dev.parent, "overrun (%u/%u remaining)\n",
 			 spi_readl(as, TCR), spi_readl(as, RCR));
 
 		/*
@@ -442,7 +449,7 @@ atmel_spi_interrupt(int irq, void *dev_id)
 		spi_readl(as, SR);
 
 		atmel_spi_msg_done(master, as, msg, -EIO, 0);
-	} else if (pending & SPI_BIT(ENDRX)) {
+	} else if (pending & (SPI_BIT(RXBUFF) | SPI_BIT(ENDRX))) {
 		ret = IRQ_HANDLED;
 
 		spi_writel(as, IDR, pending);
diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c
index 072c4a595334..87b73e0169c5 100644
--- a/drivers/spi/au1550_spi.c
+++ b/drivers/spi/au1550_spi.c
@@ -26,6 +26,7 @@
 #include <linux/errno.h>
 #include <linux/device.h>
 #include <linux/platform_device.h>
+#include <linux/resource.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_bitbang.h>
 #include <linux/dma-mapping.h>
@@ -81,6 +82,7 @@ struct au1550_spi {
 	struct spi_master *master;
 	struct device *dev;
 	struct au1550_spi_info *pdata;
+	struct resource *ioarea;
 };
 
 
@@ -96,6 +98,8 @@ static dbdev_tab_t au1550_spi_mem_dbdev =
 	.dev_intpolarity	= 0
 };
 
+static int ddma_memid;	/* id to above mem dma device */
+
 static void au1550_spi_bits_handlers_set(struct au1550_spi *hw, int bpw);
 
 
@@ -330,7 +334,7 @@ static int au1550_spi_dma_rxtmp_alloc(struct au1550_spi *hw, unsigned size)
 	hw->dma_rx_tmpbuf_size = size;
 	hw->dma_rx_tmpbuf_addr = dma_map_single(hw->dev, hw->dma_rx_tmpbuf,
 			size, DMA_FROM_DEVICE);
-	if (dma_mapping_error(hw->dma_rx_tmpbuf_addr)) {
+	if (dma_mapping_error(hw->dev, hw->dma_rx_tmpbuf_addr)) {
 		kfree(hw->dma_rx_tmpbuf);
 		hw->dma_rx_tmpbuf = 0;
 		hw->dma_rx_tmpbuf_size = 0;
@@ -374,7 +378,7 @@ static int au1550_spi_dma_txrxb(struct spi_device *spi, struct spi_transfer *t)
 			dma_rx_addr = dma_map_single(hw->dev,
 					(void *)t->rx_buf,
 					t->len, DMA_FROM_DEVICE);
-			if (dma_mapping_error(dma_rx_addr))
+			if (dma_mapping_error(hw->dev, dma_rx_addr))
 				dev_err(hw->dev, "rx dma map error\n");
 		}
 	} else {
@@ -397,7 +401,7 @@ static int au1550_spi_dma_txrxb(struct spi_device *spi, struct spi_transfer *t)
 			dma_tx_addr = dma_map_single(hw->dev,
 					(void *)t->tx_buf,
 					t->len, DMA_TO_DEVICE);
-			if (dma_mapping_error(dma_tx_addr))
+			if (dma_mapping_error(hw->dev, dma_tx_addr))
 				dev_err(hw->dev, "tx dma map error\n");
 		}
 	} else {
@@ -480,9 +484,13 @@ static irqreturn_t au1550_spi_dma_irq_callback(struct au1550_spi *hw)
 		au1xxx_dbdma_reset(hw->dma_tx_ch);
 		au1550_spi_reset_fifos(hw);
 
-		dev_err(hw->dev,
-			"Unexpected SPI error: event=0x%x stat=0x%x!\n",
-			evnt, stat);
+		if (evnt == PSC_SPIEVNT_RO)
+			dev_err(hw->dev,
+				"dma transfer: receive FIFO overflow!\n");
+		else
+			dev_err(hw->dev,
+				"dma transfer: unexpected SPI error "
+				"(event=0x%x stat=0x%x)!\n", evnt, stat);
 
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
@@ -592,17 +600,17 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 
 	if ((evnt & (PSC_SPIEVNT_MM | PSC_SPIEVNT_RO
 				| PSC_SPIEVNT_RU | PSC_SPIEVNT_TO
-				| PSC_SPIEVNT_TU | PSC_SPIEVNT_SD))
+				| PSC_SPIEVNT_SD))
 			!= 0) {
-		dev_err(hw->dev,
-			"Unexpected SPI error: event=0x%x stat=0x%x!\n",
-			evnt, stat);
 		/*
 		 * due to an error we consider transfer as done,
 		 * so mask all events until before next transfer start
 		 */
 		au1550_spi_mask_ack_all(hw);
 		au1550_spi_reset_fifos(hw);
+		dev_err(hw->dev,
+			"pio transfer: unexpected SPI error "
+			"(event=0x%x stat=0x%x)!\n", evnt, stat);
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
 	}
@@ -616,27 +624,50 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 		stat = hw->regs->psc_spistat;
 		au_sync();
 
-		if ((stat & PSC_SPISTAT_RE) == 0 && hw->rx_count < hw->len) {
+		/*
+		 * Take care to not let the Rx FIFO overflow.
+		 *
+		 * We only write a byte if we have read one at least. Initially,
+		 * the write fifo is full, so we should read from the read fifo
+		 * first.
+		 * In case we miss a word from the read fifo, we should get a
+		 * RO event and should back out.
+		 */
+		if (!(stat & PSC_SPISTAT_RE) && hw->rx_count < hw->len) {
 			hw->rx_word(hw);
-			/* ack the receive request event */
-			hw->regs->psc_spievent = PSC_SPIEVNT_RR;
-			au_sync();
 			busy = 1;
-		}
 
-		if ((stat & PSC_SPISTAT_TF) == 0 && hw->tx_count < hw->len) {
-			hw->tx_word(hw);
-			/* ack the transmit request event */
-			hw->regs->psc_spievent = PSC_SPIEVNT_TR;
-			au_sync();
-			busy = 1;
+			if (!(stat & PSC_SPISTAT_TF) && hw->tx_count < hw->len)
+				hw->tx_word(hw);
 		}
 	} while (busy);
 
-	evnt = hw->regs->psc_spievent;
+	hw->regs->psc_spievent = PSC_SPIEVNT_RR | PSC_SPIEVNT_TR;
 	au_sync();
 
-	if (hw->rx_count >= hw->len || (evnt & PSC_SPIEVNT_MD) != 0) {
+	/*
+	 * Restart the SPI transmission in case of a transmit underflow.
+	 * This seems to work despite the notes in the Au1550 data book
+	 * of Figure 8-4 with flowchart for SPI master operation:
+	 *
+	 * """Note 1: An XFR Error Interrupt occurs, unless masked,
+	 * for any of the following events: Tx FIFO Underflow,
+	 * Rx FIFO Overflow, or Multiple-master Error
+	 *    Note 2: In case of a Tx Underflow Error, all zeroes are
+	 * transmitted."""
+	 *
+	 * By simply restarting the spi transfer on Tx Underflow Error,
+	 * we assume that spi transfer was paused instead of zeroes
+	 * transmittion mentioned in the Note 2 of Au1550 data book.
+	 */
+	if (evnt & PSC_SPIEVNT_TU) {
+		hw->regs->psc_spievent = PSC_SPIEVNT_TU | PSC_SPIEVNT_MD;
+		au_sync();
+		hw->regs->psc_spipcr = PSC_SPIPCR_MS;
+		au_sync();
+	}
+
+	if (hw->rx_count >= hw->len) {
 		/* transfer completed successfully */
 		au1550_spi_mask_ack_all(hw);
 		complete(&hw->master_done);
@@ -725,6 +756,8 @@ static void __init au1550_spi_setup_psc_as_spi(struct au1550_spi *hw)
 		stat = hw->regs->psc_spistat;
 		au_sync();
 	} while ((stat & PSC_SPISTAT_DR) == 0);
+
+	au1550_spi_reset_fifos(hw);
 }
 
 
@@ -732,6 +765,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 {
 	struct au1550_spi *hw;
 	struct spi_master *master;
+	struct resource *r;
 	int err = 0;
 
 	master = spi_alloc_master(&pdev->dev, sizeof(struct au1550_spi));
@@ -753,76 +787,64 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 		goto err_no_pdata;
 	}
 
-	platform_set_drvdata(pdev, hw);
-
-	init_completion(&hw->master_done);
-
-	hw->bitbang.master = hw->master;
-	hw->bitbang.setup_transfer = au1550_spi_setupxfer;
-	hw->bitbang.chipselect = au1550_spi_chipsel;
-	hw->bitbang.master->setup = au1550_spi_setup;
-	hw->bitbang.txrx_bufs = au1550_spi_txrx_bufs;
+	r = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!r) {
+		dev_err(&pdev->dev, "no IRQ\n");
+		err = -ENODEV;
+		goto err_no_iores;
+	}
+	hw->irq = r->start;
+
+	hw->usedma = 0;
+	r = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+	if (r) {
+		hw->dma_tx_id = r->start;
+		r = platform_get_resource(pdev, IORESOURCE_DMA, 1);
+		if (r) {
+			hw->dma_rx_id = r->start;
+			if (usedma && ddma_memid) {
+				if (pdev->dev.dma_mask == NULL)
+					dev_warn(&pdev->dev, "no dma mask\n");
+				else
+					hw->usedma = 1;
+			}
+		}
+	}
 
-	switch (hw->pdata->bus_num) {
-	case 0:
-		hw->irq = AU1550_PSC0_INT;
-		hw->regs = (volatile psc_spi_t *)PSC0_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC0_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC0_TX;
-		break;
-	case 1:
-		hw->irq = AU1550_PSC1_INT;
-		hw->regs = (volatile psc_spi_t *)PSC1_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC1_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC1_TX;
-		break;
-	case 2:
-		hw->irq = AU1550_PSC2_INT;
-		hw->regs = (volatile psc_spi_t *)PSC2_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC2_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC2_TX;
-		break;
-	case 3:
-		hw->irq = AU1550_PSC3_INT;
-		hw->regs = (volatile psc_spi_t *)PSC3_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC3_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC3_TX;
-		break;
-	default:
-		dev_err(&pdev->dev, "Wrong bus_num of SPI\n");
-		err = -ENOENT;
-		goto err_no_pdata;
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!r) {
+		dev_err(&pdev->dev, "no mmio resource\n");
+		err = -ENODEV;
+		goto err_no_iores;
 	}
 
-	if (request_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t),
-			pdev->name) == NULL) {
+	hw->ioarea = request_mem_region(r->start, sizeof(psc_spi_t),
+					pdev->name);
+	if (!hw->ioarea) {
 		dev_err(&pdev->dev, "Cannot reserve iomem region\n");
 		err = -ENXIO;
 		goto err_no_iores;
 	}
 
-
-	if (usedma) {
-		if (pdev->dev.dma_mask == NULL)
-			dev_warn(&pdev->dev, "no dma mask\n");
-		else
-			hw->usedma = 1;
+	hw->regs = (psc_spi_t __iomem *)ioremap(r->start, sizeof(psc_spi_t));
+	if (!hw->regs) {
+		dev_err(&pdev->dev, "cannot ioremap\n");
+		err = -ENXIO;
+		goto err_ioremap;
 	}
 
-	if (hw->usedma) {
-		/*
-		 * create memory device with 8 bits dev_devwidth
-		 * needed for proper byte ordering to spi fifo
-		 */
-		int memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev);
-		if (!memid) {
-			dev_err(&pdev->dev,
-				"Cannot create dma 8 bit mem device\n");
-			err = -ENXIO;
-			goto err_dma_add_dev;
-		}
+	platform_set_drvdata(pdev, hw);
+
+	init_completion(&hw->master_done);
+
+	hw->bitbang.master = hw->master;
+	hw->bitbang.setup_transfer = au1550_spi_setupxfer;
+	hw->bitbang.chipselect = au1550_spi_chipsel;
+	hw->bitbang.master->setup = au1550_spi_setup;
+	hw->bitbang.txrx_bufs = au1550_spi_txrx_bufs;
 
-		hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(memid,
+	if (hw->usedma) {
+		hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(ddma_memid,
 			hw->dma_tx_id, NULL, (void *)hw);
 		if (hw->dma_tx_ch == 0) {
 			dev_err(&pdev->dev,
@@ -841,7 +863,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 
 
 		hw->dma_rx_ch = au1xxx_dbdma_chan_alloc(hw->dma_rx_id,
-			memid, NULL, (void *)hw);
+			ddma_memid, NULL, (void *)hw);
 		if (hw->dma_rx_ch == 0) {
 			dev_err(&pdev->dev,
 				"Cannot allocate rx dma channel\n");
@@ -874,7 +896,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 		goto err_no_irq;
 	}
 
-	master->bus_num = hw->pdata->bus_num;
+	master->bus_num = pdev->id;
 	master->num_chipselect = hw->pdata->num_chipselect;
 
 	/*
@@ -924,8 +946,11 @@ err_no_txdma_descr:
 		au1xxx_dbdma_chan_free(hw->dma_tx_ch);
 
 err_no_txdma:
-err_dma_add_dev:
-	release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t));
+	iounmap((void __iomem *)hw->regs);
+
+err_ioremap:
+	release_resource(hw->ioarea);
+	kfree(hw->ioarea);
 
 err_no_iores:
 err_no_pdata:
@@ -944,7 +969,9 @@ static int __exit au1550_spi_remove(struct platform_device *pdev)
 
 	spi_bitbang_stop(&hw->bitbang);
 	free_irq(hw->irq, hw);
-	release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t));
+	iounmap((void __iomem *)hw->regs);
+	release_resource(hw->ioarea);
+	kfree(hw->ioarea);
 
 	if (hw->usedma) {
 		au1550_spi_dma_rxtmp_free(hw);
@@ -971,12 +998,24 @@ static struct platform_driver au1550_spi_drv = {
 
 static int __init au1550_spi_init(void)
 {
+	/*
+	 * create memory device with 8 bits dev_devwidth
+	 * needed for proper byte ordering to spi fifo
+	 */
+	if (usedma) {
+		ddma_memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev);
+		if (!ddma_memid)
+			printk(KERN_ERR "au1550-spi: cannot add memory"
+					"dbdma device\n");
+	}
 	return platform_driver_probe(&au1550_spi_drv, au1550_spi_probe);
 }
 module_init(au1550_spi_init);
 
 static void __exit au1550_spi_exit(void)
 {
+	if (usedma && ddma_memid)
+		au1xxx_ddma_del_device(ddma_memid);
 	platform_driver_unregister(&au1550_spi_drv);
 }
 module_exit(au1550_spi_exit);
diff --git a/drivers/spi/mpc52xx_psc_spi.c b/drivers/spi/mpc52xx_psc_spi.c
index 681d62325d3d..25eda71f4bf4 100644
--- a/drivers/spi/mpc52xx_psc_spi.c
+++ b/drivers/spi/mpc52xx_psc_spi.c
@@ -17,7 +17,7 @@
 #include <linux/interrupt.h>
 
 #if defined(CONFIG_PPC_MERGE)
-#include <asm/of_platform.h>
+#include <linux/of_platform.h>
 #else
 #include <linux/platform_device.h>
 #endif
@@ -148,7 +148,6 @@ static int mpc52xx_psc_spi_transfer_rxtx(struct spi_device *spi,
 	unsigned rfalarm;
 	unsigned send_at_once = MPC52xx_PSC_BUFSIZE;
 	unsigned recv_at_once;
-	unsigned bpw = mps->bits_per_word / 8;
 
 	if (!t->tx_buf && !t->rx_buf && t->len)
 		return -EINVAL;
@@ -164,22 +163,15 @@ static int mpc52xx_psc_spi_transfer_rxtx(struct spi_device *spi,
 		}
 
 		dev_dbg(&spi->dev, "send %d bytes...\n", send_at_once);
-		if (tx_buf) {
-			for (; send_at_once; sb++, send_at_once--) {
-				/* set EOF flag */
-				if (mps->bits_per_word
-						&& (sb + 1) % bpw == 0)
-					out_8(&psc->ircr2, 0x01);
+		for (; send_at_once; sb++, send_at_once--) {
+			/* set EOF flag before the last word is sent */
+			if (send_at_once == 1)
+				out_8(&psc->ircr2, 0x01);
+
+			if (tx_buf)
 				out_8(&psc->mpc52xx_psc_buffer_8, tx_buf[sb]);
-			}
-		} else {
-			for (; send_at_once; sb++, send_at_once--) {
-				/* set EOF flag */
-				if (mps->bits_per_word
-						&& ((sb + 1) % bpw) == 0)
-					out_8(&psc->ircr2, 0x01);
+			else
 				out_8(&psc->mpc52xx_psc_buffer_8, 0);
-			}
 		}
 
 
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index b1cc148036c1..9d2186fd74aa 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -35,8 +35,8 @@
 
 #include <linux/spi/spi.h>
 
-#include <asm/arch/dma.h>
-#include <asm/arch/clock.h>
+#include <mach/dma.h>
+#include <mach/clock.h>
 
 
 #define OMAP2_MCSPI_MAX_FREQ		48000000
@@ -836,7 +836,7 @@ static int omap2_mcspi_transfer(struct spi_device *spi, struct spi_message *m)
 		if (tx_buf != NULL) {
 			t->tx_dma = dma_map_single(&spi->dev, (void *) tx_buf,
 					len, DMA_TO_DEVICE);
-			if (dma_mapping_error(t->tx_dma)) {
+			if (dma_mapping_error(&spi->dev, t->tx_dma)) {
 				dev_dbg(&spi->dev, "dma %cX %d bytes error\n",
 						'T', len);
 				return -EINVAL;
@@ -845,7 +845,7 @@ static int omap2_mcspi_transfer(struct spi_device *spi, struct spi_message *m)
 		if (rx_buf != NULL) {
 			t->rx_dma = dma_map_single(&spi->dev, rx_buf, t->len,
 					DMA_FROM_DEVICE);
-			if (dma_mapping_error(t->rx_dma)) {
+			if (dma_mapping_error(&spi->dev, t->rx_dma)) {
 				dev_dbg(&spi->dev, "dma %cX %d bytes error\n",
 						'R', len);
 				if (tx_buf != NULL)
diff --git a/drivers/spi/omap_uwire.c b/drivers/spi/omap_uwire.c
index d9ae111c27ae..5515eb97d7c5 100644
--- a/drivers/spi/omap_uwire.c
+++ b/drivers/spi/omap_uwire.c
@@ -47,12 +47,12 @@
 
 #include <asm/system.h>
 #include <asm/irq.h>
-#include <asm/hardware.h>
+#include <mach/hardware.h>
 #include <asm/io.h>
 #include <asm/mach-types.h>
 
-#include <asm/arch/mux.h>
-#include <asm/arch/omap730.h>	/* OMAP730_IO_CONF registers */
+#include <mach/mux.h>
+#include <mach/omap730.h>	/* OMAP730_IO_CONF registers */
 
 
 /* FIXME address is now a platform device resource,
diff --git a/drivers/spi/orion_spi.c b/drivers/spi/orion_spi.c
new file mode 100644
index 000000000000..b872bfaf4bd2
--- /dev/null
+++ b/drivers/spi/orion_spi.c
@@ -0,0 +1,574 @@
+/*
+ * orion_spi.c -- Marvell Orion SPI controller driver
+ *
+ * Author: Shadi Ammouri <shadi@marvell.com>
+ * Copyright (C) 2007-2008 Marvell Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/platform_device.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/orion_spi.h>
+#include <asm/unaligned.h>
+
+#define DRIVER_NAME			"orion_spi"
+
+#define ORION_NUM_CHIPSELECTS		1 /* only one slave is supported*/
+#define ORION_SPI_WAIT_RDY_MAX_LOOP	2000 /* in usec */
+
+#define ORION_SPI_IF_CTRL_REG		0x00
+#define ORION_SPI_IF_CONFIG_REG		0x04
+#define ORION_SPI_DATA_OUT_REG		0x08
+#define ORION_SPI_DATA_IN_REG		0x0c
+#define ORION_SPI_INT_CAUSE_REG		0x10
+
+#define ORION_SPI_IF_8_16_BIT_MODE	(1 << 5)
+#define ORION_SPI_CLK_PRESCALE_MASK	0x1F
+
+struct orion_spi {
+	struct work_struct	work;
+
+	/* Lock access to transfer list.	*/
+	spinlock_t		lock;
+
+	struct list_head	msg_queue;
+	struct spi_master	*master;
+	void __iomem		*base;
+	unsigned int		max_speed;
+	unsigned int		min_speed;
+	struct orion_spi_info	*spi_info;
+};
+
+static struct workqueue_struct *orion_spi_wq;
+
+static inline void __iomem *spi_reg(struct orion_spi *orion_spi, u32 reg)
+{
+	return orion_spi->base + reg;
+}
+
+static inline void
+orion_spi_setbits(struct orion_spi *orion_spi, u32 reg, u32 mask)
+{
+	void __iomem *reg_addr = spi_reg(orion_spi, reg);
+	u32 val;
+
+	val = readl(reg_addr);
+	val |= mask;
+	writel(val, reg_addr);
+}
+
+static inline void
+orion_spi_clrbits(struct orion_spi *orion_spi, u32 reg, u32 mask)
+{
+	void __iomem *reg_addr = spi_reg(orion_spi, reg);
+	u32 val;
+
+	val = readl(reg_addr);
+	val &= ~mask;
+	writel(val, reg_addr);
+}
+
+static int orion_spi_set_transfer_size(struct orion_spi *orion_spi, int size)
+{
+	if (size == 16) {
+		orion_spi_setbits(orion_spi, ORION_SPI_IF_CONFIG_REG,
+				  ORION_SPI_IF_8_16_BIT_MODE);
+	} else if (size == 8) {
+		orion_spi_clrbits(orion_spi, ORION_SPI_IF_CONFIG_REG,
+				  ORION_SPI_IF_8_16_BIT_MODE);
+	} else {
+		pr_debug("Bad bits per word value %d (only 8 or 16 are "
+			 "allowed).\n", size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int orion_spi_baudrate_set(struct spi_device *spi, unsigned int speed)
+{
+	u32 tclk_hz;
+	u32 rate;
+	u32 prescale;
+	u32 reg;
+	struct orion_spi *orion_spi;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+
+	tclk_hz = orion_spi->spi_info->tclk;
+
+	/*
+	 * the supported rates are: 4,6,8...30
+	 * round up as we look for equal or less speed
+	 */
+	rate = DIV_ROUND_UP(tclk_hz, speed);
+	rate = roundup(rate, 2);
+
+	/* check if requested speed is too small */
+	if (rate > 30)
+		return -EINVAL;
+
+	if (rate < 4)
+		rate = 4;
+
+	/* Convert the rate to SPI clock divisor value.	*/
+	prescale = 0x10 + rate/2;
+
+	reg = readl(spi_reg(orion_spi, ORION_SPI_IF_CONFIG_REG));
+	reg = ((reg & ~ORION_SPI_CLK_PRESCALE_MASK) | prescale);
+	writel(reg, spi_reg(orion_spi, ORION_SPI_IF_CONFIG_REG));
+
+	return 0;
+}
+
+/*
+ * called only when no transfer is active on the bus
+ */
+static int
+orion_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
+{
+	struct orion_spi *orion_spi;
+	unsigned int speed = spi->max_speed_hz;
+	unsigned int bits_per_word = spi->bits_per_word;
+	int	rc;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+
+	if ((t != NULL) && t->speed_hz)
+		speed = t->speed_hz;
+
+	if ((t != NULL) && t->bits_per_word)
+		bits_per_word = t->bits_per_word;
+
+	rc = orion_spi_baudrate_set(spi, speed);
+	if (rc)
+		return rc;
+
+	return orion_spi_set_transfer_size(orion_spi, bits_per_word);
+}
+
+static void orion_spi_set_cs(struct orion_spi *orion_spi, int enable)
+{
+	if (enable)
+		orion_spi_setbits(orion_spi, ORION_SPI_IF_CTRL_REG, 0x1);
+	else
+		orion_spi_clrbits(orion_spi, ORION_SPI_IF_CTRL_REG, 0x1);
+}
+
+static inline int orion_spi_wait_till_ready(struct orion_spi *orion_spi)
+{
+	int i;
+
+	for (i = 0; i < ORION_SPI_WAIT_RDY_MAX_LOOP; i++) {
+		if (readl(spi_reg(orion_spi, ORION_SPI_INT_CAUSE_REG)))
+			return 1;
+		else
+			udelay(1);
+	}
+
+	return -1;
+}
+
+static inline int
+orion_spi_write_read_8bit(struct spi_device *spi,
+			  const u8 **tx_buf, u8 **rx_buf)
+{
+	void __iomem *tx_reg, *rx_reg, *int_reg;
+	struct orion_spi *orion_spi;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+	tx_reg = spi_reg(orion_spi, ORION_SPI_DATA_OUT_REG);
+	rx_reg = spi_reg(orion_spi, ORION_SPI_DATA_IN_REG);
+	int_reg = spi_reg(orion_spi, ORION_SPI_INT_CAUSE_REG);
+
+	/* clear the interrupt cause register */
+	writel(0x0, int_reg);
+
+	if (tx_buf && *tx_buf)
+		writel(*(*tx_buf)++, tx_reg);
+	else
+		writel(0, tx_reg);
+
+	if (orion_spi_wait_till_ready(orion_spi) < 0) {
+		dev_err(&spi->dev, "TXS timed out\n");
+		return -1;
+	}
+
+	if (rx_buf && *rx_buf)
+		*(*rx_buf)++ = readl(rx_reg);
+
+	return 1;
+}
+
+static inline int
+orion_spi_write_read_16bit(struct spi_device *spi,
+			   const u16 **tx_buf, u16 **rx_buf)
+{
+	void __iomem *tx_reg, *rx_reg, *int_reg;
+	struct orion_spi *orion_spi;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+	tx_reg = spi_reg(orion_spi, ORION_SPI_DATA_OUT_REG);
+	rx_reg = spi_reg(orion_spi, ORION_SPI_DATA_IN_REG);
+	int_reg = spi_reg(orion_spi, ORION_SPI_INT_CAUSE_REG);
+
+	/* clear the interrupt cause register */
+	writel(0x0, int_reg);
+
+	if (tx_buf && *tx_buf)
+		writel(__cpu_to_le16(get_unaligned((*tx_buf)++)), tx_reg);
+	else
+		writel(0, tx_reg);
+
+	if (orion_spi_wait_till_ready(orion_spi) < 0) {
+		dev_err(&spi->dev, "TXS timed out\n");
+		return -1;
+	}
+
+	if (rx_buf && *rx_buf)
+		put_unaligned(__le16_to_cpu(readl(rx_reg)), (*rx_buf)++);
+
+	return 1;
+}
+
+static unsigned int
+orion_spi_write_read(struct spi_device *spi, struct spi_transfer *xfer)
+{
+	struct orion_spi *orion_spi;
+	unsigned int count;
+	int word_len;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+	word_len = spi->bits_per_word;
+	count = xfer->len;
+
+	if (word_len == 8) {
+		const u8 *tx = xfer->tx_buf;
+		u8 *rx = xfer->rx_buf;
+
+		do {
+			if (orion_spi_write_read_8bit(spi, &tx, &rx) < 0)
+				goto out;
+			count--;
+		} while (count);
+	} else if (word_len == 16) {
+		const u16 *tx = xfer->tx_buf;
+		u16 *rx = xfer->rx_buf;
+
+		do {
+			if (orion_spi_write_read_16bit(spi, &tx, &rx) < 0)
+				goto out;
+			count -= 2;
+		} while (count);
+	}
+
+out:
+	return xfer->len - count;
+}
+
+
+static void orion_spi_work(struct work_struct *work)
+{
+	struct orion_spi *orion_spi =
+		container_of(work, struct orion_spi, work);
+
+	spin_lock_irq(&orion_spi->lock);
+	while (!list_empty(&orion_spi->msg_queue)) {
+		struct spi_message *m;
+		struct spi_device *spi;
+		struct spi_transfer *t = NULL;
+		int par_override = 0;
+		int status = 0;
+		int cs_active = 0;
+
+		m = container_of(orion_spi->msg_queue.next, struct spi_message,
+				 queue);
+
+		list_del_init(&m->queue);
+		spin_unlock_irq(&orion_spi->lock);
+
+		spi = m->spi;
+
+		/* Load defaults */
+		status = orion_spi_setup_transfer(spi, NULL);
+
+		if (status < 0)
+			goto msg_done;
+
+		list_for_each_entry(t, &m->transfers, transfer_list) {
+			if (par_override || t->speed_hz || t->bits_per_word) {
+				par_override = 1;
+				status = orion_spi_setup_transfer(spi, t);
+				if (status < 0)
+					break;
+				if (!t->speed_hz && !t->bits_per_word)
+					par_override = 0;
+			}
+
+			if (!cs_active) {
+				orion_spi_set_cs(orion_spi, 1);
+				cs_active = 1;
+			}
+
+			if (t->len)
+				m->actual_length +=
+					orion_spi_write_read(spi, t);
+
+			if (t->delay_usecs)
+				udelay(t->delay_usecs);
+
+			if (t->cs_change) {
+				orion_spi_set_cs(orion_spi, 0);
+				cs_active = 0;
+			}
+		}
+
+msg_done:
+		if (cs_active)
+			orion_spi_set_cs(orion_spi, 0);
+
+		m->status = status;
+		m->complete(m->context);
+
+		spin_lock_irq(&orion_spi->lock);
+	}
+
+	spin_unlock_irq(&orion_spi->lock);
+}
+
+static int __init orion_spi_reset(struct orion_spi *orion_spi)
+{
+	/* Verify that the CS is deasserted */
+	orion_spi_set_cs(orion_spi, 0);
+
+	return 0;
+}
+
+static int orion_spi_setup(struct spi_device *spi)
+{
+	struct orion_spi *orion_spi;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+
+	if (spi->mode) {
+		dev_err(&spi->dev, "setup: unsupported mode bits %x\n",
+			spi->mode);
+		return -EINVAL;
+	}
+
+	if (spi->bits_per_word == 0)
+		spi->bits_per_word = 8;
+
+	if ((spi->max_speed_hz == 0)
+			|| (spi->max_speed_hz > orion_spi->max_speed))
+		spi->max_speed_hz = orion_spi->max_speed;
+
+	if (spi->max_speed_hz < orion_spi->min_speed) {
+		dev_err(&spi->dev, "setup: requested speed too low %d Hz\n",
+			spi->max_speed_hz);
+		return -EINVAL;
+	}
+
+	/*
+	 * baudrate & width will be set orion_spi_setup_transfer
+	 */
+	return 0;
+}
+
+static int orion_spi_transfer(struct spi_device *spi, struct spi_message *m)
+{
+	struct orion_spi *orion_spi;
+	struct spi_transfer *t = NULL;
+	unsigned long flags;
+
+	m->actual_length = 0;
+	m->status = 0;
+
+	/* reject invalid messages and transfers */
+	if (list_empty(&m->transfers) || !m->complete)
+		return -EINVAL;
+
+	orion_spi = spi_master_get_devdata(spi->master);
+
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		unsigned int bits_per_word = spi->bits_per_word;
+
+		if (t->tx_buf == NULL && t->rx_buf == NULL && t->len) {
+			dev_err(&spi->dev,
+				"message rejected : "
+				"invalid transfer data buffers\n");
+			goto msg_rejected;
+		}
+
+		if ((t != NULL) && t->bits_per_word)
+			bits_per_word = t->bits_per_word;
+
+		if ((bits_per_word != 8) && (bits_per_word != 16)) {
+			dev_err(&spi->dev,
+				"message rejected : "
+				"invalid transfer bits_per_word (%d bits)\n",
+				bits_per_word);
+			goto msg_rejected;
+		}
+		/*make sure buffer length is even when working in 16 bit mode*/
+		if ((t != NULL) && (t->bits_per_word == 16) && (t->len & 1)) {
+			dev_err(&spi->dev,
+				"message rejected : "
+				"odd data length (%d) while in 16 bit mode\n",
+				t->len);
+			goto msg_rejected;
+		}
+
+		if (t->speed_hz && t->speed_hz < orion_spi->min_speed) {
+			dev_err(&spi->dev,
+				"message rejected : "
+				"device min speed (%d Hz) exceeds "
+				"required transfer speed (%d Hz)\n",
+				orion_spi->min_speed, t->speed_hz);
+			goto msg_rejected;
+		}
+	}
+
+
+	spin_lock_irqsave(&orion_spi->lock, flags);
+	list_add_tail(&m->queue, &orion_spi->msg_queue);
+	queue_work(orion_spi_wq, &orion_spi->work);
+	spin_unlock_irqrestore(&orion_spi->lock, flags);
+
+	return 0;
+msg_rejected:
+	/* Message rejected and not queued */
+	m->status = -EINVAL;
+	if (m->complete)
+		m->complete(m->context);
+	return -EINVAL;
+}
+
+static int __init orion_spi_probe(struct platform_device *pdev)
+{
+	struct spi_master *master;
+	struct orion_spi *spi;
+	struct resource *r;
+	struct orion_spi_info *spi_info;
+	int status = 0;
+
+	spi_info = pdev->dev.platform_data;
+
+	master = spi_alloc_master(&pdev->dev, sizeof *spi);
+	if (master == NULL) {
+		dev_dbg(&pdev->dev, "master allocation failed\n");
+		return -ENOMEM;
+	}
+
+	if (pdev->id != -1)
+		master->bus_num = pdev->id;
+
+	master->setup = orion_spi_setup;
+	master->transfer = orion_spi_transfer;
+	master->num_chipselect = ORION_NUM_CHIPSELECTS;
+
+	dev_set_drvdata(&pdev->dev, master);
+
+	spi = spi_master_get_devdata(master);
+	spi->master = master;
+	spi->spi_info = spi_info;
+
+	spi->max_speed = DIV_ROUND_UP(spi_info->tclk, 4);
+	spi->min_speed = DIV_ROUND_UP(spi_info->tclk, 30);
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (r == NULL) {
+		status = -ENODEV;
+		goto out;
+	}
+
+	if (!request_mem_region(r->start, (r->end - r->start) + 1,
+				pdev->dev.bus_id)) {
+		status = -EBUSY;
+		goto out;
+	}
+	spi->base = ioremap(r->start, SZ_1K);
+
+	INIT_WORK(&spi->work, orion_spi_work);
+
+	spin_lock_init(&spi->lock);
+	INIT_LIST_HEAD(&spi->msg_queue);
+
+	if (orion_spi_reset(spi) < 0)
+		goto out_rel_mem;
+
+	status = spi_register_master(master);
+	if (status < 0)
+		goto out_rel_mem;
+
+	return status;
+
+out_rel_mem:
+	release_mem_region(r->start, (r->end - r->start) + 1);
+
+out:
+	spi_master_put(master);
+	return status;
+}
+
+
+static int __exit orion_spi_remove(struct platform_device *pdev)
+{
+	struct spi_master *master;
+	struct orion_spi *spi;
+	struct resource *r;
+
+	master = dev_get_drvdata(&pdev->dev);
+	spi = spi_master_get_devdata(master);
+
+	cancel_work_sync(&spi->work);
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(r->start, (r->end - r->start) + 1);
+
+	spi_unregister_master(master);
+
+	return 0;
+}
+
+MODULE_ALIAS("platform:" DRIVER_NAME);
+
+static struct platform_driver orion_spi_driver = {
+	.driver = {
+		.name	= DRIVER_NAME,
+		.owner	= THIS_MODULE,
+	},
+	.remove		= __exit_p(orion_spi_remove),
+};
+
+static int __init orion_spi_init(void)
+{
+	orion_spi_wq = create_singlethread_workqueue(
+				orion_spi_driver.driver.name);
+	if (orion_spi_wq == NULL)
+		return -ENOMEM;
+
+	return platform_driver_probe(&orion_spi_driver, orion_spi_probe);
+}
+module_init(orion_spi_init);
+
+static void __exit orion_spi_exit(void)
+{
+	flush_workqueue(orion_spi_wq);
+	platform_driver_unregister(&orion_spi_driver);
+
+	destroy_workqueue(orion_spi_wq);
+}
+module_exit(orion_spi_exit);
+
+MODULE_DESCRIPTION("Orion SPI driver");
+MODULE_AUTHOR("Shadi Ammouri <shadi@marvell.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index 0c452c46ab07..d47d3636227f 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -31,15 +31,14 @@
 
 #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/regs-ssp.h>
-#include <asm/arch/ssp.h>
-#include <asm/arch/pxa2xx_spi.h>
+#include <mach/hardware.h>
+#include <mach/pxa-regs.h>
+#include <mach/regs-ssp.h>
+#include <mach/ssp.h>
+#include <mach/pxa2xx_spi.h>
 
 MODULE_AUTHOR("Stephen Street");
 MODULE_DESCRIPTION("PXA2xx SSP SPI Controller");
@@ -48,9 +47,10 @@ MODULE_ALIAS("platform:pxa2xx-spi");
 
 #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 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 MAX_DMA_LEN		8191
 
 /*
  * for testing SSCR1 changes that require SSP restart, basically
@@ -145,7 +145,6 @@ struct driver_data {
 	size_t tx_map_len;
 	u8 n_bytes;
 	u32 dma_width;
-	int cs_change;
 	int (*write)(struct driver_data *drv_data);
 	int (*read)(struct driver_data *drv_data);
 	irqreturn_t (*transfer_handler)(struct driver_data *drv_data);
@@ -353,7 +352,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 	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))
+	if (dma_mapping_error(dev, drv_data->rx_dma))
 		return 0;
 
 	/* Stream map the tx buffer */
@@ -361,7 +360,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 						drv_data->tx_map_len,
 						DMA_TO_DEVICE);
 
-	if (dma_mapping_error(drv_data->tx_dma)) {
+	if (dma_mapping_error(dev, drv_data->tx_dma)) {
 		dma_unmap_single(dev, drv_data->rx_dma,
 					drv_data->rx_map_len, DMA_FROM_DEVICE);
 		return 0;
@@ -407,8 +406,45 @@ static void giveback(struct driver_data *drv_data)
 					struct spi_transfer,
 					transfer_list);
 
+	/* Delay if requested before any change in chip select */
+	if (last_transfer->delay_usecs)
+		udelay(last_transfer->delay_usecs);
+
+	/* Drop chip select UNLESS cs_change is true or we are returning
+	 * a message with an error, or next message is for another chip
+	 */
 	if (!last_transfer->cs_change)
 		drv_data->cs_control(PXA2XX_CS_DEASSERT);
+	else {
+		struct spi_message *next_msg;
+
+		/* Holding of cs was hinted, but we need to make sure
+		 * the next message is for the same chip.  Don't waste
+		 * time with the following tests unless this was hinted.
+		 *
+		 * We cannot postpone this until pump_messages, because
+		 * after calling msg->complete (below) the driver that
+		 * 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(&drv_data->lock, flags);
+		if (list_empty(&drv_data->queue))
+			next_msg = NULL;
+		else
+			next_msg = list_entry(drv_data->queue.next,
+					struct spi_message, queue);
+		spin_unlock_irqrestore(&drv_data->lock, flags);
+
+		/* see if the next and current messages point
+		 * to the same chip
+		 */
+		if (next_msg && next_msg->spi != msg->spi)
+			next_msg = NULL;
+		if (!next_msg || msg->state == ERROR_STATE)
+			drv_data->cs_control(PXA2XX_CS_DEASSERT);
+	}
 
 	msg->state = NULL;
 	if (msg->complete)
@@ -491,10 +527,9 @@ static void dma_transfer_complete(struct driver_data *drv_data)
 	msg->actual_length += drv_data->len -
 				(drv_data->rx_end - drv_data->rx);
 
-	/* Release chip select if requested, transfer delays are
-	 * handled in pump_transfers */
-	if (drv_data->cs_change)
-		drv_data->cs_control(PXA2XX_CS_DEASSERT);
+	/* Transfer delays and chip select release are
+	 * handled in pump_transfers or giveback
+	 */
 
 	/* Move to next transfer */
 	msg->state = next_transfer(drv_data);
@@ -603,10 +638,9 @@ static void int_transfer_complete(struct driver_data *drv_data)
 	drv_data->cur_msg->actual_length += drv_data->len -
 				(drv_data->rx_end - drv_data->rx);
 
-	/* Release chip select if requested, transfer delays are
-	 * handled in pump_transfers */
-	if (drv_data->cs_change)
-		drv_data->cs_control(PXA2XX_CS_DEASSERT);
+	/* Transfer delays and chip select release are
+	 * handled in pump_transfers or giveback
+	 */
 
 	/* Move to next transfer */
 	drv_data->cur_msg->state = next_transfer(drv_data);
@@ -841,23 +875,40 @@ static void pump_transfers(unsigned long data)
 		return;
 	}
 
-	/* Delay if requested at end of transfer*/
+	/* Delay if requested at end of transfer before CS change */
 	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);
+
+		/* Drop chip select only if cs_change is requested */
+		if (previous->cs_change)
+			drv_data->cs_control(PXA2XX_CS_DEASSERT);
 	}
 
-	/* Check transfer length */
-	if (transfer->len > 8191)
-	{
-		dev_warn(&drv_data->pdev->dev, "pump_transfers: transfer "
-				"length greater than 8191\n");
-		message->status = -EINVAL;
-		giveback(drv_data);
-		return;
+	/* Check for transfers that need multiple DMA segments */
+	if (transfer->len > MAX_DMA_LEN && chip->enable_dma) {
+
+		/* reject already-mapped transfers; PIO won't always work */
+		if (message->is_dma_mapped
+				|| transfer->rx_dma || transfer->tx_dma) {
+			dev_err(&drv_data->pdev->dev,
+				"pump_transfers: mapped transfer length "
+				"of %u is greater than %d\n",
+				transfer->len, MAX_DMA_LEN);
+			message->status = -EINVAL;
+			giveback(drv_data);
+			return;
+		}
+
+		/* warn ... we force this to PIO mode */
+		if (printk_ratelimit())
+			dev_warn(&message->spi->dev, "pump_transfers: "
+				"DMA disabled for transfer length %ld "
+				"greater than %d\n",
+				(long)drv_data->len, MAX_DMA_LEN);
 	}
 
 	/* Setup the transfer state based on the type of transfer */
@@ -879,7 +930,6 @@ static void pump_transfers(unsigned long data)
 	drv_data->len = transfer->len & DCMD_LENGTH;
 	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 */
 	cr0 = chip->cr0;
@@ -926,7 +976,7 @@ static void pump_transfers(unsigned long data)
 							&dma_thresh))
 				if (printk_ratelimit())
 					dev_warn(&message->spi->dev,
-						"pump_transfer: "
+						"pump_transfers: "
 						"DMA burst size reduced to "
 						"match bits_per_word\n");
 		}
@@ -940,8 +990,23 @@ static void pump_transfers(unsigned long data)
 
 	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))) {
+	/* Try to map dma buffer and do a dma transfer if successful, but
+	 * only if the length is non-zero and less than MAX_DMA_LEN.
+	 *
+	 * Zero-length non-descriptor DMA is illegal on PXA2xx; force use
+	 * of PIO instead.  Care is needed above because the transfer may
+	 * have have been passed with buffers that are already dma mapped.
+	 * A zero-length transfer in PIO mode will not try to write/read
+	 * to/from the buffers
+	 *
+	 * REVISIT large transfers are exactly where we most want to be
+	 * using DMA.  If this happens much, split those transfers into
+	 * multiple DMA segments rather than forcing PIO.
+	 */
+	drv_data->dma_mapped = 0;
+	if (drv_data->len > 0 && drv_data->len <= MAX_DMA_LEN)
+		drv_data->dma_mapped = map_dma_buffers(drv_data);
+	if (drv_data->dma_mapped) {
 
 		/* Ensure we have the correct interrupt handler */
 		drv_data->transfer_handler = dma_transfer;
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 1ad12afc6ba0..75e86865234c 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -178,6 +178,114 @@ struct boardinfo {
 static LIST_HEAD(board_list);
 static DEFINE_MUTEX(board_lock);
 
+/**
+ * spi_alloc_device - Allocate a new SPI device
+ * @master: Controller to which device is connected
+ * Context: can sleep
+ *
+ * Allows a driver to allocate and initialize a spi_device without
+ * registering it immediately.  This allows a driver to directly
+ * fill the spi_device with device parameters before calling
+ * spi_add_device() on it.
+ *
+ * Caller is responsible to call spi_add_device() on the returned
+ * spi_device structure to add it to the SPI master.  If the caller
+ * needs to discard the spi_device without adding it, then it should
+ * call spi_dev_put() on it.
+ *
+ * Returns a pointer to the new device, or NULL.
+ */
+struct spi_device *spi_alloc_device(struct spi_master *master)
+{
+	struct spi_device	*spi;
+	struct device		*dev = master->dev.parent;
+
+	if (!spi_master_get(master))
+		return NULL;
+
+	spi = kzalloc(sizeof *spi, GFP_KERNEL);
+	if (!spi) {
+		dev_err(dev, "cannot alloc spi_device\n");
+		spi_master_put(master);
+		return NULL;
+	}
+
+	spi->master = master;
+	spi->dev.parent = dev;
+	spi->dev.bus = &spi_bus_type;
+	spi->dev.release = spidev_release;
+	device_initialize(&spi->dev);
+	return spi;
+}
+EXPORT_SYMBOL_GPL(spi_alloc_device);
+
+/**
+ * spi_add_device - Add spi_device allocated with spi_alloc_device
+ * @spi: spi_device to register
+ *
+ * Companion function to spi_alloc_device.  Devices allocated with
+ * spi_alloc_device can be added onto the spi bus with this function.
+ *
+ * Returns 0 on success; negative errno on failure
+ */
+int spi_add_device(struct spi_device *spi)
+{
+	static DEFINE_MUTEX(spi_add_lock);
+	struct device *dev = spi->master->dev.parent;
+	int status;
+
+	/* Chipselects are numbered 0..max; validate. */
+	if (spi->chip_select >= spi->master->num_chipselect) {
+		dev_err(dev, "cs%d >= max %d\n",
+			spi->chip_select,
+			spi->master->num_chipselect);
+		return -EINVAL;
+	}
+
+	/* Set the bus ID string */
+	snprintf(spi->dev.bus_id, sizeof spi->dev.bus_id,
+			"%s.%u", spi->master->dev.bus_id,
+			spi->chip_select);
+
+
+	/* We need to make sure there's no other device with this
+	 * chipselect **BEFORE** we call setup(), else we'll trash
+	 * its configuration.  Lock against concurrent add() calls.
+	 */
+	mutex_lock(&spi_add_lock);
+
+	if (bus_find_device_by_name(&spi_bus_type, NULL, spi->dev.bus_id)
+			!= NULL) {
+		dev_err(dev, "chipselect %d already in use\n",
+				spi->chip_select);
+		status = -EBUSY;
+		goto done;
+	}
+
+	/* Drivers may modify this initial i/o setup, but will
+	 * normally rely on the device being setup.  Devices
+	 * using SPI_CS_HIGH can't coexist well otherwise...
+	 */
+	status = spi->master->setup(spi);
+	if (status < 0) {
+		dev_err(dev, "can't %s %s, status %d\n",
+				"setup", spi->dev.bus_id, status);
+		goto done;
+	}
+
+	/* Device may be bound to an active driver when this returns */
+	status = device_add(&spi->dev);
+	if (status < 0)
+		dev_err(dev, "can't %s %s, status %d\n",
+				"add", spi->dev.bus_id, status);
+	else
+		dev_dbg(dev, "registered child %s\n", spi->dev.bus_id);
+
+done:
+	mutex_unlock(&spi_add_lock);
+	return status;
+}
+EXPORT_SYMBOL_GPL(spi_add_device);
 
 /**
  * spi_new_device - instantiate one new SPI device
@@ -197,7 +305,6 @@ struct spi_device *spi_new_device(struct spi_master *master,
 				  struct spi_board_info *chip)
 {
 	struct spi_device	*proxy;
-	struct device		*dev = master->dev.parent;
 	int			status;
 
 	/* NOTE:  caller did any chip->bus_num checks necessary.
@@ -207,64 +314,28 @@ struct spi_device *spi_new_device(struct spi_master *master,
 	 * suggests syslogged diagnostics are best here (ugh).
 	 */
 
-	/* Chipselects are numbered 0..max; validate. */
-	if (chip->chip_select >= master->num_chipselect) {
-		dev_err(dev, "cs%d > max %d\n",
-			chip->chip_select,
-			master->num_chipselect);
+	proxy = spi_alloc_device(master);
+	if (!proxy)
 		return NULL;
-	}
 
-	if (!spi_master_get(master))
-		return NULL;
+	WARN_ON(strlen(chip->modalias) >= sizeof(proxy->modalias));
 
-	proxy = kzalloc(sizeof *proxy, GFP_KERNEL);
-	if (!proxy) {
-		dev_err(dev, "can't alloc dev for cs%d\n",
-			chip->chip_select);
-		goto fail;
-	}
-	proxy->master = master;
 	proxy->chip_select = chip->chip_select;
 	proxy->max_speed_hz = chip->max_speed_hz;
 	proxy->mode = chip->mode;
 	proxy->irq = chip->irq;
-	proxy->modalias = chip->modalias;
-
-	snprintf(proxy->dev.bus_id, sizeof proxy->dev.bus_id,
-			"%s.%u", master->dev.bus_id,
-			chip->chip_select);
-	proxy->dev.parent = dev;
-	proxy->dev.bus = &spi_bus_type;
+	strlcpy(proxy->modalias, chip->modalias, sizeof(proxy->modalias));
 	proxy->dev.platform_data = (void *) chip->platform_data;
 	proxy->controller_data = chip->controller_data;
 	proxy->controller_state = NULL;
-	proxy->dev.release = spidev_release;
 
-	/* drivers may modify this initial i/o setup */
-	status = master->setup(proxy);
+	status = spi_add_device(proxy);
 	if (status < 0) {
-		dev_err(dev, "can't %s %s, status %d\n",
-				"setup", proxy->dev.bus_id, status);
-		goto fail;
+		spi_dev_put(proxy);
+		return NULL;
 	}
 
-	/* driver core catches callers that misbehave by defining
-	 * devices that already exist.
-	 */
-	status = device_register(&proxy->dev);
-	if (status < 0) {
-		dev_err(dev, "can't %s %s, status %d\n",
-				"add", proxy->dev.bus_id, status);
-		goto fail;
-	}
-	dev_dbg(dev, "registered child %s\n", proxy->dev.bus_id);
 	return proxy;
-
-fail:
-	spi_master_put(master);
-	kfree(proxy);
-	return NULL;
 }
 EXPORT_SYMBOL_GPL(spi_new_device);
 
@@ -502,7 +573,7 @@ struct spi_master *spi_busnum_to_master(u16 bus_num)
 	struct device		*dev;
 	struct spi_master	*master = NULL;
 
-	dev = class_find_device(&spi_master_class, &bus_num,
+	dev = class_find_device(&spi_master_class, NULL, &bus_num,
 				__spi_master_match);
 	if (dev)
 		master = container_of(dev, struct spi_master, dev);
diff --git a/drivers/spi/spi_imx.c b/drivers/spi/spi_imx.c
index c730d05bfeb6..61ba147e384d 100644
--- a/drivers/spi/spi_imx.c
+++ b/drivers/spi/spi_imx.c
@@ -29,15 +29,15 @@
 #include <linux/spi/spi.h>
 #include <linux/workqueue.h>
 #include <linux/delay.h>
+#include <linux/clk.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/hardware.h>
 #include <asm/delay.h>
 
-#include <asm/arch/hardware.h>
-#include <asm/arch/imx-dma.h>
-#include <asm/arch/spi_imx.h>
+#include <mach/hardware.h>
+#include <mach/imx-dma.h>
+#include <mach/spi_imx.h>
 
 /*-------------------------------------------------------------------------*/
 /* SPI Registers offsets from peripheral base address */
@@ -250,6 +250,8 @@ struct driver_data {
 	int tx_dma_needs_unmap;
 	size_t tx_map_len;
 	u32 dummy_dma_buf ____cacheline_aligned;
+
+	struct clk *clk;
 };
 
 /* Runtime state */
@@ -488,7 +490,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 							buf,
 							drv_data->tx_map_len,
 							DMA_TO_DEVICE);
-			if (dma_mapping_error(drv_data->tx_dma))
+			if (dma_mapping_error(dev, drv_data->tx_dma))
 				return -1;
 
 			drv_data->tx_dma_needs_unmap = 1;
@@ -513,7 +515,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 					buf,
 					drv_data->len,
 					DMA_FROM_DEVICE);
-		if (dma_mapping_error(drv_data->rx_dma))
+		if (dma_mapping_error(dev, drv_data->rx_dma))
 			return -1;
 		drv_data->rx_dma_needs_unmap = 1;
 	}
@@ -531,7 +533,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 					buf,
 					drv_data->tx_map_len,
 					DMA_TO_DEVICE);
-	if (dma_mapping_error(drv_data->tx_dma)) {
+	if (dma_mapping_error(dev, drv_data->tx_dma)) {
 		if (drv_data->rx_dma) {
 			dma_unmap_single(dev,
 					drv_data->rx_dma,
@@ -855,15 +857,15 @@ static irqreturn_t spi_int(int irq, void *dev_id)
 	return drv_data->transfer_handler(drv_data);
 }
 
-static inline u32 spi_speed_hz(u32 data_rate)
+static inline u32 spi_speed_hz(struct driver_data *drv_data, u32 data_rate)
 {
-	return imx_get_perclk2() / (4 << ((data_rate) >> 13));
+	return clk_get_rate(drv_data->clk) / (4 << ((data_rate) >> 13));
 }
 
-static u32 spi_data_rate(u32 speed_hz)
+static u32 spi_data_rate(struct driver_data *drv_data, u32 speed_hz)
 {
 	u32 div;
-	u32 quantized_hz = imx_get_perclk2() >> 2;
+	u32 quantized_hz = clk_get_rate(drv_data->clk) >> 2;
 
 	for (div = SPI_PERCLK2_DIV_MIN;
 		div <= SPI_PERCLK2_DIV_MAX;
@@ -947,7 +949,7 @@ static void pump_transfers(unsigned long data)
 	tmp = transfer->speed_hz;
 	if (tmp == 0)
 		tmp = chip->max_speed_hz;
-	tmp = spi_data_rate(tmp);
+	tmp = spi_data_rate(drv_data, tmp);
 	u32_EDIT(control, SPI_CONTROL_DATARATE, tmp);
 
 	writel(control, regs + SPI_CONTROL);
@@ -1109,7 +1111,7 @@ static int transfer(struct spi_device *spi, struct spi_message *msg)
 	msg->actual_length = 0;
 
 	/* Per transfer setup check */
-	min_speed_hz = spi_speed_hz(SPI_CONTROL_DATARATE_MIN);
+	min_speed_hz = spi_speed_hz(drv_data, SPI_CONTROL_DATARATE_MIN);
 	max_speed_hz = spi->max_speed_hz;
 	list_for_each_entry(trans, &msg->transfers, transfer_list) {
 		tmp = trans->bits_per_word;
@@ -1176,6 +1178,7 @@ msg_rejected:
    applied and notified to the calling driver. */
 static int setup(struct spi_device *spi)
 {
+	struct driver_data *drv_data = spi_master_get_devdata(spi->master);
 	struct spi_imx_chip *chip_info;
 	struct chip_data *chip;
 	int first_setup = 0;
@@ -1304,14 +1307,14 @@ static int setup(struct spi_device *spi)
 	chip->n_bytes = (tmp <= 8) ? 1 : 2;
 
 	/* SPI datarate */
-	tmp = spi_data_rate(spi->max_speed_hz);
+	tmp = spi_data_rate(drv_data, spi->max_speed_hz);
 	if (tmp == SPI_CONTROL_DATARATE_BAD) {
 		status = -EINVAL;
 		dev_err(&spi->dev,
 			"setup - "
 			"HW min speed (%d Hz) exceeds required "
 			"max speed (%d Hz)\n",
-			spi_speed_hz(SPI_CONTROL_DATARATE_MIN),
+			spi_speed_hz(drv_data, SPI_CONTROL_DATARATE_MIN),
 			spi->max_speed_hz);
 		if (first_setup)
 			goto err_first_setup;
@@ -1321,7 +1324,7 @@ static int setup(struct spi_device *spi)
 	} else {
 		u32_EDIT(chip->control, SPI_CONTROL_DATARATE, tmp);
 		/* Actual rounded max_speed_hz */
-		tmp = spi_speed_hz(tmp);
+		tmp = spi_speed_hz(drv_data, tmp);
 		spi->max_speed_hz = tmp;
 		chip->max_speed_hz = tmp;
 	}
@@ -1352,7 +1355,7 @@ static int setup(struct spi_device *spi)
 		chip->period & SPI_PERIOD_WAIT,
 		spi->mode,
 		spi->bits_per_word,
-		spi_speed_hz(SPI_CONTROL_DATARATE_MIN),
+		spi_speed_hz(drv_data, SPI_CONTROL_DATARATE_MIN),
 		spi->max_speed_hz);
 	return status;
 
@@ -1465,6 +1468,14 @@ static int __init spi_imx_probe(struct platform_device *pdev)
 		goto err_no_pdata;
 	}
 
+	drv_data->clk = clk_get(&pdev->dev, "perclk2");
+	if (IS_ERR(drv_data->clk)) {
+		dev_err(&pdev->dev, "probe - cannot get get\n");
+		status = PTR_ERR(drv_data->clk);
+		goto err_no_clk;
+	}
+	clk_enable(drv_data->clk);
+
 	/* Allocate master with space for drv_data */
 	master = spi_alloc_master(dev, sizeof(struct driver_data));
 	if (!master) {
@@ -1526,24 +1537,24 @@ static int __init spi_imx_probe(struct platform_device *pdev)
 	drv_data->rx_channel = -1;
 	if (platform_info->enable_dma) {
 		/* Get rx DMA channel */
-		status = imx_dma_request_by_prio(&drv_data->rx_channel,
-			"spi_imx_rx", DMA_PRIO_HIGH);
-		if (status < 0) {
+		drv_data->rx_channel = imx_dma_request_by_prio("spi_imx_rx",
+							       DMA_PRIO_HIGH);
+		if (drv_data->rx_channel < 0) {
 			dev_err(dev,
 				"probe - problem (%d) requesting rx channel\n",
-				status);
+				drv_data->rx_channel);
 			goto err_no_rxdma;
 		} else
 			imx_dma_setup_handlers(drv_data->rx_channel, NULL,
 						dma_err_handler, drv_data);
 
 		/* Get tx DMA channel */
-		status = imx_dma_request_by_prio(&drv_data->tx_channel,
-						"spi_imx_tx", DMA_PRIO_MEDIUM);
-		if (status < 0) {
+		drv_data->tx_channel = imx_dma_request_by_prio("spi_imx_tx",
+							       DMA_PRIO_MEDIUM);
+		if (drv_data->tx_channel < 0) {
 			dev_err(dev,
 				"probe - problem (%d) requesting tx channel\n",
-				status);
+				drv_data->tx_channel);
 			imx_dma_free(drv_data->rx_channel);
 			goto err_no_txdma;
 		} else
@@ -1623,6 +1634,9 @@ err_no_iores:
 	spi_master_put(master);
 
 err_no_pdata:
+	clk_disable(drv_data->clk);
+	clk_put(drv_data->clk);
+err_no_clk:
 err_no_mem:
 	return status;
 }
@@ -1662,6 +1676,9 @@ static int __exit spi_imx_remove(struct platform_device *pdev)
 	if (irq >= 0)
 		free_irq(irq, drv_data);
 
+	clk_disable(drv_data->clk);
+	clk_put(drv_data->clk);
+
 	/* Release map resources */
 	iounmap(drv_data->regs);
 	release_resource(drv_data->ioarea);
diff --git a/drivers/spi/spi_mpc83xx.c b/drivers/spi/spi_mpc83xx.c
index 6832da6f7109..ac0e3e4b3c54 100644
--- a/drivers/spi/spi_mpc83xx.c
+++ b/drivers/spi/spi_mpc83xx.c
@@ -266,21 +266,21 @@ int mpc83xx_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
 
 	cs->hw_mode |= SPMODE_LEN(bits_per_word);
 
-	if ((mpc83xx_spi->spibrg / hz) >= 64) {
-		pm = mpc83xx_spi->spibrg / (hz * 64) - 1;
-		if (pm > 0x0f) {
+	if ((mpc83xx_spi->spibrg / hz) > 64) {
+		cs->hw_mode |= SPMODE_DIV16;
+		pm = mpc83xx_spi->spibrg / (hz * 64);
+		if (pm > 16) {
 			dev_err(&spi->dev, "Requested speed is too "
 				"low: %d Hz. Will use %d Hz instead.\n",
 				hz, mpc83xx_spi->spibrg / 1024);
-			pm = 0x0f;
+			pm = 16;
 		}
-		cs->hw_mode |= SPMODE_PM(pm) | SPMODE_DIV16;
-	} else {
+	} else
 		pm = mpc83xx_spi->spibrg / (hz * 4);
-		if (pm)
-			pm--;
-		cs->hw_mode |= SPMODE_PM(pm);
-	}
+	if (pm)
+		pm--;
+
+	cs->hw_mode |= SPMODE_PM(pm);
 	regval =  mpc83xx_spi_read_reg(&mpc83xx_spi->base->mode);
 	if (cs->hw_mode != regval) {
 		unsigned long flags;
@@ -312,11 +312,20 @@ static int mpc83xx_spi_bufs(struct spi_device *spi, struct spi_transfer *t)
 	if (t->bits_per_word)
 		bits_per_word = t->bits_per_word;
 	len = t->len;
-	if (bits_per_word > 8)
+	if (bits_per_word > 8) {
+		/* invalid length? */
+		if (len & 1)
+			return -EINVAL;
 		len /= 2;
-	if (bits_per_word > 16)
+	}
+	if (bits_per_word > 16) {
+		/* invalid length? */
+		if (len & 1)
+			return -EINVAL;
 		len /= 2;
+	}
 	mpc83xx_spi->count = len;
+
 	INIT_COMPLETION(mpc83xx_spi->done);
 
 	/* enable rx ints */
diff --git a/drivers/spi/spi_s3c24xx.c b/drivers/spi/spi_s3c24xx.c
index 0885cc357a37..3eb414b84a9d 100644
--- a/drivers/spi/spi_s3c24xx.c
+++ b/drivers/spi/spi_s3c24xx.c
@@ -25,11 +25,11 @@
 
 #include <asm/io.h>
 #include <asm/dma.h>
-#include <asm/hardware.h>
+#include <mach/hardware.h>
 
-#include <asm/arch/regs-gpio.h>
+#include <mach/regs-gpio.h>
 #include <asm/plat-s3c24xx/regs-spi.h>
-#include <asm/arch/spi.h>
+#include <mach/spi.h>
 
 struct s3c24xx_spi {
 	/* bitbang has to be first */
@@ -236,6 +236,19 @@ static irqreturn_t s3c24xx_spi_irq(int irq, void *dev)
 	return IRQ_HANDLED;
 }
 
+static void s3c24xx_spi_initialsetup(struct s3c24xx_spi *hw)
+{
+	/* for the moment, permanently enable the clock */
+
+	clk_enable(hw->clk);
+
+	/* program defaults into the registers */
+
+	writeb(0xff, hw->regs + S3C2410_SPPRE);
+	writeb(SPPIN_DEFAULT, hw->regs + S3C2410_SPPIN);
+	writeb(SPCON_DEFAULT, hw->regs + S3C2410_SPCON);
+}
+
 static int __init s3c24xx_spi_probe(struct platform_device *pdev)
 {
 	struct s3c2410_spi_info *pdata;
@@ -270,6 +283,7 @@ static int __init s3c24xx_spi_probe(struct platform_device *pdev)
 	/* setup the master state. */
 
 	master->num_chipselect = hw->pdata->num_cs;
+	master->bus_num = pdata->bus_num;
 
 	/* setup the state for the bitbang driver */
 
@@ -326,15 +340,7 @@ static int __init s3c24xx_spi_probe(struct platform_device *pdev)
 		goto err_no_clk;
 	}
 
-	/* for the moment, permanently enable the clock */
-
-	clk_enable(hw->clk);
-
-	/* program defaults into the registers */
-
-	writeb(0xff, hw->regs + S3C2410_SPPRE);
-	writeb(SPPIN_DEFAULT, hw->regs + S3C2410_SPPIN);
-	writeb(SPCON_DEFAULT, hw->regs + S3C2410_SPCON);
+	s3c24xx_spi_initialsetup(hw);
 
 	/* setup any gpio we can */
 
@@ -414,7 +420,7 @@ static int s3c24xx_spi_resume(struct platform_device *pdev)
 {
 	struct s3c24xx_spi *hw = platform_get_drvdata(pdev);
 
-	clk_enable(hw->clk);
+	s3c24xx_spi_initialsetup(hw);
 	return 0;
 }
 
@@ -424,7 +430,7 @@ static int s3c24xx_spi_resume(struct platform_device *pdev)
 #endif
 
 MODULE_ALIAS("platform:s3c2410-spi");
-static struct platform_driver s3c24xx_spidrv = {
+static struct platform_driver s3c24xx_spi_driver = {
 	.remove		= __exit_p(s3c24xx_spi_remove),
 	.suspend	= s3c24xx_spi_suspend,
 	.resume		= s3c24xx_spi_resume,
@@ -436,12 +442,12 @@ static struct platform_driver s3c24xx_spidrv = {
 
 static int __init s3c24xx_spi_init(void)
 {
-        return platform_driver_probe(&s3c24xx_spidrv, s3c24xx_spi_probe);
+        return platform_driver_probe(&s3c24xx_spi_driver, s3c24xx_spi_probe);
 }
 
 static void __exit s3c24xx_spi_exit(void)
 {
-        platform_driver_unregister(&s3c24xx_spidrv);
+        platform_driver_unregister(&s3c24xx_spi_driver);
 }
 
 module_init(s3c24xx_spi_init);
diff --git a/drivers/spi/spi_s3c24xx_gpio.c b/drivers/spi/spi_s3c24xx_gpio.c
index e33f6145c560..cc1f647f579b 100644
--- a/drivers/spi/spi_s3c24xx_gpio.c
+++ b/drivers/spi/spi_s3c24xx_gpio.c
@@ -21,9 +21,9 @@
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_bitbang.h>
 
-#include <asm/arch/regs-gpio.h>
-#include <asm/arch/spi-gpio.h>
-#include <asm/hardware.h>
+#include <mach/regs-gpio.h>
+#include <mach/spi-gpio.h>
+#include <mach/hardware.h>
 
 struct s3c2410_spigpio {
 	struct spi_bitbang		 bitbang;
diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index 799337f7fde1..e5e0cfed5e3b 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -30,6 +30,7 @@
 #include <linux/errno.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
+#include <linux/smp_lock.h>
 
 #include <linux/spi/spi.h>
 #include <linux/spi/spidev.h>
@@ -167,14 +168,14 @@ spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos)
 
 	mutex_lock(&spidev->buf_lock);
 	status = spidev_sync_read(spidev, count);
-	if (status == 0) {
+	if (status > 0) {
 		unsigned long	missing;
 
-		missing = copy_to_user(buf, spidev->buffer, count);
-		if (count && missing == count)
+		missing = copy_to_user(buf, spidev->buffer, status);
+		if (missing == status)
 			status = -EFAULT;
 		else
-			status = count - missing;
+			status = status - missing;
 	}
 	mutex_unlock(&spidev->buf_lock);
 
@@ -200,8 +201,6 @@ spidev_write(struct file *filp, const char __user *buf,
 	missing = copy_from_user(spidev->buffer, buf, count);
 	if (missing == 0) {
 		status = spidev_sync_write(spidev, count);
-		if (status == 0)
-			status = count;
 	} else
 		status = -EFAULT;
 	mutex_unlock(&spidev->buf_lock);
@@ -229,7 +228,6 @@ static int spidev_message(struct spidev_data *spidev,
 	 * We walk the array of user-provided transfers, using each one
 	 * to initialize a kernel version of the same transfer.
 	 */
-	mutex_lock(&spidev->buf_lock);
 	buf = spidev->buffer;
 	total = 0;
 	for (n = n_xfers, k_tmp = k_xfers, u_tmp = u_xfers;
@@ -297,14 +295,12 @@ static int spidev_message(struct spidev_data *spidev,
 	status = total;
 
 done:
-	mutex_unlock(&spidev->buf_lock);
 	kfree(k_xfers);
 	return status;
 }
 
-static int
-spidev_ioctl(struct inode *inode, struct file *filp,
-		unsigned int cmd, unsigned long arg)
+static long
+spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	int			err = 0;
 	int			retval = 0;
@@ -342,6 +338,14 @@ spidev_ioctl(struct inode *inode, struct file *filp,
 	if (spi == NULL)
 		return -ESHUTDOWN;
 
+	/* use the buffer lock here for triple duty:
+	 *  - prevent I/O (from us) so calling spi_setup() is safe;
+	 *  - prevent concurrent SPI_IOC_WR_* from morphing
+	 *    data fields while SPI_IOC_RD_* reads them;
+	 *  - SPI_IOC_MESSAGE needs the buffer locked "normally".
+	 */
+	mutex_lock(&spidev->buf_lock);
+
 	switch (cmd) {
 	/* read requests */
 	case SPI_IOC_RD_MODE:
@@ -457,6 +461,8 @@ spidev_ioctl(struct inode *inode, struct file *filp,
 		kfree(ioc);
 		break;
 	}
+
+	mutex_unlock(&spidev->buf_lock);
 	spi_dev_put(spi);
 	return retval;
 }
@@ -466,6 +472,7 @@ static int spidev_open(struct inode *inode, struct file *filp)
 	struct spidev_data	*spidev;
 	int			status = -ENXIO;
 
+	lock_kernel();
 	mutex_lock(&device_list_lock);
 
 	list_for_each_entry(spidev, &device_list, device_entry) {
@@ -491,6 +498,7 @@ static int spidev_open(struct inode *inode, struct file *filp)
 		pr_debug("spidev: nothing for minor %d\n", iminor(inode));
 
 	mutex_unlock(&device_list_lock);
+	unlock_kernel();
 	return status;
 }
 
@@ -532,7 +540,7 @@ static struct file_operations spidev_fops = {
 	 */
 	.write =	spidev_write,
 	.read =		spidev_read,
-	.ioctl =	spidev_ioctl,
+	.unlocked_ioctl = spidev_ioctl,
 	.open =		spidev_open,
 	.release =	spidev_release,
 };
@@ -575,7 +583,8 @@ static int spidev_probe(struct spi_device *spi)
 		struct device *dev;
 
 		spidev->devt = MKDEV(SPIDEV_MAJOR, minor);
-		dev = device_create(spidev_class, &spi->dev, spidev->devt,
+		dev = device_create_drvdata(spidev_class, &spi->dev,
+				spidev->devt, spidev,
 				"spidev%d.%d",
 				spi->master->bus_num, spi->chip_select);
 		status = IS_ERR(dev) ? PTR_ERR(dev) : 0;
@@ -585,7 +594,6 @@ static int spidev_probe(struct spi_device *spi)
 	}
 	if (status == 0) {
 		set_bit(minor, minors);
-		spi_set_drvdata(spi, spidev);
 		list_add(&spidev->device_entry, &device_list);
 	}
 	mutex_unlock(&device_list_lock);
diff --git a/drivers/spi/xilinx_spi.c b/drivers/spi/xilinx_spi.c
index 113a0468ffcb..68d6f4988fb5 100644
--- a/drivers/spi/xilinx_spi.c
+++ b/drivers/spi/xilinx_spi.c
@@ -353,11 +353,12 @@ static int __init xilinx_spi_probe(struct platform_device *dev)
 		goto put_master;
 	}
 
-	xspi->irq = platform_get_irq(dev, 0);
-	if (xspi->irq < 0) {
+	ret = platform_get_irq(dev, 0);
+	if (ret < 0) {
 		ret = -ENXIO;
 		goto unmap_io;
 	}
+	xspi->irq = ret;
 
 	master->bus_num = pdata->bus_num;
 	master->num_chipselect = pdata->num_chipselect;