Merge branch 'next-spi' of git://git.secretlab.ca/git/linux-2.6

* 'next-spi' of git://git.secretlab.ca/git/linux-2.6: (53 commits)
  spi/omap2_mcspi: Verify TX reg is empty after TX only xfer with DMA
  spi/omap2_mcspi: disable channel after TX_ONLY transfer in PIO mode
  spi/bfin_spi: namespace local structs
  spi/bfin_spi: init early
  spi/bfin_spi: check per-transfer bits_per_word
  spi/bfin_spi: warn when CS is driven by hardware (CPHA=0)
  spi/bfin_spi: cs should be always low when a new transfer begins
  spi/bfin_spi: fix typo in comment
  spi/bfin_spi: reject unsupported SPI modes
  spi/bfin_spi: use dma_disable_irq_nosync() in irq handler
  spi/bfin_spi: combine duplicate SPI_CTL read/write logic
  spi/bfin_spi: reset ctl_reg bits when setup is run again on a device
  spi/bfin_spi: push all size checks into the transfer function
  spi/bfin_spi: use nosync when disabling the IRQ from the IRQ handler
  spi/bfin_spi: sync hardware state before reprogramming everything
  spi/bfin_spi: save/restore state when suspending/resuming
  spi/bfin_spi: redo GPIO CS handling
  Blackfin: SPI: expand SPI bitmasks
  spi/bfin_spi: use the SPI namespaced bit names
  spi/bfin_spi: drop extra memory we don't need
  ...

24 files changed, 3842 insertions, 1232 deletions

diff --git a/Documentation/powerpc/dts-bindings/fsl/spi.txt b/Documentation/powerpc/dts-bindings/fsl/spi.txt
index 80510c018ee..777abd7399d 100644
--- a/Documentation/powerpc/dts-bindings/fsl/spi.txt
+++ b/Documentation/powerpc/dts-bindings/fsl/spi.txt
@@ -1,7 +1,9 @@
 * SPI (Serial Peripheral Interface)
 
 Required properties:
-- cell-index : SPI controller index.
+- cell-index : QE SPI subblock index.
+		0: QE subblock SPI1
+		1: QE subblock SPI2
 - compatible : should be "fsl,spi".
 - mode : the SPI operation mode, it can be "cpu" or "cpu-qe".
 - reg : Offset and length of the register set for the device
@@ -29,3 +31,23 @@ Example:
 		gpios = <&gpio 18 1	// device reg=<0>
 			 &gpio 19 1>;	// device reg=<1>
 	};
+
+
+* eSPI (Enhanced Serial Peripheral Interface)
+
+Required properties:
+- compatible : should be "fsl,mpc8536-espi".
+- reg : Offset and length of the register set for the device.
+- interrupts : should contain eSPI interrupt, the device has one interrupt.
+- fsl,espi-num-chipselects : the number of the chipselect signals.
+
+Example:
+	spi@110000 {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		compatible = "fsl,mpc8536-espi";
+		reg = <0x110000 0x1000>;
+		interrupts = <53 0x2>;
+		interrupt-parent = <&mpic>;
+		fsl,espi-num-chipselects = <4>;
+	};
diff --git a/arch/arm/mach-lpc32xx/phy3250.c b/arch/arm/mach-lpc32xx/phy3250.c
index bc9a42da214..0c936cf5675 100644
--- a/arch/arm/mach-lpc32xx/phy3250.c
+++ b/arch/arm/mach-lpc32xx/phy3250.c
@@ -172,18 +172,12 @@ static void phy3250_spi_cs_set(u32 control)
 }
 
 static struct pl022_config_chip spi0_chip_info = {
-	.lbm			= LOOPBACK_DISABLED,
 	.com_mode		= INTERRUPT_TRANSFER,
 	.iface			= SSP_INTERFACE_MOTOROLA_SPI,
 	.hierarchy		= SSP_MASTER,
 	.slave_tx_disable	= 0,
-	.endian_tx		= SSP_TX_LSB,
-	.endian_rx		= SSP_RX_LSB,
-	.data_size		= SSP_DATA_BITS_8,
 	.rx_lev_trig		= SSP_RX_4_OR_MORE_ELEM,
 	.tx_lev_trig		= SSP_TX_4_OR_MORE_EMPTY_LOC,
-	.clk_phase		= SSP_CLK_FIRST_EDGE,
-	.clk_pol		= SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len		= SSP_BITS_8,
 	.wait_state		= SSP_MWIRE_WAIT_ZERO,
 	.duplex			= SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -239,6 +233,7 @@ static int __init phy3250_spi_board_register(void)
 			.max_speed_hz = 5000000,
 			.bus_num = 0,
 			.chip_select = 0,
+			.mode = SPI_MODE_0,
 			.platform_data = &eeprom,
 			.controller_data = &spi0_chip_info,
 		},
diff --git a/arch/arm/mach-u300/dummyspichip.c b/arch/arm/mach-u300/dummyspichip.c
index 5f55012b7c9..03f79361259 100644
--- a/arch/arm/mach-u300/dummyspichip.c
+++ b/arch/arm/mach-u300/dummyspichip.c
@@ -46,7 +46,6 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * struct, this is just used here to alter the behaviour of the chip
 	 * in order to perform tests.
 	 */
-	struct pl022_config_chip *chip_info = spi->controller_data;
 	int status;
 	u8 txbuf[14] = {0xDE, 0xAD, 0xBE, 0xEF, 0x2B, 0xAD,
 			0xCA, 0xFE, 0xBA, 0xBE, 0xB1, 0x05,
@@ -72,7 +71,7 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * Force chip to 8 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_8;
+	spi->bits_per_word = 8;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
@@ -159,7 +158,7 @@ static ssize_t dummy_looptest(struct device *dev,
 	 * Force chip to 16 bit mode
 	 * WARNING: NEVER DO THIS IN REAL DRIVER CODE, THIS SHOULD BE STATIC!
 	 */
-	chip_info->data_size = SSP_DATA_BITS_16;
+	spi->bits_per_word = 16;
 	/* You should NOT DO THIS EITHER */
 	spi->master->setup(spi);
 
diff --git a/arch/arm/mach-u300/spi.c b/arch/arm/mach-u300/spi.c
index f0e887bea30..edb2c0d255c 100644
--- a/arch/arm/mach-u300/spi.c
+++ b/arch/arm/mach-u300/spi.c
@@ -30,8 +30,6 @@ static void select_dummy_chip(u32 chipselect)
 }
 
 struct pl022_config_chip dummy_chip_info = {
-	/* Nominally this is LOOPBACK_DISABLED, but this is our dummy chip! */
-	.lbm = LOOPBACK_ENABLED,
 	/*
 	 * available POLLING_TRANSFER and INTERRUPT_TRANSFER,
 	 * DMA_TRANSFER does not work
@@ -42,14 +40,8 @@ struct pl022_config_chip dummy_chip_info = {
 	.hierarchy = SSP_MASTER,
 	/* 0 = drive TX even as slave, 1 = do not drive TX as slave */
 	.slave_tx_disable = 0,
-	/* LSB first */
-	.endian_tx = SSP_TX_LSB,
-	.endian_rx = SSP_RX_LSB,
-	.data_size = SSP_DATA_BITS_8, /* used to be 12 in some default */
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_LOW,
 	.ctrl_len = SSP_BITS_12,
 	.wait_state = SSP_MWIRE_WAIT_ZERO,
 	.duplex = SSP_MICROWIRE_CHANNEL_FULL_DUPLEX,
@@ -75,7 +67,7 @@ static struct spi_board_info u300_spi_devices[] = {
 		.bus_num        = 0, /* Only one bus on this chip */
 		.chip_select    = 0,
 		/* Means SPI_CS_HIGH, change if e.g low CS */
-		.mode           = 0,
+		.mode           = SPI_MODE_1 | SPI_LSB_FIRST | SPI_LOOP,
 	},
 #endif
 };
diff --git a/arch/arm/mach-ux500/board-mop500.c b/arch/arm/mach-ux500/board-mop500.c
index 0e8fd135a57..219ae0ca4ee 100644
--- a/arch/arm/mach-ux500/board-mop500.c
+++ b/arch/arm/mach-ux500/board-mop500.c
@@ -55,19 +55,13 @@ static void ab4500_spi_cs_control(u32 command)
 }
 
 struct pl022_config_chip ab4500_chip_info = {
-	.lbm = LOOPBACK_DISABLED,
 	.com_mode = INTERRUPT_TRANSFER,
 	.iface = SSP_INTERFACE_MOTOROLA_SPI,
 	/* we can act as master only */
 	.hierarchy = SSP_MASTER,
 	.slave_tx_disable = 0,
-	.endian_rx = SSP_RX_MSB,
-	.endian_tx = SSP_TX_MSB,
-	.data_size = SSP_DATA_BITS_24,
 	.rx_lev_trig = SSP_RX_1_OR_MORE_ELEM,
 	.tx_lev_trig = SSP_TX_1_OR_MORE_EMPTY_LOC,
-	.clk_phase = SSP_CLK_SECOND_EDGE,
-	.clk_pol = SSP_CLK_POL_IDLE_HIGH,
 	.cs_control = ab4500_spi_cs_control,
 };
 
@@ -83,7 +77,7 @@ static struct spi_board_info u8500_spi_devices[] = {
 		.max_speed_hz = 12000000,
 		.bus_num = 0,
 		.chip_select = 0,
-		.mode = SPI_MODE_0,
+		.mode = SPI_MODE_3,
 		.irq = IRQ_DB8500_AB8500,
 	},
 };
diff --git a/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h b/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
index e5aba8f95b7..b226f7405e6 100644
--- a/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
+++ b/arch/arm/plat-samsung/include/plat/s3c64xx-spi.h
@@ -32,6 +32,8 @@ struct s3c64xx_spi_csinfo {
  * struct s3c64xx_spi_info - SPI Controller defining structure
  * @src_clk_nr: Clock source index for the CLK_CFG[SPI_CLKSEL] field.
  * @src_clk_name: Platform name of the corresponding clock.
+ * @clk_from_cmu: If the SPI clock/prescalar control block is present
+ *     by the platform's clock-management-unit and not in SPI controller.
  * @num_cs: Number of CS this controller emulates.
  * @cfg_gpio: Configure pins for this SPI controller.
  * @fifo_lvl_mask: All tx fifo_lvl fields start at offset-6
@@ -41,6 +43,7 @@ struct s3c64xx_spi_csinfo {
 struct s3c64xx_spi_info {
 	int src_clk_nr;
 	char *src_clk_name;
+	bool clk_from_cmu;
 
 	int num_cs;
 
diff --git a/arch/blackfin/include/asm/bfin5xx_spi.h b/arch/blackfin/include/asm/bfin5xx_spi.h
index ed4f8c6db0c..4223cf08ce8 100644
--- a/arch/blackfin/include/asm/bfin5xx_spi.h
+++ b/arch/blackfin/include/asm/bfin5xx_spi.h
@@ -11,26 +11,17 @@
 
 #define MIN_SPI_BAUD_VAL	2
 
-#define SPI_READ              0
-#define SPI_WRITE             1
-
-#define SPI_CTRL_OFF            0x0
-#define SPI_FLAG_OFF            0x4
-#define SPI_STAT_OFF            0x8
-#define SPI_TXBUFF_OFF          0xc
-#define SPI_RXBUFF_OFF          0x10
-#define SPI_BAUD_OFF            0x14
-#define SPI_SHAW_OFF            0x18
-
-
 #define BIT_CTL_ENABLE      0x4000
 #define BIT_CTL_OPENDRAIN   0x2000
 #define BIT_CTL_MASTER      0x1000
-#define BIT_CTL_POLAR       0x0800
-#define BIT_CTL_PHASE       0x0400
-#define BIT_CTL_BITORDER    0x0200
+#define BIT_CTL_CPOL        0x0800
+#define BIT_CTL_CPHA        0x0400
+#define BIT_CTL_LSBF        0x0200
 #define BIT_CTL_WORDSIZE    0x0100
-#define BIT_CTL_MISOENABLE  0x0020
+#define BIT_CTL_EMISO       0x0020
+#define BIT_CTL_PSSE        0x0010
+#define BIT_CTL_GM          0x0008
+#define BIT_CTL_SZ          0x0004
 #define BIT_CTL_RXMOD       0x0000
 #define BIT_CTL_TXMOD       0x0001
 #define BIT_CTL_TIMOD_DMA_TX 0x0003
@@ -50,61 +41,7 @@
 #define BIT_STU_SENDOVER    0x0001
 #define BIT_STU_RECVFULL    0x0020
 
-#define CFG_SPI_ENABLE      1
-#define CFG_SPI_DISABLE     0
-
-#define CFG_SPI_OUTENABLE   1
-#define CFG_SPI_OUTDISABLE  0
-
-#define CFG_SPI_ACTLOW      1
-#define CFG_SPI_ACTHIGH     0
-
-#define CFG_SPI_PHASESTART  1
-#define CFG_SPI_PHASEMID    0
-
-#define CFG_SPI_MASTER      1
-#define CFG_SPI_SLAVE       0
-
-#define CFG_SPI_SENELAST    0
-#define CFG_SPI_SENDZERO    1
-
-#define CFG_SPI_RCVFLUSH    1
-#define CFG_SPI_RCVDISCARD  0
-
-#define CFG_SPI_LSBFIRST    1
-#define CFG_SPI_MSBFIRST    0
-
-#define CFG_SPI_WORDSIZE16  1
-#define CFG_SPI_WORDSIZE8   0
-
-#define CFG_SPI_MISOENABLE   1
-#define CFG_SPI_MISODISABLE  0
-
-#define CFG_SPI_READ      0x00
-#define CFG_SPI_WRITE     0x01
-#define CFG_SPI_DMAREAD   0x02
-#define CFG_SPI_DMAWRITE  0x03
-
-#define CFG_SPI_CSCLEARALL  0
-#define CFG_SPI_CHIPSEL1    1
-#define CFG_SPI_CHIPSEL2    2
-#define CFG_SPI_CHIPSEL3    3
-#define CFG_SPI_CHIPSEL4    4
-#define CFG_SPI_CHIPSEL5    5
-#define CFG_SPI_CHIPSEL6    6
-#define CFG_SPI_CHIPSEL7    7
-
-#define CFG_SPI_CS1VALUE    1
-#define CFG_SPI_CS2VALUE    2
-#define CFG_SPI_CS3VALUE    3
-#define CFG_SPI_CS4VALUE    4
-#define CFG_SPI_CS5VALUE    5
-#define CFG_SPI_CS6VALUE    6
-#define CFG_SPI_CS7VALUE    7
-
-#define CMD_SPI_SET_BAUDRATE  2
-#define CMD_SPI_GET_SYSTEMCLOCK   25
-#define CMD_SPI_SET_WRITECONTINUOUS     26
+#define MAX_CTRL_CS          8  /* cs in spi controller */
 
 /* device.platform_data for SSP controller devices */
 struct bfin5xx_spi_master {
@@ -120,9 +57,7 @@ struct bfin5xx_spi_chip {
 	u16 ctl_reg;
 	u8 enable_dma;
 	u8 bits_per_word;
-	u8 cs_change_per_word;
 	u16 cs_chg_udelay; /* Some devices require 16-bit delays */
-	u32 cs_gpio;
 	/* Value to send if no TX value is supplied, usually 0x0 or 0xFFFF */
 	u16 idle_tx_val;
 	u8 pio_interrupt; /* Enable spi data irq */
diff --git a/arch/powerpc/boot/dts/mpc8536ds.dts b/arch/powerpc/boot/dts/mpc8536ds.dts
index 815cebb2e3e..a75c10eed26 100644
--- a/arch/powerpc/boot/dts/mpc8536ds.dts
+++ b/arch/powerpc/boot/dts/mpc8536ds.dts
@@ -108,6 +108,58 @@
 			};
 		};
 
+		spi@7000 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc8536-espi";
+			reg = <0x7000 0x1000>;
+			interrupts = <59 0x2>;
+			interrupt-parent = <&mpic>;
+			fsl,espi-num-chipselects = <4>;
+
+			flash@0 {
+				#address-cells = <1>;
+				#size-cells = <1>;
+				compatible = "spansion,s25sl12801";
+				reg = <0>;
+				spi-max-frequency = <40000000>;
+				partition@u-boot {
+					label = "u-boot";
+					reg = <0x00000000 0x00100000>;
+					read-only;
+				};
+				partition@kernel {
+					label = "kernel";
+					reg = <0x00100000 0x00500000>;
+					read-only;
+				};
+				partition@dtb {
+					label = "dtb";
+					reg = <0x00600000 0x00100000>;
+					read-only;
+				};
+				partition@fs {
+					label = "file system";
+					reg = <0x00700000 0x00900000>;
+				};
+			};
+			flash@1 {
+				compatible = "spansion,s25sl12801";
+				reg = <1>;
+				spi-max-frequency = <40000000>;
+			};
+			flash@2 {
+				compatible = "spansion,s25sl12801";
+				reg = <2>;
+				spi-max-frequency = <40000000>;
+			};
+			flash@3 {
+				compatible = "spansion,s25sl12801";
+				reg = <3>;
+				spi-max-frequency = <40000000>;
+			};
+		};
+
 		dma@21300 {
 			#address-cells = <1>;
 			#size-cells = <1>;
diff --git a/arch/powerpc/boot/dts/p4080ds.dts b/arch/powerpc/boot/dts/p4080ds.dts
index 2f0de24e382..5b7fc29dd6c 100644
--- a/arch/powerpc/boot/dts/p4080ds.dts
+++ b/arch/powerpc/boot/dts/p4080ds.dts
@@ -236,22 +236,19 @@
 		};
 
 		spi@110000 {
-			cell-index = <0>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			compatible = "fsl,espi";
+			compatible = "fsl,p4080-espi", "fsl,mpc8536-espi";
 			reg = <0x110000 0x1000>;
 			interrupts = <53 0x2>;
 			interrupt-parent = <&mpic>;
-			espi,num-ss-bits = <4>;
-			mode = "cpu";
+			fsl,espi-num-chipselects = <4>;
 
-			fsl_m25p80@0 {
+			flash@0 {
 				#address-cells = <1>;
 				#size-cells = <1>;
-				compatible = "fsl,espi-flash";
+				compatible = "spansion,s25sl12801";
 				reg = <0>;
-				linux,modalias = "fsl_m25p80";
 				spi-max-frequency = <40000000>; /* input clock */
 				partition@u-boot {
 					label = "u-boot";
diff --git a/drivers/mfd/ab8500-spi.c b/drivers/mfd/ab8500-spi.c
index e1c8b62b086..01b6d584442 100644
--- a/drivers/mfd/ab8500-spi.c
+++ b/drivers/mfd/ab8500-spi.c
@@ -83,6 +83,11 @@ static int __devinit ab8500_spi_probe(struct spi_device *spi)
 	struct ab8500 *ab8500;
 	int ret;
 
+	spi->bits_per_word = 24;
+	ret = spi_setup(spi);
+	if (ret < 0)
+		return ret;
+
 	ab8500 = kzalloc(sizeof *ab8500, GFP_KERNEL);
 	if (!ab8500)
 		return -ENOMEM;
diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 91c2f4f3af1..9949c252c23 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -182,12 +182,27 @@ config SPI_MPC512x_PSC
 	  This enables using the Freescale MPC5121 Programmable Serial
 	  Controller in SPI master mode.
 
-config SPI_MPC8xxx
-	tristate "Freescale MPC8xxx SPI controller"
+config SPI_FSL_LIB
+	tristate
 	depends on FSL_SOC
+
+config SPI_FSL_SPI
+	tristate "Freescale SPI controller"
+	depends on FSL_SOC
+	select SPI_FSL_LIB
 	help
-	  This enables using the Freescale MPC8xxx SPI controllers in master
-	  mode.
+	  This enables using the Freescale SPI controllers in master mode.
+	  MPC83xx platform uses the controller in cpu mode or CPM/QE mode.
+	  MPC8569 uses the controller in QE mode, MPC8610 in cpu mode.
+
+config SPI_FSL_ESPI
+	tristate "Freescale eSPI controller"
+	depends on FSL_SOC
+	select SPI_FSL_LIB
+	help
+	  This enables using the Freescale eSPI controllers in master mode.
+	  From MPC8536, 85xx platform uses the controller, and all P10xx,
+	  P20xx, P30xx,P40xx, P50xx uses this controller.
 
 config SPI_OMAP_UWIRE
 	tristate "OMAP1 MicroWire"
@@ -298,6 +313,13 @@ config SPI_STMP3XXX
 	help
 	  SPI driver for Freescale STMP37xx/378x SoC SSP interface
 
+config SPI_TOPCLIFF_PCH
+	tristate "Topcliff PCH SPI Controller"
+	depends on PCI
+	help
+	  SPI driver for the Topcliff PCH (Platform Controller Hub) SPI bus
+	  used in some x86 embedded processors.
+
 config SPI_TXX9
 	tristate "Toshiba TXx9 SPI controller"
 	depends on GENERIC_GPIO && CPU_TX49XX
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index e9cbd18217a..557aaadf56b 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -2,9 +2,7 @@
 # Makefile for kernel SPI drivers.
 #
 
-ifeq ($(CONFIG_SPI_DEBUG),y)
-EXTRA_CFLAGS += -DDEBUG
-endif
+ccflags-$(CONFIG_SPI_DEBUG) := -DDEBUG
 
 # small core, mostly translating board-specific
 # config declarations into driver model code
@@ -34,11 +32,14 @@ obj-$(CONFIG_SPI_PL022)			+= amba-pl022.o
 obj-$(CONFIG_SPI_MPC512x_PSC)		+= mpc512x_psc_spi.o
 obj-$(CONFIG_SPI_MPC52xx_PSC)		+= mpc52xx_psc_spi.o
 obj-$(CONFIG_SPI_MPC52xx)		+= mpc52xx_spi.o
-obj-$(CONFIG_SPI_MPC8xxx)		+= spi_mpc8xxx.o
+obj-$(CONFIG_SPI_FSL_LIB)		+= spi_fsl_lib.o
+obj-$(CONFIG_SPI_FSL_ESPI)		+= spi_fsl_espi.o
+obj-$(CONFIG_SPI_FSL_SPI)		+= spi_fsl_spi.o
 obj-$(CONFIG_SPI_PPC4xx)		+= spi_ppc4xx.o
 obj-$(CONFIG_SPI_S3C24XX_GPIO)		+= spi_s3c24xx_gpio.o
 obj-$(CONFIG_SPI_S3C24XX)		+= spi_s3c24xx_hw.o
 obj-$(CONFIG_SPI_S3C64XX)		+= spi_s3c64xx.o
+obj-$(CONFIG_SPI_TOPCLIFF_PCH)		+= spi_topcliff_pch.o
 obj-$(CONFIG_SPI_TXX9)			+= spi_txx9.o
 obj-$(CONFIG_SPI_XILINX)		+= xilinx_spi.o
 obj-$(CONFIG_SPI_XILINX_OF)		+= xilinx_spi_of.o
diff --git a/drivers/spi/amba-pl022.c b/drivers/spi/amba-pl022.c
index 4c37c4e2864..fb3d1b31772 100644
--- a/drivers/spi/amba-pl022.c
+++ b/drivers/spi/amba-pl022.c
@@ -27,7 +27,6 @@
 /*
  * TODO:
  * - add timeout on polled transfers
- * - add generic DMA framework support
  */
 
 #include <linux/init.h>
@@ -45,6 +44,9 @@
 #include <linux/amba/pl022.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/scatterlist.h>
 
 /*
  * This macro is used to define some register default values.
@@ -381,6 +383,14 @@ struct pl022 {
 	enum ssp_reading		read;
 	enum ssp_writing		write;
 	u32				exp_fifo_level;
+	/* DMA settings */
+#ifdef CONFIG_DMA_ENGINE
+	struct dma_chan			*dma_rx_channel;
+	struct dma_chan			*dma_tx_channel;
+	struct sg_table			sgt_rx;
+	struct sg_table			sgt_tx;
+	char				*dummypage;
+#endif
 };
 
 /**
@@ -406,7 +416,7 @@ struct chip_data {
 	u16 dmacr;
 	u16 cpsr;
 	u8 n_bytes;
-	u8 enable_dma:1;
+	bool enable_dma;
 	enum ssp_reading read;
 	enum ssp_writing write;
 	void (*cs_control) (u32 command);
@@ -763,6 +773,371 @@ static void *next_transfer(struct pl022 *pl022)
 	}
 	return STATE_DONE;
 }
+
+/*
+ * This DMA functionality is only compiled in if we have
+ * access to the generic DMA devices/DMA engine.
+ */
+#ifdef CONFIG_DMA_ENGINE
+static void unmap_free_dma_scatter(struct pl022 *pl022)
+{
+	/* Unmap and free the SG tables */
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	sg_free_table(&pl022->sgt_rx);
+	sg_free_table(&pl022->sgt_tx);
+}
+
+static void dma_callback(void *data)
+{
+	struct pl022 *pl022 = data;
+	struct spi_message *msg = pl022->cur_msg;
+
+	BUG_ON(!pl022->sgt_rx.sgl);
+
+#ifdef VERBOSE_DEBUG
+	/*
+	 * Optionally dump out buffers to inspect contents, this is
+	 * good if you want to convince yourself that the loopback
+	 * read/write contents are the same, when adopting to a new
+	 * DMA engine.
+	 */
+	{
+		struct scatterlist *sg;
+		unsigned int i;
+
+		dma_sync_sg_for_cpu(&pl022->adev->dev,
+				    pl022->sgt_rx.sgl,
+				    pl022->sgt_rx.nents,
+				    DMA_FROM_DEVICE);
+
+		for_each_sg(pl022->sgt_rx.sgl, sg, pl022->sgt_rx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI RX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI RX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+		for_each_sg(pl022->sgt_tx.sgl, sg, pl022->sgt_tx.nents, i) {
+			dev_dbg(&pl022->adev->dev, "SPI TX SG ENTRY: %d", i);
+			print_hex_dump(KERN_ERR, "SPI TX: ",
+				       DUMP_PREFIX_OFFSET,
+				       16,
+				       1,
+				       sg_virt(sg),
+				       sg_dma_len(sg),
+				       1);
+		}
+	}
+#endif
+
+	unmap_free_dma_scatter(pl022);
+
+	/* Update total bytes transfered */
+	msg->actual_length += pl022->cur_transfer->len;
+	if (pl022->cur_transfer->cs_change)
+		pl022->cur_chip->
+			cs_control(SSP_CHIP_DESELECT);
+
+	/* Move to next transfer */
+	msg->state = next_transfer(pl022);
+	tasklet_schedule(&pl022->pump_transfers);
+}
+
+static void setup_dma_scatter(struct pl022 *pl022,
+			      void *buffer,
+			      unsigned int length,
+			      struct sg_table *sgtab)
+{
+	struct scatterlist *sg;
+	int bytesleft = length;
+	void *bufp = buffer;
+	int mapbytes;
+	int i;
+
+	if (buffer) {
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			/*
+			 * If there are less bytes left than what fits
+			 * in the current page (plus page alignment offset)
+			 * we just feed in this, else we stuff in as much
+			 * as we can.
+			 */
+			if (bytesleft < (PAGE_SIZE - offset_in_page(bufp)))
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE - offset_in_page(bufp);
+			sg_set_page(sg, virt_to_page(bufp),
+				    mapbytes, offset_in_page(bufp));
+			bufp += mapbytes;
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX target page @ %p, %d bytes, %d left\n",
+				bufp, mapbytes, bytesleft);
+		}
+	} else {
+		/* Map the dummy buffer on every page */
+		for_each_sg(sgtab->sgl, sg, sgtab->nents, i) {
+			if (bytesleft < PAGE_SIZE)
+				mapbytes = bytesleft;
+			else
+				mapbytes = PAGE_SIZE;
+			sg_set_page(sg, virt_to_page(pl022->dummypage),
+				    mapbytes, 0);
+			bytesleft -= mapbytes;
+			dev_dbg(&pl022->adev->dev,
+				"set RX/TX to dummy page %d bytes, %d left\n",
+				mapbytes, bytesleft);
+
+		}
+	}
+	BUG_ON(bytesleft);
+}
+
+/**
+ * configure_dma - configures the channels for the next transfer
+ * @pl022: SSP driver's private data structure
+ */
+static int configure_dma(struct pl022 *pl022)
+{
+	struct dma_slave_config rx_conf = {
+		.src_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_FROM_DEVICE,
+		.src_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	struct dma_slave_config tx_conf = {
+		.dst_addr = SSP_DR(pl022->phybase),
+		.direction = DMA_TO_DEVICE,
+		.dst_maxburst = pl022->vendor->fifodepth >> 1,
+	};
+	unsigned int pages;
+	int ret;
+	int sglen;
+	struct dma_chan *rxchan = pl022->dma_rx_channel;
+	struct dma_chan *txchan = pl022->dma_tx_channel;
+	struct dma_async_tx_descriptor *rxdesc;
+	struct dma_async_tx_descriptor *txdesc;
+	dma_cookie_t cookie;
+
+	/* Check that the channels are available */
+	if (!rxchan || !txchan)
+		return -ENODEV;
+
+	switch (pl022->read) {
+	case READING_NULL:
+		/* Use the same as for writing */
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case READING_U8:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case READING_U16:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case READING_U32:
+		rx_conf.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		break;
+	}
+
+	switch (pl022->write) {
+	case WRITING_NULL:
+		/* Use the same as for reading */
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+		break;
+	case WRITING_U8:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
+		break;
+	case WRITING_U16:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
+		break;
+	case WRITING_U32:
+		tx_conf.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;;
+		break;
+	}
+
+	/* SPI pecularity: we need to read and write the same width */
+	if (rx_conf.src_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		rx_conf.src_addr_width = tx_conf.dst_addr_width;
+	if (tx_conf.dst_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+		tx_conf.dst_addr_width = rx_conf.src_addr_width;
+	BUG_ON(rx_conf.src_addr_width != tx_conf.dst_addr_width);
+
+	rxchan->device->device_control(rxchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &rx_conf);
+	txchan->device->device_control(txchan, DMA_SLAVE_CONFIG,
+				       (unsigned long) &tx_conf);
+
+	/* Create sglists for the transfers */
+	pages = (pl022->cur_transfer->len >> PAGE_SHIFT) + 1;
+	dev_dbg(&pl022->adev->dev, "using %d pages for transfer\n", pages);
+
+	ret = sg_alloc_table(&pl022->sgt_rx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_rx_sg;
+
+	ret = sg_alloc_table(&pl022->sgt_tx, pages, GFP_KERNEL);
+	if (ret)
+		goto err_alloc_tx_sg;
+
+	/* Fill in the scatterlists for the RX+TX buffers */
+	setup_dma_scatter(pl022, pl022->rx,
+			  pl022->cur_transfer->len, &pl022->sgt_rx);
+	setup_dma_scatter(pl022, pl022->tx,
+			  pl022->cur_transfer->len, &pl022->sgt_tx);
+
+	/* Map DMA buffers */
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+			   pl022->sgt_rx.nents, DMA_FROM_DEVICE);
+	if (!sglen)
+		goto err_rx_sgmap;
+
+	sglen = dma_map_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+			   pl022->sgt_tx.nents, DMA_TO_DEVICE);
+	if (!sglen)
+		goto err_tx_sgmap;
+
+	/* Send both scatterlists */
+	rxdesc = rxchan->device->device_prep_slave_sg(rxchan,
+				      pl022->sgt_rx.sgl,
+				      pl022->sgt_rx.nents,
+				      DMA_FROM_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!rxdesc)
+		goto err_rxdesc;
+
+	txdesc = txchan->device->device_prep_slave_sg(txchan,
+				      pl022->sgt_tx.sgl,
+				      pl022->sgt_tx.nents,
+				      DMA_TO_DEVICE,
+				      DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!txdesc)
+		goto err_txdesc;
+
+	/* Put the callback on the RX transfer only, that should finish last */
+	rxdesc->callback = dma_callback;
+	rxdesc->callback_param = pl022;
+
+	/* Submit and fire RX and TX with TX last so we're ready to read! */
+	cookie = rxdesc->tx_submit(rxdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_rx;
+	cookie = txdesc->tx_submit(txdesc);
+	if (dma_submit_error(cookie))
+		goto err_submit_tx;
+	rxchan->device->device_issue_pending(rxchan);
+	txchan->device->device_issue_pending(txchan);
+
+	return 0;
+
+err_submit_tx:
+err_submit_rx:
+err_txdesc:
+	txchan->device->device_control(txchan, DMA_TERMINATE_ALL, 0);
+err_rxdesc:
+	rxchan->device->device_control(rxchan, DMA_TERMINATE_ALL, 0);
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_tx.sgl,
+		     pl022->sgt_tx.nents, DMA_TO_DEVICE);
+err_tx_sgmap:
+	dma_unmap_sg(&pl022->adev->dev, pl022->sgt_rx.sgl,
+		     pl022->sgt_tx.nents, DMA_FROM_DEVICE);
+err_rx_sgmap:
+	sg_free_table(&pl022->sgt_tx);
+err_alloc_tx_sg:
+	sg_free_table(&pl022->sgt_rx);
+err_alloc_rx_sg:
+	return -ENOMEM;
+}
+
+static int __init pl022_dma_probe(struct pl022 *pl022)
+{
+	dma_cap_mask_t mask;
+
+	/* Try to acquire a generic DMA engine slave channel */
+	dma_cap_zero(mask);
+	dma_cap_set(DMA_SLAVE, mask);
+	/*
+	 * We need both RX and TX channels to do DMA, else do none
+	 * of them.
+	 */
+	pl022->dma_rx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_rx_param);
+	if (!pl022->dma_rx_channel) {
+		dev_err(&pl022->adev->dev, "no RX DMA channel!\n");
+		goto err_no_rxchan;
+	}
+
+	pl022->dma_tx_channel = dma_request_channel(mask,
+					    pl022->master_info->dma_filter,
+					    pl022->master_info->dma_tx_param);
+	if (!pl022->dma_tx_channel) {
+		dev_err(&pl022->adev->dev, "no TX DMA channel!\n");
+		goto err_no_txchan;
+	}
+
+	pl022->dummypage = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (!pl022->dummypage) {
+		dev_err(&pl022->adev->dev, "no DMA dummypage!\n");
+		goto err_no_dummypage;
+	}
+
+	dev_info(&pl022->adev->dev, "setup for DMA on RX %s, TX %s\n",
+		 dma_chan_name(pl022->dma_rx_channel),
+		 dma_chan_name(pl022->dma_tx_channel));
+
+	return 0;
+<