[PATCHv2 0/3] spi: support for Socionext Synquacer platform

[PATCHv2 0/3] spi: support for Socionext Synquacer platform

[jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w]

From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>

Hello,

 Support for Socionext's FIP controller intended for flash device interfacing.
The controller can operate in 'direct' or 'command' mode. One mode directly
talks and provide a read/write i/f to the flash device. Other works as plain
SPI mode. This driver runs the controller as a SPI controller.

Changes since v1:
	# Changed licence header to C++ style comment.
	# Removed redundant lock and transfer_mode backup member.
	# Fixed divisor to allow upto 254.

Jassi Brar (3):
  dt-bindings: spi: Add DT bindings for Synquacer
  spi: Add spi driver for Socionext Synquacer platform
  MAINTAINERS: Add entry for Synquacer SPI driver

 .../devicetree/bindings/spi/spi-synquacer.txt      |  24 +
 MAINTAINERS                                        |   7 +
 drivers/spi/Kconfig                                |  11 +
 drivers/spi/Makefile                               |   1 +
 drivers/spi/spi-synquacer.c                        | 649 +++++++++++++++++++++
 5 files changed, 682 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
 create mode 100644 drivers/spi/spi-synquacer.c

-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w]

From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>

This patch adds documentation for Device-Tree bindings for the
Socionext Synquacer spi driver.

Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
 1 file changed, 24 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt

diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
new file mode 100644
index 0000000..d013cfd
--- /dev/null
+++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
@@ -0,0 +1,24 @@
+* Socionext Synquacer HS-SPI bindings
+
+Required Properties:
+- compatible: should be "socionext,synquacer-spi"
+- reg: physical base address of the controller and length of memory mapped
+       region.
+- clocks: Must contain an entry for rate source clock(s).
+- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
+
+Optional Properties:
+- num-cs: total number of chipselects
+- socionext,use-rtm: boolean, if required to use "retimed clock" for RX
+- socionext,set-aces: boolean, if same active clock edges field to be set.
+
+Example:
+
+	spi0: spi@ff110000 {
+		compatible = "socionext,synquacer-spi";
+		reg = <0xff110000 0x1000>;
+		clocks = <&clk_fip006_spi>;
+		clock-names = "iHCLK";
+		socionext,use-rtm;
+		socionext,set-aces;
+	};
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCHv2 2/3] spi: Add spi driver for Socionext Synquacer platform

[jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w]

From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>

This patch adds support for controller found on synquacer platforms.

Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 drivers/spi/Kconfig         |  11 +
 drivers/spi/Makefile        |   1 +
 drivers/spi/spi-synquacer.c | 639 ++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 651 insertions(+)
 create mode 100644 drivers/spi/spi-synquacer.c

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 6037839..9e04bbe 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -659,6 +659,17 @@ config SPI_SUN6I
 	help
 	  This enables using the SPI controller on the Allwinner A31 SoCs.
 
+config SPI_SYNQUACER
+	tristate "Socionext's Synquacer HighSpeed SPI controller"
+	depends on ARCH_SYNQUACER || COMPILE_TEST
+	select SPI_BITBANG
+	help
+	  SPI driver for Socionext's High speed SPI controller which provides
+	  various operating modes for interfacing to serial peripheral devices
+	  that use the de-facto standard SPI protocol.
+
+	  It also supports the new dual-bit and quad-bit SPI protocol.
+
 config SPI_MXS
 	tristate "Freescale MXS SPI controller"
 	depends on ARCH_MXS
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 34c5f28..7c222f2 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -96,6 +96,7 @@ obj-$(CONFIG_SPI_STM32) 		+= spi-stm32.o
 obj-$(CONFIG_SPI_ST_SSC4)		+= spi-st-ssc4.o
 obj-$(CONFIG_SPI_SUN4I)			+= spi-sun4i.o
 obj-$(CONFIG_SPI_SUN6I)			+= spi-sun6i.o
+obj-$(CONFIG_SPI_SYNQUACER)		+= spi-synquacer.o
 obj-$(CONFIG_SPI_TEGRA114)		+= spi-tegra114.o
 obj-$(CONFIG_SPI_TEGRA20_SFLASH)	+= spi-tegra20-sflash.o
 obj-$(CONFIG_SPI_TEGRA20_SLINK)		+= spi-tegra20-slink.o
diff --git a/drivers/spi/spi-synquacer.c b/drivers/spi/spi-synquacer.c
new file mode 100644
index 0000000..4c68c9a
--- /dev/null
+++ b/drivers/spi/spi-synquacer.c
@@ -0,0 +1,649 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Synquacer HSSPI controller driver
+//
+// Copyright (c) 2015-2018 Socionext Inc.
+// Copyright (c) 2018 Linaro Ltd.
+//
+
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/scatterlist.h>
+#include <linux/slab.h>
+#include <linux/spi/spi.h>
+#include <linux/spinlock.h>
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+
+#define MCTRL		0x0
+#define MEN	BIT(0)
+#define CSEN	BIT(1)
+#define IPCLK	BIT(3)
+#define MES	BIT(4)
+#define SYNCON	BIT(5)
+
+#define PCC0		0x4
+#define PCC(n)		(PCC0 + (n) * 4)
+#define RTM	BIT(3)
+#define ACES	BIT(2)
+#define SAFESYNC	BIT(16)
+#define CPHA	BIT(0)
+#define CPOL	BIT(1)
+#define SSPOL	BIT(4)
+#define SDIR	BIT(7)
+#define SS2CD	5
+#define SENDIAN	BIT(8)
+#define CDRS_SHIFT	9
+#define CDRS_MASK	0x7f
+
+#define TXF		0x14
+#define TXE		0x18
+#define TXC		0x1c
+#define RXF		0x20
+#define RXE		0x24
+#define RXC		0x28
+
+#define FAULTF		0x2c
+#define FAULTC		0x30
+
+#define DMCFG		0x34
+#define SSDC		BIT(1)
+#define MSTARTEN	BIT(2)
+
+#define	DMSTART		0x38
+#define TRIGGER		BIT(0)
+#define DMSTOP		BIT(8)
+#define CS_MASK		3
+#define CS_SHIFT	16
+#define DATA_TXRX	0
+#define DATA_RX		1
+#define DATA_TX		2
+#define DATA_MASK	3
+#define DATA_SHIFT	26
+#define BUS_WIDTH	24
+
+#define	DMBCC		0x3c
+#define DMSTATUS	0x40
+#define RX_DATA_MASK	0x1f
+#define RX_DATA_SHIFT	8
+#define TX_DATA_MASK	0x1f
+#define TX_DATA_SHIFT	16
+
+#define TXBITCNT	0x44
+
+#define FIFOCFG		0x4c
+#define BPW_MASK	0x3
+#define BPW_SHIFT	8
+#define RX_FLUSH	BIT(11)
+#define TX_FLUSH	BIT(12)
+#define RX_TRSHLD_MASK		0xf
+#define RX_TRSHLD_SHIFT		0
+#define TX_TRSHLD_MASK		0xf
+#define TX_TRSHLD_SHIFT		4
+
+#define TXFIFO		0x50
+#define RXFIFO		0x90
+#define MID		0xfc
+
+#define FIFO_DEPTH	16
+#define TX_TRSHLD	4
+#define RX_TRSHLD	(FIFO_DEPTH - TX_TRSHLD)
+
+#define TXBIT	BIT(1)
+#define RXBIT	BIT(2)
+
+struct synquacer_spi {
+	struct device *dev;
+	struct spi_master *master;
+
+	unsigned int cs;
+	unsigned int bpw;
+	unsigned int mode;
+	unsigned int speed;
+	bool aces, rtm;
+	void *rx_buf;
+	const void *tx_buf;
+	struct clk *clk;
+	void __iomem *regs;
+	unsigned int tx_words, rx_words;
+	unsigned int bus_width;
+};
+
+static void read_fifo(struct synquacer_spi *sspi)
+{
+	u32 len = readl_relaxed(sspi->regs + DMSTATUS);
+	int i;
+
+	len = (len >> RX_DATA_SHIFT) & RX_DATA_MASK;
+	len = min_t(unsigned int, len, sspi->rx_words);
+
+	switch (sspi->bpw) {
+	case 8:
+		{
+		u8 *buf = sspi->rx_buf;
+
+		for (i = 0; i < len; i++)
+			*buf++ = readb_relaxed(sspi->regs + RXFIFO);
+		sspi->rx_buf = buf;
+		break;
+		}
+	case 16:
+		{
+		u16 *buf = sspi->rx_buf;
+
+		for (i = 0; i < len; i++)
+			*buf++ = readw_relaxed(sspi->regs + RXFIFO);
+		sspi->rx_buf = buf;
+		break;
+		}
+	default:
+		{
+		u32 *buf = sspi->rx_buf;
+
+		for (i = 0; i < len; i++)
+			*buf++ = readl_relaxed(sspi->regs + RXFIFO);
+		sspi->rx_buf = buf;
+		break;
+		}
+	}
+
+	sspi->rx_words -= len;
+}
+
+static void write_fifo(struct synquacer_spi *sspi)
+{
+	u32 len = readl_relaxed(sspi->regs + DMSTATUS);
+	int i;
+
+	len = (len >> TX_DATA_SHIFT) & TX_DATA_MASK;
+	len = min_t(unsigned int, FIFO_DEPTH - len, sspi->tx_words);
+
+	switch (sspi->bpw) {
+	case 8:
+		{
+		const u8 *buf = sspi->tx_buf;
+
+		for (i = 0; i < len; i++)
+			writeb_relaxed(*buf++, sspi->regs + TXFIFO);
+		sspi->tx_buf = buf;
+		break;
+		}
+	case 16:
+		{
+		const u16 *buf = sspi->tx_buf;
+
+		for (i = 0; i < len; i++)
+			writew_relaxed(*buf++, sspi->regs + TXFIFO);
+		sspi->tx_buf = buf;
+		break;
+		}
+	default:
+		{
+		const u32 *buf = sspi->tx_buf;
+
+		for (i = 0; i < len; i++)
+			writel_relaxed(*buf++, sspi->regs + TXFIFO);
+		sspi->tx_buf = buf;
+		break;
+		}
+	}
+	sspi->tx_words -= len;
+}
+
+static int synquacer_spi_config(struct spi_master *master,
+				struct spi_device *spi,
+				struct spi_transfer *xfer)
+{
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+	unsigned int speed, mode, bpw, cs, bus_width;
+	unsigned long rate;
+	u32 val, div;
+
+	/* Full Duplex only on 1bit wide bus */
+	if (xfer->rx_buf && xfer->tx_buf &&
+	    (xfer->rx_nbits != 1 || xfer->tx_nbits != 1)) {
+		dev_err(sspi->dev,
+			"RX and TX bus widths must match for Full-Duplex!\n");
+		return -EINVAL;
+	}
+
+	if (xfer->tx_buf)
+		bus_width = xfer->tx_nbits;
+	else
+		bus_width = xfer->rx_nbits;
+
+	mode = spi->mode;
+	cs = spi->chip_select;
+	speed = xfer->speed_hz;
+	bpw = xfer->bits_per_word;
+
+	/* return if nothing to change */
+	if (speed == sspi->speed &&
+	    bus_width == sspi->bus_width && bpw == sspi->bpw &&
+	    mode == sspi->mode && cs == sspi->cs) {
+		return 0;
+	}
+
+	rate = clk_get_rate(sspi->clk);
+
+	div = DIV_ROUND_UP(rate, speed);
+	if (div > 254) {
+		dev_err(sspi->dev, "Requested rate too low (%u)\n",
+			sspi->speed);
+		return -EINVAL;
+	}
+
+	val = readl_relaxed(sspi->regs + PCC(cs));
+	val &= ~SAFESYNC;
+	if (bpw == 8 &&	(mode & (SPI_TX_DUAL | SPI_RX_DUAL)) && div < 3)
+		val |= SAFESYNC;
+	if (bpw == 8 &&	(mode & (SPI_TX_QUAD | SPI_RX_QUAD)) && div < 6)
+		val |= SAFESYNC;
+	if (bpw == 16 && (mode & (SPI_TX_QUAD | SPI_RX_QUAD)) && div < 3)
+		val |= SAFESYNC;
+
+	if (mode & SPI_CPHA)
+		val |= CPHA;
+	else
+		val &= ~CPHA;
+
+	if (mode & SPI_CPOL)
+		val |= CPOL;
+	else
+		val &= ~CPOL;
+
+	if (mode & SPI_CS_HIGH)
+		val |= SSPOL;
+	else
+		val &= ~SSPOL;
+
+	if (mode & SPI_LSB_FIRST)
+		val |= SDIR;
+	else
+		val &= ~SDIR;
+
+	if (sspi->aces)
+		val |= ACES;
+	else
+		val &= ~ACES;
+
+	if (sspi->rtm)
+		val |= RTM;
+	else
+		val &= ~RTM;
+
+	val |= (3 << SS2CD);
+	val |= SENDIAN;
+
+	val &= ~(CDRS_MASK << CDRS_SHIFT);
+	val |= ((div >> 1) << CDRS_SHIFT);
+
+	writel_relaxed(val, sspi->regs + PCC(cs));
+
+	val = readl_relaxed(sspi->regs + FIFOCFG);
+	val &= ~(BPW_MASK << BPW_SHIFT);
+	val |= ((bpw / 8 - 1) << BPW_SHIFT);
+	writel_relaxed(val, sspi->regs + FIFOCFG);
+
+	val = readl_relaxed(sspi->regs + DMSTART);
+	val &= ~(DATA_MASK << DATA_SHIFT);
+
+	if (xfer->tx_buf && xfer->rx_buf)
+		val |= (DATA_TXRX << DATA_SHIFT);
+	else if (xfer->rx_buf)
+		val |= (DATA_RX << DATA_SHIFT);
+	else
+		val |= (DATA_TX << DATA_SHIFT);
+
+	val &= ~(3 << BUS_WIDTH);
+	val |= ((bus_width >> 1) << BUS_WIDTH);
+	writel_relaxed(val, sspi->regs + DMSTART);
+
+	sspi->bpw = bpw;
+	sspi->mode = mode;
+	sspi->speed = speed;
+	sspi->cs = spi->chip_select;
+	sspi->bus_width = bus_width;
+
+	return 0;
+}
+
+static int synquacer_spi_transfer_one(struct spi_master *master,
+				      struct spi_device *spi,
+				      struct spi_transfer *xfer)
+{
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+	unsigned long bpw, flags;
+	int ret, words, busy = 0;
+	u32 val;
+
+	val = readl_relaxed(sspi->regs + FIFOCFG);
+	val |= RX_FLUSH;
+	val |= TX_FLUSH;
+	writel_relaxed(val, sspi->regs + FIFOCFG);
+
+	/* See if we can transfer 4-bytes as 1 word even if not asked */
+	bpw = xfer->bits_per_word;
+	if (bpw == 8 && !(xfer->len % 4) && !(spi->mode & SPI_LSB_FIRST))
+		xfer->bits_per_word = 32;
+
+	ret = synquacer_spi_config(master, spi, xfer);
+
+	/* restore */
+	xfer->bits_per_word = bpw;
+
+	if (ret)
+		return ret;
+
+	sspi->tx_buf = xfer->tx_buf;
+	sspi->rx_buf = xfer->rx_buf;
+
+	switch (sspi->bpw) {
+	case 8:
+		words = xfer->len;
+		break;
+	case 16:
+		words = xfer->len / 2;
+		break;
+	default:
+		words = xfer->len / 4;
+		break;
+	}
+
+	if (xfer->tx_buf) {
+		busy |= TXBIT;
+		sspi->tx_words = words;
+	} else {
+		sspi->tx_words = 0;
+	}
+
+	if (xfer->rx_buf) {
+		busy |= RXBIT;
+		sspi->rx_words = words;
+	} else {
+		sspi->rx_words = 0;
+	}
+
+	if (xfer->tx_buf)
+		write_fifo(sspi);
+
+	if (xfer->rx_buf) {
+		val = readl_relaxed(sspi->regs + FIFOCFG);
+		val &= ~(RX_TRSHLD_MASK << RX_TRSHLD_SHIFT);
+		val |= ((sspi->rx_words > FIFO_DEPTH ?
+			RX_TRSHLD : sspi->rx_words) << RX_TRSHLD_SHIFT);
+		writel_relaxed(val, sspi->regs + FIFOCFG);
+	}
+
+	writel_relaxed(~0, sspi->regs + TXC);
+	writel_relaxed(~0, sspi->regs + RXC);
+
+	/* Trigger */
+	val = readl_relaxed(sspi->regs + DMSTART);
+	val |= TRIGGER;
+	writel_relaxed(val, sspi->regs + DMSTART);
+
+	while (busy & (RXBIT | TXBIT)) {
+		if (sspi->rx_words)
+			read_fifo(sspi);
+		else
+			busy &= ~RXBIT;
+
+		if (sspi->tx_words) {
+			write_fifo(sspi);
+		} else {
+			u32 len;
+
+			do { /* wait for shifter to empty out */
+				cpu_relax();
+				len = readl_relaxed(sspi->regs + DMSTATUS);
+				len = (len >> TX_DATA_SHIFT) & TX_DATA_MASK;
+			} while (xfer->tx_buf && len);
+			busy &= ~TXBIT;
+		}
+	}
+
+	return 0;
+}
+
+static void synquacer_spi_set_cs(struct spi_device *spi, bool enable)
+{
+	struct synquacer_spi *sspi = spi_master_get_devdata(spi->master);
+	u32 val;
+
+	val = readl_relaxed(sspi->regs + DMSTART);
+	val &= ~(CS_MASK << CS_SHIFT);
+	val |= spi->chip_select << CS_SHIFT;
+
+	if (!enable) {
+		writel_relaxed(val, sspi->regs + DMSTART);
+
+		val = readl_relaxed(sspi->regs + DMSTART);
+		val &= ~DMSTOP;
+		writel_relaxed(val, sspi->regs + DMSTART);
+	} else {
+		val |= DMSTOP;
+		writel_relaxed(val, sspi->regs + DMSTART);
+
+		if (sspi->rx_buf) {
+			u32 buf[16];
+
+			sspi->rx_buf = buf;
+			sspi->rx_words = 16;
+			read_fifo(sspi);
+		}
+	}
+}
+
+static int synquacer_spi_enable(struct spi_master *master)
+{
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+	u32 val;
+
+	/* Disable module */
+	writel_relaxed(0, sspi->regs + MCTRL);
+	val = 0xfffff;
+	while (--val && (readl_relaxed(sspi->regs + MCTRL) & MES))
+		cpu_relax();
+	if (!val)
+		return -EBUSY;
+
+	writel_relaxed(0, sspi->regs + TXE);
+	writel_relaxed(0, sspi->regs + RXE);
+	val = readl_relaxed(sspi->regs + TXF);
+	writel_relaxed(val, sspi->regs + TXC);
+	val = readl_relaxed(sspi->regs + RXF);
+	writel_relaxed(val, sspi->regs + RXC);
+	val = readl_relaxed(sspi->regs + FAULTF);
+	writel_relaxed(val, sspi->regs + FAULTC);
+
+	val = readl_relaxed(sspi->regs + DMCFG);
+	val &= ~SSDC;
+	val &= ~MSTARTEN;
+	writel_relaxed(val, sspi->regs + DMCFG);
+
+	val = readl_relaxed(sspi->regs + MCTRL);
+	if (!strcmp(__clk_get_name(sspi->clk), "iHCLK"))
+		val &= ~IPCLK;
+	else
+		val |= IPCLK;
+
+	val &= ~CSEN;
+	val |= MEN;
+	val |= SYNCON;
+	writel_relaxed(val, sspi->regs + MCTRL);
+
+	return 0;
+}
+
+static int synquacer_spi_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct spi_master *master;
+	struct synquacer_spi *sspi;
+	struct resource *res;
+	u32 num_cs = 1;
+	int ret;
+
+	master = spi_alloc_master(&pdev->dev, sizeof(*sspi));
+	if (!master)
+		return -ENOMEM;
+	platform_set_drvdata(pdev, master);
+
+	sspi = spi_master_get_devdata(master);
+	sspi->dev = &pdev->dev;
+	sspi->master = master;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	sspi->regs = devm_ioremap_resource(sspi->dev, res);
+	if (IS_ERR(sspi->regs)) {
+		ret = PTR_ERR(sspi->regs);
+		goto put_spi;
+	}
+
+	sspi->clk = devm_clk_get(sspi->dev, "iHCLK");
+	if (IS_ERR(sspi->clk)) {
+		sspi->clk = devm_clk_get(sspi->dev, "iPCLK");
+		if (IS_ERR(sspi->clk)) {
+			dev_err(&pdev->dev, "No source clock\n");
+			ret = PTR_ERR(sspi->clk);
+			goto put_spi;
+		}
+	}
+
+	sspi->aces = of_property_read_bool(np, "socionext,set-aces");
+	sspi->rtm = of_property_read_bool(np, "socionext,use-rtm");
+
+	of_property_read_u32(np, "num-cs", &num_cs);
+	master->num_chipselect = num_cs;
+
+	ret = clk_prepare_enable(sspi->clk);
+	if (ret)
+		goto put_spi;
+
+	master->dev.of_node = np;
+	master->auto_runtime_pm = true;
+	master->bus_num = pdev->id;
+
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_TX_DUAL | SPI_RX_DUAL |
+				SPI_TX_QUAD | SPI_RX_QUAD;
+	master->bits_per_word_mask = SPI_BPW_MASK(32) | SPI_BPW_MASK(24)
+					 | SPI_BPW_MASK(16) | SPI_BPW_MASK(8);
+	master->max_speed_hz = clk_get_rate(sspi->clk);
+	master->min_speed_hz = master->max_speed_hz / 254;
+
+	master->set_cs = synquacer_spi_set_cs;
+	master->transfer_one = synquacer_spi_transfer_one;
+
+	ret = synquacer_spi_enable(master);
+	if (ret)
+		goto fail_enable;
+
+	pm_runtime_set_active(sspi->dev);
+	pm_runtime_enable(sspi->dev);
+
+	ret = devm_spi_register_master(sspi->dev, master);
+	if (ret)
+		goto disable_pm;
+
+	return 0;
+
+disable_pm:
+	pm_runtime_disable(sspi->dev);
+fail_enable:
+	clk_disable_unprepare(sspi->clk);
+put_spi:
+	spi_master_put(master);
+
+	return ret;
+}
+
+static int synquacer_spi_remove(struct platform_device *pdev)
+{
+	struct spi_master *master = platform_get_drvdata(pdev);
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+
+	pm_runtime_disable(sspi->dev);
+	clk_disable_unprepare(sspi->clk);
+	spi_master_put(master);
+
+	return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int synquacer_spi_suspend(struct device *dev)
+{
+	struct spi_master *master = dev_get_drvdata(dev);
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+	int ret;
+
+	ret = spi_master_suspend(master);
+	if (ret)
+		return ret;
+
+	if (!pm_runtime_suspended(dev))
+		clk_disable_unprepare(sspi->clk);
+
+	return ret;
+}
+
+static int synquacer_spi_resume(struct device *dev)
+{
+	struct spi_master *master = dev_get_drvdata(dev);
+	struct synquacer_spi *sspi = spi_master_get_devdata(master);
+	int ret;
+
+	if (!pm_runtime_suspended(dev)) {
+		/* Ensure reconfigure during next xfer */
+		sspi->speed = 0;
+
+		ret = clk_prepare_enable(sspi->clk);
+		if (ret < 0) {
+			dev_err(dev, "failed to enable clk (%d)\n", ret);
+			return ret;
+		}
+
+		ret = synquacer_spi_enable(master);
+		if (ret) {
+			dev_err(dev, "failed to enable spi (%d)\n", ret);
+			return ret;
+		}
+	}
+
+	ret = spi_master_resume(master);
+	if (ret < 0)
+		clk_disable_unprepare(sspi->clk);
+
+	return ret;
+}
+#endif /* CONFIG_PM_SLEEP */
+
+static const struct dev_pm_ops synquacer_spi_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(synquacer_spi_suspend, synquacer_spi_resume)
+};
+
+static const struct of_device_id synquacer_spi_of_match[] = {
+	{.compatible = "socionext,synquacer-spi",},
+	{},
+};
+MODULE_DEVICE_TABLE(of, synquacer_spi_of_match);
+
+static struct platform_driver synquacer_spi_driver = {
+	.driver = {
+		.name = "synquacer-spi",
+		.pm = &synquacer_spi_pm_ops,
+		.of_match_table = of_match_ptr(synquacer_spi_of_match),
+	},
+	.probe = synquacer_spi_probe,
+	.remove = synquacer_spi_remove,
+};
+module_platform_driver(synquacer_spi_driver);
+
+MODULE_DESCRIPTION("Socionext Synquacer HS-SPI controller driver");
+MODULE_AUTHOR("Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>");
+MODULE_LICENSE("GPL v2");
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCHv2 3/3] MAINTAINERS: Add entry for Synquacer SPI driver

[jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w]

From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>

Add entry for the Synquacer spi driver and DT bindings.

Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 MAINTAINERS | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 1899480..da79a2a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -12627,6 +12627,13 @@ F:	drivers/md/raid*
 F:	include/linux/raid/
 F:	include/uapi/linux/raid/
 
+SOCIONEXT (SNI) Synquacer SPI DRIVER
+M:	Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
+L:	linux-spi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
+S:	Maintained
+F:	drivers/spi/spi-synquacer.c
+F:	Documentation/devicetree/bindings/spi/spi-synquacer.txt
+
 SONIC NETWORK DRIVER
 M:	Thomas Bogendoerfer <tsbogend-I1c7kopa9pxLokYuJOExCg@public.gmane.org>
 L:	netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Rob Herring]

On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>
> This patch adds documentation for Device-Tree bindings for the
> Socionext Synquacer spi driver.
>
> Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> ---
>  .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
>  1 file changed, 24 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
>
> diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
> new file mode 100644
> index 0000000..d013cfd
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
> @@ -0,0 +1,24 @@
> +* Socionext Synquacer HS-SPI bindings
> +
> +Required Properties:
> +- compatible: should be "socionext,synquacer-spi"
> +- reg: physical base address of the controller and length of memory mapped
> +       region.
> +- clocks: Must contain an entry for rate source clock(s).
> +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK

Huh? The clock binding should reflect all clocks connected to a block,
not a selection of which one you want to use.

Rob
--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Jassi Brar]

On Mon, Jan 15, 2018 at 8:45 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
> On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>> From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>
>> This patch adds documentation for Device-Tree bindings for the
>> Socionext Synquacer spi driver.
>>
>> Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>> ---
>>  .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
>>  1 file changed, 24 insertions(+)
>>  create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>
>> diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>> new file mode 100644
>> index 0000000..d013cfd
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>> @@ -0,0 +1,24 @@
>> +* Socionext Synquacer HS-SPI bindings
>> +
>> +Required Properties:
>> +- compatible: should be "socionext,synquacer-spi"
>> +- reg: physical base address of the controller and length of memory mapped
>> +       region.
>> +- clocks: Must contain an entry for rate source clock(s).
>> +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
>
> Huh? The clock binding should reflect all clocks connected to a block,
> not a selection of which one you want to use.
>
Both the clocks are internal to the block and derived from the same source.
Instead of defining a new "use-ipclk" property, the driver uses the
clock-names to choose the appropriate divider.
I am open to any better option.

Thanks
--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Rob Herring]

On Mon, Jan 15, 2018 at 9:53 AM, Jassi Brar <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> On Mon, Jan 15, 2018 at 8:45 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
>> On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>>> From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>>
>>> This patch adds documentation for Device-Tree bindings for the
>>> Socionext Synquacer spi driver.
>>>
>>> Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>> ---
>>>  .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
>>>  1 file changed, 24 insertions(+)
>>>  create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>
>>> diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>> new file mode 100644
>>> index 0000000..d013cfd
>>> --- /dev/null
>>> +++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>> @@ -0,0 +1,24 @@
>>> +* Socionext Synquacer HS-SPI bindings
>>> +
>>> +Required Properties:
>>> +- compatible: should be "socionext,synquacer-spi"
>>> +- reg: physical base address of the controller and length of memory mapped
>>> +       region.
>>> +- clocks: Must contain an entry for rate source clock(s).
>>> +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
>>
>> Huh? The clock binding should reflect all clocks connected to a block,
>> not a selection of which one you want to use.
>>
> Both the clocks are internal to the block and derived from the same source.
> Instead of defining a new "use-ipclk" property, the driver uses the
> clock-names to choose the appropriate divider.
> I am open to any better option.

If one is preferred, then why not always use it? Or how does one
decide which clock to use?

Rob
--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Jassi Brar]

On Mon, Jan 15, 2018 at 10:47 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
> On Mon, Jan 15, 2018 at 9:53 AM, Jassi Brar <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>> On Mon, Jan 15, 2018 at 8:45 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
>>> On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>>>> From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>>>
>>>> This patch adds documentation for Device-Tree bindings for the
>>>> Socionext Synquacer spi driver.
>>>>
>>>> Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>>> ---
>>>>  .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
>>>>  1 file changed, 24 insertions(+)
>>>>  create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>> new file mode 100644
>>>> index 0000000..d013cfd
>>>> --- /dev/null
>>>> +++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>> @@ -0,0 +1,24 @@
>>>> +* Socionext Synquacer HS-SPI bindings
>>>> +
>>>> +Required Properties:
>>>> +- compatible: should be "socionext,synquacer-spi"
>>>> +- reg: physical base address of the controller and length of memory mapped
>>>> +       region.
>>>> +- clocks: Must contain an entry for rate source clock(s).
>>>> +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
>>>
>>> Huh? The clock binding should reflect all clocks connected to a block,
>>> not a selection of which one you want to use.
>>>
>> Both the clocks are internal to the block and derived from the same source.
>> Instead of defining a new "use-ipclk" property, the driver uses the
>> clock-names to choose the appropriate divider.
>> I am open to any better option.
>
> If one is preferred, then why not always use it? Or how does one
> decide which clock to use?
>
A slight correction, there is a mux inside the block which selects
clock from two input ports (iPCLK and iHCLK) and send that to the
divider. Depending upon the spi slave speed requirements the platform
may choose to connect either PCLK or HCLK (or maybe both but switching
is said to be not feasible without block reset and DT can't suggest
switch in runtime so we ask DT to provide only the source clock).

Now the idea dawns that DT provide both clocks and the driver select
from the mux looking at speed requirements of the slave at runtime.
However, then, there will be a wide range of speeds that both could
provide. Not to forget the block drives 4 slaves and optimising speed
for one could potentially break other slaves. Practically the h/w
designer would have already thought of speed requirements of the
slaves and made sure that port is populated .... the driver could
simple count upon that good design.

Thanks
--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Trent Piepho]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 2390 bytes --]

On Tue, 2018-01-16 at 08:59 +0530, Jassi Brar wrote:
> On Mon, Jan 15, 2018 at 10:47 PM, Rob Herring <robh+dt@kernel.org> wrote:
> > On Mon, Jan 15, 2018 at 9:53 AM, Jassi Brar <jassisinghbrar@gmail.com> wrote:
> > > On Mon, Jan 15, 2018 at 8:45 PM, Rob Herring <robh+dt@kernel.org> wrote:
> > > > On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar@gmail.com> wrote:
> > > > > 
> > > > > +- clocks: Must contain an entry for rate source clock(s).
> > > > > +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
> > > > 
> > > > Huh? The clock binding should reflect all clocks connected to a block,
> > > > not a selection of which one you want to use.
> > > > 
> > > 
> > > Both the clocks are internal to the block and derived from the same source.
> > > Instead of defining a new "use-ipclk" property, the driver uses the
> > > clock-names to choose the appropriate divider.
> > > I am open to any better option.
> > 
> > If one is preferred, then why not always use it? Or how does one
> > decide which clock to use?
> > 
> 
> A slight correction, there is a mux inside the block which selects
> clock from two input ports (iPCLK and iHCLK) and send that to the
> divider. Depending upon the spi slave speed requirements the platform

It seems like the DT is a reasonable description of how the hardware is
designed.  It's possible to choose different clocks with different
rates to be wired to PCLK and/or HCLK, via some internal clock tree in
the SoC?  That's what the DT would indicate.

You could maybe allow one or both to be connected, and if both are
connected, have the driver prefer one based on some driver logic. 
Which would allow one to then...

> Now the idea dawns that DT provide both clocks and the driver select
> from the mux looking at speed requirements of the slave at runtime.

add this feature at some point.  It might be simpler to just prefer
HCLK if both are there.  The idea being that the DT should describe the
hardware rather than tell the driver what to do.

Speed can be changed on a xfer by xfer basis on a single message to a
single slave.  Trying to switch clocks during a message, if it requires
resetting the whole block, sounds rather difficult.

N‹§²æìr¸›yúèšØb²X¬¶Ç§vØ^–)Þº{.nÇ+‰·zøœzÚÞz)í…æèw*\x1fjg¬±¨\x1e¶‰šŽŠÝ¢j.ïÛ°\½½MŽúgjÌæa×\x02››–' ™©Þ¢¸\f¢·¦j:+v‰¨ŠwèjØm¶Ÿÿ¾\a«‘êçzZ+ƒùšŽŠÝ¢j"ú!¶i

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Jassi Brar]

On Mon, Jan 15, 2018 at 10:47 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
> On Mon, Jan 15, 2018 at 9:53 AM, Jassi Brar <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>> On Mon, Jan 15, 2018 at 8:45 PM, Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> wrote:
>>> On Mon, Jan 15, 2018 at 7:05 AM,  <jassisinghbrar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
>>>> From: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>>>
>>>> This patch adds documentation for Device-Tree bindings for the
>>>> Socionext Synquacer spi driver.
>>>>
>>>> Signed-off-by: Jassi Brar <jaswinder.singh-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>>>> ---
>>>>  .../devicetree/bindings/spi/spi-synquacer.txt      | 24 ++++++++++++++++++++++
>>>>  1 file changed, 24 insertions(+)
>>>>  create mode 100644 Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/spi/spi-synquacer.txt b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>> new file mode 100644
>>>> index 0000000..d013cfd
>>>> --- /dev/null
>>>> +++ b/Documentation/devicetree/bindings/spi/spi-synquacer.txt
>>>> @@ -0,0 +1,24 @@
>>>> +* Socionext Synquacer HS-SPI bindings
>>>> +
>>>> +Required Properties:
>>>> +- compatible: should be "socionext,synquacer-spi"
>>>> +- reg: physical base address of the controller and length of memory mapped
>>>> +       region.
>>>> +- clocks: Must contain an entry for rate source clock(s).
>>>> +- clock-names: Shall be "iHCLK" or "iPCLK". iHCLK is preferred over iPCLK
>>>
>>> Huh? The clock binding should reflect all clocks connected to a block,
>>> not a selection of which one you want to use.
>>>
>> Both the clocks are internal to the block and derived from the same source.
>> Instead of defining a new "use-ipclk" property, the driver uses the
>> clock-names to choose the appropriate divider.
>> I am open to any better option.
>
> If one is preferred, then why not always use it? Or how does one
> decide which clock to use?
>
So I got my native speaker friend to interpret the Japanese datasheet.

The iHCLK clocks the whole block and can also be the source for
internal dividers. An external, optional, source iPCLK is also
provided for producing lower (Fpclk < Fhclk) rates from the dividers.
That is, iHCLK is always needed, while iPCLK can be provided if that
is to be fed to the dividers.

So the dt bindings should define one required clock (iHCLK) and
another optional clock (iPCLK). The driver would use iPCLK if
provided, otherwise the default iHCLK.

Will update DT and driver for next submission.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCHv2 1/3] dt-bindings: spi: Add DT bindings for Synquacer

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 795 bytes --]

On Tue, Jan 16, 2018 at 08:02:06PM +0000, Trent Piepho wrote:
> On Tue, 2018-01-16 at 08:59 +0530, Jassi Brar wrote:

> > Now the idea dawns that DT provide both clocks and the driver select
> > from the mux looking at speed requirements of the slave at runtime.

> add this feature at some point.  It might be simpler to just prefer
> HCLK if both are there.  The idea being that the DT should describe the
> hardware rather than tell the driver what to do.

Or only try one if the other fails to give something.

> Speed can be changed on a xfer by xfer basis on a single message to a
> single slave.  Trying to switch clocks during a message, if it requires
> resetting the whole block, sounds rather difficult.

In practice that's so vanishingly rare it's probably not worth worrying
about.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]