Merge tag 'spi-v5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi

Pull spi updates from Mark Brown:
 "The big update this time around is some excellent work from David
  Jander who went through the fast path and really eliminated overheads,
  meaning that we are seeing a huge reduction in the time spent between
  transfers for single threaded clients.

  Benchmarking has been coming out at about a halving of overhead which
  is clearly visible in system level usage that stresses SPI like some
  CAN and IIO applications, especially with small transfers. Thanks to
  David for taking the time to drill down into this and push the work
  upstream.

  Otherwise there's been a bunch of new device support and the usual
  updates.

   - Optimisation of the fast path, particularly around the number and
     types of locking operations, from David Jander.

   - Support for Arbel NPCM845, HP GXP, Intel Meteor Lake and Thunder
     Bay, MediaTek MT8188 and MT8365, Microchip FPGAs, nVidia Tegra 241
     and Samsung Exynos Auto v9 and 4210"

* tag 'spi-v5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (97 commits)
  MAINTAINERS: add spi support to GXP
  spi: dt-bindings: add documentation for hpe,gxp-spifi
  spi: spi-gxp: Add support for HPE GXP SoCs
  spi: a3700: support BE for AC5 SPI driver
  spi/panel: dt-bindings: drop CPHA and CPOL from common properties
  spi: bcm2835: enable shared interrupt support
  spi: dt-bindings: spi-controller: correct example indentation
  spi: dt-bindings: qcom,spi-geni-qcom: allow three interconnects
  spi: npcm-fiu: Add NPCM8XX support
  dt-binding: spi: Add npcm845 compatible to npcm-fiu document
  spi: npcm-fiu: Modify direct read dummy configuration
  spi: atmel: remove #ifdef CONFIG_{PM, SLEEP}
  spi: dt-bindings: Add compatible for MediaTek MT8188
  spi: dt-bindings: mediatek,spi-mtk-nor: Update bindings for nor flash
  spi: dt-bindings: atmel,at91rm9200-spi: convert to json-schema
  spi: tegra20-slink: fix UAF in tegra_slink_remove()
  spi: Fix simplification of devm_spi_register_controller
  spi: microchip-core: switch to use dev_err_probe()
  spi: microchip-core: switch to use devm_spi_alloc_master()
  spi: microchip-core: fix UAF in mchp_corespi_remove()
  ...

31 files changed, 1807 insertions, 578 deletions

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 35ce57878b27..e32f6a2058ae 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -371,6 +371,13 @@ config SPI_FSL_QUADSPI
 	  This controller does not support generic SPI messages. It only
 	  supports the high-level SPI memory interface.
 
+config SPI_GXP
+	tristate "GXP SPI driver"
+	depends on ARCH_HPE || COMPILE_TEST
+	help
+	  This enables support for the driver for GXP bus attached SPI
+	  controllers.
+
 config SPI_HISI_KUNPENG
 	tristate "HiSilicon SPI Controller for Kunpeng SoCs"
 	depends on (ARM64 && ACPI) || COMPILE_TEST
@@ -575,6 +582,15 @@ config SPI_MESON_SPIFC
 	  This enables master mode support for the SPIFC (SPI flash
 	  controller) available in Amlogic Meson SoCs.
 
+config SPI_MICROCHIP_CORE
+	tristate "Microchip FPGA SPI controllers"
+	depends on SPI_MASTER
+	help
+	  This enables the SPI driver for Microchip FPGA SPI controllers.
+	  Say Y or M here if you want to use the "hard" controllers on
+	  PolarFire SoC.
+	  If built as a module, it will be called spi-microchip-core.
+
 config SPI_MT65XX
 	tristate "MediaTek SPI controller"
 	depends on ARCH_MEDIATEK || COMPILE_TEST
diff --git a/drivers/spi/Makefile b/drivers/spi/Makefile
index 0f44eb6083a5..15d2f3835e45 100644
--- a/drivers/spi/Makefile
+++ b/drivers/spi/Makefile
@@ -57,6 +57,7 @@ obj-$(CONFIG_SPI_FSL_LPSPI)		+= spi-fsl-lpspi.o
 obj-$(CONFIG_SPI_FSL_QUADSPI)		+= spi-fsl-qspi.o
 obj-$(CONFIG_SPI_FSL_SPI)		+= spi-fsl-spi.o
 obj-$(CONFIG_SPI_GPIO)			+= spi-gpio.o
+obj-$(CONFIG_SPI_GXP)			+= spi-gxp.o
 obj-$(CONFIG_SPI_HISI_KUNPENG)		+= spi-hisi-kunpeng.o
 obj-$(CONFIG_SPI_HISI_SFC_V3XX)		+= spi-hisi-sfc-v3xx.o
 obj-$(CONFIG_SPI_IMG_SPFI)		+= spi-img-spfi.o
@@ -71,6 +72,7 @@ obj-$(CONFIG_SPI_LM70_LLP)		+= spi-lm70llp.o
 obj-$(CONFIG_SPI_LP8841_RTC)		+= spi-lp8841-rtc.o
 obj-$(CONFIG_SPI_MESON_SPICC)		+= spi-meson-spicc.o
 obj-$(CONFIG_SPI_MESON_SPIFC)		+= spi-meson-spifc.o
+obj-$(CONFIG_SPI_MICROCHIP_CORE)	+= spi-microchip-core.o
 obj-$(CONFIG_SPI_MPC512x_PSC)		+= spi-mpc512x-psc.o
 obj-$(CONFIG_SPI_MPC52xx_PSC)		+= spi-mpc52xx-psc.o
 obj-$(CONFIG_SPI_MPC52xx)		+= spi-mpc52xx.o
diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c
index 480c0c8c18e4..976a217e356d 100644
--- a/drivers/spi/atmel-quadspi.c
+++ b/drivers/spi/atmel-quadspi.c
@@ -21,6 +21,7 @@
 #include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 #include <linux/spi/spi-mem.h>
 
 /* QSPI register offsets */
@@ -285,7 +286,7 @@ static bool atmel_qspi_supports_op(struct spi_mem *mem,
 
 	/* special case not supported by hardware */
 	if (op->addr.nbytes == 2 && op->cmd.buswidth != op->addr.buswidth &&
-		op->dummy.nbytes == 0)
+	    op->dummy.nbytes == 0)
 		return false;
 
 	return true;
@@ -417,9 +418,13 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 	if (op->addr.val + op->data.nbytes > aq->mmap_size)
 		return -ENOTSUPP;
 
+	err = pm_runtime_resume_and_get(&aq->pdev->dev);
+	if (err < 0)
+		return err;
+
 	err = atmel_qspi_set_cfg(aq, op, &offset);
 	if (err)
-		return err;
+		goto pm_runtime_put;
 
 	/* Skip to the final steps if there is no data */
 	if (op->data.nbytes) {
@@ -441,7 +446,7 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 	/* Poll INSTRuction End status */
 	sr = atmel_qspi_read(aq, QSPI_SR);
 	if ((sr & QSPI_SR_CMD_COMPLETED) == QSPI_SR_CMD_COMPLETED)
-		return err;
+		goto pm_runtime_put;
 
 	/* Wait for INSTRuction End interrupt */
 	reinit_completion(&aq->cmd_completion);
@@ -452,6 +457,9 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 		err = -ETIMEDOUT;
 	atmel_qspi_write(QSPI_SR_CMD_COMPLETED, aq, QSPI_IDR);
 
+pm_runtime_put:
+	pm_runtime_mark_last_busy(&aq->pdev->dev);
+	pm_runtime_put_autosuspend(&aq->pdev->dev);
 	return err;
 }
 
@@ -472,6 +480,7 @@ static int atmel_qspi_setup(struct spi_device *spi)
 	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
 	unsigned long src_rate;
 	u32 scbr;
+	int ret;
 
 	if (ctrl->busy)
 		return -EBUSY;
@@ -488,9 +497,16 @@ static int atmel_qspi_setup(struct spi_device *spi)
 	if (scbr > 0)
 		scbr--;
 
+	ret = pm_runtime_resume_and_get(ctrl->dev.parent);
+	if (ret < 0)
+		return ret;
+
 	aq->scr = QSPI_SCR_SCBR(scbr);
 	atmel_qspi_write(aq->scr, aq, QSPI_SCR);
 
+	pm_runtime_mark_last_busy(ctrl->dev.parent);
+	pm_runtime_put_autosuspend(ctrl->dev.parent);
+
 	return 0;
 }
 
@@ -621,11 +637,24 @@ static int atmel_qspi_probe(struct platform_device *pdev)
 	if (err)
 		goto disable_qspick;
 
+	pm_runtime_set_autosuspend_delay(&pdev->dev, 500);
+	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
+	pm_runtime_enable(&pdev->dev);
+	pm_runtime_get_noresume(&pdev->dev);
+
 	atmel_qspi_init(aq);
 
 	err = spi_register_controller(ctrl);
-	if (err)
+	if (err) {
+		pm_runtime_put_noidle(&pdev->dev);
+		pm_runtime_disable(&pdev->dev);
+		pm_runtime_set_suspended(&pdev->dev);
+		pm_runtime_dont_use_autosuspend(&pdev->dev);
 		goto disable_qspick;
+	}
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_put_autosuspend(&pdev->dev);
 
 	return 0;
 
@@ -641,9 +670,18 @@ static int atmel_qspi_remove(struct platform_device *pdev)
 {
 	struct spi_controller *ctrl = platform_get_drvdata(pdev);
 	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
+	int ret;
+
+	ret = pm_runtime_resume_and_get(&pdev->dev);
+	if (ret < 0)
+		return ret;
 
 	spi_unregister_controller(ctrl);
 	atmel_qspi_write(QSPI_CR_QSPIDIS, aq, QSPI_CR);
+
+	pm_runtime_disable(&pdev->dev);
+	pm_runtime_put_noidle(&pdev->dev);
+
 	clk_disable_unprepare(aq->qspick);
 	clk_disable_unprepare(aq->pclk);
 	return 0;
@@ -653,10 +691,19 @@ static int __maybe_unused atmel_qspi_suspend(struct device *dev)
 {
 	struct spi_controller *ctrl = dev_get_drvdata(dev);
 	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
+	int ret;
+
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret < 0)
+		return ret;
 
 	atmel_qspi_write(QSPI_CR_QSPIDIS, aq, QSPI_CR);
-	clk_disable_unprepare(aq->qspick);
-	clk_disable_unprepare(aq->pclk);
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_force_suspend(dev);
+
+	clk_unprepare(aq->qspick);
+	clk_unprepare(aq->pclk);
 
 	return 0;
 }
@@ -665,19 +712,54 @@ static int __maybe_unused atmel_qspi_resume(struct device *dev)
 {
 	struct spi_controller *ctrl = dev_get_drvdata(dev);
 	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
+	int ret;
 
-	clk_prepare_enable(aq->pclk);
-	clk_prepare_enable(aq->qspick);
+	clk_prepare(aq->pclk);
+	clk_prepare(aq->qspick);
+
+	ret = pm_runtime_force_resume(dev);
+	if (ret < 0)
+		return ret;
 
 	atmel_qspi_init(aq);
 
 	atmel_qspi_write(aq->scr, aq, QSPI_SCR);
 
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	return 0;
+}
+
+static int __maybe_unused atmel_qspi_runtime_suspend(struct device *dev)
+{
+	struct spi_controller *ctrl = dev_get_drvdata(dev);
+	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
+
+	clk_disable(aq->qspick);
+	clk_disable(aq->pclk);
+
 	return 0;
 }
 
-static SIMPLE_DEV_PM_OPS(atmel_qspi_pm_ops, atmel_qspi_suspend,
-			 atmel_qspi_resume);
+static int __maybe_unused atmel_qspi_runtime_resume(struct device *dev)
+{
+	struct spi_controller *ctrl = dev_get_drvdata(dev);
+	struct atmel_qspi *aq = spi_controller_get_devdata(ctrl);
+	int ret;
+
+	ret = clk_enable(aq->pclk);
+	if (ret)
+		return ret;
+
+	return clk_enable(aq->qspick);
+}
+
+static const struct dev_pm_ops __maybe_unused atmel_qspi_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(atmel_qspi_suspend, atmel_qspi_resume)
+	SET_RUNTIME_PM_OPS(atmel_qspi_runtime_suspend,
+			   atmel_qspi_runtime_resume, NULL)
+};
 
 static const struct atmel_qspi_caps atmel_sama5d2_qspi_caps = {};
 
@@ -704,7 +786,7 @@ static struct platform_driver atmel_qspi_driver = {
 	.driver = {
 		.name	= "atmel_qspi",
 		.of_match_table	= atmel_qspi_dt_ids,
-		.pm	= &atmel_qspi_pm_ops,
+		.pm	= pm_ptr(&atmel_qspi_pm_ops),
 	},
 	.probe		= atmel_qspi_probe,
 	.remove		= atmel_qspi_remove,
diff --git a/drivers/spi/spi-altera-dfl.c b/drivers/spi/spi-altera-dfl.c
index ca40923258af..596e181ae136 100644
--- a/drivers/spi/spi-altera-dfl.c
+++ b/drivers/spi/spi-altera-dfl.c
@@ -128,9 +128,9 @@ static int dfl_spi_altera_probe(struct dfl_device *dfl_dev)
 	struct spi_master *master;
 	struct altera_spi *hw;
 	void __iomem *base;
-	int err = -ENODEV;
+	int err;
 
-	master = spi_alloc_master(dev, sizeof(struct altera_spi));
+	master = devm_spi_alloc_master(dev, sizeof(struct altera_spi));
 	if (!master)
 		return -ENOMEM;
 
@@ -159,10 +159,9 @@ static int dfl_spi_altera_probe(struct dfl_device *dfl_dev)
 	altera_spi_init_master(master);
 
 	err = devm_spi_register_master(dev, master);
-	if (err) {
-		dev_err(dev, "%s failed to register spi master %d\n", __func__, err);
-		goto exit;
-	}
+	if (err)
+		return dev_err_probe(dev, err, "%s failed to register spi master\n",
+				     __func__);
 
 	if (dfl_dev->revision == FME_FEATURE_REV_MAX10_SPI_N5010)
 		strscpy(board_info.modalias, "m10-n5010", SPI_NAME_SIZE);
@@ -179,9 +178,6 @@ static int dfl_spi_altera_probe(struct dfl_device *dfl_dev)
 	}
 
 	return 0;
-exit:
-	spi_master_put(master);
-	return err;
 }
 
 static const struct dfl_device_id dfl_spi_altera_ids[] = {
diff --git a/drivers/spi/spi-amd.c b/drivers/spi/spi-amd.c
index efdcbe6c4c26..08df4f8d0531 100644
--- a/drivers/spi/spi-amd.c
+++ b/drivers/spi/spi-amd.c
@@ -40,14 +40,23 @@
 #define AMD_SPI_XFER_TX		1
 #define AMD_SPI_XFER_RX		2
 
+/**
+ * enum amd_spi_versions - SPI controller versions
+ * @AMD_SPI_V1:		AMDI0061 hardware version
+ * @AMD_SPI_V2:		AMDI0062 hardware version
+ */
 enum amd_spi_versions {
-	AMD_SPI_V1 = 1,	/* AMDI0061 */
-	AMD_SPI_V2,	/* AMDI0062 */
+	AMD_SPI_V1 = 1,
+	AMD_SPI_V2,
 };
 
+/**
+ * struct amd_spi - SPI driver instance
+ * @io_remap_addr:	Start address of the SPI controller registers
+ * @version:		SPI controller hardware version
+ */
 struct amd_spi {
 	void __iomem *io_remap_addr;
-	unsigned long io_base_addr;
 	enum amd_spi_versions version;
 };
 
@@ -281,22 +290,19 @@ static int amd_spi_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct spi_master *master;
 	struct amd_spi *amd_spi;
-	int err = 0;
+	int err;
 
 	/* Allocate storage for spi_master and driver private data */
-	master = spi_alloc_master(dev, sizeof(struct amd_spi));
-	if (!master) {
-		dev_err(dev, "Error allocating SPI master\n");
-		return -ENOMEM;
-	}
+	master = devm_spi_alloc_master(dev, sizeof(struct amd_spi));
+	if (!master)
+		return dev_err_probe(dev, -ENOMEM, "Error allocating SPI master\n");
 
 	amd_spi = spi_master_get_devdata(master);
 	amd_spi->io_remap_addr = devm_platform_ioremap_resource(pdev, 0);
-	if (IS_ERR(amd_spi->io_remap_addr)) {
-		err = PTR_ERR(amd_spi->io_remap_addr);
-		dev_err(dev, "error %d ioremap of SPI registers failed\n", err);
-		goto err_free_master;
-	}
+	if (IS_ERR(amd_spi->io_remap_addr))
+		return dev_err_probe(dev, PTR_ERR(amd_spi->io_remap_addr),
+				     "ioremap of SPI registers failed\n");
+
 	dev_dbg(dev, "io_remap_address: %p\n", amd_spi->io_remap_addr);
 
 	amd_spi->version = (enum amd_spi_versions) device_get_match_data(dev);
@@ -313,17 +319,10 @@ static int amd_spi_probe(struct platform_device *pdev)
 
 	/* Register the controller with SPI framework */
 	err = devm_spi_register_master(dev, master);
-	if (err) {
-		dev_err(dev, "error %d registering SPI controller\n", err);
-		goto err_free_master;
-	}
+	if (err)
+		return dev_err_probe(dev, err, "error registering SPI controller\n");
 
 	return 0;
-
-err_free_master:
-	spi_master_put(master);
-
-	return err;
 }
 
 #ifdef CONFIG_ACPI
diff --git a/drivers/spi/spi-armada-3700.c b/drivers/spi/spi-armada-3700.c
index d8cc4b270644..9df9fc40b783 100644
--- a/drivers/spi/spi-armada-3700.c
+++ b/drivers/spi/spi-armada-3700.c
@@ -497,7 +497,7 @@ static int a3700_spi_fifo_write(struct a3700_spi *a3700_spi)
 
 	while (!a3700_is_wfifo_full(a3700_spi) && a3700_spi->buf_len) {
 		val = *(u32 *)a3700_spi->tx_buf;
-		spireg_write(a3700_spi, A3700_SPI_DATA_OUT_REG, val);
+		spireg_write(a3700_spi, A3700_SPI_DATA_OUT_REG, cpu_to_le32(val));
 		a3700_spi->buf_len -= 4;
 		a3700_spi->tx_buf += 4;
 	}
@@ -519,7 +519,7 @@ static int a3700_spi_fifo_read(struct a3700_spi *a3700_spi)
 	while (!a3700_is_rfifo_empty(a3700_spi) && a3700_spi->buf_len) {
 		val = spireg_read(a3700_spi, A3700_SPI_DATA_IN_REG);
 		if (a3700_spi->buf_len >= 4) {
-
+			val = le32_to_cpu(val);
 			memcpy(a3700_spi->rx_buf, &val, 4);
 
 			a3700_spi->buf_len -= 4;
diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c
index 9e300a932699..c4f22d50dba5 100644
--- a/drivers/spi/spi-atmel.c
+++ b/drivers/spi/spi-atmel.c
@@ -1631,7 +1631,6 @@ static int atmel_spi_remove(struct platform_device *pdev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
 static int atmel_spi_runtime_suspend(struct device *dev)
 {
 	struct spi_master *master = dev_get_drvdata(dev);
@@ -1653,7 +1652,6 @@ static int atmel_spi_runtime_resume(struct device *dev)
 	return clk_prepare_enable(as->clk);
 }
 
-#ifdef CONFIG_PM_SLEEP
 static int atmel_spi_suspend(struct device *dev)
 {
 	struct spi_master *master = dev_get_drvdata(dev);
@@ -1693,17 +1691,12 @@ static int atmel_spi_resume(struct device *dev)
 	/* Start the queue running */
 	return spi_master_resume(master);
 }
-#endif
 
 static const struct dev_pm_ops atmel_spi_pm_ops = {
-	SET_SYSTEM_SLEEP_PM_OPS(atmel_spi_suspend, atmel_spi_resume)
-	SET_RUNTIME_PM_OPS(atmel_spi_runtime_suspend,
-			   atmel_spi_runtime_resume, NULL)
+	SYSTEM_SLEEP_PM_OPS(atmel_spi_suspend, atmel_spi_resume)
+	RUNTIME_PM_OPS(atmel_spi_runtime_suspend,
+		       atmel_spi_runtime_resume, NULL)
 };
-#define ATMEL_SPI_PM_OPS	(&atmel_spi_pm_ops)
-#else
-#define ATMEL_SPI_PM_OPS	NULL
-#endif
 
 static const struct of_device_id atmel_spi_dt_ids[] = {
 	{ .compatible = "atmel,at91rm9200-spi" },
@@ -1715,7 +1708,7 @@ MODULE_DEVICE_TABLE(of, atmel_spi_dt_ids);
 static struct platform_driver atmel_spi_driver = {
 	.driver		= {
 		.name	= "atmel_spi",
-		.pm	= ATMEL_SPI_PM_OPS,
+		.pm	= pm_ptr(&atmel_spi_pm_ops),
 		.of_match_table	= atmel_spi_dt_ids,
 	},
 	.probe		= atmel_spi_probe,
diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c
index 0933948d7df3..747e03228c48 100644
--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -372,6 +372,10 @@ static irqreturn_t bcm2835_spi_interrupt(int irq, void *dev_id)
 	struct bcm2835_spi *bs = dev_id;
 	u32 cs = bcm2835_rd(bs, BCM2835_SPI_CS);
 
+	/* Bail out early if interrupts are not enabled */
+	if (!(cs & BCM2835_SPI_CS_INTR))
+		return IRQ_NONE;
+
 	/*
 	 * An interrupt is signaled either if DONE is set (TX FIFO empty)
 	 * or if RXR is set (RX FIFO >= ¾ full).
@@ -1369,8 +1373,8 @@ static int bcm2835_spi_probe(struct platform_device *pdev)
 	bcm2835_wr(bs, BCM2835_SPI_CS,
 		   BCM2835_SPI_CS_CLEAR_RX | BCM2835_SPI_CS_CLEAR_TX);
 
-	err = devm_request_irq(&pdev->dev, bs->irq, bcm2835_spi_interrupt, 0,
-			       dev_name(&pdev->dev), bs);
+	err = devm_request_irq(&pdev->dev, bs->irq, bcm2835_spi_interrupt,
+			       IRQF_SHARED, dev_name(&pdev->dev), bs);
 	if (err) {
 		dev_err(&pdev->dev, "could not request IRQ: %d\n", err);
 		goto out_dma_release;
diff --git a/drivers/spi/spi-dw-core.c b/drivers/spi/spi-dw-core.c
index ecea471ff42c..f87d97ccd2d6 100644
--- a/drivers/spi/spi-dw-core.c
+++ b/drivers/spi/spi-dw-core.c
@@ -307,8 +307,9 @@ static u32 dw_spi_prepare_cr0(struct dw_spi *dws, struct spi_device *spi)
 		if (spi->mode & SPI_LOOP)
 			cr0 |= DW_HSSI_CTRLR0_SRL;
 
-		if (dws->caps & DW_SPI_CAP_KEEMBAY_MST)
-			cr0 |= DW_HSSI_CTRLR0_KEEMBAY_MST;
+		/* CTRLR0[31] MST */
+		if (dw_spi_ver_is_ge(dws, HSSI, 102A))
+			cr0 |= DW_HSSI_CTRLR0_MST;
 	}
 
 	return cr0;
@@ -942,7 +943,9 @@ int dw_spi_add_host(struct device *dev, struct dw_spi *dws)
 
 	if (dws->dma_ops && dws->dma_ops->dma_init) {
 		ret = dws->dma_ops->dma_init(dev, dws);
-		if (ret) {
+		if (ret == -EPROBE_DEFER) {
+			goto err_free_irq;
+		} else if (ret) {
 			dev_warn(dev, "DMA init failed\n");
 		} else {
 			master->can_dma = dws->dma_ops->can_dma;
@@ -963,6 +966,7 @@ err_dma_exit:
 	if (dws->dma_ops && dws->dma_ops->dma_exit)
 		dws->dma_ops->dma_exit(dws);
 	dw_spi_enable_chip(dws, 0);
+err_free_irq:
 	free_irq(dws->irq, master);
 err_free_master:
 	spi_controller_put(master);
diff --git a/drivers/spi/spi-dw-dma.c b/drivers/spi/spi-dw-dma.c
index 63e5260100ec..1322b8cce5b7 100644
--- a/drivers/spi/spi-dw-dma.c
+++ b/drivers/spi/spi-dw-dma.c
@@ -139,15 +139,20 @@ err_exit:
 
 static int dw_spi_dma_init_generic(struct device *dev, struct dw_spi *dws)
 {
-	dws->rxchan = dma_request_slave_channel(dev, "rx");
-	if (!dws->rxchan)
-		return -ENODEV;
+	int ret;
 
-	dws->txchan = dma_request_slave_channel(dev, "tx");
-	if (!dws->txchan) {
-		dma_release_channel(dws->rxchan);
+	dws->rxchan = dma_request_chan(dev, "rx");
+	if (IS_ERR(dws->rxchan)) {
+		ret = PTR_ERR(dws->rxchan);
 		dws->rxchan = NULL;
-		return -ENODEV;
+		goto err_exit;
+	}
+
+	dws->txchan = dma_request_chan(dev, "tx");
+	if (IS_ERR(dws->txchan)) {
+		ret = PTR_ERR(dws->txchan);
+		dws->txchan = NULL;
+		goto free_rxchan;
 	}
 
 	dws->master->dma_rx = dws->rxchan;
@@ -160,6 +165,12 @@ static int dw_spi_dma_init_generic(struct device *dev, struct dw_spi *dws)
 	dw_spi_dma_sg_burst_init(dws);
 
 	return 0;
+
+free_rxchan:
+	dma_release_channel(dws->rxchan);
+	dws->rxchan = NULL;
+err_exit:
+	return ret;
 }
 
 static void dw_spi_dma_exit(struct dw_spi *dws)
diff --git a/drivers/spi/spi-dw-mmio.c b/drivers/spi/spi-dw-mmio.c
index 5101c4c6017b..26c40ea6dd12 100644
--- a/drivers/spi/spi-dw-mmio.c
+++ b/drivers/spi/spi-dw-mmio.c
@@ -214,11 +214,10 @@ static int dw_spi_hssi_init(struct platform_device *pdev,
 	return 0;
 }
 
-static int dw_spi_keembay_init(struct platform_device *pdev,
-			       struct dw_spi_mmio *dwsmmio)
+static int dw_spi_intel_init(struct platform_device *pdev,
+			     struct dw_spi_mmio *dwsmmio)
 {
 	dwsmmio->dws.ip = DW_HSSI_ID;
-	dwsmmio->dws.caps = DW_SPI_CAP_KEEMBAY_MST;
 
 	return 0;
 }
@@ -349,7 +348,8 @@ static const struct of_device_id dw_spi_mmio_of_match[] = {
 	{ .compatible = "amazon,alpine-dw-apb-ssi", .data = dw_spi_alpine_init},
 	{ .compatible = "renesas,rzn1-spi", .data = dw_spi_pssi_init},
 	{ .compatible = "snps,dwc-ssi-1.01a", .data = dw_spi_hssi_init},
-	{ .compatible = "intel,keembay-ssi", .data = dw_spi_keembay_init},
+	{ .compatible = "intel,keembay-ssi", .data = dw_spi_intel_init},
+	{ .compatible = "intel,thunderbay-ssi", .data = dw_spi_intel_init},
 	{ .compatible = "microchip,sparx5-spi", dw_spi_mscc_sparx5_init},
 	{ .compatible = "canaan,k210-spi", dw_spi_canaan_k210_init},
 	{ /* end of table */}
diff --git a/drivers/spi/spi-dw.h b/drivers/spi/spi-dw.h
index d5ee5130601e..9e8eb2b52d5c 100644
--- a/drivers/spi/spi-dw.h
+++ b/drivers/spi/spi-dw.h
@@ -23,7 +23,7 @@
 	((_dws)->ip == DW_ ## _ip ## _ID)
 
 #define __dw_spi_ver_cmp(_dws, _ip, _ver, _op) \
-	(dw_spi_ip_is(_dws, _ip) && (_dws)->ver _op DW_ ## _ip ## _ver)
+	(dw_spi_ip_is(_dws, _ip) && (_dws)->ver _op DW_ ## _ip ## _ ## _ver)
 
 #define dw_spi_ver_is(_dws, _ip, _ver) __dw_spi_ver_cmp(_dws, _ip, _ver, ==)
 
@@ -31,8 +31,7 @@
 
 /* DW SPI controller capabilities */
 #define DW_SPI_CAP_CS_OVERRIDE		BIT(0)
-#define DW_SPI_CAP_KEEMBAY_MST		BIT(1)
-#define DW_SPI_CAP_DFS32		BIT(2)
+#define DW_SPI_CAP_DFS32		BIT(1)
 
 /* Register offsets (Generic for both DWC APB SSI and DWC SSI IP-cores) */
 #define DW_SPI_CTRLR0			0x00
@@ -94,13 +93,7 @@
 #define DW_HSSI_CTRLR0_SCPOL			BIT(9)
 #define DW_HSSI_CTRLR0_TMOD_MASK		GENMASK(11, 10)
 #define DW_HSSI_CTRLR0_SRL			BIT(13)
-
-/*
- * For Keem Bay, CTRLR0[31] is used to select controller mode.
- * 0: SSI is slave
- * 1: SSI is master
- */
-#define DW_HSSI_CTRLR0_KEEMBAY_MST		BIT(31)
+#define DW_HSSI_CTRLR0_MST			BIT(31)
 
 /* Bit fields in CTRLR1 */
 #define DW_SPI_NDF_MASK				GENMASK(15, 0)
diff --git a/drivers/spi/spi-fsi.c b/drivers/spi/spi-fsi.c
index 72ab066ce552..cf1e4f9ebd72 100644
--- a/drivers/spi/spi-fsi.c
+++ b/drivers/spi/spi-fsi.c
@@ -24,8 +24,7 @@
 #define FSI2SPI_IRQ			0x20
 
 #define SPI_FSI_BASE			0x70000
-#define SPI_FSI_INIT_TIMEOUT_MS		1000
-#define SPI_FSI_STATUS_TIMEOUT_MS	100
+#define SPI_FSI_TIMEOUT_MS		1000
 #define SPI_FSI_MAX_RX_SIZE		8
 #define SPI_FSI_MAX_TX_SIZE		40
 
@@ -299,6 +298,7 @@ static void fsi_spi_sequence_init(struct fsi_spi_sequence *seq)
 static int fsi_spi_transfer_data(struct fsi_spi *ctx,
 				 struct spi_transfer *transfer)
 {
+	int loops;
 	int rc = 0;
 	unsigned long end;
 	u64 status = 0ULL;
@@ -317,9 +317,10 @@ static int fsi_spi_transfer_data(struct fsi_spi *ctx,
 			if (rc)
 				return rc;
 
-			end = jiffies + msecs_to_jiffies(SPI_FSI_STATUS_TIMEOUT_MS);
+			loops = 0;
+			end = jiffies + msecs_to_jiffies(SPI_FSI_TIMEOUT_MS);
 			do {
-				if (time_after(jiffies, end))
+				if (loops++ && time_after(jiffies, end))
 					return -ETIMEDOUT;
 
 				rc = fsi_spi_status(ctx, &status, "TX");
@@ -335,9 +336,10 @@ static int fsi_spi_transfer_data(struct fsi_spi *ctx,
 		u8 *rx = transfer->rx_buf;
 
 		while (transfer->len > recv) {
-			end = jiffies + msecs_to_jiffies(SPI_FSI_STATUS_TIMEOUT_MS);
+			loops = 0;
+			end = jiffies + msecs_to_jiffies(SPI_FSI_TIMEOUT_MS);
 			do {
-				if (time_after(jiffies, end))
+				if (loops++ && time_after(jiffies, end))
 					return -ETIMEDOUT;
 
 				rc = fsi_spi_status(ctx, &status, "RX");
@@ -359,6 +361,7 @@ static int fsi_spi_transfer_data(struct fsi_spi *ctx,
 
 static int fsi_spi_transfer_init(struct fsi_spi *ctx)
 {
+	int loops = 0;
 	int rc;
 	bool reset = false;
 	unsigned long end;
@@ -369,9 +372,9 @@ static int fsi_spi_transfer_init(struct fsi_spi *ctx)
 		SPI_FSI_CLOCK_CFG_SCK_NO_DEL |
 		FIELD_PREP(SPI_FSI_CLOCK_CFG_SCK_DIV, 19);
 
-	end = jiffies + msecs_to_jiffies(SPI_FSI_INIT_TIMEOUT_MS);
+	end = jiffies + msecs_to_jiffies(SPI_FSI_TIMEOUT_MS);
 	do {
-		if (time_after(jiffies, end))
+		if (loops++ && time_after(jiffies, end))
 			return -ETIMEDOUT;
 
 		rc = fsi_spi_read_reg(ctx, SPI_FSI_STATUS, &status);
diff --git a/drivers/spi/spi-gxp.c b/drivers/spi/spi-gxp.c
new file mode 100644
index 000000000000..9ea355f7d64f
--- /dev/null
+++ b/drivers/spi/spi-gxp.c
@@ -0,0 +1,325 @@
+// SPDX-License-Identifier: GPL-2.0=or-later
+/* Copyright (C) 2022 Hewlett-Packard Development Company, L.P. */
+
+#include <linux/iopoll.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/spi-mem.h>
+
+#define GXP_SPI0_MAX_CHIPSELECT	2
+#define GXP_SPI_SLEEP_TIME	1
+#define GXP_SPI_TIMEOUT (130 * 1000000 / GXP_SPI_SLEEP_TIME)
+
+#define MANUAL_MODE		0
+#define DIRECT_MODE		1
+#define SPILDAT_LEN		256
+
+#define OFFSET_SPIMCFG		0x0
+#define OFFSET_SPIMCTRL		0x4
+#define OFFSET_SPICMD		0x5
+#define OFFSET_SPIDCNT		0x6
+#define OFFSET_SPIADDR		0x8
+#define OFFSET_SPIINTSTS	0xc
+
+#define SPIMCTRL_START		0x01
+#define SPIMCTRL_BUSY		0x02
+#define SPIMCTRL_DIR		0x08
+
+struct gxp_spi;
+
+struct gxp_spi_chip {
+	struct gxp_spi *spifi;
+	u32 cs;
+};
+
+struct gxp_spi_data {
+	u32 max_cs;
+	u32 mode_bits;
+};
+
+struct gxp_spi {
+	const struct gxp_spi_data *data;
+	void __iomem *reg_base;
+	void __iomem *dat_base;
+	void __iomem *dir_base;
+	struct device *dev;
+	struct gxp_spi_chip chips[GXP_SPI0_MAX_CHIPSELECT];
+};
+
+static void gxp_spi_set_mode(struct gxp_spi *spifi, int mode)
+{
+	u8 value;
+	void __iomem *reg_base = spifi->reg_base;
+
+	value = readb(reg_base + OFFSET_SPIMCTRL);
+
+	if (mode == MANUAL_MODE) {
+		writeb(0x55, reg_base + OFFSET_SPICMD);
+		writeb(0xaa, reg_base + OFFSET_SPICMD);
+		value &= ~0x30;
+	} else {
+		value |= 0x30;
+	}
+	writeb(value, reg_base + OFFSET_SPIMCTRL);
+}
+
+static int gxp_spi_read_reg(struct gxp_spi_chip *chip, const struct spi_mem_op *op)
+{
+	int ret;
+	struct gxp_spi *spifi = chip->spifi;
+	void __iomem *reg_base = spifi->reg_base;
+	u32 value;
+
+	value = readl(reg_base + OFFSET_SPIMCFG);
+	value &= ~(1 << 24);
+	value |= (chip->cs << 24);
+	value &= ~(0x07 << 16);
+	value &= ~(0x1f << 19);
+	writel(value, reg_base + OFFSET_SPIMCFG);
+
+	writel(0, reg_base + OFFSET_SPIADDR);
+
+	writeb(op->cmd.opcode, reg_base + OFFSET_SPICMD);
+
+	writew(op->data.nbytes, reg_base + OFFSET_SPIDCNT);
+
+	value = readb(reg_base + OFFSET_SPIMCTRL);
+	value &= ~SPIMCTRL_DIR;
+	value |= SPIMCTRL_START;
+
+	writeb(value, reg_base + OFFSET_SPIMCTRL);
+
+	ret = readb_poll_timeout(reg_base + OFFSET_SPIMCTRL, value,
+				 !(value & SPIMCTRL_BUSY),
+				 GXP_SPI_SLEEP_TIME, GXP_SPI_TIMEOUT);
+	if (ret) {
+		dev_warn(spifi->dev, "read reg busy time out\n");
+		return ret;
+	}
+
+	memcpy_fromio(op->data.buf.in, spifi->dat_base, op->data.nbytes);
+	return ret;
+}
+
+static int gxp_spi_write_reg(struct gxp_spi_chip *chip, const struct spi_mem_op *op)
+{
+	int ret;
+	struct gxp_spi *spifi = chip->spifi;
+	void __iomem *reg_base = spifi->reg_base;
+	u32 value;
+
+	value = readl(reg_base + OFFSET_SPIMCFG);
+	value &= ~(1 << 24);
+	value |= (chip->cs << 24);
+	value &= ~(0x07 << 16);
+	value &= ~(0x1f << 19);
+	writel(value, reg_base + OFFSET_SPIMCFG);
+
+	writel(0, reg_base + OFFSET_SPIADDR);
+
+	writeb(op->cmd.opcode, reg_base + OFFSET_SPICMD);
+
+	memcpy_toio(spifi->dat_base, op->data.buf.in, op->data.nbytes);
+
+	writew(op->data.nbytes, reg_base + OFFSET_SPIDCNT);
+
+	value = readb(reg_base + OFFSET_SPIMCTRL);
+	value |= SPIMCTRL_DIR;
+	value |= SPIMCTRL_START;
+
+	writeb(value, reg_base + OFFSET_SPIMCTRL);
+
+	ret = readb_poll_timeout(reg_base + OFFSET_SPIMCTRL, value,
+				 !(value & SPIMCTRL_BUSY),
+				 GXP_SPI_SLEEP_TIME, GXP_SPI_TIMEOUT);
+	if (ret)
+		dev_warn(spifi->dev, "write reg busy time out\n");
+
+	return ret;
+}
+
+static ssize_t gxp_spi_read(struct gxp_spi_chip *chip, const struct spi_mem_op *op)
+{
+	struct gxp_spi *spifi = chip->spifi;
+	u32 offset = op->addr.val;
+
+	if (chip->cs == 0)
+		offset += 0x4000000;
+
+	memcpy_fromio(op->data.buf.in, spifi->dir_base + offset, op->data.nbytes);
+
+	return 0;
+}
+
+static ssize_t gxp_spi_write(struct gxp_spi_chip *chip, const struct spi_mem_op *op)
+{
+	struct gxp_spi *spifi = chip->spifi;
+	void __iomem *reg_base = spifi->reg_base;
+	u32 write_len;
+	u32 value;
+	int ret;
+
+	write_len = op->data.nbytes;
+	if (write_len > SPILDAT_LEN)
+		write_len = SPILDAT_LEN;
+
+	value = readl(reg_base + OFFSET_SPIMCFG);
+	value &= ~(1 << 24);
+	value |= (chip->cs << 24);
+	value &= ~(0x07 << 16);
+	value |= (op->addr.nbytes << 16);
+	value &= ~(0x1f << 19);
+	writel(value, reg_base + OFFSET_SPIMCFG);
+
+	writel(op->addr.val, reg_base + OFFSET_SPIADDR);
+
+	writeb(op->cmd.opcode, reg_base + OFFSET_SPICMD);
+
+	writew(write_len, reg_base + OFFSET_SPIDCNT);
+
+	memcpy_toio(spifi->dat_base, op->data.buf.in, write_len);
+
+	value = readb(reg_base + OFFSET_SPIMCTRL);
+	value |= SPIMCTRL_DIR;
+	value |= SPIMCTRL_START;
+
+	writeb(value, reg_base + OFFSET_SPIMCTRL);
+
+	ret = readb_poll_timeout(reg_base + OFFSET_SPIMCTRL, value,
+				 !(value & SPIMCTRL_BUSY),
+				 GXP_SPI_SLEEP_TIME, GXP_SPI_TIMEOUT);
+	if (ret) {
+		dev_warn(spifi->dev, "write busy time out\n");
+		return ret;
+	}
+
+	return write_len;
+}
+
+static int do_gxp_exec_mem_op(struct spi_mem *mem, const struct spi_mem_op *op)
+{
+	struct gxp_spi *spifi = spi_controller_get_devdata(mem->spi->master);
+	struct gxp_spi_chip *chip = &spifi->chips[mem->spi->chip_select];
+	int ret;
+
+	if (op->data.dir == SPI_MEM_DATA_IN) {
+		if (!op->addr.nbytes)
+			ret = gxp_spi_read_reg(chip, op);
+		else
+			ret = gxp_spi_read(chip, op);
+	} else {
+		if (!op->addr.nbytes)
+			ret = gxp_spi_write_reg(chip, op);
+		else
+			ret = gxp_spi_write(chip, op);
+	}
+
+	return ret;
+}
+
+static int gxp_exec_mem_op(struct spi_mem *mem, const struct spi_mem_op *op)
+{
+	int ret;
+
+	ret = do_gxp_exec_mem_op(mem, op);
+	if (ret)
+		dev_err(&mem->spi->dev, "operation failed: %d", ret);
+
+	return ret;
+}
+
+static const struct spi_controller_mem_ops gxp_spi_mem_ops = {
+	.exec_op = gxp_exec_mem_op,
+};
+
+static int gxp_spi_setup(struct spi_device *spi)
+{
+	struct gxp_spi *spifi = spi_controller_get_devdata(spi->master);
+	unsigned int cs = spi->chip_select;
+	struct gxp_spi_chip *chip = &spifi->chips[cs];
+
+	chip->spifi = spifi;
+	chip->cs = cs;
+
+	gxp_spi_set_mode(spifi, MANUAL_MODE);
+
+	return 0;
+}
+
+static int gxp_spifi_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	const struct gxp_spi_data *data;
+	struct spi_controller *ctlr;
+	struct gxp_spi *spifi;
+	struct resource *res;
+	int ret;
+
+	data = of_device_get_match_data(&pdev->dev);
+
+	ctlr = devm_spi_alloc_master(dev, sizeof(*spifi));
+	if (!ctlr)
+		return -ENOMEM;
+
+	spifi = spi_controller_get_devdata(ctlr);
+
+	platform_set_drvdata(pdev, spifi);
+	spifi->data = data;
+	spifi->dev = dev;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	spifi->reg_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(spifi->reg_base))
+		return PTR_ERR(spifi->reg_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	spifi->dat_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(spifi->dat_base))
+		return PTR_ERR(spifi->dat_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 2);
+	spifi->dir_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(spifi->dir_base))
+		return PTR_ERR(spifi->dir_base);
+
+	ctlr->mode_bits = data->mode_bits;
+	ctlr->bus_num = pdev->id;
+	ctlr->mem_ops = &gxp_spi_mem_ops;
+	ctlr->setup = gxp_spi_setup;
+	ctlr->num_chipselect = data->max_cs;
+	ctlr->dev.of_node = dev->of_node;
+
+	ret = devm_spi_register_controller(dev, ctlr);
+	if (ret) {
+		return dev_err_probe(&pdev->dev, ret,
+				     "failed to register spi controller\n");
+	}
+
+	return 0;
+}
+
+static const struct gxp_spi_data gxp_spifi_data = {
+	.max_cs	= 2,
+	.mode_bits = 0,
+};
+
+static const struct of_device_id gxp_spifi_match[] = {
+	{.compatible = "hpe,gxp-spifi", .data = &gxp_spifi_data },
+	{ /* null */ }
+};
+MODULE_DEVICE_TABLE(of, gxp_spifi_match);
+
+static struct platform_driver gxp_spifi_driver = {
+	.probe = gxp_spifi_probe,
+	.driver = {
+		.name = "gxp-spifi",
+		.of_match_table = gxp_spifi_match,
+	},
+};
+module_platform_driver(gxp_spifi_driver);
+
+MODULE_DESCRIPTION("HPE GXP SPI Flash Interface driver");
+MODULE_AUTHOR("Nick Hawkins <nick.hawkins@hpe.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi-intel-pci.c b/drivers/spi/spi-intel-pci.c
index f6eec7a869b6..f0d532ea40e8 100644
--- a/drivers/spi/spi-intel-pci.c
+++ b/drivers/spi/spi-intel-pci.c
@@ -74,6 +74,7 @@ static const struct pci_device_id intel_spi_pci_ids[] = {
 	{ PCI_VDEVICE(INTEL, 0x54a4), (unsigned long)&cnl_info },
 	{ PCI_VDEVICE(INTEL, 0x7a24), (unsigned long)&cnl_info },
 	{ PCI_VDEVICE(INTEL, 0x7aa4), (unsigned long)&cnl_info },
+	{ PCI_VDEVICE(INTEL, 0x7e23), (unsigned long)&cnl_info },
 	{ PCI_VDEVICE(INTEL, 0xa0a4), (unsigned long)&bxt_info },
 	{ PCI_VDEVICE(INTEL, 0xa1a4), (unsigned long)&bxt_info },
 	{ PCI_VDEVICE(INTEL, 0xa224), (unsigned long)&bxt_info },
diff --git a/drivers/spi/spi-intel.c b/drivers/spi/spi-intel.c
index 50f42983b950..66063687ae27 100644
--- a/drivers/spi/spi-intel.c
+++ b/drivers/spi/spi-intel.c
@@ -1236,8 +1236,8 @@ static int intel_spi_populate_chip(struct intel_spi *ispi)
 		return -ENOMEM;
 
 	pdata->nr_parts = 1;
-	pdata->parts = devm_kcalloc(ispi->dev, sizeof(*pdata->parts),
-				    pdata->nr_parts, GFP_KERNEL);
+	pdata->parts = devm_kcalloc(ispi->dev, pdata->nr_parts,
+				    sizeof(*pdata->parts), GFP_KERNEL);
 	if (!pdata->parts)
 		return -ENOMEM;
 
diff --git a/drivers/spi/spi-microchip-core.c b/drivers/spi/spi-microchip-core.c
new file mode 100644
index 000000000000..ce4385330b19
--- /dev/null
+++ b/drivers/spi/spi-microchip-core.c
@@ -0,0 +1,617 @@
+// SPDX-License-Identifier: (GPL-2.0)
+/*
+ * Microchip CoreSPI SPI controller driver
+ *
+ * Copyright (c) 2018-2022 Microchip Technology Inc. and its subsidiaries
+ *
+ * Author: Daire McNamara <daire.mcnamara@microchip.com>
+ * Author: Conor Dooley <conor.dooley@microchip.com>
+ *
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/spi/spi.h>
+
+#define MAX_LEN				(0xffff)
+#define MAX_CS				(8)
+#define DEFAULT_FRAMESIZE		(8)
+#define FIFO_DEPTH			(32)
+#define CLK_GEN_MODE1_MAX		(255)
+#define CLK_GEN_MODE0_MAX		(15)
+#define CLK_GEN_MIN			(0)
+#define MODE_X_MASK_SHIFT		(24)
+
+#define CONTROL_ENABLE			BIT(0)
+#define CONTROL_MASTER			BIT(1)
+#define CONTROL_RX_DATA_INT		BIT(4)
+#define CONTROL_TX_DATA_INT		BIT(5)
+#define CONTROL_RX_OVER_INT		BIT(6)
+#define CONTROL_TX_UNDER_INT		BIT(7)
+#define CONTROL_SPO			BIT(24)
+#define CONTROL_SPH			BIT(25)
+#define CONTROL_SPS			BIT(26)
+#define CONTROL_FRAMEURUN		BIT(27)
+#define CONTROL_CLKMODE			BIT(28)
+#define CONTROL_BIGFIFO			BIT(29)
+#define CONTROL_OENOFF			BIT(30)
+#define CONTROL_RESET			BIT(31)
+
+#define CONTROL_MODE_MASK		GENMASK(3, 2)
+#define  MOTOROLA_MODE			(0)
+#define CONTROL_FRAMECNT_MASK		GENMASK(23, 8)
+#define CONTROL_FRAMECNT_SHIFT		(8)
+
+#define STATUS_ACTIVE			BIT(14)
+#define STATUS_SSEL			BIT(13)
+#define STATUS_FRAMESTART		BIT(12)
+#define STATUS_TXFIFO_EMPTY_NEXT_READ	BIT(11)
+#define STATUS_TXFIFO_EMPTY		BIT(10)
+#define STATUS_TXFIFO_FULL_NEXT_WRITE	BIT(9)
+#define STATUS_TXFIFO_FULL		BIT(8)
+#define STATUS_RXFIFO_EMPTY_NEXT_READ	BIT(7)
+#define STATUS_RXFIFO_EMPTY		BIT(6)
+#define STATUS_RXFIFO_FULL_NEXT_WRITE	BIT(5)
+#define STATUS_RXFIFO_FULL		BIT(4)
+#define STATUS_TX_UNDERRUN		BIT(3)
+#define STATUS_RX_OVERFLOW		BIT(2)
+#define STATUS_RXDAT_RXED		BIT(1)
+#define STATUS_TXDAT_SENT		BIT(0)
+
+#define INT_TXDONE			BIT(0)
+#define INT_RXRDY			BIT(1)
+#define INT_RX_CHANNEL_OVERFLOW		BIT(2)
+#define INT_TX_CHANNEL_UNDERRUN		BIT(3)
+
+#define INT_ENABLE_MASK (CONTROL_RX_DATA_INT | CONTROL_TX_DATA_INT | \
+			 CONTROL_RX_OVER_INT | CONTROL_TX_UNDER_INT)
+
+#define REG_CONTROL		(0x00)
+#define REG_FRAME_SIZE		(0x04)
+#define REG_STATUS		(0x08)
+#define REG_INT_CLEAR		(0x0c)
+#define REG_RX_DATA		(0x10)
+#define REG_TX_DATA		(0x14)
+#define REG_CLK_GEN		(0x18)
+#define REG_SLAVE_SELECT	(0x1c)
+#define  SSEL_MASK		GENMASK(7, 0)
+#define  SSEL_DIRECT		BIT(8)
+#define  SSELOUT_SHIFT		9
+#define  SSELOUT		BIT(SSELOUT_SHIFT)
+#define REG_MIS			(0x20)
+#define REG_RIS			(0x24)
+#define REG_CONTROL2		(0x28)
+#define REG_COMMAND		(0x2c)
+#define REG_PKTSIZE		(0x30)
+#define REG_CMD_SIZE		(0x34)
+#define REG_HWSTATUS		(0x38)
+#define REG_STAT8		(0x3c)
+#define REG_CTRL2		(0x48)
+#define REG_FRAMESUP		(0x50)
+
+struct mchp_corespi {
+	void __iomem *regs;
+	struct clk *clk;
+	const u8 *tx_buf;
+	u8 *rx_buf;
+	u32 clk_gen; /* divider for spi output clock generated by the controller */
+	u32 clk_mode;
+	int irq;
+	int tx_len;
+	int rx_len;
+	int pending;
+};
+
+static inline u32 mchp_corespi_read(struct mchp_corespi *spi, unsigned int reg)
+{
+	return readl(spi->regs + reg);
+}
+
+static inline void mchp_corespi_write(struct mchp_corespi *spi, unsigned int reg, u32 val)
+{
+	writel(val, spi->regs + reg);
+}
+
+static inline void mchp_corespi_enable(struct mchp_corespi *spi)
+{
+	u32 control = mchp_corespi_read(spi, REG_CONTROL);
+
+	control |= CONTROL_ENABLE;
+
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static inline void mchp_corespi_disable(struct mchp_corespi *spi)
+{
+	u32 control = mchp_corespi_read(spi, REG_CONTROL);
+
+	control &= ~CONTROL_ENABLE;
+
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static inline void mchp_corespi_read_fifo(struct mchp_corespi *spi)
+{
+	u8 data;
+	int fifo_max, i = 0;
+
+	fifo_max = min(spi->rx_len, FIFO_DEPTH);
+
+	while ((i < fifo_max) && !(mchp_corespi_read(spi, REG_STATUS) & STATUS_RXFIFO_EMPTY)) {
+		data = mchp_corespi_read(spi, REG_RX_DATA);
+
+		if (spi->rx_buf)
+			*spi->rx_buf++ = data;
+		i++;
+	}
+	spi->rx_len -= i;
+	spi->pending -= i;
+}
+
+static void mchp_corespi_enable_ints(struct mchp_corespi *spi)
+{
+	u32 control, mask = INT_ENABLE_MASK;
+
+	mchp_corespi_disable(spi);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+
+	control |= mask;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+
+	control |= CONTROL_ENABLE;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static void mchp_corespi_disable_ints(struct mchp_corespi *spi)
+{
+	u32 control, mask = INT_ENABLE_MASK;
+
+	mchp_corespi_disable(spi);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	control &= ~mask;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+
+	control |= CONTROL_ENABLE;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static inline void mchp_corespi_set_xfer_size(struct mchp_corespi *spi, int len)
+{
+	u32 control;
+	u16 lenpart;
+
+	/*
+	 * Disable the SPI controller. Writes to transfer length have
+	 * no effect when the controller is enabled.
+	 */
+	mchp_corespi_disable(spi);
+
+	/*
+	 * The lower 16 bits of the frame count are stored in the control reg
+	 * for legacy reasons, but the upper 16 written to a different register:
+	 * FRAMESUP. While both the upper and lower bits can be *READ* from the
+	 * FRAMESUP register, writing to the lower 16 bits is a NOP
+	 */
+	lenpart = len & 0xffff;
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	control &= ~CONTROL_FRAMECNT_MASK;
+	control |= lenpart << CONTROL_FRAMECNT_SHIFT;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+
+	lenpart = len & 0xffff0000;
+	mchp_corespi_write(spi, REG_FRAMESUP, lenpart);
+
+	control |= CONTROL_ENABLE;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static inline void mchp_corespi_write_fifo(struct mchp_corespi *spi)
+{
+	u8 byte;
+	int fifo_max, i = 0;
+
+	fifo_max = min(spi->tx_len, FIFO_DEPTH);
+	mchp_corespi_set_xfer_size(spi, fifo_max);
+
+	while ((i < fifo_max) && !(mchp_corespi_read(spi, REG_STATUS) & STATUS_TXFIFO_FULL)) {
+		byte = spi->tx_buf ? *spi->tx_buf++ : 0xaa;
+		mchp_corespi_write(spi, REG_TX_DATA, byte);
+		i++;
+	}
+
+	spi->tx_len -= i;
+	spi->pending += i;
+}
+
+static inline void mchp_corespi_set_framesize(struct mchp_corespi *spi, int bt)
+{
+	u32 control;
+
+	/*
+	 * Disable the SPI controller. Writes to the frame size have
+	 * no effect when the controller is enabled.
+	 */
+	mchp_corespi_disable(spi);
+
+	mchp_corespi_write(spi, REG_FRAME_SIZE, bt);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	control |= CONTROL_ENABLE;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static void mchp_corespi_set_cs(struct spi_device *spi, bool disable)
+{
+	u32 reg;
+	struct mchp_corespi *corespi = spi_master_get_devdata(spi->master);
+
+	reg = mchp_corespi_read(corespi, REG_SLAVE_SELECT);
+	reg &= ~BIT(spi->chip_select);
+	reg |= !disable << spi->chip_select;
+
+	mchp_corespi_write(corespi, REG_SLAVE_SELECT, reg);
+}
+
+static int mchp_corespi_setup(struct spi_device *spi)
+{
+	struct mchp_corespi *corespi = spi_master_get_devdata(spi->master);
+	u32 reg;
+
+	/*
+	 * Active high slaves need to be specifically set to their inactive
+	 * states during probe by adding them to the "control group" & thus
+	 * driving their select line low.
+	 */
+	if (spi->mode & SPI_CS_HIGH) {
+		reg = mchp_corespi_read(corespi, REG_SLAVE_SELECT);
+		reg |= BIT(spi->chip_select);
+		mchp_corespi_write(corespi, REG_SLAVE_SELECT, reg);
+	}
+	return 0;
+}
+
+static void mchp_corespi_init(struct spi_master *master, struct mchp_corespi *spi)
+{
+	unsigned long clk_hz;
+	u32 control = mchp_corespi_read(spi, REG_CONTROL);
+
+	control |= CONTROL_MASTER;
+
+	control &= ~CONTROL_MODE_MASK;
+	control |= MOTOROLA_MODE;
+
+	mchp_corespi_set_framesize(spi, DEFAULT_FRAMESIZE);
+
+	/* max. possible spi clock rate is the apb clock rate */
+	clk_hz = clk_get_rate(spi->clk);
+	master->max_speed_hz = clk_hz;
+
+	/*
+	 * The controller must be configured so that it doesn't remove Chip
+	 * Select until the entire message has been transferred, even if at
+	 * some points TX FIFO becomes empty.
+	 *
+	 * BIGFIFO mode is also enabled, which sets the fifo depth to 32 frames
+	 * for the 8 bit transfers that this driver uses.
+	 */
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	control |= CONTROL_SPS | CONTROL_BIGFIFO;
+
+	mchp_corespi_write(spi, REG_CONTROL, control);
+
+	mchp_corespi_enable_ints(spi);
+
+	/*
+	 * It is required to enable direct mode, otherwise control over the chip
+	 * select is relinquished to the hardware. SSELOUT is enabled too so we
+	 * can deal with active high slaves.
+	 */
+	mchp_corespi_write(spi, REG_SLAVE_SELECT, SSELOUT | SSEL_DIRECT);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+
+	control &= ~CONTROL_RESET;
+	control |= CONTROL_ENABLE;
+
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static inline void mchp_corespi_set_clk_gen(struct mchp_corespi *spi)
+{
+	u32 control;
+
+	mchp_corespi_disable(spi);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	if (spi->clk_mode)
+		control |= CONTROL_CLKMODE;
+	else
+		control &= ~CONTROL_CLKMODE;
+
+	mchp_corespi_write(spi, REG_CLK_GEN, spi->clk_gen);
+	mchp_corespi_write(spi, REG_CONTROL, control);
+	mchp_corespi_write(spi, REG_CONTROL, control | CONTROL_ENABLE);
+}
+
+static inline void mchp_corespi_set_mode(struct mchp_corespi *spi, unsigned int mode)
+{
+	u32 control, mode_val;
+
+	switch (mode & SPI_MODE_X_MASK) {
+	case SPI_MODE_0:
+		mode_val = 0;
+		break;
+	case SPI_MODE_1:
+		mode_val = CONTROL_SPH;
+		break;
+	case SPI_MODE_2:
+		mode_val = CONTROL_SPO;
+		break;
+	case SPI_MODE_3:
+		mode_val = CONTROL_SPH | CONTROL_SPO;
+		break;
+	}
+
+	/*
+	 * Disable the SPI controller. Writes to the frame size have
+	 * no effect when the controller is enabled.
+	 */
+	mchp_corespi_disable(spi);
+
+	control = mchp_corespi_read(spi, REG_CONTROL);
+	control &= ~(SPI_MODE_X_MASK << MODE_X_MASK_SHIFT);
+	control |= mode_val;
+
+	mchp_corespi_write(spi, REG_CONTROL, control);
+
+	control |= CONTROL_ENABLE;
+	mchp_corespi_write(spi, REG_CONTROL, control);
+}
+
+static irqreturn_t mchp_corespi_interrupt(int irq, void *dev_id)
+{
+	struct spi_master *master = dev_id;
+	struct mchp_corespi *spi = spi_master_get_devdata(master);
+	u32 intfield = mchp_corespi_read(spi, REG_MIS) & 0xf;
+	bool finalise = false;
+
+	/* Interrupt line may be shared and not for us at all */
+	if (intfield == 0)
+		return IRQ_NONE;
+
+	if (intfield & INT_TXDONE) {
+		mchp_corespi_write(spi, REG_INT_CLEAR, INT_TXDONE);
+
+		if (spi->rx_len)
+			mchp_corespi_read_fifo(spi);
+
+		if (spi->tx_len)
+			mchp_corespi_write_fifo(spi);
+
+		if (!spi->rx_len)
+			finalise = true;
+	}
+
+	if (intfield & INT_RXRDY)
+		mchp_corespi_write(spi, REG_INT_CLEAR, INT_RXRDY);
+
+	if (intfield & INT_RX_CHANNEL_OVERFLOW) {
+		mchp_corespi_write(spi, REG_INT_CLEAR, INT_RX_CHANNEL_OVERFLOW);
+		finalise = true;
+		dev_err(&master->dev,
+			"%s: RX OVERFLOW: rxlen: %d, txlen: %d\n", __func__,
+			spi->rx_len, spi->tx_len);
+	}
+
+	if (intfield & INT_TX_CHANNEL_UNDERRUN) {
+		mchp_corespi_write(spi, REG_INT_CLEAR, INT_TX_CHANNEL_UNDERRUN);
+		finalise = true;
+		dev_err(&master->dev,
+			"%s: TX UNDERFLOW: rxlen: %d, txlen: %d\n", __func__,
+			spi->rx_len, spi->tx_len);
+	}
+
+	if (finalise)
+		spi_finalize_current_transfer(master);
+
+	return IRQ_HANDLED;
+}
+
+static int mchp_corespi_calculate_clkgen(struct mchp_corespi *spi,
+					 unsigned long target_hz)
+{
+	unsigned long clk_hz, spi_hz, clk_gen;
+
+	clk_hz = clk_get_rate(spi->clk);
+	if (!clk_hz)
+		return -EINVAL;
+	spi_hz = min(target_hz, clk_hz);
+
+	/*
+	 * There are two possible clock modes for the controller generated
+	 * clock's division ratio:
+	 * CLK_MODE = 0: 1 / (2^(CLK_GEN + 1)) where CLK_GEN = 0 to 15.
+	 * CLK_MODE = 1: 1 / (2 * CLK_GEN + 1) where CLK_GEN = 0 to 255.
+	 * First try mode 1, fall back to 0 and if we have tried both modes and
+	 * we /still/ can't get a good setting, we then throw the toys out of
+	 * the pram and give up
+	 * clk_gen is the register name for the clock divider on MPFS.
+	 */
+	clk_gen = DIV_ROUND_UP(clk_hz, 2 * spi_hz) - 1;
+	if (clk_gen > CLK_GEN_MODE1_MAX || clk_gen <= CLK_GEN_MIN) {
+		clk_gen = DIV_ROUND_UP(clk_hz, spi_hz);
+		clk_gen = fls(clk_gen) - 1;
+
+		if (clk_gen > CLK_GEN_MODE0_MAX)
+			return -EINVAL;
+
+		spi->clk_mode = 0;
+	} else {
+		spi->clk_mode = 1;
+	}
+
+	spi->clk_gen = clk_gen;
+	return 0;
+}
+
+static int mchp_corespi_transfer_one(struct spi_master *master,
+				     struct spi_device *spi_dev,
+				     struct spi_transfer *xfer)
+{
+	struct mchp_corespi *spi = spi_master_get_devdata(master);
+	int ret;
+
+	ret = mchp_corespi_calculate_clkgen(spi, (unsigned long)xfer->speed_hz);
+	if (ret) {
+		dev_err(&master->dev, "failed to set clk_gen for target %u Hz\n", xfer->speed_hz);
+		return ret;
+	}
+
+	mchp_corespi_set_clk_gen(spi);
+
+	spi->tx_buf = xfer->tx_buf;
+	spi->rx_buf = xfer->rx_buf;
+	spi->tx_len = xfer->len;
+	spi->rx_len = xfer->len;
+	spi->pending = 0;
+
+	mchp_corespi_set_xfer_size(spi, (spi->tx_len > FIFO_DEPTH)
+				   ? FIFO_DEPTH : spi->tx_len);
+
+	if (spi->tx_len)
+		mchp_corespi_write_fifo(spi);
+	return 1;
+}
+
+static int mchp_corespi_prepare_message(struct spi_master *master,
+					struct spi_message *msg)
+{
+	struct spi_device *spi_dev = msg->spi;
+	struct mchp_corespi *spi = spi_master_get_devdata(master);
+
+	mchp_corespi_set_framesize(spi, DEFAULT_FRAMESIZE);
+	mchp_corespi_set_mode(spi, spi_dev->mode);
+
+	return 0;
+}
+
+static int mchp_corespi_probe(struct platform_device *pdev)
+{
+	struct spi_master *master;
+	struct mchp_corespi *spi;
+	struct resource *res;
+	u32 num_cs;
+	int ret = 0;
+
+	master = devm_spi_alloc_master(&pdev->dev, sizeof(*spi));
+	if (!master)
+		return dev_err_probe(&pdev->dev, -ENOMEM,
+				     "unable to allocate master for SPI controller\n");
+
+	platform_set_drvdata(pdev, master);
+
+	if (of_property_read_u32(pdev->dev.of_node, "num-cs", &num_cs))
+		num_cs = MAX_CS;
+
+	master->num_chipselect = num_cs;
+	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH;
+	master->setup = mchp_corespi_setup;
+	master->bits_per_word_mask = SPI_BPW_MASK(8);
+	master->transfer_one = mchp_corespi_transfer_one;
+	master->prepare_message = mchp_corespi_prepare_message;
+	master->set_cs = mchp_corespi_set_cs;
+	master->dev.of_node = pdev->dev.of_node;
+
+	spi = spi_master_get_devdata(master);
+
+	spi->regs = devm_platform_get_and_ioremap_resource(pdev, 0, &res);
+	if (IS_ERR(spi->regs))
+		return PTR_ERR(spi->regs);
+
+	spi->irq = platform_get_irq(pdev, 0);
+	if (spi->irq <= 0)
+		return dev_err_probe(&pdev->dev, -ENXIO,
+				     "invalid IRQ %d for SPI controller\n",
+				     spi->irq);
+
+	ret = devm_request_irq(&pdev->dev, spi->irq, mchp_corespi_interrupt,
+			       IRQF_SHARED, dev_name(&pdev->dev), master);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret,
+				     "could not request irq: %d\n", ret);
+
+	spi->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(spi->clk))
+		return dev_err_probe(&pdev->dev, PTR_ERR(spi->clk),
+				     "could not get clk: %d\n", ret);
+
+	ret = clk_prepare_enable(spi->clk);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret,
+				     "failed to enable clock\n");
+
+	mchp_corespi_init(master, spi);
+
+	ret = devm_spi_register_master(&pdev->dev, master);
+	if (ret) {
+		mchp_corespi_disable(spi);
+		clk_disable_unprepare(spi->clk);
+		return dev_err_probe(&pdev->dev, ret,
+				     "unable to register master for SPI controller\n");
+	}
+
+	dev_info(&pdev->dev, "Registered SPI controller %d\n", master->bus_num);
+
+	return 0;
+}
+
+static int mchp_corespi_remove(struct platform_device *pdev)
+{
+	struct spi_master *master  = platform_get_drvdata(pdev);
+	struct mchp_corespi *spi = spi_master_get_devdata(master);
+
+	mchp_corespi_disable_ints(spi);
+	clk_disable_unprepare(spi->clk);
+	mchp_corespi_disable(spi);
+
+	return 0;
+}
+
+#define MICROCHIP_SPI_PM_OPS (NULL)
+
+/*
+ * Platform driver data structure
+ */
+
+#if defined(CONFIG_OF)
+static const struct of_device_id mchp_corespi_dt_ids[] = {
+	{ .compatible = "microchip,mpfs-spi" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, mchp_corespi_dt_ids);
+#endif
+
+static struct platform_driver mchp_corespi_driver = {
+	.probe = mchp_corespi_probe,
+	.driver = {
+		.name = "microchip-corespi",
+		.pm = MICROCHIP_SPI_PM_OPS,
+		.of_match_table = of_match_ptr(mchp_corespi_dt_ids),
+	},
+	.remove = mchp_corespi_remove,
+};
+module_platform_driver(mchp_corespi_driver);
+MODULE_DESCRIPTION("Microchip coreSPI SPI controller driver");
+MODULE_AUTHOR("Daire McNamara <daire.mcnamara@microchip.com>");
+MODULE_AUTHOR("Conor Dooley <conor.dooley@microchip.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/spi/spi-mpc52xx-psc.c b/drivers/spi/spi-mpc52xx-psc.c
index 7654736c2c0e..609311231e64 100644
--- a/drivers/spi/spi-mpc52xx-psc.c
+++ b/drivers/spi/spi-mpc52xx-psc.c
@@ -37,12 +37,6 @@ struct mpc52xx_psc_spi {
 	struct mpc52xx_psc_fifo __iomem *fifo;
 	unsigned int irq;
 	u8 bits_per_word;
-	u8 busy;
-
-	struct work_struct work;
-
-	struct list_head queue;
-	spinlock_t lock;
 
 	struct completion done;
 };
@@ -198,69 +192,53 @@ static int mpc52xx_psc_spi_transfer_rxtx(struct spi_device *spi,
 	return 0;
 }
 
-static void mpc52xx_psc_spi_work(struct work_struct *work)
+int mpc52xx_psc_spi_transfer_one_message(struct spi_controller *ctlr,
+					 struct spi_message *m)
 {
-	struct mpc52xx_psc_spi *mps =
-		container_of(work, struct mpc52xx_psc_spi, work);
-
-	spin_lock_irq(&mps->lock);
-	mps->busy = 1;
-	while (!list_empty(&mps->queue)) {
-		struct spi_message *m;
-		struct spi_device *spi;
-		struct spi_transfer *t = NULL;
-		unsigned cs_change;
-		int status;
-
-		m = container_of(mps->queue.next, struct spi_message, queue);
-		list_del_init(&m->queue);
-		spin_unlock_irq(&mps->lock);
-
-		spi = m->spi;
-		cs_change = 1;
-		status = 0;
-		list_for_each_entry (t, &m->transfers, transfer_list) {
-			if (t->bits_per_word || t->speed_hz) {
-				status = mpc52xx_psc_spi_transfer_setup(spi, t);
-				if (status < 0)
-					break;
-			}
-
-			if (cs_change)
-				mpc52xx_psc_spi_activate_cs(spi);
-			cs_change = t->cs_change;
-
-			status = mpc52xx_psc_spi_transfer_rxtx(spi, t);
-			if (status)
+	struct spi_device *spi;
+	struct spi_transfer *t = NULL;
+	unsigned cs_change;
+	int status;
+
+	spi = m->spi;
+	cs_change = 1;
+	status = 0;
+	list_for_each_entry (t, &m->transfers, transfer_list) {
+		if (t->bits_per_word || t->speed_hz) {
+			status = mpc52xx_psc_spi_transfer_setup(spi, t);
+			if (status < 0)
 				break;
-			m->actual_length += t->len;
+		}
 
-			spi_transfer_delay_exec(t);
+		if (cs_change)
+			mpc52xx_psc_spi_activate_cs(spi);
+		cs_change = t->cs_change;
 
-			if (cs_change)
-				mpc52xx_psc_spi_deactivate_cs(spi);
-		}
+		status = mpc52xx_psc_spi_transfer_rxtx(spi, t);
+		if (status)
+			break;
+		m->actual_length += t->len;
 
-		m->status = status;
-		if (m->complete)
-			m->complete(m->context);
+		spi_transfer_delay_exec(t);
 
-		if (status || !cs_change)
+		if (cs_change)
 			mpc52xx_psc_spi_deactivate_cs(spi);
+	}
 
-		mpc52xx_psc_spi_transfer_setup(spi, NULL);
+	m->status = status;
+	if (status || !cs_change)
+		mpc52xx_psc_spi_deactivate_cs(spi);
 
-		spin_lock_irq(&mps->lock);
-	}
-	mps->busy = 0;
-	spin_unlock_irq(&mps->lock);
+	mpc52xx_psc_spi_transfer_setup(spi, NULL);
+
+	spi_finalize_current_message(ctlr);
+
+	return 0;
 }
 
 static int mpc52xx_psc_spi_setup(struct spi_device *spi)
 {
-	struct mpc52xx_psc_spi *mps = spi_master_get_devdata(spi->master);
 	struct mpc52xx_psc_spi_cs *cs = spi->controller_state;
-	unsigned long flags;
 
 	if (spi->bits_per_word%8)
 		return -EINVAL;
@@ -275,28 +253,6 @@ static int mpc52xx_psc_spi_setup(struct spi_device *spi)
 	cs->bits_per_word = spi->bits_per_word;
 	cs->speed_hz = spi->max_speed_hz;
 
-	spin_lock_irqsave(&mps->lock, flags);
-	if (!mps->busy)
-		mpc52xx_psc_spi_deactivate_cs(spi);
-	spin_unlock_irqrestore(&mps->lock, flags);
-
-	return 0;
-}
-
-static int mpc52xx_psc_spi_transfer(struct spi_device *spi,
-		struct spi_message *m)
-{
-	struct mpc52xx_psc_spi *mps = spi_master_get_devdata(spi->master);
-	unsigned long flags;
-
-	m->actual_length = 0;
-	m->status = -EINPROGRESS;
-
-	spin_lock_irqsave(&mps->lock, flags);
-	list_add_tail(&m->queue, &mps->queue);
-	schedule_work(&mps->work);
-	spin_unlock_irqrestore(&mps->lock, flags);
-
 	return 0;
 }
 
@@ -391,7 +347,7 @@ static int mpc52xx_psc_spi_do_probe(struct device *dev, u32 regaddr,
 		master->num_chipselect = pdata->max_chipselect;
 	}
 	master->setup = mpc52xx_psc_spi_setup;
-	master->transfer = mpc52xx_psc_spi_transfer;
+	master->transfer_one_message = mpc52xx_psc_spi_transfer_one_message;
 	master->cleanup = mpc52xx_psc_spi_cleanup;
 	master->dev.of_node = dev->of_node;
 
@@ -415,10 +371,7 @@ static int mpc52xx_psc_spi_do_probe(struct device *dev, u32 regaddr,
 		goto free_irq;
 	}
 
-	spin_lock_init(&mps->lock);
 	init_completion(&mps->done);
-	INIT_WORK(&mps->work, mpc52xx_psc_spi_work);
-	INIT_LIST_HEAD(&mps->queue);
 
 	ret = spi_register_master(master);
 	if (ret < 0)
@@ -470,7 +423,6 @@ static int mpc52xx_psc_spi_of_remove(struct platform_device *op)
 	struct spi_master *master = spi_master_get(platform_get_drvdata(op));
 	struct mpc52xx_psc_spi *mps = spi_master_get_devdata(master);
 
-	flush_work(&mps->work);
 	spi_unregister_master(master);
 	free_irq(mps->irq, mps);
 	if (mps->psc)
diff --git a/drivers/spi/spi-npcm-fiu.c b/drivers/spi/spi-npcm-fiu.c
index ba67dbed9fb8..49f6424e35af 100644
--- a/drivers/spi/spi-npcm-fiu.c
+++ b/drivers/spi/spi-npcm-fiu.c
@@ -36,6 +36,7 @@
 #define NPCM_FIU_UMA_DR1		0x34
 #define NPCM_FIU_UMA_DR2		0x38
 #define NPCM_FIU_UMA_DR3		0x3C
+#define NPCM_FIU_CFG			0x78
 #define NPCM_FIU_MAX_REG_LIMIT		0x80
 
 /* FIU Direct Read Configuration Register */
@@ -151,6 +152,9 @@
 #define NPCM_FIU_UMA_DR3_RB13		GENMASK(15, 8)
 #define NPCM_FIU_UMA_DR3_RB12		GENMASK(7, 0)
 
+/* FIU Configuration Register */
+#define NPCM_FIU_CFG_FIU_FIX		BIT(31)
+
 /* FIU Read Mode */
 enum {
 	DRD_SINGLE_WIRE_MODE	= 0,
@@ -187,6 +191,7 @@ enum {
 	FIU0 = 0,
 	FIU3,
 	FIUX,
+	FIU1,
 };
 
 struct npcm_fiu_info {
@@ -214,6 +219,21 @@ static const struct fiu_data npcm7xx_fiu_data = {
 	.fiu_max = 3,
 };
 
+static const struct npcm_fiu_info npxm8xx_fiu_info[] = {
+	{.name = "FIU0", .fiu_id = FIU0,
+		.max_map_size = MAP_SIZE_128MB, .max_cs = 2},
+	{.name = "FIU3", .fiu_id = FIU3,
+		.max_map_size = MAP_SIZE_128MB, .max_cs = 4},
+	{.name = "FIUX", .fiu_id = FIUX,
+		.max_map_size = MAP_SIZE_16MB, .max_cs = 2},
+	{.name = "FIU1", .fiu_id = FIU1,
+		.max_map_size = MAP_SIZE_16MB, .max_cs = 4} };
+
+static const struct fiu_data npxm8xx_fiu_data = {
+	.npcm_fiu_data_info = npxm8xx_fiu_info,
+	.fiu_max = 4,
+};
+
 struct npcm_fiu_spi;
 
 struct npcm_fiu_chip {
@@ -252,8 +272,7 @@ static void npcm_fiu_set_drd(struct npcm_fiu_spi *fiu,
 	fiu->drd_op.addr.buswidth = op->addr.buswidth;
 	regmap_update_bits(fiu->regmap, NPCM_FIU_DRD_CFG,
 			   NPCM_FIU_DRD_CFG_DBW,
-			   ((op->dummy.nbytes * ilog2(op->addr.buswidth)) / BITS_PER_BYTE)
-			   << NPCM_FIU_DRD_DBW_SHIFT);
+			   op->dummy.nbytes << NPCM_FIU_DRD_DBW_SHIFT);
 	fiu->drd_op.dummy.nbytes = op->dummy.nbytes;
 	regmap_update_bits(fiu->regmap, NPCM_FIU_DRD_CFG,
 			   NPCM_FIU_DRD_CFG_RDCMD, op->cmd.opcode);
@@ -625,6 +644,10 @@ static int npcm_fiu_dirmap_create(struct spi_mem_dirmap_desc *desc)
 		regmap_update_bits(gcr_regmap, NPCM7XX_INTCR3_OFFSET,
 				   NPCM7XX_INTCR3_FIU_FIX,
 				   NPCM7XX_INTCR3_FIU_FIX);
+	} else {
+		regmap_update_bits(fiu->regmap, NPCM_FIU_CFG,
+				   NPCM_FIU_CFG_FIU_FIX,
+				   NPCM_FIU_CFG_FIU_FIX);
 	}
 
 	if (desc->info.op_tmpl.data.dir == SPI_MEM_DATA_IN) {
@@ -665,6 +688,7 @@ static const struct spi_controller_mem_ops npcm_fiu_mem_ops = {
 
 static const struct of_device_id npcm_fiu_dt_ids[] = {
 	{ .compatible = "nuvoton,npcm750-fiu", .data = &npcm7xx_fiu_data  },
+	{ .compatible = "nuvoton,npcm845-fiu", .data = &npxm8xx_fiu_data  },
 	{ /* sentinel */ }
 };
 
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index edb42d08857d..838d12e65144 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -1404,6 +1404,10 @@ static const struct pci_device_id pxa2xx_spi_pci_compound_match[] = {
 	{ PCI_VDEVICE(INTEL, 0x7aab), LPSS_CNL_SSP },
 	{ PCI_VDEVICE(INTEL, 0x7af9), LPSS_CNL_SSP },
 	{ PCI_VDEVICE(INTEL, 0x7afb), LPSS_CNL_SSP },
+	/* MTL-P */
+	{ PCI_VDEVICE(INTEL, 0x7e27), LPSS_CNL_SSP },
+	{ PCI_VDEVICE(INTEL, 0x7e30), LPSS_CNL_SSP },
+	{ PCI_VDEVICE(INTEL, 0x7e46), LPSS_CNL_SSP },
 	/* CNL-LP */
 	{ PCI_VDEVICE(INTEL, 0x9daa), LPSS_CNL_SSP },
 	{ PCI_VDEVICE(INTEL, 0x9dab), LPSS_CNL_SSP },
diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c
index c26440e9058d..7f346866614a 100644
--- a/drivers/spi/spi-s3c64xx.c
+++ b/drivers/spi/spi-s3c64xx.c
@@ -18,7 +18,7 @@
 
 #include <linux/platform_data/spi-s3c64xx.h>
 
-#define MAX_SPI_PORTS		6
+#define MAX_SPI_PORTS		12
 #define S3C64XX_SPI_QUIRK_POLL		(1 << 0)
 #define S3C64XX_SPI_QUIRK_CS_AUTO	(1 << 1)
 #define AUTOSUSPEND_TIMEOUT	2000
@@ -59,6 +59,7 @@
 #define S3C64XX_SPI_MODE_BUS_TSZ_HALFWORD	(1<<17)
 #define S3C64XX_SPI_MODE_BUS_TSZ_WORD		(2<<17)
 #define S3C64XX_SPI_MODE_BUS_TSZ_MASK		(3<<17)
+#define S3C64XX_SPI_MODE_SELF_LOOPBACK		(1<<3)
 #define S3C64XX_SPI_MODE_RXDMA_ON		(1<<2)
 #define S3C64XX_SPI_MODE_TXDMA_ON		(1<<1)
 #define S3C64XX_SPI_MODE_4BURST			(1<<0)
@@ -130,11 +131,13 @@ struct s3c64xx_spi_dma_data {
  * @fifo_lvl_mask: Bit-mask for {TX|RX}_FIFO_LVL bits in SPI_STATUS register.
  * @rx_lvl_offset: Bit offset of RX_FIFO_LVL bits in SPI_STATUS regiter.
  * @tx_st_done: Bit offset of TX_DONE bit in SPI_STATUS regiter.
+ * @clk_div: Internal clock divider
  * @quirks: Bitmask of known quirks
  * @high_speed: True, if the controller supports HIGH_SPEED_EN bit.
  * @clk_from_cmu: True, if the controller does not include a clock mux and
  *	prescaler unit.
  * @clk_ioclk: True if clock is present on this device
+ * @has_loopback: True if loopback mode can be supported
  *
  * The Samsung s3c64xx SPI controller are used on various Samsung SoC's but
  * differ in some aspects such as the size of the fifo and spi bus clock
@@ -146,9 +149,11 @@ struct s3c64xx_spi_port_config {
 	int	rx_lvl_offset;
 	int	tx_st_done;
 	int	quirks;
+	int	clk_div;
 	bool	high_speed;
 	bool	clk_from_cmu;
 	bool	clk_ioclk;
+	bool	has_loopback;
 };
 
 /**
@@ -350,19 +355,59 @@ static int s3c64xx_spi_prepare_transfer(struct spi_master *spi)
 	if (is_polling(sdd))
 		return 0;
 
+	/* Requests DMA channels */
+	sdd->rx_dma.ch = dma_request_chan(&sdd->pdev->dev, "rx");
+	if (IS_ERR(sdd->rx_dma.ch)) {
+		dev_err(&sdd->pdev->dev, "Failed to get RX DMA channel\n");
+		sdd->rx_dma.ch = NULL;
+		return 0;
+	}
+
+	sdd->tx_dma.ch = dma_request_chan(&sdd->pdev->dev, "tx");
+	if (IS_ERR(sdd->tx_dma.ch)) {
+		dev_err(&sdd->pdev->dev, "Failed to get TX DMA channel\n");
+		dma_release_channel(sdd->rx_dma.ch);
+		sdd->tx_dma.ch = NULL;
+		sdd->rx_dma.ch = NULL;
+		return 0;
+	}
+
 	spi->dma_rx = sdd->rx_dma.ch;
 	spi->dma_tx = sdd->tx_dma.ch;
 
 	return 0;
 }
 
+static int s3c64xx_spi_unprepare_transfer(struct spi_master *spi)
+{
+	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(spi);
+
+	if (is_polling(sdd))
+		return 0;
+
+	/* Releases DMA channels if they are allocated */
+	if (sdd->rx_dma.ch && sdd->tx_dma.ch) {
+		dma_release_channel(sdd->rx_dma.ch);
+		dma_release_channel(sdd->tx_dma.ch);
+		sdd->rx_dma.ch = 0;
+		sdd->tx_dma.ch = 0;
+	}
+
+	return 0;
+}
+
 static bool s3c64xx_spi_can_dma(struct spi_master *master,
 				struct spi_device *spi,
 				struct spi_transfer *xfer)
 {
 	struct s3c64xx_spi_driver_data *sdd = spi_master_get_devdata(master);
 
-	return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1;
+	if (sdd->rx_dma.ch && sdd->tx_dma.ch) {
+		return xfer->len > (FIFO_LVL_MASK(sdd) >> 1) + 1;
+	} else {
+		return false;
+	}
+
 }
 
 static int s3c64xx_enable_datapath(struct s3c64xx_spi_driver_data *sdd,
@@ -577,6 +622,7 @@ static int s3c64xx_spi_config(struct s3c64xx_spi_driver_data *sdd)
 	void __iomem *regs = sdd->regs;
 	int ret;
 	u32 val;
+	int div = sdd->port_conf->clk_div;
 
 	/* Disable Clock */
 	if (!sdd->port_conf->clk_from_cmu) {
@@ -619,19 +665,21 @@ static int s3c64xx_spi_config(struct s3c64xx_spi_driver_data *sdd)
 		break;
 	}
 
+	if ((sdd->cur_mode & SPI_LOOP) && sdd->port_conf->has_loopback)
+		val |= S3C64XX_SPI_MODE_SELF_LOOPBACK;
+
 	writel(val, regs + S3C64XX_SPI_MODE_CFG);
 
 	if (sdd->port_conf->clk_from_cmu) {
-		/* The src_clk clock is divided internally by 2 */
-		ret = clk_set_rate(sdd->src_clk, sdd->cur_speed * 2);
+		ret = clk_set_rate(sdd->src_clk, sdd->cur_speed * div);
 		if (ret)
 			return ret;
-		sdd->cur_speed = clk_get_rate(sdd->src_clk) / 2;
+		sdd->cur_speed = clk_get_rate(sdd->src_clk) / div;
 	} else {
 		/* Configure Clock */
 		val = readl(regs + S3C64XX_SPI_CLK_CFG);
 		val &= ~S3C64XX_SPI_PSR_MASK;
-		val |= ((clk_get_rate(sdd->src_clk) / sdd->cur_speed / 2 - 1)
+		val |= ((clk_get_rate(sdd->src_clk) / sdd->cur_speed / div - 1)
 				& S3C64XX_SPI_PSR_MASK);
 		writel(val, regs + S3C64XX_SPI_CLK_CFG);
 
@@ -697,7 +745,7 @@ static int s3c64xx_spi_transfer_one(struct spi_master *master,
 	    sdd->rx_dma.ch && sdd->tx_dma.ch) {
 		use_dma = 1;
 
-	} else if (is_polling(sdd) && xfer->len > fifo_len) {
+	} else if (xfer->len > fifo_len) {
 		tx_buf = xfer->tx_buf;
 		rx_buf = xfer->rx_buf;
 		origin_len = xfer->len;
@@ -825,6 +873,7 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 	struct s3c64xx_spi_csinfo *cs = spi->controller_data;
 	struct s3c64xx_spi_driver_data *sdd;
 	int err;
+	int div;
 
 	sdd = spi_master_get_devdata(spi->master);
 	if (spi->dev.of_node) {
@@ -843,22 +892,24 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 
 	pm_runtime_get_sync(&sdd->pdev->dev);
 
+	div = sdd->port_conf->clk_div;
+
 	/* Check if we can provide the requested rate */
 	if (!sdd->port_conf->clk_from_cmu) {
 		u32 psr, speed;
 
 		/* Max possible */
-		speed = clk_get_rate(sdd->src_clk) / 2 / (0 + 1);
+		speed = clk_get_rate(sdd->src_clk) / div / (0 + 1);
 
 		if (spi->max_speed_hz > speed)
 			spi->max_speed_hz = speed;
 
-		psr = clk_get_rate(sdd->src_clk) / 2 / spi->max_speed_hz - 1;
+		psr = clk_get_rate(sdd->src_clk) / div / spi->max_speed_hz - 1;
 		psr &= S3C64XX_SPI_PSR_MASK;
 		if (psr == S3C64XX_SPI_PSR_MASK)
 			psr--;
 
-		speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
+		speed = clk_get_rate(sdd->src_clk) / div / (psr + 1);
 		if (spi->max_speed_hz < speed) {
 			if (psr+1 < S3C64XX_SPI_PSR_MASK) {
 				psr++;
@@ -868,7 +919,7 @@ static int s3c64xx_spi_setup(struct spi_device *spi)
 			}
 		}
 
-		speed = clk_get_rate(sdd->src_clk) / 2 / (psr + 1);
+		speed = clk_get_rate(sdd->src_clk) / div / (psr + 1);
 		if (spi->max_speed_hz >= speed) {
 			spi->max_speed_hz = speed;
 		} else {
@@ -1098,6 +1149,7 @@ static int s3c64xx_spi_probe(struct platform_device *pdev)
 	master->setup = s3c64xx_spi_setup;
 	master->cleanup = s3c64xx_spi_cleanup;
 	master->prepare_transfer_hardware = s3c64xx_spi_prepare_transfer;
+	master->unprepare_transfer_hardware = s3c64xx_spi_unprepare_transfer;
 	master->prepare_message = s3c64xx_spi_prepare_message;
 	master->transfer_one = s3c64xx_spi_transfer_one;
 	master->num_chipselect = sci->num_cs;
@@ -1107,6 +1159,8 @@ static int s3c64xx_spi_probe(struct platform_device *pdev)
 					SPI_BPW_MASK(8);
 	/* the spi->mode bits understood by this driver: */
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH;
+	if (sdd->port_conf->has_loopback)
+		master->mode_bits |= SPI_LOOP;
 	master->auto_runtime_pm = true;
 	if (!is_polling(sdd))
 		master->can_dma = s3c64xx_spi_can_dma;
@@ -1167,22 +1221,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev)
 		}
 	}
 
-	if (!is_polling(sdd)) {
-		/* Acquire DMA channels */
-		sdd->rx_dma.ch = dma_request_chan(&pdev->dev, "rx");
-		if (IS_ERR(sdd->rx_dma.ch)) {
-			dev_err(&pdev->dev, "Failed to get RX DMA channel\n");
-			ret = PTR_ERR(sdd->rx_dma.ch);
-			goto err_disable_io_clk;
-		}
-		sdd->tx_dma.ch = dma_request_chan(&pdev->dev, "tx");
-		if (IS_ERR(sdd->tx_dma.ch)) {
-			dev_err(&pdev->dev, "Failed to get TX DMA channel\n");
-			ret = PTR_ERR(sdd->tx_dma.ch);
-			goto err_release_rx_dma;
-		}
-	}
-
 	pm_runtime_set_autosuspend_delay(&pdev->dev, AUTOSUSPEND_TIMEOUT);
 	pm_runtime_use_autosuspend(&pdev->dev);
 	pm_runtime_set_active(&pdev->dev);
@@ -1228,12 +1266,6 @@ err_pm_put:
 	pm_runtime_disable(&pdev->dev);
 	pm_runtime_set_suspended(&pdev->dev);
 
-	if (!is_polling(sdd))
-		dma_release_channel(sdd->tx_dma.ch);
-err_release_rx_dma:
-	if (!is_polling(sdd))
-		dma_release_channel(sdd->rx_dma.ch);
-err_disable_io_clk:
 	clk_disable_unprepare(sdd->ioclk);
 err_disable_src_clk:
 	clk_disable_unprepare(sdd->src_clk);
@@ -1369,6 +1401,7 @@ static const struct s3c64xx_spi_port_config s3c2443_spi_port_config = {
 	.fifo_lvl_mask	= { 0x7f },
 	.rx_lvl_offset	= 13,
 	.tx_st_done	= 21,
+	.clk_div	= 2,
 	.high_speed	= true,
 };
 
@@ -1376,12 +1409,14 @@ static const struct s3c64xx_spi_port_config s3c6410_spi_port_config = {
 	.fifo_lvl_mask	= { 0x7f, 0x7F },
 	.rx_lvl_offset	= 13,
 	.tx_st_done	= 21,
+	.clk_div	= 2,
 };
 
 static const struct s3c64xx_spi_port_config s5pv210_spi_port_config = {
 	.fifo_lvl_mask	= { 0x1ff, 0x7F },
 	.rx_lvl_offset	= 15,
 	.tx_st_done	= 25,
+	.clk_div	= 2,
 	.high_speed	= true,
 };
 
@@ -1389,6 +1424,7 @@ static const struct s3c64xx_spi_port_config exynos4_spi_port_config = {
 	.fifo_lvl_mask	= { 0x1ff, 0x7F, 0x7F },
 	.rx_lvl_offset	= 15,
 	.tx_st_done	= 25,
+	.clk_div	= 2,
 	.high_speed	= true,
 	.clk_from_cmu	= true,
 	.quirks		= S3C64XX_SPI_QUIRK_CS_AUTO,
@@ -1398,6 +1434,7 @@ static const struct s3c64xx_spi_port_config exynos7_spi_port_config = {
 	.fifo_lvl_mask	= { 0x1ff, 0x7F, 0x7F, 0x7F, 0x7F, 0x1ff},
 	.rx_lvl_offset	= 15,
 	.tx_st_done	= 25,
+	.clk_div	= 2,
 	.high_speed	= true,
 	.clk_from_cmu	= true,
 	.quirks		= S3C64XX_SPI_QUIRK_CS_AUTO,
@@ -1407,16 +1444,31 @@ static const struct s3c64xx_spi_port_config exynos5433_spi_port_config = {
 	.fifo_lvl_mask	= { 0x1ff, 0x7f, 0x7f, 0x7f, 0x7f, 0x1ff},
 	.rx_lvl_offset	= 15,
 	.tx_st_done	= 25,
+	.clk_div	= 2,
+	.high_speed	= true,
+	.clk_from_cmu	= true,
+	.clk_ioclk	= true,
+	.quirks		= S3C64XX_SPI_QUIRK_CS_AUTO,
+};
+
+static const struct s3c64xx_spi_port_config exynosautov9_spi_port_config = {
+	.fifo_lvl_mask	= { 0x1ff, 0x1ff, 0x7f, 0x7f, 0x7f, 0x7f, 0x1ff, 0x7f,
+			    0x7f, 0x7f, 0x7f, 0x7f},
+	.rx_lvl_offset	= 15,
+	.tx_st_done	= 25,
+	.clk_div	= 4,
 	.high_speed	= true,
 	.clk_from_cmu	= true,
 	.clk_ioclk	= true,
+	.has_loopback	= true,
 	.quirks		= S3C64XX_SPI_QUIRK_CS_AUTO,
 };
 
-static struct s3c64xx_spi_port_config fsd_spi_port_config = {
+static const struct s3c64xx_spi_port_config fsd_spi_port_config = {
 	.fifo_lvl_mask	= { 0x7f, 0x7f, 0x7f, 0x7f, 0x7f},
 	.rx_lvl_offset	= 15,
 	.tx_st_done	= 25,
+	.clk_div	= 2,
 	.high_speed	= true,
 	.clk_from_cmu	= true,
 	.clk_ioclk	= false,
@@ -1453,6 +1505,9 @@ static const struct of_device_id s3c64xx_spi_dt_match[] = {
 	{ .compatible = "samsung,exynos5433-spi",
 			.data = (void *)&exynos5433_spi_port_config,
 	},
+	{ .compatible = "samsung,exynosautov9-spi",
+			.data = (void *)&exynosautov9_spi_port_config,
+	},
 	{ .compatible = "tesla,fsd-spi",
 			.data = (void *)&fsd_spi_port_config,
 	},
diff --git a/drivers/spi/spi-sh.c b/drivers/spi/spi-sh.c
index 45f304935332..3e72fad99adf 100644
--- a/drivers/spi/spi-sh.c
+++ b/drivers/spi/spi-sh.c
@@ -73,11 +73,8 @@ struct spi_sh_data {
 	void __iomem *addr;
 	int irq;
 	struct spi_master *master;
-	struct list_head queue;
-	struct work_struct ws;
 	unsigned long cr1;
 	wait_queue_head_t wait;
-	spinlock_t lock;
 	int width;
 };
 
@@ -271,47 +268,39 @@ static int spi_sh_receive(struct spi_sh_data *ss, struct spi_message *mesg,
 	return 0;
 }
 
-static void spi_sh_work(struct work_struct *work)
+static int spi_sh_transfer_one_message(struct spi_controller *ctlr,
+					struct spi_message *mesg)
 {
-	struct spi_sh_data *ss = container_of(work, struct spi_sh_data, ws);
-	struct spi_message *mesg;
+	struct spi_sh_data *ss = spi_controller_get_devdata(ctlr);
 	struct spi_transfer *t;
-	unsigned long flags;
 	int ret;
 
 	pr_debug("%s: enter\n", __func__);
 
-	spin_lock_irqsave(&ss->lock, flags);
-	while (!list_empty(&ss->queue)) {
-		mesg = list_entry(ss->queue.next, struct spi_message, queue);
-		list_del_init(&mesg->queue);
-
-		spin_unlock_irqrestore(&ss->lock, flags);
-		list_for_each_entry(t, &mesg->transfers, transfer_list) {
-			pr_debug("tx_buf = %p, rx_buf = %p\n",
-					t->tx_buf, t->rx_buf);
-			pr_debug("len = %d, delay.value = %d\n",
-					t->len, t->delay.value);
-
-			if (t->tx_buf) {
-				ret = spi_sh_send(ss, mesg, t);
-				if (ret < 0)
-					goto error;
-			}
-			if (t->rx_buf) {
-				ret = spi_sh_receive(ss, mesg, t);
-				if (ret < 0)
-					goto error;
-			}
-			mesg->actual_length += t->len;
-		}
-		spin_lock_irqsave(&ss->lock, flags);
+	spi_sh_clear_bit(ss, SPI_SH_SSA, SPI_SH_CR1);
 
-		mesg->status = 0;
-		if (mesg->complete)
-			mesg->complete(mesg->context);
+	list_for_each_entry(t, &mesg->transfers, transfer_list) {
+		pr_debug("tx_buf = %p, rx_buf = %p\n",
+			 t->tx_buf, t->rx_buf);
+		pr_debug("len = %d, delay.value = %d\n",
+			 t->len, t->delay.value);
+
+		if (t->tx_buf) {
+			ret = spi_sh_send(ss, mesg, t);
+			if (ret < 0)
+				goto error;
+		}
+		if (t->rx_buf) {
+			ret = spi_sh_receive(ss, mesg, t);
+			if (ret < 0)
+				goto error;
+		}
+		mesg->actual_length += t->len;
 	}
 
+	mesg->status = 0;
+	spi_finalize_current_message(ctlr);
+
 	clear_fifo(ss);
 	spi_sh_set_bit(ss, SPI_SH_SSD, SPI_SH_CR1);
 	udelay(100);
@@ -321,12 +310,11 @@ static void spi_sh_work(struct work_struct *work)
 
 	clear_fifo(ss);
 
-	spin_unlock_irqrestore(&ss->lock, flags);
-
-	return;
+	return 0;
 
  error:
 	mesg->status = ret;
+	spi_finalize_current_message(ctlr);
 	if (mesg->complete)
 		mesg->complete(mesg->context);
 
@@ -334,6 +322,7 @@ static void spi_sh_work(struct work_struct *work)
 			 SPI_SH_CR1);
 	clear_fifo(ss);
 
+	return ret;
 }
 
 static int spi_sh_setup(struct spi_device *spi)
@@ -355,29 +344,6 @@ static int spi_sh_setup(struct spi_device *spi)
 	return 0;
 }
 
-static int spi_sh_transfer(struct spi_device *spi, struct spi_message *mesg)
-{
-	struct spi_sh_data *ss = spi_master_get_devdata(spi->master);
-	unsigned long flags;
-
-	pr_debug("%s: enter\n", __func__);
-	pr_debug("\tmode = %02x\n", spi->mode);
-
-	spin_lock_irqsave(&ss->lock, flags);
-
-	mesg->actual_length = 0;
-	mesg->status = -EINPROGRESS;
-
-	spi_sh_clear_bit(ss, SPI_SH_SSA, SPI_SH_CR1);
-
-	list_add_tail(&mesg->queue, &ss->queue);
-	schedule_work(&ss->ws);
-
-	spin_unlock_irqrestore(&ss->lock, flags);
-
-	return 0;
-}
-
 static void spi_sh_cleanup(struct spi_device *spi)
 {
 	struct spi_sh_data *ss = spi_master_get_devdata(spi->master);
@@ -416,7 +382,6 @@ static int spi_sh_remove(struct platform_device *pdev)
 	struct spi_sh_data *ss = platform_get_drvdata(pdev);
 
 	spi_unregister_master(ss->master);
-	flush_work(&ss->ws);
 	free_irq(ss->irq, ss);
 
 	return 0;
@@ -467,9 +432,6 @@ static int spi_sh_probe(struct platform_device *pdev)
 		dev_err(&pdev->dev, "ioremap error.\n");
 		return -ENOMEM;
 	}
-	INIT_LIST_HEAD(&ss->queue);
-	spin_lock_init(&ss->lock);
-	INIT_WORK(&ss->ws, spi_sh_work);
 	init_waitqueue_head(&ss->wait);
 
 	ret = request_irq(irq, spi_sh_irq, 0, "spi_sh", ss);
@@ -481,7 +443,7 @@ static int spi_sh_probe(struct platform_device *pdev)
 	master->num_chipselect = 2;
 	master->bus_num = pdev->id;
 	master->setup = spi_sh_setup;
-	master->transfer = spi_sh_transfer;
+	master->transfer_one_message = spi_sh_transfer_one_message;
 	master->cleanup = spi_sh_cleanup;
 
 	ret = spi_register_master(master);
diff --git a/drivers/spi/spi-sifive.c b/drivers/spi/spi-sifive.c
index f7c1e20432e0..e29e85cee88a 100644
--- a/drivers/spi/spi-sifive.c
+++ b/drivers/spi/spi-sifive.c
@@ -427,6 +427,44 @@ static int sifive_spi_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static int sifive_spi_suspend(struct device *dev)
+{
+	struct spi_master *master = dev_get_drvdata(dev);
+	struct sifive_spi *spi = spi_master_get_devdata(master);
+	int ret;
+
+	ret = spi_master_suspend(master);
+	if (ret)
+		return ret;
+
+	/* Disable all the interrupts just in case */
+	sifive_spi_write(spi, SIFIVE_SPI_REG_IE, 0);
+
+	clk_disable_unprepare(spi->clk);
+
+	return ret;
+}
+
+static int sifive_spi_resume(struct device *dev)
+{
+	struct spi_master *master = dev_get_drvdata(dev);
+	struct sifive_spi *spi = spi_master_get_devdata(master);
+	int ret;
+
+	ret = clk_prepare_enable(spi->clk);
+	if (ret)
+		return ret;
+	ret = spi_master_resume(master);
+	if (ret)
+		clk_disable_unprepare(spi->clk);
+
+	return ret;
+}
+
+static DEFINE_SIMPLE_DEV_PM_OPS(sifive_spi_pm_ops,
+				sifive_spi_suspend, sifive_spi_resume);
+
+
 static const struct of_device_id sifive_spi_of_match[] = {
 	{ .compatible = "sifive,spi0", },
 	{}
@@ -438,6 +476,7 @@ static struct platform_driver sifive_spi_driver = {
 	.remove = sifive_spi_remove,
 	.driver = {
 		.name = SIFIVE_SPI_DRIVER_NAME,
+		.pm = &sifive_spi_pm_ops,
 		.of_match_table = sifive_spi_of_match,
 	},
 };
diff --git a/drivers/spi/spi-stm32-qspi.c b/drivers/spi/spi-stm32-qspi.c
index c0239e405c39..f3fe92300639 100644
--- a/drivers/spi/spi-stm32-qspi.c
+++ b/drivers/spi/spi-stm32-qspi.c
@@ -299,8 +299,7 @@ static int stm32_qspi_wait_nobusy(struct stm32_qspi *qspi)
 						 STM32_BUSY_TIMEOUT_US);
 }
 
-static int stm32_qspi_wait_cmd(struct stm32_qspi *qspi,
-			       const struct spi_mem_op *op)
+static int stm32_qspi_wait_cmd(struct stm32_qspi *qspi)
 {
 	u32 cr, sr;
 	int err = 0;
@@ -331,8 +330,7 @@ out:
 	return err;
 }
 
-static int stm32_qspi_wait_poll_status(struct stm32_qspi *qspi,
-				       const struct spi_mem_op *op)
+static int stm32_qspi_wait_poll_status(struct stm32_qspi *qspi)
 {
 	u32 cr;
 
@@ -349,7 +347,7 @@ static int stm32_qspi_wait_poll_status(struct stm32_qspi *qspi,
 	return 0;
 }
 
-static int stm32_qspi_get_mode(struct stm32_qspi *qspi, u8 buswidth)
+static int stm32_qspi_get_mode(u8 buswidth)
 {
 	if (buswidth == 4)
 		return CCR_BUSWIDTH_4;
@@ -382,11 +380,11 @@ static int stm32_qspi_send(struct spi_mem *mem, const struct spi_mem_op *op)
 	ccr = qspi->fmode;
 	ccr |= FIELD_PREP(CCR_INST_MASK, op->cmd.opcode);
 	ccr |= FIELD_PREP(CCR_IMODE_MASK,
-			  stm32_qspi_get_mode(qspi, op->cmd.buswidth));
+			  stm32_qspi_get_mode(op->cmd.buswidth));
 
 	if (op->addr.nbytes) {
 		ccr |= FIELD_PREP(CCR_ADMODE_MASK,
-				  stm32_qspi_get_mode(qspi, op->addr.buswidth));
+				  stm32_qspi_get_mode(op->addr.buswidth));
 		ccr |= FIELD_PREP(CCR_ADSIZE_MASK, op->addr.nbytes - 1);
 	}
 
@@ -396,7 +394,7 @@ static int stm32_qspi_send(struct spi_mem *mem, const struct spi_mem_op *op)
 
 	if (op->data.nbytes) {
 		ccr |= FIELD_PREP(CCR_DMODE_MASK,
-				  stm32_qspi_get_mode(qspi, op->data.buswidth));
+				  stm32_qspi_get_mode(op->data.buswidth));
 	}
 
 	writel_relaxed(ccr, qspi->io_base + QSPI_CCR);
@@ -405,7 +403,7 @@ static int stm32_qspi_send(struct spi_mem *mem, const struct spi_mem_op *op)
 		writel_relaxed(op->addr.val, qspi->io_base + QSPI_AR);
 
 	if (qspi->fmode == CCR_FMODE_APM)
-		err_poll_status = stm32_qspi_wait_poll_status(qspi, op);
+		err_poll_status = stm32_qspi_wait_poll_status(qspi);
 
 	err = stm32_qspi_tx(qspi, op);
 
@@ -420,7 +418,7 @@ static int stm32_qspi_send(struct spi_mem *mem, const struct spi_mem_op *op)
 		goto abort;
 
 	/* wait end of tx in indirect mode */
-	err = stm32_qspi_wait_cmd(qspi, op);
+	err = stm32_qspi_wait_cmd(qspi);
 	if (err)
 		goto abort;
 
diff --git a/drivers/spi/spi-synquacer.c b/drivers/spi/spi-synquacer.c
index ea706d9629cb..47cbe73137c2 100644
--- a/drivers/spi/spi-synquacer.c
+++ b/drivers/spi/spi-synquacer.c
@@ -783,6 +783,7 @@ static int __maybe_unused synquacer_spi_resume(struct device *dev)
 
 		ret = synquacer_spi_enable(master);
 		if (ret) {
+			clk_disable_unprepare(sspi->clk);
 			dev_err(dev, "failed to enable spi (%d)\n", ret);
 			return ret;
 		}
diff --git a/drivers/spi/spi-tegra20-slink.c b/drivers/spi/spi-tegra20-slink.c
index 38360434d6e9..148043d0c2b8 100644
--- a/drivers/spi/spi-tegra20-slink.c
+++ b/drivers/spi/spi-tegra20-slink.c
@@ -1136,7 +1136,7 @@ exit_free_master:
 
 static int tegra_slink_remove(struct platform_device *pdev)
 {
-	struct spi_master *master = platform_get_drvdata(pdev);
+	struct spi_master *master = spi_master_get(platform_get_drvdata(pdev));
 	struct tegra_slink_data	*tspi = spi_master_get_devdata(master);
 
 	spi_unregister_master(master);
@@ -1151,6 +1151,7 @@ static int tegra_slink_remove(struct platform_device *pdev)
 	if (tspi->rx_dma_chan)
 		tegra_slink_deinit_dma_param(tspi, true);
 
+	spi_master_put(master);
 	return 0;
 }
 
diff --git a/drivers/spi/spi-tegra210-quad.c b/drivers/spi/spi-tegra210-quad.c
index 66f647f32876..c89592b21ffc 100644
--- a/drivers/spi/spi-tegra210-quad.c
+++ b/drivers/spi/spi-tegra210-quad.c
@@ -37,6 +37,16 @@
 #define QSPI_RX_EN				BIT(12)
 #define QSPI_CS_SW_VAL				BIT(20)
 #define QSPI_CS_SW_HW				BIT(21)
+
+#define QSPI_CS_POL_INACTIVE(n)			(1 << (22 + (n)))
+#define QSPI_CS_POL_INACTIVE_MASK		(0xF << 22)
+#define QSPI_CS_SEL_0				(0 << 26)
+#define QSPI_CS_SEL_1				(1 << 26)
+#define QSPI_CS_SEL_2				(2 << 26)
+#define QSPI_CS_SEL_3				(3 << 26)
+#define QSPI_CS_SEL_MASK			(3 << 26)
+#define QSPI_CS_SEL(x)				(((x) & 0x3) << 26)
+
 #define QSPI_CONTROL_MODE_0			(0 << 28)
 #define QSPI_CONTROL_MODE_3			(3 << 28)
 #define QSPI_CONTROL_MODE_MASK			(3 << 28)
@@ -154,6 +164,7 @@
 struct tegra_qspi_soc_data {
 	bool has_dma;
 	bool cmb_xfer_capable;
+	unsigned int cs_count;
 };
 
 struct tegra_qspi_client_data {
@@ -812,6 +823,7 @@ static u32 tegra_qspi_setup_transfer_one(struct spi_device *spi, struct spi_tran
 		tegra_qspi_mask_clear_irq(tqspi);
 
 		command1 = tqspi->def_command1_reg;
+		command1 |= QSPI_CS_SEL(spi->chip_select);
 		command1 |= QSPI_BIT_LENGTH(bits_per_word - 1);
 
 		command1 &= ~QSPI_CONTROL_MODE_MASK;
@@ -941,10 +953,11 @@ static int tegra_qspi_setup(struct spi_device *spi)
 
 	/* keep default cs state to inactive */
 	val = tqspi->def_command1_reg;
+	val |= QSPI_CS_SEL(spi->chip_select);
 	if (spi->mode & SPI_CS_HIGH)
-		val &= ~QSPI_CS_SW_VAL;
+		val &= ~QSPI_CS_POL_INACTIVE(spi->chip_select);
 	else
-		val |= QSPI_CS_SW_VAL;
+		val |= QSPI_CS_POL_INACTIVE(spi->chip_select);
 
 	tqspi->def_command1_reg = val;
 	tegra_qspi_writel(tqspi, tqspi->def_command1_reg, QSPI_COMMAND1);
@@ -1425,16 +1438,25 @@ static irqreturn_t tegra_qspi_isr_thread(int irq, void *context_data)
 static struct tegra_qspi_soc_data tegra210_qspi_soc_data = {
 	.has_dma = true,
 	.cmb_xfer_capable = false,
+	.cs_count = 1,
 };
 
 static struct tegra_qspi_soc_data tegra186_qspi_soc_data = {
 	.has_dma = true,
 	.cmb_xfer_capable = true,
+	.cs_count = 1,
 };
 
 static struct tegra_qspi_soc_data tegra234_qspi_soc_data = {
 	.has_dma = false,
 	.cmb_xfer_capable = true,
+	.cs_count = 1,
+};
+
+static struct tegra_qspi_soc_data tegra241_qspi_soc_data = {
+	.has_dma = false,
+	.cmb_xfer_capable = true,
+	.cs_count = 4,
 };
 
 static const struct of_device_id tegra_qspi_of_match[] = {
@@ -1450,6 +1472,9 @@ static const struct of_device_id tegra_qspi_of_match[] = {
 	}, {
 		.compatible = "nvidia,tegra234-qspi",
 		.data	    = &tegra234_qspi_soc_data,
+	}, {
+		.compatible = "nvidia,tegra241-qspi",
+		.data	    = &tegra241_qspi_soc_data,
 	},
 	{}
 };
@@ -1467,6 +1492,9 @@ static const struct acpi_device_id tegra_qspi_acpi_match[] = {
 	}, {
 		.id = "NVDA1413",
 		.driver_data = (kernel_ulong_t)&tegra234_qspi_soc_data,
+	}, {
+		.id = "NVDA1513",
+		.driver_data = (kernel_ulong_t)&tegra241_qspi_soc_data,
 	},
 	{}
 };
@@ -1506,6 +1534,7 @@ static int tegra_qspi_probe(struct platform_device *pdev)
 	spin_lock_init(&tqspi->lock);
 
 	tqspi->soc_data = device_get_match_data(&pdev->dev);
+	master->num_chipselect = tqspi->soc_data->cs_count;
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	tqspi->base = devm_ioremap_resource(&pdev->dev, r);
 	if (IS_ERR(tqspi->base))
diff --git a/drivers/spi/spi-ti-qspi.c b/drivers/spi/spi-ti-qspi.c
index b5b65d882d7a..60086869bcae 100644
--- a/drivers/spi/spi-ti-qspi.c
+++ b/drivers/spi/spi-ti-qspi.c
@@ -57,7 +57,6 @@ struct ti_qspi {
 	void			*rx_bb_addr;
 	struct dma_chan		*rx_chan;
 
-	u32 spi_max_frequency;
 	u32 cmd;
 	u32 dc;
 
@@ -140,37 +139,19 @@ static inline void ti_qspi_write(struct ti_qspi *qspi,
 static int ti_qspi_setup(struct spi_device *spi)
 {
 	struct ti_qspi	*qspi = spi_master_get_devdata(spi->master);
-	struct ti_qspi_regs *ctx_reg = &qspi->ctx_reg;
-	int clk_div = 0, ret;
-	u32 clk_ctrl_reg, clk_rate, clk_mask;
+	int ret;
 
 	if (spi->master->busy) {
 		dev_dbg(qspi->dev, "master busy doing other transfers\n");
 		return -EBUSY;
 	}
 
-	if (!qspi->spi_max_frequency) {
+	if (!qspi->master->max_speed_hz) {
 		dev_err(qspi->dev, "spi max frequency not defined\n");
 		return -EINVAL;
 	}
 
-	clk_rate = clk_get_rate(qspi->fclk);
-
-	clk_div = DIV_ROUND_UP(clk_rate, qspi->spi_max_frequency) - 1;
-
-	if (clk_div < 0) {
-		dev_dbg(qspi->dev, "clock divider < 0, using /1 divider\n");
-		return -EINVAL;
-	}
-
-	if (clk_div > QSPI_CLK_DIV_MAX) {
-		dev_dbg(qspi->dev, "clock divider >%d , using /%d divider\n",
-				QSPI_CLK_DIV_MAX, QSPI_CLK_DIV_MAX + 1);
-		return -EINVAL;
-	}
-
-	dev_dbg(qspi->dev, "hz: %d, clock divider %d\n",
-			qspi->spi_max_frequency, clk_div);
+	spi->max_speed_hz = min(spi->max_speed_hz, qspi->master->max_speed_hz);
 
 	ret = pm_runtime_resume_and_get(qspi->dev);
 	if (ret < 0) {
@@ -178,18 +159,6 @@ static int ti_qspi_setup(struct spi_device *spi)
 		return ret;
 	}
 
-	clk_ctrl_reg = ti_qspi_read(qspi, QSPI_SPI_CLOCK_CNTRL_REG);
-
-	clk_ctrl_reg &= ~QSPI_CLK_EN;
-
-	/* disable SCLK */
-	ti_qspi_write(qspi, clk_ctrl_reg, QSPI_SPI_CLOCK_CNTRL_REG);
-
-	/* enable SCLK */
-	clk_mask = QSPI_CLK_EN | clk_div;
-	ti_qspi_write(qspi, clk_mask, QSPI_SPI_CLOCK_CNTRL_REG);
-	ctx_reg->clkctrl = clk_mask;
-
 	pm_runtime_mark_last_busy(qspi->dev);
 	ret = pm_runtime_put_autosuspend(qspi->dev);
 	if (ret < 0) {
@@ -200,6 +169,37 @@ static int ti_qspi_setup(struct spi_device *spi)
 	return 0;
 }
 
+static void ti_qspi_setup_clk(struct ti_qspi *qspi, u32 speed_hz)
+{
+	struct ti_qspi_regs *ctx_reg = &qspi->ctx_reg;
+	int clk_div;
+	u32 clk_ctrl_reg, clk_rate, clk_ctrl_new;
+
+	clk_rate = clk_get_rate(qspi->fclk);
+	clk_div = DIV_ROUND_UP(clk_rate, speed_hz) - 1;
+	clk_div = clamp(clk_div, 0, QSPI_CLK_DIV_MAX);
+	dev_dbg(qspi->dev, "hz: %d, clock divider %d\n", speed_hz, clk_div);
+
+	pm_runtime_resume_and_get(qspi->dev);
+
+	clk_ctrl_new = QSPI_CLK_EN | clk_div;
+	if (ctx_reg->clkctrl != clk_ctrl_new) {
+		clk_ctrl_reg = ti_qspi_read(qspi, QSPI_SPI_CLOCK_CNTRL_REG);
+
+		clk_ctrl_reg &= ~QSPI_CLK_EN;
+
+		/* disable SCLK */
+		ti_qspi_write(qspi, clk_ctrl_reg, QSPI_SPI_CLOCK_CNTRL_REG);
+
+		/* enable SCLK */
+		ti_qspi_write(qspi, clk_ctrl_new, QSPI_SPI_CLOCK_CNTRL_REG);
+		ctx_reg->clkctrl = clk_ctrl_new;
+	}
+
+	pm_runtime_mark_last_busy(qspi->dev);
+	pm_runtime_put_autosuspend(qspi->dev);
+}
+
 static void ti_qspi_restore_ctx(struct ti_qspi *qspi)
 {
 	struct ti_qspi_regs *ctx_reg = &qspi->ctx_reg;
@@ -623,8 +623,10 @@ static int ti_qspi_exec_mem_op(struct spi_mem *mem,
 
 	mutex_lock(&qspi->list_lock);
 
-	if (!qspi->mmap_enabled || qspi->current_cs != mem->spi->chip_select)
+	if (!qspi->mmap_enabled || qspi->current_cs != mem->spi->chip_select) {
+		ti_qspi_setup_clk(qspi, mem->spi->max_speed_hz);
 		ti_qspi_enable_memory_map(mem->spi);
+	}
 	ti_qspi_setup_mmap_read(mem->spi, op->cmd.opcode, op->data.buswidth,
 				op->addr.nbytes, op->dummy.nbytes);
 
@@ -701,6 +703,7 @@ static int ti_qspi_start_transfer_one(struct spi_master *master,
 		wlen = t->bits_per_word >> 3;
 		transfer_len_words = min(t->len / wlen, frame_len_words);
 
+		ti_qspi_setup_clk(qspi, t->speed_hz);
 		ret = qspi_transfer_msg(qspi, t, transfer_len_words * wlen);
 		if (ret) {
 			dev_dbg(qspi->dev, "transfer message failed\n");
@@ -851,7 +854,7 @@ static int ti_qspi_probe(struct platform_device *pdev)
 	pm_runtime_enable(&pdev->dev);
 
 	if (!of_property_read_u32(np, "spi-max-frequency", &max_freq))
-		qspi->spi_max_frequency = max_freq;
+		master->max_speed_hz = max_freq;
 
 	dma_cap_zero(mask);
 	dma_cap_set(DMA_MEMCPY, mask);
diff --git a/drivers/spi/spi-topcliff-pch.c b/drivers/spi/spi-topcliff-pch.c
index dfaa1d79a78b..cbb60198a7f0 100644
--- a/drivers/spi/spi-topcliff-pch.c
+++ b/drivers/spi/spi-topcliff-pch.c
@@ -455,35 +455,10 @@ static void pch_spi_reset(struct spi_master *master)
 
 static int pch_spi_transfer(struct spi_device *pspi, struct spi_message *pmsg)
 {
-
-	struct spi_transfer *transfer;
 	struct pch_spi_data *data = spi_master_get_devdata(pspi->master);
 	int retval;
 	unsigned long flags;
 
-	spin_lock_irqsave(&data->lock, flags);
-	/* validate Tx/Rx buffers and Transfer length */
-	list_for_each_entry(transfer, &pmsg->transfers, transfer_list) {
-		if (!transfer->tx_buf && !transfer->rx_buf) {
-			dev_err(&pspi->dev,
-				"%s Tx and Rx buffer NULL\n", __func__);
-			retval = -EINVAL;
-			goto err_return_spinlock;
-		}
-
-		if (!transfer->len) {
-			dev_err(&pspi->dev, "%s Transfer length invalid\n",
-				__func__);
-			retval = -EINVAL;
-			goto err_return_spinlock;
-		}
-
-		dev_dbg(&pspi->dev,
-			"%s Tx/Rx buffer valid. Transfer length valid\n",
-			__func__);
-	}
-	spin_unlock_irqrestore(&data->lock, flags);
-
 	/* We won't process any messages if we have been asked to terminate */
 	if (data->status == STATUS_EXITING) {
 		dev_err(&pspi->dev, "%s status = STATUS_EXITING.\n", __func__);
@@ -518,10 +493,6 @@ static int pch_spi_transfer(struct spi_device *pspi, struct spi_message *pmsg)
 err_out:
 	dev_dbg(&pspi->dev, "%s RETURN=%d\n", __func__, retval);
 	return retval;
-err_return_spinlock:
-	dev_dbg(&pspi->dev, "%s RETURN=%d\n", __func__, retval);
-	spin_unlock_irqrestore(&data->lock, flags);
-	return retval;
 }
 
 static inline void pch_spi_select_chip(struct pch_spi_data *data,
@@ -1365,6 +1336,7 @@ static int pch_spi_pd_probe(struct platform_device *plat_dev)
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_LSB_FIRST;
 	master->bits_per_word_mask = SPI_BPW_MASK(8) | SPI_BPW_MASK(16);
 	master->max_speed_hz = PCH_MAX_BAUDRATE;
+	master->flags = SPI_CONTROLLER_MUST_RX | SPI_CONTROLLER_MUST_TX;
 
 	data->board_dat = board_dat;
 	data->plat_dev = plat_dev;
diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 2b5afae8ff7f..c760aac070e5 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -134,6 +134,8 @@
 #define GQSPI_DMA_UNALIGN		0x3
 #define GQSPI_DEFAULT_NUM_CS	1	/* Default number of chip selects */
 
+#define GQSPI_MAX_NUM_CS	2	/* Maximum number of chip selects */
+
 #define SPI_AUTOSUSPEND_TIMEOUT		3000
 enum mode_type {GQSPI_MODE_IO, GQSPI_MODE_DMA};
 
@@ -363,8 +365,13 @@ static void zynqmp_qspi_chipselect(struct spi_device *qspi, bool is_high)
 	genfifoentry |= GQSPI_GENFIFO_MODE_SPI;
 
 	if (!is_high) {
-		xqspi->genfifobus = GQSPI_GENFIFO_BUS_LOWER;
-		xqspi->genfifocs = GQSPI_GENFIFO_CS_LOWER;
+		if (!qspi->chip_select) {
+			xqspi->genfifobus = GQSPI_GENFIFO_BUS_LOWER;
+			xqspi->genfifocs = GQSPI_GENFIFO_CS_LOWER;
+		} else {
+			xqspi->genfifobus = GQSPI_GENFIFO_BUS_UPPER;
+			xqspi->genfifocs = GQSPI_GENFIFO_CS_UPPER;
+		}
 		genfifoentry |= xqspi->genfifobus;
 		genfifoentry |= xqspi->genfifocs;
 		genfifoentry |= GQSPI_GENFIFO_CS_SETUP;
@@ -1099,6 +1106,7 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 	struct zynqmp_qspi *xqspi;
 	struct device *dev = &pdev->dev;
 	struct device_node *np = dev->of_node;
+	u32 num_cs;
 
 	ctlr = spi_alloc_master(&pdev->dev, sizeof(*xqspi));
 	if (!ctlr)
@@ -1176,8 +1184,19 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 	if (ret)
 		goto clk_dis_all;
 
+	ret = of_property_read_u32(np, "num-cs", &num_cs);
+	if (ret < 0) {
+		ctlr->num_chipselect = GQSPI_DEFAULT_NUM_CS;
+	} else if (num_cs > GQSPI_MAX_NUM_CS) {
+		ret = -EINVAL;
+		dev_err(&pdev->dev, "only %d chip selects are available\n",
+			GQSPI_MAX_NUM_CS);
+		goto clk_dis_all;
+	} else {
+		ctlr->num_chipselect = num_cs;
+	}
+
 	ctlr->bits_per_word_mask = SPI_BPW_MASK(8);
-	ctlr->num_chipselect = GQSPI_DEFAULT_NUM_CS;
 	ctlr->mem_ops = &zynqmp_qspi_mem_ops;
 	ctlr->setup = zynqmp_qspi_setup_op;
 	ctlr->max_speed_hz = clk_get_rate(xqspi->refclk) / 2;
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index ea09d1b42bf6..1c14d682ffed 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -33,6 +33,7 @@
 #include <linux/idr.h>
 #include <linux/platform_data/x86/apple.h>
 #include <linux/ptp_clock_kernel.h>
+#include <linux/percpu.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/spi.h>
@@ -49,6 +50,7 @@ static void spidev_release(struct device *dev)
 
 	spi_controller_put(spi->controller);
 	kfree(spi->driver_override);
+	free_percpu(spi->pcpu_statistics);
 	kfree(spi);
 }
 
@@ -93,6 +95,47 @@ static ssize_t driver_override_show(struct device *dev,
 }
 static DEVICE_ATTR_RW(driver_override);
 
+static struct spi_statistics *spi_alloc_pcpu_stats(struct device *dev)
+{
+	struct spi_statistics __percpu *pcpu_stats;
+
+	if (dev)
+		pcpu_stats = devm_alloc_percpu(dev, struct spi_statistics);
+	else
+		pcpu_stats = alloc_percpu_gfp(struct spi_statistics, GFP_KERNEL);
+
+	if (pcpu_stats) {
+		int cpu;
+
+		for_each_possible_cpu(cpu) {
+			struct spi_statistics *stat;
+
+			stat = per_cpu_ptr(pcpu_stats, cpu);
+			u64_stats_init(&stat->syncp);
+		}
+	}
+	return pcpu_stats;
+}
+
+#define spi_pcpu_stats_totalize(ret, in, field)				\
+do {									\
+	int i;								\
+	ret = 0;							\
+	for_each_possible_cpu(i) {					\
+		const struct spi_statistics *pcpu_stats;		\
+		u64 inc;						\
+		unsigned int start;					\
+		pcpu_stats = per_cpu_ptr(in, i);			\
+		do {							\
+			start = u64_stats_fetch_begin_irq(		\
+					&pcpu_stats->syncp);		\
+			inc = u64_stats_read(&pcpu_stats->field);	\
+		} while (u64_stats_fetch_retry_irq(			\
+					&pcpu_stats->syncp, start));	\
+		ret += inc;						\
+	}								\
+} while (0)
+
 #define SPI_STATISTICS_ATTRS(field, file)				\
 static ssize_t spi_controller_##field##_show(struct device *dev,	\
 					     struct device_attribute *attr, \
@@ -100,7 +143,7 @@ static ssize_t spi_controller_##field##_show(struct device *dev,	\
 {									\
 	struct spi_controller *ctlr = container_of(dev,			\
 					 struct spi_controller, dev);	\
-	return spi_statistics_##field##_show(&ctlr->statistics, buf);	\
+	return spi_statistics_##field##_show(ctlr->pcpu_statistics, buf); \
 }									\
 static struct device_attribute dev_attr_spi_controller_##field = {	\
 	.attr = { .name = file, .mode = 0444 },				\
@@ -111,47 +154,46 @@ static ssize_t spi_device_##field##_show(struct device *dev,		\
 					char *buf)			\
 {									\
 	struct spi_device *spi = to_spi_device(dev);			\
-	return spi_statistics_##field##_show(&spi->statistics, buf);	\
+	return spi_statistics_##field##_show(spi->pcpu_statistics, buf); \
 }									\
 static struct device_attribute dev_attr_spi_device_##field = {		\
 	.attr = { .name = file, .mode = 0444 },				\
 	.show = spi_device_##field##_show,				\
 }
 
-#define SPI_STATISTICS_SHOW_NAME(name, file, field, format_string)	\
+#define SPI_STATISTICS_SHOW_NAME(name, file, field)			\
 static ssize_t spi_statistics_##name##_show(struct spi_statistics *stat, \
 					    char *buf)			\
 {									\
-	unsigned long flags;						\
 	ssize_t len;							\
-	spin_lock_irqsave(&stat->lock, flags);				\
-	len = sysfs_emit(buf, format_string "\n", stat->field);		\
-	spin_unlock_irqrestore(&stat->lock, flags);			\
+	u64 val;							\
+	spi_pcpu_stats_totalize(val, stat, field);			\
+	len = sysfs_emit(buf, "%llu\n", val);				\
 	return len;							\
 }									\
 SPI_STATISTICS_ATTRS(name, file)
 
-#define SPI_STATISTICS_SHOW(field, format_string)			\
+#define SPI_STATISTICS_SHOW(field)					\
 	SPI_STATISTICS_SHOW_NAME(field, __stringify(field),		\
-				 field, format_string)
+				 field)
 
-SPI_STATISTICS_SHOW(messages, "%lu");
-SPI_STATISTICS_SHOW(transfers, "%lu");
-SPI_STATISTICS_SHOW(errors, "%lu");
-SPI_STATISTICS_SHOW(timedout, "%lu");
+SPI_STATISTICS_SHOW(messages);
+SPI_STATISTICS_SHOW(transfers);
+SPI_STATISTICS_SHOW(errors);
+SPI_STATISTICS_SHOW(timedout);
 
-SPI_STATISTICS_SHOW(spi_sync, "%lu");
-SPI_STATISTICS_SHOW(spi_sync_immediate, "%lu");
-SPI_STATISTICS_SHOW(spi_async, "%lu");
+SPI_STATISTICS_SHOW(spi_sync);
+SPI_STATISTICS_SHOW(spi_sync_immediate);
+SPI_STATISTICS_SHOW(spi_async);
 
-SPI_STATISTICS_SHOW(bytes, "%llu");
-SPI_STATISTICS_SHOW(bytes_rx, "%llu");
-SPI_STATISTICS_SHOW(bytes_tx, "%llu");
+SPI_STATISTICS_SHOW(bytes);
+SPI_STATISTICS_SHOW(bytes_rx);
+SPI_STATISTICS_SHOW(bytes_tx);
 
 #define SPI_STATISTICS_TRANSFER_BYTES_HISTO(index, number)		\
 	SPI_STATISTICS_SHOW_NAME(transfer_bytes_histo##index,		\
 				 "transfer_bytes_histo_" number,	\
-				 transfer_bytes_histo[index],  "%lu")
+				 transfer_bytes_histo[index])
 SPI_STATISTICS_TRANSFER_BYTES_HISTO(0,  "0-1");
 SPI_STATISTICS_TRANSFER_BYTES_HISTO(1,  "2-3");
 SPI_STATISTICS_TRANSFER_BYTES_HISTO(2,  "4-7");
@@ -170,7 +212,7 @@ SPI_STATISTICS_TRANSFER_BYTES_HISTO(14, "16384-32767");
 SPI_STATISTICS_TRANSFER_BYTES_HISTO(15, "32768-65535");
 SPI_STATISTICS_TRANSFER_BYTES_HISTO(16, "65536+");
 
-SPI_STATISTICS_SHOW(transfers_split_maxsize, "%lu");
+SPI_STATISTICS_SHOW(transfers_split_maxsize);
 
 static struct attribute *spi_dev_attrs[] = {
 	&dev_attr_modalias.attr,
@@ -267,30 +309,33 @@ static const struct attribute_group *spi_master_groups[] = {
 	NULL,
 };
 
-static void spi_statistics_add_transfer_stats(struct spi_statistics *stats,
+static void spi_statistics_add_transfer_stats(struct spi_statistics *pcpu_stats,
 					      struct spi_transfer *xfer,
 					      struct spi_controller *ctlr)
 {
-	unsigned long flags;
 	int l2len = min(fls(xfer->len), SPI_STATISTICS_HISTO_SIZE) - 1;
+	struct spi_statistics *stats;
 
 	if (l2len < 0)
 		l2len = 0;
 
-	spin_lock_irqsave(&stats->lock, flags);
+	get_cpu();
+	stats = this_cpu_ptr(pcpu_stats);
+	u64_stats_update_begin(&stats->syncp);
 
-	stats->transfers++;
-	stats->transfer_bytes_histo[l2len]++;
+	u64_stats_inc(&stats->transfers);
+	u64_stats_inc(&stats->transfer_bytes_histo[l2len]);
 
-	stats->bytes += xfer->len;
+	u64_stats_add(&stats->bytes, xfer->len);
 	if ((xfer->tx_buf) &&
 	    (xfer->tx_buf != ctlr->dummy_tx))
-		stats->bytes_tx += xfer->len;
+		u64_stats_add(&stats->bytes_tx, xfer->len);
 	if ((xfer->rx_buf) &&
 	    (xfer->rx_buf != ctlr->dummy_rx))
-		stats->bytes_rx += xfer->len;
+		u64_stats_add(&stats->bytes_rx, xfer->len);
 
-	spin_unlock_irqrestore(&stats->lock, flags);
+	u64_stats_update_end(&stats->syncp);
+	put_cpu();
 }
 
 /*
@@ -519,14 +564,19 @@ struct spi_device *spi_alloc_device(struct spi_controller *ctlr)
 		return NULL;
 	}
 
+	spi->pcpu_statistics = spi_alloc_pcpu_stats(NULL);
+	if (!spi->pcpu_statistics) {
+		kfree(spi);
+		spi_controller_put(ctlr);
+		return NULL;
+	}
+
 	spi->master = spi->controller = ctlr;
 	spi->dev.parent = &ctlr->dev;
 	spi->dev.bus = &spi_bus_type;
 	spi->dev.release = spidev_release;
 	spi->mode = ctlr->buswidth_override_bits;
 
-	spin_lock_init(&spi->statistics.lock);
-
 	device_initialize(&spi->dev);
 	return spi;
 }
@@ -1225,8 +1275,8 @@ static int spi_transfer_wait(struct spi_controller *ctlr,
 			     struct spi_message *msg,
 			     struct spi_transfer *xfer)
 {
-	struct spi_statistics *statm = &ctlr->statistics;
-	struct spi_statistics *stats = &msg->spi->statistics;
+	struct spi_statistics *statm = ctlr->pcpu_statistics;
+	struct spi_statistics *stats = msg->spi->pcpu_statistics;
 	u32 speed_hz = xfer->speed_hz;
 	unsigned long long ms;
 
@@ -1304,7 +1354,7 @@ int spi_delay_to_ns(struct spi_delay *_delay, struct spi_transfer *xfer)
 		/* Nothing to do here */
 		break;
 	case SPI_DELAY_UNIT_SCK:
-		/* clock cycles need to be obtained from spi_transfer */
+		/* Clock cycles need to be obtained from spi_transfer */
 		if (!xfer)
 			return -EINVAL;
 		/*
@@ -1353,7 +1403,7 @@ static void _spi_transfer_cs_change_delay(struct spi_message *msg,
 	u32 unit = xfer->cs_change_delay.unit;
 	int ret;
 
-	/* return early on "fast" mode - for everything but USECS */
+	/* Return early on "fast" mode - for everything but USECS */
 	if (!delay) {
 		if (unit == SPI_DELAY_UNIT_USECS)
 			_spi_transfer_delay_ns(default_delay_ns);
@@ -1382,8 +1432,8 @@ static int spi_transfer_one_message(struct spi_controller *ctlr,
 	struct spi_transfer *xfer;
 	bool keep_cs = false;
 	int ret = 0;
-	struct spi_statistics *statm = &ctlr->statistics;
-	struct spi_statistics *stats = &msg->spi->statistics;
+	struct spi_statistics *statm = ctlr->pcpu_statistics;
+	struct spi_statistics *stats = msg->spi->pcpu_statistics;
 
 	spi_set_cs(msg->spi, true, false);
 
@@ -1499,6 +1549,103 @@ static void spi_idle_runtime_pm(struct spi_controller *ctlr)
 	}
 }
 
+static int __spi_pump_transfer_message(struct spi_controller *ctlr,
+		struct spi_message *msg, bool was_busy)
+{
+	struct spi_transfer *xfer;
+	int ret;
+
+	if (!was_busy && ctlr->auto_runtime_pm) {
+		ret = pm_runtime_get_sync(ctlr->dev.parent);
+		if (ret < 0) {
+			pm_runtime_put_noidle(ctlr->dev.parent);
+			dev_err(&ctlr->dev, "Failed to power device: %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	if (!was_busy)
+		trace_spi_controller_busy(ctlr);
+
+	if (!was_busy && ctlr->prepare_transfer_hardware) {
+		ret = ctlr->prepare_transfer_hardware(ctlr);
+		if (ret) {
+			dev_err(&ctlr->dev,
+				"failed to prepare transfer hardware: %d\n",
+				ret);
+
+			if (ctlr->auto_runtime_pm)
+				pm_runtime_put(ctlr->dev.parent);
+
+			msg->status = ret;
+			spi_finalize_current_message(ctlr);
+
+			return ret;
+		}
+	}
+
+	trace_spi_message_start(msg);
+
+	if (ctlr->prepare_message) {
+		ret = ctlr->prepare_message(ctlr, msg);
+		if (ret) {
+			dev_err(&ctlr->dev, "failed to prepare message: %d\n",
+				ret);
+			msg->status = ret;
+			spi_finalize_current_message(ctlr);
+			return ret;
+		}
+		msg->prepared = true;
+	}
+
+	ret = spi_map_msg(ctlr, msg);
+	if (ret) {
+		msg->status = ret;
+		spi_finalize_current_message(ctlr);
+		return ret;
+	}
+
+	if (!ctlr->ptp_sts_supported && !ctlr->transfer_one) {
+		list_for_each_entry(xfer, &msg->transfers, transfer_list) {
+			xfer->ptp_sts_word_pre = 0;
+			ptp_read_system_prets(xfer->ptp_sts);
+		}
+	}
+
+	/*
+	 * Drivers implementation of transfer_one_message() must arrange for
+	 * spi_finalize_current_message() to get called. Most drivers will do
+	 * this in the calling context, but some don't. For those cases, a
+	 * completion is used to guarantee that this function does not return
+	 * until spi_finalize_current_message() is done accessing
+	 * ctlr->cur_msg.
+	 * Use of the following two flags enable to opportunistically skip the
+	 * use of the completion since its use involves expensive spin locks.
+	 * In case of a race with the context that calls
+	 * spi_finalize_current_message() the completion will always be used,
+	 * due to strict ordering of these flags using barriers.
+	 */
+	WRITE_ONCE(ctlr->cur_msg_incomplete, true);
+	WRITE_ONCE(ctlr->cur_msg_need_completion, false);
+	reinit_completion(&ctlr->cur_msg_completion);
+	smp_wmb(); /* Make these available to spi_finalize_current_message() */
+
+	ret = ctlr->transfer_one_message(ctlr, msg);
+	if (ret) {
+		dev_err(&ctlr->dev,
+			"failed to transfer one message from queue\n");
+		return ret;
+	}
+
+	WRITE_ONCE(ctlr->cur_msg_need_completion, true);
+	smp_mb(); /* See spi_finalize_current_message()... */
+	if (READ_ONCE(ctlr->cur_msg_incomplete))
+		wait_for_completion(&ctlr->cur_msg_completion);
+
+	return 0;
+}
+
 /**
  * __spi_pump_messages - function which processes spi message queue
  * @ctlr: controller to process queue for
@@ -1514,34 +1661,25 @@ static void spi_idle_runtime_pm(struct spi_controller *ctlr)
  */
 static void __spi_pump_messages(struct spi_controller *ctlr, bool in_kthread)
 {
-	struct spi_transfer *xfer;
 	struct spi_message *msg;
 	bool was_busy = false;
 	unsigned long flags;
 	int ret;
 
+	/* Take the IO mutex */
+	mutex_lock(&ctlr->io_mutex);
+
 	/* Lock queue */
 	spin_lock_irqsave(&ctlr->queue_lock, flags);
 
 	/* Make sure we are not already running a message */
-	if (ctlr->cur_msg) {
-		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
-		return;
-	}
-
-	/* If another context is idling the device then defer */
-	if (ctlr->idling) {
-		kthread_queue_work(ctlr->kworker, &ctlr->pump_messages);
-		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
-		return;
-	}
+	if (ctlr->cur_msg)
+		goto out_unlock;
 
 	/* Check if the queue is idle */
 	if (list_empty(&ctlr->queue) || !ctlr->running) {
-		if (!ctlr->busy) {
-			spin_unlock_irqrestore(&ctlr->queue_lock, flags);
-			return;
-		}
+		if (!ctlr->busy)
+			goto out_unlock;
 
 		/* Defer any non-atomic teardown to the thread */
 		if (!in_kthread) {
@@ -1549,17 +1687,16 @@ static void __spi_pump_messages(struct spi_controller *ctlr, bool in_kthread)
 			    !ctlr->unprepare_transfer_hardware) {
 				spi_idle_runtime_pm(ctlr);
 				ctlr->busy = false;
+				ctlr->queue_empty = true;
 				trace_spi_controller_idle(ctlr);
 			} else {
 				kthread_queue_work(ctlr->kworker,
 						   &ctlr->pump_messages);
 			}
-			spin_unlock_irqrestore(&ctlr->queue_lock, flags);
-			return;
+			goto out_unlock;
 		}
 
 		ctlr->busy = false;
-		ctlr->idling = true;
 		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
 
 		kfree(ctlr->dummy_rx);
@@ -1574,9 +1711,8 @@ static void __spi_pump_messages(struct spi_controller *ctlr, bool in_kthread)
 		trace_spi_controller_idle(ctlr);
 
 		spin_lock_irqsave(&ctlr->queue_lock, flags);
-		ctlr->idling = false;
-		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
-		return;
+		ctlr->queue_empty = true;
+		goto out_unlock;
 	}
 
 	/* Extract head of queue */
@@ -1590,81 +1726,23 @@ static void __spi_pump_messages(struct spi_controller *ctlr, bool in_kthread)
 		ctlr->busy = true;
 	spin_unlock_irqrestore(&ctlr->queue_lock, flags);
 
-	mutex_lock(&ctlr->io_mutex);
-
-	if (!was_busy && ctlr->auto_runtime_pm) {
-		ret = pm_runtime_resume_and_get(ctlr->dev.parent);
-		if (ret < 0) {
-			dev_err(&ctlr->dev, "Failed to power device: %d\n",
-				ret);
-			mutex_unlock(&ctlr->io_mutex);
-			return;
-		}
-	}
-
-	if (!was_busy)
-		trace_spi_controller_busy(ctlr);
-
-	if (!was_busy && ctlr->prepare_transfer_hardware) {
-		ret = ctlr->prepare_transfer_hardware(ctlr);
-		if (ret) {
-			dev_err(&ctlr->dev,
-				"failed to prepare transfer hardware: %d\n",
-				ret);
-
-			if (ctlr->auto_runtime_pm)
-				pm_runtime_put(ctlr->dev.parent);
-
-			msg->status = ret;
-			spi_finalize_current_message(ctlr);
-
-			mutex_unlock(&ctlr->io_mutex);
-			return;
-		}
-	}
-
-	trace_spi_message_start(msg);
-
-	if (ctlr->prepare_message) {
-		ret = ctlr->prepare_message(ctlr, msg);
-		if (ret) {
-			dev_err(&ctlr->dev, "failed to prepare message: %d\n",
-				ret);
-			msg->status = ret;
-			spi_finalize_current_message(ctlr);
-			goto out;
-		}
-		ctlr->cur_msg_prepared = true;
-	}
-
-	ret = spi_map_msg(ctlr, msg);
-	if (ret) {
-		msg->status = ret;
-		spi_finalize_current_message(ctlr);
-		goto out;
-	}
-
-	if (!ctlr->ptp_sts_supported && !ctlr->transfer_one) {
-		list_for_each_entry(xfer, &msg->transfers, transfer_list) {
-			xfer->ptp_sts_word_pre = 0;
-			ptp_read_system_prets(xfer->ptp_sts);
-		}
-	}
+	ret = __spi_pump_transfer_message(ctlr, msg, was_busy);
+	if (!ret)
+		kthread_queue_work(ctlr->kworker, &ctlr->pump_messages);
 
-	ret = ctlr->transfer_one_message(ctlr, msg);
-	if (ret) {
-		dev_err(&ctlr->dev,
-			"failed to transfer one message from queue: %d\n",
-			ret);
-		goto out;
-	}
+	ctlr->cur_msg = NULL;
+	ctlr->fallback = false;
 
-out:
 	mutex_unlock(&ctlr->io_mutex);
 
 	/* Prod the scheduler in case transfer_one() was busy waiting */
 	if (!ret)
 		cond_resched();
+	return;
+
+out_unlock:
+	spin_unlock_irqrestore(&ctlr->queue_lock, flags);
+	mutex_unlock(&ctlr->io_mutex);
 }
 
 /**
@@ -1789,6 +1867,7 @@ static int spi_init_queue(struct spi_controller *ctlr)
 {
 	ctlr->running = false;
 	ctlr->busy = false;
+	ctlr->queue_empty = true;
 
 	ctlr->kworker = kthread_create_worker(0, dev_name(&ctlr->dev));
 	if (IS_ERR(ctlr->kworker)) {
@@ -1826,7 +1905,7 @@ struct spi_message *spi_get_next_queued_message(struct spi_controller *ctlr)
 	struct spi_message *next;
 	unsigned long flags;
 
-	/* get a pointer to the next message, if any */
+	/* Get a pointer to the next message, if any */
 	spin_lock_irqsave(&ctlr->queue_lock, flags);
 	next = list_first_entry_or_null(&ctlr->queue, struct spi_message,
 					queue);
@@ -1847,12 +1926,9 @@ void spi_finalize_current_message(struct spi_controller *ctlr)
 {
 	struct spi_transfer *xfer;
 	struct spi_message *mesg;
-	unsigned long flags;
 	int ret;
 
-	spin_lock_irqsave(&ctlr->queue_lock, flags);
 	mesg = ctlr->cur_msg;
-	spin_unlock_irqrestore(&ctlr->queue_lock, flags);
 
 	if (!ctlr->ptp_sts_supported && !ctlr->transfer_one) {
 		list_for_each_entry(xfer, &mesg->transfers, transfer_list) {
@@ -1876,7 +1952,7 @@ void spi_finalize_current_message(struct spi_controller *ctlr)
 	 */
 	spi_res_release(ctlr, mesg);
 
-	if (ctlr->cur_msg_prepared && ctlr->unprepare_message) {
+	if (mesg->prepared && ctlr->unprepare_message) {
 		ret = ctlr->unprepare_message(ctlr, mesg);
 		if (ret) {
 			dev_err(&ctlr->dev, "failed to unprepare message: %d\n",
@@ -1884,12 +1960,12 @@ void spi_finalize_current_message(struct spi_controller *ctlr)
 		}
 	}
 
-	spin_lock_irqsave(&ctlr->queue_lock, flags);
-	ctlr->cur_msg = NULL;
-	ctlr->cur_msg_prepared = false;
-	ctlr->fallback = false;
-	kthread_queue_work(ctlr->kworker, &ctlr->pump_messages);
-	spin_unlock_irqrestore(&ctlr->queue_lock, flags);
+	mesg->prepared = false;
+
+	WRITE_ONCE(ctlr->cur_msg_incomplete, false);
+	smp_mb(); /* See __spi_pump_transfer_message()... */
+	if (READ_ONCE(ctlr->cur_msg_need_completion))
+		complete(&ctlr->cur_msg_completion);
 
 	trace_spi_message_done(mesg);
 
@@ -1992,6 +2068,7 @@ static int __spi_queued_transfer(struct spi_device *spi,
 	msg->status = -EINPROGRESS;
 
 	list_add_tail(&msg->queue, &ctlr->queue);
+	ctlr->queue_empty = false;
 	if (!ctlr->busy && need_pump)
 		kthread_queue_work(ctlr->kworker, &ctlr->pump_messages);
 
@@ -2376,9 +2453,6 @@ static int acpi_spi_add_resource(struct acpi_resource *ares, void *data)
 			if (lookup->index != -1 && lookup->n++ != lookup->index)
 				return 1;
 
-			if (lookup->index == -1 && !ctlr)
-				return -ENODEV;
-
 			status = acpi_get_handle(NULL,
 						 sb->resource_source.string_ptr,
 						 &parent_handle);
@@ -2398,7 +2472,7 @@ static int acpi_spi_add_resource(struct acpi_resource *ares, void *data)
 
 				ctlr = acpi_spi_find_controller_by_adev(adev);
 				if (!ctlr)
-					return -ENODEV;
+					return -EPROBE_DEFER;
 
 				lookup->ctlr = ctlr;
 			}
@@ -2481,8 +2555,8 @@ struct spi_device *acpi_spi_device_alloc(struct spi_controller *ctlr,
 	acpi_dev_free_resource_list(&resource_list);
 
 	if (ret < 0)
-		/* found SPI in _CRS but it points to another controller */
-		return ERR_PTR(-ENODEV);
+		/* Found SPI in _CRS but it points to another controller */
+		return ERR_PTR(ret);
 
 	if (!lookup.max_speed_hz &&
 	    ACPI_SUCCESS(acpi_get_parent(adev->handle, &parent_handle)) &&
@@ -2937,7 +3011,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 		return status;
 
 	if (ctlr->bus_num >= 0) {
-		/* devices with a fixed bus num must check-in with the num */
+		/* Devices with a fixed bus num must check-in with the num */
 		mutex_lock(&board_lock);
 		id = idr_alloc(&spi_master_idr, ctlr, ctlr->bus_num,
 			ctlr->bus_num + 1, GFP_KERNEL);
@@ -2946,7 +3020,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 			return id == -ENOSPC ? -EBUSY : id;
 		ctlr->bus_num = id;
 	} else if (ctlr->dev.of_node) {
-		/* allocate dynamic bus number using Linux idr */
+		/* Allocate dynamic bus number using Linux idr */
 		id = of_alias_get_id(ctlr->dev.of_node, "spi");
 		if (id >= 0) {
 			ctlr->bus_num = id;
@@ -2975,6 +3049,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 	}
 	ctlr->bus_lock_flag = 0;
 	init_completion(&ctlr->xfer_completion);
+	init_completion(&ctlr->cur_msg_completion);
 	if (!ctlr->max_dma_len)
 		ctlr->max_dma_len = INT_MAX;
 
@@ -3004,7 +3079,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 		goto free_bus_id;
 	}
 
-	/* setting last_cs to -1 means no chip selected */
+	/* Setting last_cs to -1 means no chip selected */
 	ctlr->last_cs = -1;
 
 	status = device_add(&ctlr->dev);
@@ -3028,8 +3103,13 @@ int spi_register_controller(struct spi_controller *ctlr)
 			goto free_bus_id;
 		}
 	}
-	/* add statistics */
-	spin_lock_init(&ctlr->statistics.lock);
+	/* Add statistics */
+	ctlr->pcpu_statistics = spi_alloc_pcpu_stats(dev);
+	if (!ctlr->pcpu_statistics) {
+		dev_err(dev, "Error allocating per-cpu statistics\n");
+		status = -ENOMEM;
+		goto destroy_queue;
+	}
 
 	mutex_lock(&board_lock);
 	list_add_tail(&ctlr->list, &spi_controller_list);
@@ -3042,6 +3122,8 @@ int spi_register_controller(struct spi_controller *ctlr)
 	acpi_register_spi_devices(ctlr);
 	return status;
 
+destroy_queue:
+	spi_destroy_queue(ctlr);
 free_bus_id:
 	mutex_lock(&board_lock);
 	idr_remove(&spi_master_idr, ctlr->bus_num);
@@ -3050,9 +3132,9 @@ free_bus_id:
 }
 EXPORT_SYMBOL_GPL(spi_register_controller);
 
-static void devm_spi_unregister(void *ctlr)
+static void devm_spi_unregister(struct device *dev, void *res)
 {
-	spi_unregister_controller(ctlr);
+	spi_unregister_controller(*(struct spi_controller **)res);
 }
 
 /**
@@ -3071,13 +3153,22 @@ static void devm_spi_unregister(void *ctlr)
 int devm_spi_register_controller(struct device *dev,
 				 struct spi_controller *ctlr)
 {
+	struct spi_controller **ptr;
 	int ret;
 
+	ptr = devres_alloc(devm_spi_unregister, sizeof(*ptr), GFP_KERNEL);
+	if (!ptr)
+		return -ENOMEM;
+
 	ret = spi_register_controller(ctlr);
-	if (ret)
-		return ret;
+	if (!ret) {
+		*ptr = ctlr;
+		devres_add(dev, ptr);
+	} else {
+		devres_free(ptr);
+	}
 
-	return devm_add_action_or_reset(dev, devm_spi_unregister, ctlr);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(devm_spi_register_controller);
 
@@ -3124,7 +3215,7 @@ void spi_unregister_controller(struct spi_controller *ctlr)
 
 	device_del(&ctlr->dev);
 
-	/* free bus id */
+	/* Free bus id */
 	mutex_lock(&board_lock);
 	if (found == ctlr)
 		idr_remove(&spi_master_idr, id);
@@ -3183,14 +3274,14 @@ static void __spi_replace_transfers_release(struct spi_controller *ctlr,
 	struct spi_replaced_transfers *rxfer = res;
 	size_t i;
 
-	/* call extra callback if requested */
+	/* Call extra callback if requested */
 	if (rxfer->release)
 		rxfer->release(ctlr, msg, res);
 
-	/* insert replaced transfers back into the message */
+	/* Insert replaced transfers back into the message */
 	list_splice(&rxfer->replaced_transfers, rxfer->replaced_after);
 
-	/* remove the formerly inserted entries */
+	/* Remove the formerly inserted entries */
 	for (i = 0; i < rxfer->inserted; i++)
 		list_del(&rxfer->inserted_transfers[i].transfer_list);
 }
@@ -3223,7 +3314,7 @@ static struct spi_replaced_transfers *spi_replace_transfers(
 	struct spi_transfer *xfer;
 	size_t i;
 
-	/* allocate the structure using spi_res */
+	/* Allocate the structure using spi_res */
 	rxfer = spi_res_alloc(msg->spi, __spi_replace_transfers_release,
 			      struct_size(rxfer, inserted_transfers, insert)
 			      + extradatasize,
@@ -3231,15 +3322,15 @@ static struct spi_replaced_transfers *spi_replace_transfers(
 	if (!rxfer)
 		return ERR_PTR(-ENOMEM);
 
-	/* the release code to invoke before running the generic release */
+	/* The release code to invoke before running the generic release */
 	rxfer->release = release;
 
-	/* assign extradata */
+	/* Assign extradata */
 	if (extradatasize)
 		rxfer->extradata =
 			&rxfer->inserted_transfers[insert];
 
-	/* init the replaced_transfers list */
+	/* Init the replaced_transfers list */
 	INIT_LIST_HEAD(&rxfer->replaced_transfers);
 
 	/*
@@ -3248,7 +3339,7 @@ static struct spi_replaced_transfers *spi_replace_transfers(
 	 */
 	rxfer->replaced_after = xfer_first->transfer_list.prev;
 
-	/* remove the requested number of transfers */
+	/* Remove the requested number of transfers */
 	for (i = 0; i < remove; i++) {
 		/*
 		 * If the entry after replaced_after it is msg->transfers
@@ -3258,14 +3349,14 @@ static struct spi_replaced_transfers *spi_replace_transfers(
 		if (rxfer->replaced_after->next == &msg->transfers) {
 			dev_err(&msg->spi->dev,
 				"requested to remove more spi_transfers than are available\n");
-			/* insert replaced transfers back into the message */
+			/* Insert replaced transfers back into the message */
 			list_splice(&rxfer->replaced_transfers,
 				    rxfer->replaced_after);
 
-			/* free the spi_replace_transfer structure */
+			/* Free the spi_replace_transfer structure... */
 			spi_res_free(rxfer);
 
-			/* and return with an error */
+			/* ...and return with an error */
 			return ERR_PTR(-EINVAL);
 		}
 
@@ -3282,26 +3373,26 @@ static struct spi_replaced_transfers *spi_replace_transfers(
 	 * based on the first transfer to get removed.
 	 */
 	for (i = 0; i < insert; i++) {
-		/* we need to run in reverse order */
+		/* We need to run in reverse order */
 		xfer = &rxfer->inserted_transfers[insert - 1 - i];
 
-		/* copy all spi_transfer data */
+		/* Copy all spi_transfer data */
 		memcpy(xfer, xfer_first, sizeof(*xfer));
 
-		/* add to list */
+		/* Add to list */
 		list_add(&xfer->transfer_list, rxfer->replaced_after);
 
-		/* clear cs_change and delay for all but the last */
+		/* Clear cs_change and delay for all but the last */
 		if (i) {
 			xfer->cs_change = false;
 			xfer->delay.value = 0;
 		}
 	}
 
-	/* set up inserted */
+	/* Set up inserted... */
 	rxfer->inserted = insert;
 
-	/* and register it with spi_res/spi_message */
+	/* ...and register it with spi_res/spi_message */
 	spi_res_add(msg, rxfer);
 
 	return rxfer;
@@ -3318,10 +3409,10 @@ static int __spi_split_transfer_maxsize(struct spi_controller *ctlr,
 	size_t offset;
 	size_t count, i;
 
-	/* calculate how many we have to replace */
+	/* Calculate how many we have to replace */
 	count = DIV_ROUND_UP(xfer->len, maxsize);
 
-	/* create replacement */
+	/* Create replacement */
 	srt = spi_replace_transfers(msg, xfer, 1, count, NULL, 0, gfp);
 	if (IS_ERR(srt))
 		return PTR_ERR(srt);
@@ -3344,9 +3435,9 @@ static int __spi_split_transfer_maxsize(struct spi_controller *ctlr,
 	 */
 	xfers[0].len = min_t(size_t, maxsize, xfer[0].len);
 
-	/* all the others need rx_buf/tx_buf also set */
+	/* All the others need rx_buf/tx_buf also set */
 	for (i = 1, offset = maxsize; i < count; offset += maxsize, i++) {
-		/* update rx_buf, tx_buf and dma */
+		/* Update rx_buf, tx_buf and dma */
 		if (xfers[i].rx_buf)
 			xfers[i].rx_buf += offset;
 		if (xfers[i].rx_dma)
@@ -3356,7 +3447,7 @@ static int __spi_split_transfer_maxsize(struct spi_controller *ctlr,
 		if (xfers[i].tx_dma)
 			xfers[i].tx_dma += offset;
 
-		/* update length */
+		/* Update length */
 		xfers[i].len = min(maxsize, xfers[i].len - offset);
 	}
 
@@ -3366,10 +3457,10 @@ static int __spi_split_transfer_maxsize(struct spi_controller *ctlr,
 	 */
 	*xferp = &xfers[count - 1];
 
-	/* increment statistics counters */
-	SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics,
+	/* Increment statistics counters */
+	SPI_STATISTICS_INCREMENT_FIELD(ctlr->pcpu_statistics,
 				       transfers_split_maxsize);
-	SPI_STATISTICS_INCREMENT_FIELD(&msg->spi->statistics,
+	SPI_STATISTICS_INCREMENT_FIELD(msg->spi->pcpu_statistics,
 				       transfers_split_maxsize);
 
 	return 0;
@@ -3628,7 +3719,7 @@ static int __spi_validate(struct spi_device *spi, struct spi_message *message)
 			return ret;
 
 		list_for_each_entry(xfer, &message->transfers, transfer_list) {
-			/* don't change cs_change on the last entry in the list */
+			/* Don't change cs_change on the last entry in the list */
 			if (list_is_last(&xfer->transfer_list, &message->transfers))
 				break;
 			xfer->cs_change = 1;
@@ -3721,7 +3812,7 @@ static int __spi_validate(struct spi_device *spi, struct spi_message *message)
 				!(spi->mode & SPI_TX_QUAD))
 				return -EINVAL;
 		}
-		/* check transfer rx_nbits */
+		/* Check transfer rx_nbits */
 		if (xfer->rx_buf) {
 			if (spi->mode & SPI_NO_RX)
 				return -EINVAL;
@@ -3760,8 +3851,8 @@ static int __spi_async(struct spi_device *spi, struct spi_message *message)
 
 	message->spi = spi;
 
-	SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics, spi_async);
-	SPI_STATISTICS_INCREMENT_FIELD(&spi->statistics, spi_async);
+	SPI_STATISTICS_INCREMENT_FIELD(ctlr->pcpu_statistics, spi_async);
+	SPI_STATISTICS_INCREMENT_FIELD(spi->pcpu_statistics, spi_async);
 
 	trace_spi_message_submit(message);
 
@@ -3880,6 +3971,39 @@ static int spi_async_locked(struct spi_device *spi, struct spi_message *message)
 
 }
 
+static void __spi_transfer_message_noqueue(struct spi_controller *ctlr, struct spi_message *msg)
+{
+	bool was_busy;
+	int ret;
+
+	mutex_lock(&ctlr->io_mutex);
+
+	was_busy = ctlr->busy;
+
+	ctlr->cur_msg = msg;
+	ret = __spi_pump_transfer_message(ctlr, msg, was_busy);
+	if (ret)
+		goto out;
+
+	ctlr->cur_msg = NULL;
+	ctlr->fallback = false;
+
+	if (!was_busy) {
+		kfree(ctlr->dummy_rx);
+		ctlr->dummy_rx = NULL;
+		kfree(ctlr->dummy_tx);
+		ctlr->dummy_tx = NULL;
+		if (ctlr->unprepare_transfer_hardware &&
+		    ctlr->unprepare_transfer_hardware(ctlr))
+			dev_err(&ctlr->dev,
+				"failed to unprepare transfer hardware\n");
+		spi_idle_runtime_pm(ctlr);
+	}
+
+out:
+	mutex_unlock(&ctlr->io_mutex);
+}
+
 /*-------------------------------------------------------------------------*/
 
 /*
@@ -3898,51 +4022,51 @@ static int __spi_sync(struct spi_device *spi, struct spi_message *message)
 	DECLARE_COMPLETION_ONSTACK(done);
 	int status;
 	struct spi_controller *ctlr = spi->controller;
-	unsigned long flags;
 
 	status = __spi_validate(spi, message);
 	if (status != 0)
 		return status;
 
-	message->complete = spi_complete;
-	message->context = &done;
 	message->spi = spi;
 
-	SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics, spi_sync);
-	SPI_STATISTICS_INCREMENT_FIELD(&spi->statistics, spi_sync);
+	SPI_STATISTICS_INCREMENT_FIELD(ctlr->pcpu_statistics, spi_sync);
+	SPI_STATISTICS_INCREMENT_FIELD(spi->pcpu_statistics, spi_sync);
 
 	/*
-	 * If we're not using the legacy transfer method then we will
-	 * try to transfer in the calling context so special case.
-	 * This code would be less tricky if we could remove the
-	 * support for driver implemented message queues.
+	 * Checking queue_empty here only guarantees async/sync message
+	 * ordering when coming from the same context. It does not need to
+	 * guard against reentrancy from a different context. The io_mutex
+	 * will catch those cases.
 	 */
-	if (ctlr->transfer == spi_queued_transfer) {
-		spin_lock_irqsave(&ctlr->bus_lock_spinlock, flags);
+	if (READ_ONCE(ctlr->queue_empty)) {
+		message->actual_length = 0;
+		message->status = -EINPROGRESS;
 
 		trace_spi_message_submit(message);
 
-		status = __spi_queued_transfer(spi, message, false);
+		SPI_STATISTICS_INCREMENT_FIELD(ctlr->pcpu_statistics, spi_sync_immediate);
+		SPI_STATISTICS_INCREMENT_FIELD(spi->pcpu_statistics, spi_sync_immediate);
 
-		spin_unlock_irqrestore(&ctlr->bus_lock_spinlock, flags);
-	} else {
-		status = spi_async_locked(spi, message);
+		__spi_transfer_message_noqueue(ctlr, message);
+
+		return message->status;
 	}
 
+	/*
+	 * There are messages in the async queue that could have originated
+	 * from the same context, so we need to preserve ordering.
+	 * Therefor we send the message to the async queue and wait until they
+	 * are completed.
+	 */
+	message->complete = spi_complete;
+	message->context = &done;
+	status = spi_async_locked(spi, message);
 	if (status == 0) {
-		/* Push out the messages in the calling context if we can */
-		if (ctlr->transfer == spi_queued_transfer) {
-			SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics,
-						       spi_sync_immediate);
-			SPI_STATISTICS_INCREMENT_FIELD(&spi->statistics,
-						       spi_sync_immediate);
-			__spi_pump_messages(ctlr, false);
-		}
-
 		wait_for_completion(&done);
 		status = message->status;
 	}
 	message->context = NULL;
+
 	return status;
 }
 
@@ -4026,7 +4150,7 @@ int spi_bus_lock(struct spi_controller *ctlr)
 	ctlr->bus_lock_flag = 1;
 	spin_unlock_irqrestore(&ctlr->bus_lock_spinlock, flags);
 
-	/* mutex remains locked until spi_bus_unlock is called */
+	/* Mutex remains locked until spi_bus_unlock() is called */
 
 	return 0;
 }
@@ -4055,7 +4179,7 @@ int spi_bus_unlock(struct spi_controller *ctlr)
 }
 EXPORT_SYMBOL_GPL(spi_bus_unlock);
 
-/* portable code must never pass more than 32 bytes */
+/* Portable code must never pass more than 32 bytes */
 #define	SPI_BUFSIZ	max(32, SMP_CACHE_BYTES)
 
 static u8	*buf;
@@ -4121,7 +4245,7 @@ int spi_write_then_read(struct spi_device *spi,
 	x[0].tx_buf = local_buf;
 	x[1].rx_buf = local_buf + n_tx;
 
-	/* do the i/o */
+	/* Do the i/o */
 	status = spi_sync(spi, &message);
 	if (status == 0)
 		memcpy(rxbuf, x[1].rx_buf, n_rx);
@@ -4138,7 +4262,7 @@ EXPORT_SYMBOL_GPL(spi_write_then_read);
 /*-------------------------------------------------------------------------*/
 
 #if IS_ENABLED(CONFIG_OF_DYNAMIC)
-/* must call put_device() when done with returned spi_device device */
+/* Must call put_device() when done with returned spi_device device */
 static struct spi_device *of_find_spi_device_by_node(struct device_node *node)
 {
 	struct device *dev = bus_find_device_by_of_node(&spi_bus_type, node);
@@ -4146,7 +4270,7 @@ static struct spi_device *of_find_spi_device_by_node(struct device_node *node)
 	return dev ? to_spi_device(dev) : NULL;
 }
 
-/* the spi controllers are not using spi_bus, so we find it with another way */
+/* The spi controllers are not using spi_bus, so we find it with another way */
 static struct spi_controller *of_find_spi_controller_by_node(struct device_node *node)
 {
 	struct device *dev;
@@ -4157,7 +4281,7 @@ static struct spi_controller *of_find_spi_controller_by_node(struct device_node
 	if (!dev)
 		return NULL;
 
-	/* reference got in class_find_device */
+	/* Reference got in class_find_device */
 	return container_of(dev, struct spi_controller, dev);
 }
 
@@ -4172,7 +4296,7 @@ static int of_spi_notify(struct notifier_block *nb, unsigned long action,
 	case OF_RECONFIG_CHANGE_ADD:
 		ctlr = of_find_spi_controller_by_node(rd->dn->parent);
 		if (ctlr == NULL)
-			return NOTIFY_OK;	/* not for us */
+			return NOTIFY_OK;	/* Not for us */
 
 		if (of_node_test_and_set_flag(rd->dn, OF_POPULATED)) {
 			put_device(&ctlr->dev);
@@ -4191,19 +4315,19 @@ static int of_spi_notify(struct notifier_block *nb, unsigned long action,
 		break;
 
 	case OF_RECONFIG_CHANGE_REMOVE:
-		/* already depopulated? */
+		/* Already depopulated? */
 		if (!of_node_check_flag(rd->dn, OF_POPULATED))
 			return NOTIFY_OK;
 
-		/* find our device by node */
+		/* Find our device by node */
 		spi = of_find_spi_device_by_node(rd->dn);
 		if (spi == NULL)
-			return NOTIFY_OK;	/* no? not meant for us */
+			return NOTIFY_OK;	/* No? not meant for us */
 
-		/* unregister takes one ref away */
+		/* Unregister takes one ref away */
 		spi_unregister_device(spi);
 
-		/* and put the reference of the find */
+		/* And put the reference of the find */
 		put_device(&spi->dev);
 		break;
 	}