diff --git a/drivers/spi/atmel-quadspi.c b/drivers/spi/atmel-quadspi.c
index fd8007ebb14589ba6ef0d5419878ab5e573687ae..13def7f78b9e99c9829af817496c794071042981 100644
--- a/drivers/spi/atmel-quadspi.c
+++ b/drivers/spi/atmel-quadspi.c
@@ -149,6 +149,7 @@ struct atmel_qspi {
 	struct clk		*qspick;
 	struct platform_device	*pdev;
 	const struct atmel_qspi_caps *caps;
+	resource_size_t		mmap_size;
 	u32			pending;
 	u32			mr;
 	u32			scr;
@@ -329,6 +330,14 @@ static int atmel_qspi_exec_op(struct spi_mem *mem, const struct spi_mem_op *op)
 	u32 sr, offset;
 	int err;
 
+	/*
+	 * Check if the address exceeds the MMIO window size. An improvement
+	 * would be to add support for regular SPI mode and fall back to it
+	 * when the flash memories overrun the controller's memory space.
+	 */
+	if (op->addr.val + op->data.nbytes > aq->mmap_size)
+		return -ENOTSUPP;
+
 	err = atmel_qspi_set_cfg(aq, op, &offset);
 	if (err)
 		return err;
@@ -480,6 +489,8 @@ static int atmel_qspi_probe(struct platform_device *pdev)
 		goto exit;
 	}
 
+	aq->mmap_size = resource_size(res);
+
 	/* Get the peripheral clock */
 	aq->pclk = devm_clk_get(&pdev->dev, "pclk");
 	if (IS_ERR(aq->pclk))
diff --git a/drivers/spi/spi-bcm63xx-hsspi.c b/drivers/spi/spi-bcm63xx-hsspi.c
index 7327309ea3d51772ad5c57edbc3d509192d93d05..6c235306c0e4a8658e31fed475e37057d94a3663 100644
--- a/drivers/spi/spi-bcm63xx-hsspi.c
+++ b/drivers/spi/spi-bcm63xx-hsspi.c
@@ -366,7 +366,6 @@ static int bcm63xx_hsspi_probe(struct platform_device *pdev)
 			goto out_disable_clk;
 
 		rate = clk_get_rate(pll_clk);
-		clk_disable_unprepare(pll_clk);
 		if (!rate) {
 			ret = -EINVAL;
 			goto out_disable_pll_clk;
diff --git a/drivers/spi/spi-omap2-mcspi.c b/drivers/spi/spi-omap2-mcspi.c
index 7e2292c11d120b44d904c93c7433724a0ea7371e..e9e256718ef4aa9806937acc29ad5554411350d2 100644
--- a/drivers/spi/spi-omap2-mcspi.c
+++ b/drivers/spi/spi-omap2-mcspi.c
@@ -130,6 +130,7 @@ struct omap2_mcspi {
 	int			fifo_depth;
 	bool			slave_aborted;
 	unsigned int		pin_dir:1;
+	size_t			max_xfer_len;
 };
 
 struct omap2_mcspi_cs {
@@ -974,20 +975,12 @@ static int omap2_mcspi_setup_transfer(struct spi_device *spi,
  * Note that we currently allow DMA only if we get a channel
  * for both rx and tx. Otherwise we'll do PIO for both rx and tx.
  */
-static int omap2_mcspi_request_dma(struct spi_device *spi)
+static int omap2_mcspi_request_dma(struct omap2_mcspi *mcspi,
+				   struct omap2_mcspi_dma *mcspi_dma)
 {
-	struct spi_master	*master = spi->master;
-	struct omap2_mcspi	*mcspi;
-	struct omap2_mcspi_dma	*mcspi_dma;
 	int ret = 0;
 
-	mcspi = spi_master_get_devdata(master);
-	mcspi_dma = mcspi->dma_channels + spi->chip_select;
-
-	init_completion(&mcspi_dma->dma_rx_completion);
-	init_completion(&mcspi_dma->dma_tx_completion);
-
-	mcspi_dma->dma_rx = dma_request_chan(&master->dev,
+	mcspi_dma->dma_rx = dma_request_chan(mcspi->dev,
 					     mcspi_dma->dma_rx_ch_name);
 	if (IS_ERR(mcspi_dma->dma_rx)) {
 		ret = PTR_ERR(mcspi_dma->dma_rx);
@@ -995,7 +988,7 @@ static int omap2_mcspi_request_dma(struct spi_device *spi)
 		goto no_dma;
 	}
 
-	mcspi_dma->dma_tx = dma_request_chan(&master->dev,
+	mcspi_dma->dma_tx = dma_request_chan(mcspi->dev,
 					     mcspi_dma->dma_tx_ch_name);
 	if (IS_ERR(mcspi_dma->dma_tx)) {
 		ret = PTR_ERR(mcspi_dma->dma_tx);
@@ -1004,20 +997,40 @@ static int omap2_mcspi_request_dma(struct spi_device *spi)
 		mcspi_dma->dma_rx = NULL;
 	}
 
+	init_completion(&mcspi_dma->dma_rx_completion);
+	init_completion(&mcspi_dma->dma_tx_completion);
+
 no_dma:
 	return ret;
 }
 
+static void omap2_mcspi_release_dma(struct spi_master *master)
+{
+	struct omap2_mcspi *mcspi = spi_master_get_devdata(master);
+	struct omap2_mcspi_dma	*mcspi_dma;
+	int i;
+
+	for (i = 0; i < master->num_chipselect; i++) {
+		mcspi_dma = &mcspi->dma_channels[i];
+
+		if (mcspi_dma->dma_rx) {
+			dma_release_channel(mcspi_dma->dma_rx);
+			mcspi_dma->dma_rx = NULL;
+		}
+		if (mcspi_dma->dma_tx) {
+			dma_release_channel(mcspi_dma->dma_tx);
+			mcspi_dma->dma_tx = NULL;
+		}
+	}
+}
+
 static int omap2_mcspi_setup(struct spi_device *spi)
 {
 	int			ret;
 	struct omap2_mcspi	*mcspi = spi_master_get_devdata(spi->master);
 	struct omap2_mcspi_regs	*ctx = &mcspi->ctx;
-	struct omap2_mcspi_dma	*mcspi_dma;
 	struct omap2_mcspi_cs	*cs = spi->controller_state;
 
-	mcspi_dma = &mcspi->dma_channels[spi->chip_select];
-
 	if (!cs) {
 		cs = kzalloc(sizeof *cs, GFP_KERNEL);
 		if (!cs)
@@ -1042,13 +1055,6 @@ static int omap2_mcspi_setup(struct spi_device *spi)
 		}
 	}
 
-	if (!mcspi_dma->dma_rx || !mcspi_dma->dma_tx) {
-		ret = omap2_mcspi_request_dma(spi);
-		if (ret)
-			dev_warn(&spi->dev, "not using DMA for McSPI (%d)\n",
-				 ret);
-	}
-
 	ret = pm_runtime_get_sync(mcspi->dev);
 	if (ret < 0) {
 		pm_runtime_put_noidle(mcspi->dev);
@@ -1065,12 +1071,8 @@ static int omap2_mcspi_setup(struct spi_device *spi)
 
 static void omap2_mcspi_cleanup(struct spi_device *spi)
 {
-	struct omap2_mcspi	*mcspi;
-	struct omap2_mcspi_dma	*mcspi_dma;
 	struct omap2_mcspi_cs	*cs;
 
-	mcspi = spi_master_get_devdata(spi->master);
-
 	if (spi->controller_state) {
 		/* Unlink controller state from context save list */
 		cs = spi->controller_state;
@@ -1079,19 +1081,6 @@ static void omap2_mcspi_cleanup(struct spi_device *spi)
 		kfree(cs);
 	}
 
-	if (spi->chip_select < spi->master->num_chipselect) {
-		mcspi_dma = &mcspi->dma_channels[spi->chip_select];
-
-		if (mcspi_dma->dma_rx) {
-			dma_release_channel(mcspi_dma->dma_rx);
-			mcspi_dma->dma_rx = NULL;
-		}
-		if (mcspi_dma->dma_tx) {
-			dma_release_channel(mcspi_dma->dma_tx);
-			mcspi_dma->dma_tx = NULL;
-		}
-	}
-
 	if (gpio_is_valid(spi->cs_gpio))
 		gpio_free(spi->cs_gpio);
 }
@@ -1302,9 +1291,24 @@ static bool omap2_mcspi_can_dma(struct spi_master *master,
 	if (spi_controller_is_slave(master))
 		return true;
 
+	master->dma_rx = mcspi_dma->dma_rx;
+	master->dma_tx = mcspi_dma->dma_tx;
+
 	return (xfer->len >= DMA_MIN_BYTES);
 }
 
+static size_t omap2_mcspi_max_xfer_size(struct spi_device *spi)
+{
+	struct omap2_mcspi *mcspi = spi_master_get_devdata(spi->master);
+	struct omap2_mcspi_dma *mcspi_dma =
+		&mcspi->dma_channels[spi->chip_select];
+
+	if (mcspi->max_xfer_len && mcspi_dma->dma_rx)
+		return mcspi->max_xfer_len;
+
+	return SIZE_MAX;
+}
+
 static int omap2_mcspi_controller_setup(struct omap2_mcspi *mcspi)
 {
 	struct spi_master	*master = mcspi->master;
@@ -1373,6 +1377,11 @@ static struct omap2_mcspi_platform_config omap4_pdata = {
 	.regs_offset = OMAP4_MCSPI_REG_OFFSET,
 };
 
+static struct omap2_mcspi_platform_config am654_pdata = {
+	.regs_offset = OMAP4_MCSPI_REG_OFFSET,
+	.max_xfer_len = SZ_4K - 1,
+};
+
 static const struct of_device_id omap_mcspi_of_match[] = {
 	{
 		.compatible = "ti,omap2-mcspi",
@@ -1382,6 +1391,10 @@ static const struct of_device_id omap_mcspi_of_match[] = {
 		.compatible = "ti,omap4-mcspi",
 		.data = &omap4_pdata,
 	},
+	{
+		.compatible = "ti,am654-mcspi",
+		.data = &am654_pdata,
+	},
 	{ },
 };
 MODULE_DEVICE_TABLE(of, omap_mcspi_of_match);
@@ -1439,6 +1452,10 @@ static int omap2_mcspi_probe(struct platform_device *pdev)
 		mcspi->pin_dir = pdata->pin_dir;
 	}
 	regs_offset = pdata->regs_offset;
+	if (pdata->max_xfer_len) {
+		mcspi->max_xfer_len = pdata->max_xfer_len;
+		master->max_transfer_size = omap2_mcspi_max_xfer_size;
+	}
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	mcspi->base = devm_ioremap_resource(&pdev->dev, r);
@@ -1464,6 +1481,11 @@ static int omap2_mcspi_probe(struct platform_device *pdev)
 	for (i = 0; i < master->num_chipselect; i++) {
 		sprintf(mcspi->dma_channels[i].dma_rx_ch_name, "rx%d", i);
 		sprintf(mcspi->dma_channels[i].dma_tx_ch_name, "tx%d", i);
+
+		status = omap2_mcspi_request_dma(mcspi,
+						 &mcspi->dma_channels[i]);
+		if (status == -EPROBE_DEFER)
+			goto free_master;
 	}
 
 	status = platform_get_irq(pdev, 0);
@@ -1501,6 +1523,7 @@ static int omap2_mcspi_probe(struct platform_device *pdev)
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 free_master:
+	omap2_mcspi_release_dma(master);
 	spi_master_put(master);
 	return status;
 }
@@ -1510,6 +1533,8 @@ static int omap2_mcspi_remove(struct platform_device *pdev)
 	struct spi_master *master = platform_get_drvdata(pdev);
 	struct omap2_mcspi *mcspi = spi_master_get_devdata(master);
 
+	omap2_mcspi_release_dma(master);
+
 	pm_runtime_dont_use_autosuspend(mcspi->dev);
 	pm_runtime_put_sync(mcspi->dev);
 	pm_runtime_disable(&pdev->dev);
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c
index 4c7a71f0fb3e3fc7a2d53f6bd076ce8ac364e171..2e318158fca97686200ff2aa6b8169158911c204 100644
--- a/drivers/spi/spi-pxa2xx.c
+++ b/drivers/spi/spi-pxa2xx.c
@@ -70,6 +70,10 @@ MODULE_ALIAS("platform:pxa2xx-spi");
 #define LPSS_CAPS_CS_EN_SHIFT			9
 #define LPSS_CAPS_CS_EN_MASK			(0xf << LPSS_CAPS_CS_EN_SHIFT)
 
+#define LPSS_PRIV_CLOCK_GATE 0x38
+#define LPSS_PRIV_CLOCK_GATE_CLK_CTL_MASK 0x3
+#define LPSS_PRIV_CLOCK_GATE_CLK_CTL_FORCE_ON 0x3
+
 struct lpss_config {
 	/* LPSS offset from drv_data->ioaddr */
 	unsigned offset;
@@ -86,6 +90,8 @@ struct lpss_config {
 	unsigned cs_sel_shift;
 	unsigned cs_sel_mask;
 	unsigned cs_num;
+	/* Quirks */
+	unsigned cs_clk_stays_gated : 1;
 };
 
 /* Keep these sorted with enum pxa_ssp_type */
@@ -156,6 +162,7 @@ static const struct lpss_config lpss_platforms[] = {
 		.tx_threshold_hi = 56,
 		.cs_sel_shift = 8,
 		.cs_sel_mask = 3 << 8,
+		.cs_clk_stays_gated = true,
 	},
 };
 
@@ -383,6 +390,22 @@ static void lpss_ssp_cs_control(struct spi_device *spi, bool enable)
 	else
 		value |= LPSS_CS_CONTROL_CS_HIGH;
 	__lpss_ssp_write_priv(drv_data, config->reg_cs_ctrl, value);
+	if (config->cs_clk_stays_gated) {
+		u32 clkgate;
+
+		/*
+		 * Changing CS alone when dynamic clock gating is on won't
+		 * actually flip CS at that time. This ruins SPI transfers
+		 * that specify delays, or have no data. Toggle the clock mode
+		 * to force on briefly to poke the CS pin to move.
+		 */
+		clkgate = __lpss_ssp_read_priv(drv_data, LPSS_PRIV_CLOCK_GATE);
+		value = (clkgate & ~LPSS_PRIV_CLOCK_GATE_CLK_CTL_MASK) |
+			LPSS_PRIV_CLOCK_GATE_CLK_CTL_FORCE_ON;
+
+		__lpss_ssp_write_priv(drv_data, LPSS_PRIV_CLOCK_GATE, value);
+		__lpss_ssp_write_priv(drv_data, LPSS_PRIV_CLOCK_GATE, clkgate);
+	}
 }
 
 static void cs_assert(struct spi_device *spi)
diff --git a/drivers/spi/spi-qup.c b/drivers/spi/spi-qup.c
index dd3434a407ea669380e770760dcac4629f7e2fe0..a364b99497e26310fb4de0e7a3f546bfb61fcbd7 100644
--- a/drivers/spi/spi-qup.c
+++ b/drivers/spi/spi-qup.c
@@ -1217,6 +1217,11 @@ static int spi_qup_suspend(struct device *device)
 	struct spi_qup *controller = spi_master_get_devdata(master);
 	int ret;
 
+	if (pm_runtime_suspended(device)) {
+		ret = spi_qup_pm_resume_runtime(device);
+		if (ret)
+			return ret;
+	}
 	ret = spi_master_suspend(master);
 	if (ret)
 		return ret;
@@ -1225,10 +1230,8 @@ static int spi_qup_suspend(struct device *device)
 	if (ret)
 		return ret;
 
-	if (!pm_runtime_suspended(device)) {
-		clk_disable_unprepare(controller->cclk);
-		clk_disable_unprepare(controller->iclk);
-	}
+	clk_disable_unprepare(controller->cclk);
+	clk_disable_unprepare(controller->iclk);
 	return 0;
 }
 
diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 60c4de4e448568f2a2b54ce1da5f21ea510401da..7412a3042a8d2c69be4c438ef4af2bf89f4f131e 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -401,9 +401,6 @@ static void zynqmp_qspi_chipselect(struct spi_device *qspi, bool is_high)
 
 	zynqmp_gqspi_write(xqspi, GQSPI_GEN_FIFO_OFST, genfifoentry);
 
-	/* Dummy generic FIFO entry */
-	zynqmp_gqspi_write(xqspi, GQSPI_GEN_FIFO_OFST, 0x0);
-
 	/* Manually start the generic FIFO command */
 	zynqmp_gqspi_write(xqspi, GQSPI_CONFIG_OFST,
 			zynqmp_gqspi_read(xqspi, GQSPI_CONFIG_OFST) |
diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 38b4c78df506c060fa49d9b86e97020e7a910b48..755221bc3745c0602bc1bb9a2e2dafbb890f3988 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -2639,7 +2639,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 		if (ctlr->use_gpio_descriptors) {
 			status = spi_get_gpio_descs(ctlr);
 			if (status)
-				return status;
+				goto free_bus_id;
 			/*
 			 * A controller using GPIO descriptors always
 			 * supports SPI_CS_HIGH if need be.
@@ -2649,7 +2649,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 			/* Legacy code path for GPIOs from DT */
 			status = of_spi_get_gpio_numbers(ctlr);
 			if (status)
-				return status;
+				goto free_bus_id;
 		}
 	}
 
@@ -2657,17 +2657,14 @@ int spi_register_controller(struct spi_controller *ctlr)
 	 * Even if it's just one always-selected device, there must
 	 * be at least one chipselect.
 	 */
-	if (!ctlr->num_chipselect)
-		return -EINVAL;
+	if (!ctlr->num_chipselect) {
+		status = -EINVAL;
+		goto free_bus_id;
+	}
 
 	status = device_add(&ctlr->dev);
-	if (status < 0) {
-		/* free bus id */
-		mutex_lock(&board_lock);
-		idr_remove(&spi_master_idr, ctlr->bus_num);
-		mutex_unlock(&board_lock);
-		goto done;
-	}
+	if (status < 0)
+		goto free_bus_id;
 	dev_dbg(dev, "registered %s %s\n",
 			spi_controller_is_slave(ctlr) ? "slave" : "master",
 			dev_name(&ctlr->dev));
@@ -2683,11 +2680,7 @@ int spi_register_controller(struct spi_controller *ctlr)
 		status = spi_controller_initialize_queue(ctlr);
 		if (status) {
 			device_del(&ctlr->dev);
-			/* free bus id */
-			mutex_lock(&board_lock);
-			idr_remove(&spi_master_idr, ctlr->bus_num);
-			mutex_unlock(&board_lock);
-			goto done;
+			goto free_bus_id;
 		}
 	}
 	/* add statistics */
@@ -2702,7 +2695,12 @@ int spi_register_controller(struct spi_controller *ctlr)
 	/* Register devices from the device tree and ACPI */
 	of_register_spi_devices(ctlr);
 	acpi_register_spi_devices(ctlr);
-done:
+	return status;
+
+free_bus_id:
+	mutex_lock(&board_lock);
+	idr_remove(&spi_master_idr, ctlr->bus_num);
+	mutex_unlock(&board_lock);
 	return status;
 }
 EXPORT_SYMBOL_GPL(spi_register_controller);
diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index 1e217e3e9486ebd9d1147a3e95700f43d5e7935f..2ab6e782f14c2527ae7c698d1975530594e33c70 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -396,6 +396,7 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		else
 			retval = get_user(tmp, (u32 __user *)arg);
 		if (retval == 0) {
+			struct spi_controller *ctlr = spi->controller;
 			u32	save = spi->mode;
 
 			if (tmp & ~SPI_MODE_MASK) {
@@ -403,6 +404,10 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 				break;
 			}
 
+			if (ctlr->use_gpio_descriptors && ctlr->cs_gpiods &&
+			    ctlr->cs_gpiods[spi->chip_select])
+				tmp |= SPI_CS_HIGH;
+
 			tmp |= spi->mode & ~SPI_MODE_MASK;
 			spi->mode = (u16)tmp;
 			retval = spi_setup(spi);
diff --git a/include/linux/platform_data/spi-omap2-mcspi.h b/include/linux/platform_data/spi-omap2-mcspi.h
index 0bf9fddb830640dd282c38ee1a001e9c9d2b4f6e..3b400b1919a9bd8a9a446da90e37a3582af15de9 100644
--- a/include/linux/platform_data/spi-omap2-mcspi.h
+++ b/include/linux/platform_data/spi-omap2-mcspi.h
@@ -11,6 +11,7 @@ struct omap2_mcspi_platform_config {
 	unsigned short	num_cs;
 	unsigned int regs_offset;
 	unsigned int pin_dir:1;
+	size_t max_xfer_len;
 };
 
 struct omap2_mcspi_device_config {