Merge tag 'dmaengine-4.18-rc1' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine updates from Vinod Koul:

 - updates to sprd, bam_dma, stm drivers

 - remove VLAs in dmatest

 - move TI drivers to their own subdir

 - switch to SPDX tags for ima/mxs dma drivers

 - simplify getting .drvdata on bunch of drivers by Wolfram Sang

* tag 'dmaengine-4.18-rc1' of git://git.infradead.org/users/vkoul/slave-dma: (32 commits)
  dmaengine: sprd: Add Spreadtrum DMA configuration
  dmaengine: sprd: Optimize the sprd_dma_prep_dma_memcpy()
  dmaengine: imx-dma: Switch to SPDX identifier
  dmaengine: mxs-dma: Switch to SPDX identifier
  dmaengine: imx-sdma: Switch to SPDX identifier
  dmaengine: usb-dmac: Document R8A7799{0,5} bindings
  dmaengine: qcom: bam_dma: fix some doc warnings.
  dmaengine: qcom: bam_dma: fix invalid assignment warning
  dmaengine: sprd: fix an NULL vs IS_ERR() bug
  dmaengine: sprd: Use devm_ioremap_resource() to map memory
  dmaengine: sprd: Fix potential NULL dereference in sprd_dma_probe()
  dmaengine: pl330: flush before wait, and add dev burst support.
  dmaengine: axi-dmac: Request IRQ with IRQF_SHARED
  dmaengine: stm32-mdma: fix spelling mistake: "avalaible" -> "available"
  dmaengine: rcar-dmac: Document R-Car D3 bindings
  dmaengine: sprd: Move DMA request mode and interrupt type into head file
  dmaengine: sprd: Define the DMA data width type
  dmaengine: sprd: Define the DMA transfer step type
  dmaengine: ti: New directory for Texas Instruments DMA drivers
  dmaengine: shdmac: Change platform check to CONFIG_ARCH_RENESAS
  ...

31 files changed, 631 insertions, 397 deletions

diff --git a/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt b/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt
index 61315eaa7660..b1ba639554c0 100644
--- a/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt
+++ b/Documentation/devicetree/bindings/dma/renesas,rcar-dmac.txt
@@ -29,6 +29,7 @@ Required Properties:
 		- "renesas,dmac-r8a77965" (R-Car M3-N)
 		- "renesas,dmac-r8a77970" (R-Car V3M)
 		- "renesas,dmac-r8a77980" (R-Car V3H)
+		- "renesas,dmac-r8a77995" (R-Car D3)
 
 - reg: base address and length of the registers block for the DMAC
 
diff --git a/Documentation/devicetree/bindings/dma/renesas,usb-dmac.txt b/Documentation/devicetree/bindings/dma/renesas,usb-dmac.txt
index 9dc935e24e55..482e54362d3e 100644
--- a/Documentation/devicetree/bindings/dma/renesas,usb-dmac.txt
+++ b/Documentation/devicetree/bindings/dma/renesas,usb-dmac.txt
@@ -12,6 +12,8 @@ Required Properties:
 	  - "renesas,r8a7795-usb-dmac" (R-Car H3)
 	  - "renesas,r8a7796-usb-dmac" (R-Car M3-W)
 	  - "renesas,r8a77965-usb-dmac" (R-Car M3-N)
+	  - "renesas,r8a77990-usb-dmac" (R-Car E3)
+	  - "renesas,r8a77995-usb-dmac" (R-Car D3)
 - reg: base address and length of the registers block for the DMAC
 - interrupts: interrupt specifiers for the DMAC, one for each entry in
   interrupt-names.
diff --git a/MAINTAINERS b/MAINTAINERS
index 3838c94a0d47..abec3381cf45 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11790,6 +11790,14 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/rkuo/linux-hexagon-kernel.g
 S:	Supported
 F:	arch/hexagon/
 
+QUALCOMM HIDMA DRIVER
+M:	Sinan Kaya <okaya@codeaurora.org>
+L:	linux-arm-kernel@lists.infradead.org
+L:	linux-arm-msm@vger.kernel.org
+L:	dmaengine@vger.kernel.org
+S:	Supported
+F:	drivers/dma/qcom/hidma*
+
 QUALCOMM IOMMU
 M:	Rob Clark <robdclark@gmail.com>
 L:	iommu@lists.linux-foundation.org
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 6d61cd023633..ca1680afa20a 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -151,13 +151,6 @@ config DMA_JZ4780
 	  If you have a board based on such a SoC and wish to use DMA for
 	  devices which can use the DMA controller, say Y or M here.
 
-config DMA_OMAP
-	tristate "OMAP DMA support"
-	depends on ARCH_OMAP || COMPILE_TEST
-	select DMA_ENGINE
-	select DMA_VIRTUAL_CHANNELS
-	select TI_DMA_CROSSBAR if (SOC_DRA7XX || COMPILE_TEST)
-
 config DMA_SA11X0
 	tristate "SA-11x0 DMA support"
 	depends on ARCH_SA1100 || COMPILE_TEST
@@ -574,28 +567,6 @@ config TIMB_DMA
 	help
 	  Enable support for the Timberdale FPGA DMA engine.
 
-config TI_CPPI41
-	tristate "CPPI 4.1 DMA support"
-	depends on (ARCH_OMAP || ARCH_DAVINCI_DA8XX)
-	select DMA_ENGINE
-	help
-	  The Communications Port Programming Interface (CPPI) 4.1 DMA engine
-	  is currently used by the USB driver on AM335x and DA8xx platforms.
-
-config TI_DMA_CROSSBAR
-	bool
-
-config TI_EDMA
-	bool "TI EDMA support"
-	depends on ARCH_DAVINCI || ARCH_OMAP || ARCH_KEYSTONE || COMPILE_TEST
-	select DMA_ENGINE
-	select DMA_VIRTUAL_CHANNELS
-	select TI_DMA_CROSSBAR if (ARCH_OMAP || COMPILE_TEST)
-	default n
-	help
-	  Enable support for the TI EDMA controller. This DMA
-	  engine is found on TI DaVinci and AM33xx parts.
-
 config XGENE_DMA
 	tristate "APM X-Gene DMA support"
 	depends on ARCH_XGENE || COMPILE_TEST
@@ -653,6 +624,8 @@ source "drivers/dma/hsu/Kconfig"
 
 source "drivers/dma/sh/Kconfig"
 
+source "drivers/dma/ti/Kconfig"
+
 # clients
 comment "DMA Clients"
 	depends on DMA_ENGINE
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 0f62a4d49aab..203a99d68315 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -24,7 +24,6 @@ obj-$(CONFIG_COH901318) += coh901318.o coh901318_lli.o
 obj-$(CONFIG_DMA_BCM2835) += bcm2835-dma.o
 obj-$(CONFIG_DMA_JZ4740) += dma-jz4740.o
 obj-$(CONFIG_DMA_JZ4780) += dma-jz4780.o
-obj-$(CONFIG_DMA_OMAP) += omap-dma.o
 obj-$(CONFIG_DMA_SA11X0) += sa11x0-dma.o
 obj-$(CONFIG_DMA_SUN4I) += sun4i-dma.o
 obj-$(CONFIG_DMA_SUN6I) += sun6i-dma.o
@@ -69,13 +68,11 @@ obj-$(CONFIG_TXX9_DMAC) += txx9dmac.o
 obj-$(CONFIG_TEGRA20_APB_DMA) += tegra20-apb-dma.o
 obj-$(CONFIG_TEGRA210_ADMA) += tegra210-adma.o
 obj-$(CONFIG_TIMB_DMA) += timb_dma.o
-obj-$(CONFIG_TI_CPPI41) += cppi41.o
-obj-$(CONFIG_TI_DMA_CROSSBAR) += ti-dma-crossbar.o
-obj-$(CONFIG_TI_EDMA) += edma.o
 obj-$(CONFIG_XGENE_DMA) += xgene-dma.o
 obj-$(CONFIG_ZX_DMA) += zx_dma.o
 obj-$(CONFIG_ST_FDMA) += st_fdma.o
 
 obj-y += mediatek/
 obj-y += qcom/
+obj-y += ti/
 obj-y += xilinx/
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index a861b5b4d443..75f38d19fcbe 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -2041,8 +2041,7 @@ static void at_dma_shutdown(struct platform_device *pdev)
 
 static int at_dma_prepare(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct at_dma *atdma = dev_get_drvdata(dev);
 	struct dma_chan *chan, *_chan;
 
 	list_for_each_entry_safe(chan, _chan, &atdma->dma_common.channels,
@@ -2076,8 +2075,7 @@ static void atc_suspend_cyclic(struct at_dma_chan *atchan)
 
 static int at_dma_suspend_noirq(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct at_dma *atdma = dev_get_drvdata(dev);
 	struct dma_chan *chan, *_chan;
 
 	/* preserve data */
@@ -2118,8 +2116,7 @@ static void atc_resume_cyclic(struct at_dma_chan *atchan)
 
 static int at_dma_resume_noirq(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct at_dma *atdma = dev_get_drvdata(dev);
 	struct dma_chan *chan, *_chan;
 
 	/* bring back DMA controller */
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 94236ec9d410..4bf72561667c 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1833,8 +1833,7 @@ static void at_xdmac_free_chan_resources(struct dma_chan *chan)
 #ifdef CONFIG_PM
 static int atmel_xdmac_prepare(struct device *dev)
 {
-	struct platform_device	*pdev = to_platform_device(dev);
-	struct at_xdmac		*atxdmac = platform_get_drvdata(pdev);
+	struct at_xdmac		*atxdmac = dev_get_drvdata(dev);
 	struct dma_chan		*chan, *_chan;
 
 	list_for_each_entry_safe(chan, _chan, &atxdmac->dma.channels, device_node) {
@@ -1853,8 +1852,7 @@ static int atmel_xdmac_prepare(struct device *dev)
 #ifdef CONFIG_PM_SLEEP
 static int atmel_xdmac_suspend(struct device *dev)
 {
-	struct platform_device	*pdev = to_platform_device(dev);
-	struct at_xdmac		*atxdmac = platform_get_drvdata(pdev);
+	struct at_xdmac		*atxdmac = dev_get_drvdata(dev);
 	struct dma_chan		*chan, *_chan;
 
 	list_for_each_entry_safe(chan, _chan, &atxdmac->dma.channels, device_node) {
@@ -1878,8 +1876,7 @@ static int atmel_xdmac_suspend(struct device *dev)
 
 static int atmel_xdmac_resume(struct device *dev)
 {
-	struct platform_device	*pdev = to_platform_device(dev);
-	struct at_xdmac		*atxdmac = platform_get_drvdata(pdev);
+	struct at_xdmac		*atxdmac = dev_get_drvdata(dev);
 	struct at_xdmac_chan	*atchan;
 	struct dma_chan		*chan, *_chan;
 	int			i;
diff --git a/drivers/dma/dma-axi-dmac.c b/drivers/dma/dma-axi-dmac.c
index 2419fe524daa..15b2453d2647 100644
--- a/drivers/dma/dma-axi-dmac.c
+++ b/drivers/dma/dma-axi-dmac.c
@@ -687,7 +687,7 @@ static int axi_dmac_probe(struct platform_device *pdev)
 	if (ret)
 		goto err_unregister_device;
 
-	ret = request_irq(dmac->irq, axi_dmac_interrupt_handler, 0,
+	ret = request_irq(dmac->irq, axi_dmac_interrupt_handler, IRQF_SHARED,
 		dev_name(&pdev->dev), dmac);
 	if (ret)
 		goto err_unregister_of;
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index b9339524d5bd..aa1712beb0cc 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -468,6 +468,8 @@ static int dmatest_func(void *data)
 	unsigned long long	total_len = 0;
 	u8			align = 0;
 	bool			is_memset = false;
+	dma_addr_t		*srcs;
+	dma_addr_t		*dma_pq;
 
 	set_freezable();
 
@@ -551,6 +553,14 @@ static int dmatest_func(void *data)
 
 	set_user_nice(current, 10);
 
+	srcs = kcalloc(src_cnt, sizeof(dma_addr_t), GFP_KERNEL);
+	if (!srcs)
+		goto err_dstbuf;
+
+	dma_pq = kcalloc(dst_cnt, sizeof(dma_addr_t), GFP_KERNEL);
+	if (!dma_pq)
+		goto err_srcs_array;
+
 	/*
 	 * src and dst buffers are freed by ourselves below
 	 */
@@ -561,7 +571,6 @@ static int dmatest_func(void *data)
 	       && !(params->iterations && total_tests >= params->iterations)) {
 		struct dma_async_tx_descriptor *tx = NULL;
 		struct dmaengine_unmap_data *um;
-		dma_addr_t srcs[src_cnt];
 		dma_addr_t *dsts;
 		unsigned int src_off, dst_off, len;
 
@@ -676,8 +685,6 @@ static int dmatest_func(void *data)
 						      srcs, src_cnt,
 						      len, flags);
 		else if (thread->type == DMA_PQ) {
-			dma_addr_t dma_pq[dst_cnt];
-
 			for (i = 0; i < dst_cnt; i++)
 				dma_pq[i] = dsts[i] + dst_off;
 			tx = dev->device_prep_dma_pq(chan, dma_pq, srcs,
@@ -779,6 +786,9 @@ static int dmatest_func(void *data)
 	runtime = ktime_to_us(ktime);
 
 	ret = 0;
+	kfree(dma_pq);
+err_srcs_array:
+	kfree(srcs);
 err_dstbuf:
 	for (i = 0; thread->udsts[i]; i++)
 		kfree(thread->udsts[i]);
diff --git a/drivers/dma/dw/platform.c b/drivers/dma/dw/platform.c
index bc31fe802061..f62dd0944908 100644
--- a/drivers/dma/dw/platform.c
+++ b/drivers/dma/dw/platform.c
@@ -293,8 +293,7 @@ MODULE_DEVICE_TABLE(acpi, dw_dma_acpi_id_table);
 
 static int dw_suspend_late(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct dw_dma_chip *chip = platform_get_drvdata(pdev);
+	struct dw_dma_chip *chip = dev_get_drvdata(dev);
 
 	dw_dma_disable(chip);
 	clk_disable_unprepare(chip->clk);
@@ -304,8 +303,7 @@ static int dw_suspend_late(struct device *dev)
 
 static int dw_resume_early(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct dw_dma_chip *chip = platform_get_drvdata(pdev);
+	struct dw_dma_chip *chip = dev_get_drvdata(dev);
 	int ret;
 
 	ret = clk_prepare_enable(chip->clk);
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 3eaece888e75..1117b5123a6f 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -1328,8 +1328,7 @@ static int fsldma_of_remove(struct platform_device *op)
 #ifdef CONFIG_PM
 static int fsldma_suspend_late(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct fsldma_device *fdev = platform_get_drvdata(pdev);
+	struct fsldma_device *fdev = dev_get_drvdata(dev);
 	struct fsldma_chan *chan;
 	int i;
 
@@ -1360,8 +1359,7 @@ out:
 
 static int fsldma_resume_early(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct fsldma_device *fdev = platform_get_drvdata(pdev);
+	struct fsldma_device *fdev = dev_get_drvdata(dev);
 	struct fsldma_chan *chan;
 	u32 mode;
 	int i;
diff --git a/drivers/dma/idma64.c b/drivers/dma/idma64.c
index 1953e57505f4..e5c911200bdb 100644
--- a/drivers/dma/idma64.c
+++ b/drivers/dma/idma64.c
@@ -670,8 +670,7 @@ static int idma64_platform_remove(struct platform_device *pdev)
 
 static int idma64_pm_suspend(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct idma64_chip *chip = platform_get_drvdata(pdev);
+	struct idma64_chip *chip = dev_get_drvdata(dev);
 
 	idma64_off(chip->idma64);
 	return 0;
@@ -679,8 +678,7 @@ static int idma64_pm_suspend(struct device *dev)
 
 static int idma64_pm_resume(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct idma64_chip *chip = platform_get_drvdata(pdev);
+	struct idma64_chip *chip = dev_get_drvdata(dev);
 
 	idma64_on(chip->idma64);
 	return 0;
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index 715b39ae5a46..75b6ff0415ee 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -1,19 +1,13 @@
-/*
- * drivers/dma/imx-dma.c
- *
- * This file contains a driver for the Freescale i.MX DMA engine
- * found on i.MX1/21/27
- *
- * Copyright 2010 Sascha Hauer, Pengutronix <s.hauer@pengutronix.de>
- * Copyright 2012 Javier Martin, Vista Silicon <javier.martin@vista-silicon.com>
- *
- * The code contained herein is licensed under the GNU General Public
- * License. You may obtain a copy of the GNU General Public License
- * Version 2 or later at the following locations:
- *
- * http://www.opensource.org/licenses/gpl-license.html
- * http://www.gnu.org/copyleft/gpl.html
- */
+// SPDX-License-Identifier: GPL-2.0+
+//
+// drivers/dma/imx-dma.c
+//
+// This file contains a driver for the Freescale i.MX DMA engine
+// found on i.MX1/21/27
+//
+// Copyright 2010 Sascha Hauer, Pengutronix <s.hauer@pengutronix.de>
+// Copyright 2012 Javier Martin, Vista Silicon <javier.martin@vista-silicon.com>
+
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/types.h>
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index ccd03c3cedfe..f077992635c2 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -1,21 +1,14 @@
-/*
- * drivers/dma/imx-sdma.c
- *
- * This file contains a driver for the Freescale Smart DMA engine
- *
- * Copyright 2010 Sascha Hauer, Pengutronix <s.hauer@pengutronix.de>
- *
- * Based on code from Freescale:
- *
- * Copyright 2004-2009 Freescale Semiconductor, Inc. All Rights Reserved.
- *
- * The code contained herein is licensed under the GNU General Public
- * License. You may obtain a copy of the GNU General Public License
- * Version 2 or later at the following locations:
- *
- * http://www.opensource.org/licenses/gpl-license.html
- * http://www.gnu.org/copyleft/gpl.html
- */
+// SPDX-License-Identifier: GPL-2.0+
+//
+// drivers/dma/imx-sdma.c
+//
+// This file contains a driver for the Freescale Smart DMA engine
+//
+// Copyright 2010 Sascha Hauer, Pengutronix <s.hauer@pengutronix.de>
+//
+// Based on code from Freescale:
+//
+// Copyright 2004-2009 Freescale Semiconductor, Inc. All Rights Reserved.
 
 #include <linux/init.h>
 #include <linux/iopoll.h>
diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
index 41d167921fab..ae5182ff0128 100644
--- a/drivers/dma/mxs-dma.c
+++ b/drivers/dma/mxs-dma.c
@@ -1,12 +1,8 @@
-/*
- * Copyright 2011 Freescale Semiconductor, Inc. All Rights Reserved.
- *
- * Refer to drivers/dma/imx-sdma.c
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
+// SPDX-License-Identifier: GPL-2.0
+//
+// Copyright 2011 Freescale Semiconductor, Inc. All Rights Reserved.
+//
+// Refer to drivers/dma/imx-sdma.c
 
 #include <linux/init.h>
 #include <linux/types.h>
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index de1fd59fe136..6237069001c4 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -27,6 +27,7 @@
 #include <linux/of_dma.h>
 #include <linux/err.h>
 #include <linux/pm_runtime.h>
+#include <linux/bug.h>
 
 #include "dmaengine.h"
 #define PL330_MAX_CHAN		8
@@ -1094,51 +1095,96 @@ static inline int _ldst_memtomem(unsigned dry_run, u8 buf[],
 	return off;
 }
 
-static inline int _ldst_devtomem(struct pl330_dmac *pl330, unsigned dry_run,
-				 u8 buf[], const struct _xfer_spec *pxs,
-				 int cyc)
+static u32 _emit_load(unsigned int dry_run, u8 buf[],
+	enum pl330_cond cond, enum dma_transfer_direction direction,
+	u8 peri)
 {
 	int off = 0;
-	enum pl330_cond cond;
 
-	if (pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP)
-		cond = BURST;
-	else
-		cond = SINGLE;
+	switch (direction) {
+	case DMA_MEM_TO_MEM:
+		/* fall through */
+	case DMA_MEM_TO_DEV:
+		off += _emit_LD(dry_run, &buf[off], cond);
+		break;
 
-	while (cyc--) {
-		off += _emit_WFP(dry_run, &buf[off], cond, pxs->desc->peri);
-		off += _emit_LDP(dry_run, &buf[off], cond, pxs->desc->peri);
-		off += _emit_ST(dry_run, &buf[off], ALWAYS);
+	case DMA_DEV_TO_MEM:
+		if (cond == ALWAYS) {
+			off += _emit_LDP(dry_run, &buf[off], SINGLE,
+				peri);
+			off += _emit_LDP(dry_run, &buf[off], BURST,
+				peri);
+		} else {
+			off += _emit_LDP(dry_run, &buf[off], cond,
+				peri);
+		}
+		break;
 
-		if (!(pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP))
-			off += _emit_FLUSHP(dry_run, &buf[off],
-					    pxs->desc->peri);
+	default:
+		/* this code should be unreachable */
+		WARN_ON(1);
+		break;
 	}
 
 	return off;
 }
 
-static inline int _ldst_memtodev(struct pl330_dmac *pl330,
+static inline u32 _emit_store(unsigned int dry_run, u8 buf[],
+	enum pl330_cond cond, enum dma_transfer_direction direction,
+	u8 peri)
+{
+	int off = 0;
+
+	switch (direction) {
+	case DMA_MEM_TO_MEM:
+		/* fall through */
+	case DMA_DEV_TO_MEM:
+		off += _emit_ST(dry_run, &buf[off], cond);
+		break;
+
+	case DMA_MEM_TO_DEV:
+		if (cond == ALWAYS) {
+			off += _emit_STP(dry_run, &buf[off], SINGLE,
+				peri);
+			off += _emit_STP(dry_run, &buf[off], BURST,
+				peri);
+		} else {
+			off += _emit_STP(dry_run, &buf[off], cond,
+				peri);
+		}
+		break;
+
+	default:
+		/* this code should be unreachable */
+		WARN_ON(1);
+		break;
+	}
+
+	return off;
+}
+
+static inline int _ldst_peripheral(struct pl330_dmac *pl330,
 				 unsigned dry_run, u8 buf[],
-				 const struct _xfer_spec *pxs, int cyc)
+				 const struct _xfer_spec *pxs, int cyc,
+				 enum pl330_cond cond)
 {
 	int off = 0;
-	enum pl330_cond cond;
 
 	if (pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP)
 		cond = BURST;
-	else
-		cond = SINGLE;
 
+	/*
+	 * do FLUSHP at beginning to clear any stale dma requests before the
+	 * first WFP.
+	 */
+	if (!(pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP))
+		off += _emit_FLUSHP(dry_run, &buf[off], pxs->desc->peri);
 	while (cyc--) {
 		off += _emit_WFP(dry_run, &buf[off], cond, pxs->desc->peri);
-		off += _emit_LD(dry_run, &buf[off], ALWAYS);
-		off += _emit_STP(dry_run, &buf[off], cond, pxs->desc->peri);
-
-		if (!(pl330->quirks & PL330_QUIRK_BROKEN_NO_FLUSHP))
-			off += _emit_FLUSHP(dry_run, &buf[off],
-					    pxs->desc->peri);
+		off += _emit_load(dry_run, &buf[off], cond, pxs->desc->rqtype,
+			pxs->desc->peri);
+		off += _emit_store(dry_run, &buf[off], cond, pxs->desc->rqtype,
+			pxs->desc->peri);
 	}
 
 	return off;
@@ -1148,19 +1194,65 @@ static int _bursts(struct pl330_dmac *pl330, unsigned dry_run, u8 buf[],
 		const struct _xfer_spec *pxs, int cyc)
 {
 	int off = 0;
+	enum pl330_cond cond = BRST_LEN(pxs->ccr) > 1 ? BURST : SINGLE;
 
 	switch (pxs->desc->rqtype) {
 	case DMA_MEM_TO_DEV:
-		off += _ldst_memtodev(pl330, dry_run, &buf[off], pxs, cyc);
-		break;
+		/* fall through */
 	case DMA_DEV_TO_MEM:
-		off += _ldst_devtomem(pl330, dry_run, &buf[off], pxs, cyc);
+		off += _ldst_peripheral(pl330, dry_run, &buf[off], pxs, cyc,
+			cond);
 		break;
+
 	case DMA_MEM_TO_MEM:
 		off += _ldst_memtomem(dry_run, &buf[off], pxs, cyc);
 		break;
+
+	default:
+		/* this code should be unreachable */
+		WARN_ON(1);
+		break;
+	}
+
+	return off;
+}
+
+/*
+ * transfer dregs with single transfers to peripheral, or a reduced size burst
+ * for mem-to-mem.
+ */
+static int _dregs(struct pl330_dmac *pl330, unsigned int dry_run, u8 buf[],
+		const struct _xfer_spec *pxs, int transfer_length)
+{
+	int off = 0;
+	int dregs_ccr;
+
+	if (transfer_length == 0)
+		return off;
+
+	switch (pxs->desc->rqtype) {
+	case DMA_MEM_TO_DEV:
+		/* fall through */
+	case DMA_DEV_TO_MEM:
+		off += _ldst_peripheral(pl330, dry_run, &buf[off], pxs,
+			transfer_length, SINGLE);
+		break;
+
+	case DMA_MEM_TO_MEM:
+		dregs_ccr = pxs->ccr;
+		dregs_ccr &= ~((0xf << CC_SRCBRSTLEN_SHFT) |
+			(0xf << CC_DSTBRSTLEN_SHFT));
+		dregs_ccr |= (((transfer_length - 1) & 0xf) <<
+			CC_SRCBRSTLEN_SHFT);
+		dregs_ccr |= (((transfer_length - 1) & 0xf) <<
+			CC_DSTBRSTLEN_SHFT);
+		off += _emit_MOV(dry_run, &buf[off], CCR, dregs_ccr);
+		off += _ldst_memtomem(dry_run, &buf[off], pxs, 1);
+		break;
+
 	default:
-		off += 0x40000000; /* Scare off the Client */
+		/* this code should be unreachable */
+		WARN_ON(1);
 		break;
 	}
 
@@ -1256,6 +1348,8 @@ static inline int _setup_loops(struct pl330_dmac *pl330,
 	struct pl330_xfer *x = &pxs->desc->px;
 	u32 ccr = pxs->ccr;
 	unsigned long c, bursts = BYTE_TO_BURST(x->bytes, ccr);
+	int num_dregs = (x->bytes - BURST_TO_BYTE(bursts, ccr)) /
+		BRST_SIZE(ccr);
 	int off = 0;
 
 	while (bursts) {
@@ -1263,6 +1357,7 @@ static inline int _setup_loops(struct pl330_dmac *pl330,
 		off += _loop(pl330, dry_run, &buf[off], &c, pxs);
 		bursts -= c;
 	}
+	off += _dregs(pl330, dry_run, &buf[off], pxs, num_dregs);
 
 	return off;
 }
@@ -1294,7 +1389,6 @@ static int _setup_req(struct pl330_dmac *pl330, unsigned dry_run,
 		      struct _xfer_spec *pxs)
 {
 	struct _pl330_req *req = &thrd->req[index];
-	struct pl330_xfer *x;
 	u8 *buf = req->mc_cpu;
 	int off = 0;
 
@@ -1303,11 +1397,6 @@ static int _setup_req(struct pl330_dmac *pl330, unsigned dry_run,
 	/* DMAMOV CCR, ccr */
 	off += _emit_MOV(dry_run, &buf[off], CCR, pxs->ccr);
 
-	x = &pxs->desc->px;
-	/* Error if xfer length is not aligned at burst size */
-	if (x->bytes % (BRST_SIZE(pxs->ccr) * BRST_LEN(pxs->ccr)))
-		return -EINVAL;
-
 	off += _setup_xfer(pl330, dry_run, &buf[off], pxs);
 
 	/* DMASEV peripheral/event */
@@ -1365,6 +1454,20 @@ static int pl330_submit_req(struct pl330_thread *thrd,
 	u32 ccr;
 	int ret = 0;
 
+	switch (desc->rqtype) {
+	case DMA_MEM_TO_DEV:
+		break;
+
+	case DMA_DEV_TO_MEM:
+		break;
+
+	case DMA_MEM_TO_MEM:
+		break;
+
+	default:
+		return -ENOTSUPP;
+	}
+
 	if (pl330->state == DYING
 		|| pl330->dmac_tbd.reset_chan & (1 << thrd->id)) {
 		dev_info(thrd->dmac->ddma.dev, "%s:%d\n",
@@ -2106,6 +2209,18 @@ static bool pl330_prep_slave_fifo(struct dma_pl330_chan *pch,
 	return true;
 }
 
+static int fixup_burst_len(int max_burst_len, int quirks)
+{
+	if (quirks & PL330_QUIRK_BROKEN_NO_FLUSHP)
+		return 1;
+	else if (max_burst_len > PL330_MAX_BURST)
+		return PL330_MAX_BURST;
+	else if (max_burst_len < 1)
+		return 1;
+	else
+		return max_burst_len;
+}
+
 static int pl330_config(struct dma_chan *chan,
 			struct dma_slave_config *slave_config)
 {
@@ -2117,15 +2232,15 @@ static int pl330_config(struct dma_chan *chan,
 			pch->fifo_addr = slave_config->dst_addr;
 		if (slave_config->dst_addr_width)
 			pch->burst_sz = __ffs(slave_config->dst_addr_width);
-		if (slave_config->dst_maxburst)
-			pch->burst_len = slave_config->dst_maxburst;
+		pch->burst_len = fixup_burst_len(slave_config->dst_maxburst,
+			pch->dmac->quirks);
 	} else if (slave_config->direction == DMA_DEV_TO_MEM) {
 		if (slave_config->src_addr)
 			pch->fifo_addr = slave_config->src_addr;
 		if (slave_config->src_addr_width)
 			pch->burst_sz = __ffs(slave_config->src_addr_width);
-		if (slave_config->src_maxburst)
-			pch->burst_len = slave_config->src_maxburst;
+		pch->burst_len = fixup_burst_len(slave_config->src_maxburst,
+			pch->dmac->quirks);
 	}
 
 	return 0;
@@ -2519,14 +2634,8 @@ static inline int get_burst_len(struct dma_pl330_desc *desc, size_t len)
 	burst_len >>= desc->rqcfg.brst_size;
 
 	/* src/dst_burst_len can't be more than 16 */
-	if (burst_len > 16)
-		burst_len = 16;
-
-	while (burst_len > 1) {
-		if (!(len % (burst_len << desc->rqcfg.brst_size)))
-			break;
-		burst_len--;
-	}
+	if (burst_len > PL330_MAX_BURST)
+		burst_len = PL330_MAX_BURST;
 
 	return burst_len;
 }
@@ -2598,7 +2707,7 @@ static struct dma_async_tx_descriptor *pl330_prep_dma_cyclic(
 
 		desc->rqtype = direction;
 		desc->rqcfg.brst_size = pch->burst_sz;
-		desc->rqcfg.brst_len = 1;
+		desc->rqcfg.brst_len = pch->burst_len;
 		desc->bytes_requested = period_len;
 		fill_px(&desc->px, dst, src, period_len);
 
@@ -2743,7 +2852,7 @@ pl330_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		}
 
 		desc->rqcfg.brst_size = pch->burst_sz;
-		desc->rqcfg.brst_len = 1;
+		desc->rqcfg.brst_len = pch->burst_len;
 		desc->rqtype = direction;
 		desc->bytes_requested = sg_dma_len(sg);
 	}
diff --git a/drivers/dma/qcom/bam_dma.c b/drivers/dma/qcom/bam_dma.c
index 4a828c18099a..1617715aa6e0 100644
--- a/drivers/dma/qcom/bam_dma.c
+++ b/drivers/dma/qcom/bam_dma.c
@@ -451,6 +451,7 @@ static void bam_reset_channel(struct bam_chan *bchan)
 /**
  * bam_chan_init_hw - Initialize channel hardware
  * @bchan: bam channel
+ * @dir: DMA transfer direction
  *
  * This function resets and initializes the BAM channel
  */
@@ -673,7 +674,7 @@ static struct dma_async_tx_descriptor *bam_prep_slave_sg(struct dma_chan *chan,
 				remainder = 0;
 			}
 
-			async_desc->length += desc->size;
+			async_desc->length += le16_to_cpu(desc->size);
 			desc++;
 		} while (remainder > 0);
 	}
@@ -687,7 +688,7 @@ err_out:
 
 /**
  * bam_dma_terminate_all - terminate all transactions on a channel
- * @bchan: bam dma channel
+ * @chan: bam dma channel
  *
  * Dequeues and frees all transactions
  * No callbacks are done
@@ -918,7 +919,8 @@ static enum dma_status bam_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 				continue;
 
 			for (i = 0; i < async_desc->num_desc; i++)
-				residue += async_desc->curr_desc[i].size;
+				residue += le16_to_cpu(
+						async_desc->curr_desc[i].size);
 		}
 	}
 
@@ -958,7 +960,7 @@ static void bam_apply_new_config(struct bam_chan *bchan,
 
 /**
  * bam_start_dma - start next transaction
- * @bchan - bam dma channel
+ * @bchan: bam dma channel
  */
 static void bam_start_dma(struct bam_chan *bchan)
 {
diff --git a/drivers/dma/qcom/hidma.c b/drivers/dma/qcom/hidma.c
index 963cc5228d05..43d4b00b8138 100644
--- a/drivers/dma/qcom/hidma.c
+++ b/drivers/dma/qcom/hidma.c
@@ -616,8 +616,7 @@ static irqreturn_t hidma_chirq_handler_msi(int chirq, void *arg)
 static ssize_t hidma_show_values(struct device *dev,
 				 struct device_attribute *attr, char *buf)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct hidma_dev *mdev = platform_get_drvdata(pdev);
+	struct hidma_dev *mdev = dev_get_drvdata(dev);
 
 	buf[0] = 0;
 
diff --git a/drivers/dma/qcom/hidma_mgmt_sys.c b/drivers/dma/qcom/hidma_mgmt_sys.c
index d61f1068a34b..cbb89eafd844 100644
--- a/drivers/dma/qcom/hidma_mgmt_sys.c
+++ b/drivers/dma/qcom/hidma_mgmt_sys.c
@@ -107,8 +107,7 @@ static struct hidma_mgmt_fileinfo hidma_mgmt_files[] = {
 static ssize_t show_values(struct device *dev, struct device_attribute *attr,
 			   char *buf)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct hidma_mgmt_dev *mdev = platform_get_drvdata(pdev);
+	struct hidma_mgmt_dev *mdev = dev_get_drvdata(dev);
 	unsigned int i;
 
 	buf[0] = 0;
@@ -125,8 +124,7 @@ static ssize_t show_values(struct device *dev, struct device_attribute *attr,
 static ssize_t set_values(struct device *dev, struct device_attribute *attr,
 			  const char *buf, size_t count)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct hidma_mgmt_dev *mdev = platform_get_drvdata(pdev);
+	struct hidma_mgmt_dev *mdev = dev_get_drvdata(dev);
 	unsigned long tmp;
 	unsigned int i;
 	int rc;
diff --git a/drivers/dma/sh/shdmac.c b/drivers/dma/sh/shdmac.c
index c94ffab0d25c..04a74e0a95b7 100644
--- a/drivers/dma/sh/shdmac.c
+++ b/drivers/dma/sh/shdmac.c
@@ -443,7 +443,6 @@ static bool sh_dmae_reset(struct sh_dmae_device *shdev)
 	return ret;
 }
 
-#if defined(CONFIG_CPU_SH4) || defined(CONFIG_ARCH_SHMOBILE)
 static irqreturn_t sh_dmae_err(int irq, void *data)
 {
 	struct sh_dmae_device *shdev = data;
@@ -454,7 +453,6 @@ static irqreturn_t sh_dmae_err(int irq, void *data)
 	sh_dmae_reset(shdev);
 	return IRQ_HANDLED;
 }
-#endif
 
 static bool sh_dmae_desc_completed(struct shdma_chan *schan,
 				   struct shdma_desc *sdesc)
@@ -686,11 +684,8 @@ static int sh_dmae_probe(struct platform_device *pdev)
 	const struct sh_dmae_pdata *pdata;
 	unsigned long chan_flag[SH_DMAE_MAX_CHANNELS] = {};
 	int chan_irq[SH_DMAE_MAX_CHANNELS];
-#if defined(CONFIG_CPU_SH4) || defined(CONFIG_ARCH_SHMOBILE)
 	unsigned long irqflags = 0;
-	int errirq;
-#endif
-	int err, i, irq_cnt = 0, irqres = 0, irq_cap = 0;
+	int err, errirq, i, irq_cnt = 0, irqres = 0, irq_cap = 0;
 	struct sh_dmae_device *shdev;
 	struct dma_device *dma_dev;
 	struct resource *chan, *dmars, *errirq_res, *chanirq_res;
@@ -792,33 +787,32 @@ static int sh_dmae_probe(struct platform_device *pdev)
 	if (err)
 		goto rst_err;
 
-#if defined(CONFIG_CPU_SH4) || defined(CONFIG_ARCH_SHMOBILE)
-	chanirq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 1);
+	if (IS_ENABLED(CONFIG_CPU_SH4) || IS_ENABLED(CONFIG_ARCH_RENESAS)) {
+		chanirq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 1);
 
-	if (!chanirq_res)
-		chanirq_res = errirq_res;
-	else
-		irqres++;
+		if (!chanirq_res)
+			chanirq_res = errirq_res;
+		else
+			irqres++;
 
-	if (chanirq_res == errirq_res ||
-	    (errirq_res->flags & IORESOURCE_BITS) == IORESOURCE_IRQ_SHAREABLE)
-		irqflags = IRQF_SHARED;
+		if (chanirq_res == errirq_res ||
+		    (errirq_res->flags & IORESOURCE_BITS) == IORESOURCE_IRQ_SHAREABLE)
+			irqflags = IRQF_SHARED;
 
-	errirq = errirq_res->start;
+		errirq = errirq_res->start;
 
-	err = devm_request_irq(&pdev->dev, errirq, sh_dmae_err, irqflags,
-			       "DMAC Address Error", shdev);
-	if (err) {
-		dev_err(&pdev->dev,
-			"DMA failed requesting irq #%d, error %d\n",
-			errirq, err);
-		goto eirq_err;
+		err = devm_request_irq(&pdev->dev, errirq, sh_dmae_err,
+				       irqflags, "DMAC Address Error", shdev);
+		if (err) {
+			dev_err(&pdev->dev,
+				"DMA failed requesting irq #%d, error %d\n",
+				errirq, err);
+			goto eirq_err;
+		}
+	} else {
+		chanirq_res = errirq_res;
 	}
 
-#else
-	chanirq_res = errirq_res;
-#endif /* CONFIG_CPU_SH4 || CONFIG_ARCH_SHMOBILE */
-
 	if (chanirq_res->start == chanirq_res->end &&
 	    !platform_get_resource(pdev, IORESOURCE_IRQ, 1)) {
 		/* Special case - all multiplexed */
@@ -884,9 +878,7 @@ edmadevreg:
 chan_probe_err:
 	sh_dmae_chan_remove(shdev);
 
-#if defined(CONFIG_CPU_SH4) || defined(CONFIG_ARCH_SHMOBILE)
 eirq_err:
-#endif
 rst_err:
 	spin_lock_irq(&sh_dmae_lock);
 	list_del_rcu(&shdev->node);
diff --git a/drivers/dma/sprd-dma.c b/drivers/dma/sprd-dma.c
index 52ebccb483be..55df0d41355b 100644
--- a/drivers/dma/sprd-dma.c
+++ b/drivers/dma/sprd-dma.c
@@ -6,6 +6,7 @@
 
 #include <linux/clk.h>
 #include <linux/dma-mapping.h>
+#include <linux/dma/sprd-dma.h>
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
@@ -116,57 +117,21 @@
 #define SPRD_DMA_SRC_TRSF_STEP_OFFSET	0
 #define SPRD_DMA_TRSF_STEP_MASK		GENMASK(15, 0)
 
+/* define the DMA transfer step type */
+#define SPRD_DMA_NONE_STEP		0
+#define SPRD_DMA_BYTE_STEP		1
+#define SPRD_DMA_SHORT_STEP		2
+#define SPRD_DMA_WORD_STEP		4
+#define SPRD_DMA_DWORD_STEP		8
+
 #define SPRD_DMA_SOFTWARE_UID		0
 
-/*
- * enum sprd_dma_req_mode: define the DMA request mode
- * @SPRD_DMA_FRAG_REQ: fragment request mode
- * @SPRD_DMA_BLK_REQ: block request mode
- * @SPRD_DMA_TRANS_REQ: transaction request mode
- * @SPRD_DMA_LIST_REQ: link-list request mode
- *
- * We have 4 types request mode: fragment mode, block mode, transaction mode
- * and linklist mode. One transaction can contain several blocks, one block can
- * contain several fragments. Link-list mode means we can save several DMA
- * configuration into one reserved memory, then DMA can fetch each DMA
- * configuration automatically to start transfer.
- */
-enum sprd_dma_req_mode {
-	SPRD_DMA_FRAG_REQ,
-	SPRD_DMA_BLK_REQ,
-	SPRD_DMA_TRANS_REQ,
-	SPRD_DMA_LIST_REQ,
-};
-
-/*
- * enum sprd_dma_int_type: define the DMA interrupt type
- * @SPRD_DMA_NO_INT: do not need generate DMA interrupts.
- * @SPRD_DMA_FRAG_INT: fragment done interrupt when one fragment request
- * is done.
- * @SPRD_DMA_BLK_INT: block done interrupt when one block request is done.
- * @SPRD_DMA_BLK_FRAG_INT: block and fragment interrupt when one fragment
- * or one block request is done.
- * @SPRD_DMA_TRANS_INT: tansaction done interrupt when one transaction
- * request is done.
- * @SPRD_DMA_TRANS_FRAG_INT: transaction and fragment interrupt when one
- * transaction request or fragment request is done.
- * @SPRD_DMA_TRANS_BLK_INT: transaction and block interrupt when one
- * transaction request or block request is done.
- * @SPRD_DMA_LIST_INT: link-list done interrupt when one link-list request
- * is done.
- * @SPRD_DMA_CFGERR_INT: configure error interrupt when configuration is
- * incorrect.
- */
-enum sprd_dma_int_type {
-	SPRD_DMA_NO_INT,
-	SPRD_DMA_FRAG_INT,
-	SPRD_DMA_BLK_INT,
-	SPRD_DMA_BLK_FRAG_INT,
-	SPRD_DMA_TRANS_INT,
-	SPRD_DMA_TRANS_FRAG_INT,
-	SPRD_DMA_TRANS_BLK_INT,
-	SPRD_DMA_LIST_INT,
-	SPRD_DMA_CFGERR_INT,
+/* dma data width values */
+enum sprd_dma_datawidth {
+	SPRD_DMA_DATAWIDTH_1_BYTE,
+	SPRD_DMA_DATAWIDTH_2_BYTES,
+	SPRD_DMA_DATAWIDTH_4_BYTES,
+	SPRD_DMA_DATAWIDTH_8_BYTES,
 };
 
 /* dma channel hardware configuration */
@@ -199,6 +164,7 @@ struct sprd_dma_desc {
 struct sprd_dma_chn {
 	struct virt_dma_chan	vc;
 	void __iomem		*chn_base;
+	struct dma_slave_config	slave_cfg;
 	u32			chn_num;
 	u32			dev_id;
 	struct sprd_dma_desc	*cur_desc;
@@ -587,52 +553,97 @@ static void sprd_dma_issue_pending(struct dma_chan *chan)
 	spin_unlock_irqrestore(&schan->vc.lock, flags);
 }
 
-static int sprd_dma_config(struct dma_chan *chan, struct sprd_dma_desc *sdesc,
-			   dma_addr_t dest, dma_addr_t src, size_t len)
+static int sprd_dma_get_datawidth(enum dma_slave_buswidth buswidth)
+{
+	switch (buswidth) {
+	case DMA_SLAVE_BUSWIDTH_1_BYTE:
+	case DMA_SLAVE_BUSWIDTH_2_BYTES:
+	case DMA_SLAVE_BUSWIDTH_4_BYTES:
+	case DMA_SLAVE_BUSWIDTH_8_BYTES:
+		return ffs(buswidth) - 1;
+
+	default:
+		return -EINVAL;
+	}
+}
+
+static int sprd_dma_get_step(enum dma_slave_buswidth buswidth)
+{
+	switch (buswidth) {
+	case DMA_SLAVE_BUSWIDTH_1_BYTE:
+	case DMA_SLAVE_BUSWIDTH_2_BYTES:
+	case DMA_SLAVE_BUSWIDTH_4_BYTES:
+	case DMA_SLAVE_BUSWIDTH_8_BYTES:
+		return buswidth;
+
+	default:
+		return -EINVAL;
+	}
+}
+
+static int sprd_dma_fill_desc(struct dma_chan *chan,
+			      struct sprd_dma_desc *sdesc,
+			      dma_addr_t src, dma_addr_t dst, u32 len,
+			      enum dma_transfer_direction dir,
+			      unsigned long flags,
+			      struct dma_slave_config *slave_cfg)
 {
 	struct sprd_dma_dev *sdev = to_sprd_dma_dev(chan);
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
 	struct sprd_dma_chn_hw *hw = &sdesc->chn_hw;
-	u32 datawidth, src_step, des_step, fragment_len;
-	u32 block_len, req_mode, irq_mode, transcation_len;
-	u32 fix_mode = 0, fix_en = 0;
-
-	if (IS_ALIGNED(len, 4)) {
-		datawidth = 2;
-		src_step = 4;
-		des_step = 4;
-	} else if (IS_ALIGNED(len, 2)) {
-		datawidth = 1;
-		src_step = 2;
-		des_step = 2;
+	u32 req_mode = (flags >> SPRD_DMA_REQ_SHIFT) & SPRD_DMA_REQ_MODE_MASK;
+	u32 int_mode = flags & SPRD_DMA_INT_MASK;
+	int src_datawidth, dst_datawidth, src_step, dst_step;
+	u32 temp, fix_mode = 0, fix_en = 0;
+
+	if (dir == DMA_MEM_TO_DEV) {
+		src_step = sprd_dma_get_step(slave_cfg->src_addr_width);
+		if (src_step < 0) {
+			dev_err(sdev->dma_dev.dev, "invalid source step\n");
+			return src_step;
+		}
+		dst_step = SPRD_DMA_NONE_STEP;
 	} else {
-		datawidth = 0;
-		src_step = 1;
-		des_step = 1;
+		dst_step = sprd_dma_get_step(slave_cfg->dst_addr_width);
+		if (dst_step < 0) {
+			dev_err(sdev->dma_dev.dev, "invalid destination step\n");
+			return dst_step;
+		}
+		src_step = SPRD_DMA_NONE_STEP;
 	}
 
-	fragment_len = SPRD_DMA_MEMCPY_MIN_SIZE;
-	if (len <= SPRD_DMA_BLK_LEN_MASK) {
-		block_len = len;
-		transcation_len = 0;
-		req_mode = SPRD_DMA_BLK_REQ;
-		irq_mode = SPRD_DMA_BLK_INT;
-	} else {
-		block_len = SPRD_DMA_MEMCPY_MIN_SIZE;
-		transcation_len = len;
-		req_mode = SPRD_DMA_TRANS_REQ;
-		irq_mode = SPRD_DMA_TRANS_INT;
+	src_datawidth = sprd_dma_get_datawidth(slave_cfg->src_addr_width);
+	if (src_datawidth < 0) {
+		dev_err(sdev->dma_dev.dev, "invalid source datawidth\n");
+		return src_datawidth;
 	}
 
+	dst_datawidth = sprd_dma_get_datawidth(slave_cfg->dst_addr_width);
+	if (dst_datawidth < 0) {
+		dev_err(sdev->dma_dev.dev, "invalid destination datawidth\n");
+		return dst_datawidth;
+	}
+
+	if (slave_cfg->slave_id)
+		schan->dev_id = slave_cfg->slave_id;
+
 	hw->cfg = SPRD_DMA_DONOT_WAIT_BDONE << SPRD_DMA_WAIT_BDONE_OFFSET;
-	hw->wrap_ptr = (u32)((src >> SPRD_DMA_HIGH_ADDR_OFFSET) &
-			     SPRD_DMA_HIGH_ADDR_MASK);
-	hw->wrap_to = (u32)((dest >> SPRD_DMA_HIGH_ADDR_OFFSET) &
-			    SPRD_DMA_HIGH_ADDR_MASK);
 
-	hw->src_addr = (u32)(src & SPRD_DMA_LOW_ADDR_MASK);
-	hw->des_addr = (u32)(dest & SPRD_DMA_LOW_ADDR_MASK);
+	/*
+	 * wrap_ptr and wrap_to will save the high 4 bits source address and
+	 * destination address.
+	 */
+	hw->wrap_ptr = (src >> SPRD_DMA_HIGH_ADDR_OFFSET) & SPRD_DMA_HIGH_ADDR_MASK;
+	hw->wrap_to = (dst >> SPRD_DMA_HIGH_ADDR_OFFSET) & SPRD_DMA_HIGH_ADDR_MASK;
+	hw->src_addr = src & SPRD_DMA_LOW_ADDR_MASK;
+	hw->des_addr = dst & SPRD_DMA_LOW_ADDR_MASK;
 
-	if ((src_step != 0 && des_step != 0) || (src_step | des_step) == 0) {
+	/*
+	 * If the src step and dst step both are 0 or both are not 0, that means
+	 * we can not enable the fix mode. If one is 0 and another one is not,
+	 * we can enable the fix mode.
+	 */
+	if ((src_step != 0 && dst_step != 0) || (src_step | dst_step) == 0) {
 		fix_en = 0;
 	} else {
 		fix_en = 1;
@@ -642,87 +653,119 @@ static int sprd_dma_config(struct dma_chan *chan, struct sprd_dma_desc *sdesc,
 			fix_mode = 0;
 	}
 
-	hw->frg_len = datawidth << SPRD_DMA_SRC_DATAWIDTH_OFFSET |
-		datawidth << SPRD_DMA_DES_DATAWIDTH_OFFSET |
-		req_mode << SPRD_DMA_REQ_MODE_OFFSET |
-		fix_mode << SPRD_DMA_FIX_SEL_OFFSET |
-		fix_en << SPRD_DMA_FIX_EN_OFFSET |
-		(fragment_len & SPRD_DMA_FRG_LEN_MASK);
-	hw->blk_len = block_len & SPRD_DMA_BLK_LEN_MASK;
-
-	hw->intc = SPRD_DMA_CFG_ERR_INT_EN;
-
-	switch (irq_mode) {
-	case SPRD_DMA_NO_INT:
-		break;
-
-	case SPRD_DMA_FRAG_INT:
-		hw->intc |= SPRD_DMA_FRAG_INT_EN;
-		break;
+	hw->intc = int_mode | SPRD_DMA_CFG_ERR_INT_EN;
 
-	case SPRD_DMA_BLK_INT:
-		hw->intc |= SPRD_DMA_BLK_INT_EN;
-		break;
+	temp = src_datawidth << SPRD_DMA_SRC_DATAWIDTH_OFFSET;
+	temp |= dst_datawidth << SPRD_DMA_DES_DATAWIDTH_OFFSET;
+	temp |= req_mode << SPRD_DMA_REQ_MODE_OFFSET;
+	temp |= fix_mode << SPRD_DMA_FIX_SEL_OFFSET;
+	temp |= fix_en << SPRD_DMA_FIX_EN_OFFSET;
+	temp |= slave_cfg->src_maxburst & SPRD_DMA_FRG_LEN_MASK;
+	hw->frg_len = temp;
 
-	case SPRD_DMA_BLK_FRAG_INT:
-		hw->intc |= SPRD_DMA_BLK_INT_EN | SPRD_DMA_FRAG_INT_EN;
-		break;
+	hw->blk_len = len & SPRD_DMA_BLK_LEN_MASK;
+	hw->trsc_len = len & SPRD_DMA_TRSC_LEN_MASK;
 
-	case SPRD_DMA_TRANS_INT:
-		hw->intc |= SPRD_DMA_TRANS_INT_EN;
-		break;
+	temp = (dst_step & SPRD_DMA_TRSF_STEP_MASK) << SPRD_DMA_DEST_TRSF_STEP_OFFSET;
+	temp |= (src_step & SPRD_DMA_TRSF_STEP_MASK) << SPRD_DMA_SRC_TRSF_STEP_OFFSET;
+	hw->trsf_step = temp;
 
-	case SPRD_DMA_TRANS_FRAG_INT:
-		hw->intc |= SPRD_DMA_TRANS_INT_EN | SPRD_DMA_FRAG_INT_EN;
-		break;
+	hw->frg_step = 0;
+	hw->src_blk_step = 0;
+	hw->des_blk_step = 0;
+	return 0;
+}
 
-	case SPRD_DMA_TRANS_BLK_INT:
-		hw->intc |= SPRD_DMA_TRANS_INT_EN | SPRD_DMA_BLK_INT_EN;
-		break;
+static struct dma_async_tx_descriptor *
+sprd_dma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+			 size_t len, unsigned long flags)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	struct sprd_dma_desc *sdesc;
+	struct sprd_dma_chn_hw *hw;
+	enum sprd_dma_datawidth datawidth;
+	u32 step, temp;
 
-	case SPRD_DMA_LIST_INT:
-		hw->intc |= SPRD_DMA_LIST_INT_EN;
-		break;
+	sdesc = kzalloc(sizeof(*sdesc), GFP_NOWAIT);
+	if (!sdesc)
+		return NULL;
 
-	case SPRD_DMA_CFGERR_INT:
-		hw->intc |= SPRD_DMA_CFG_ERR_INT_EN;
-		break;
+	hw = &sdesc->chn_hw;
 
-	default:
-		dev_err(sdev->dma_dev.dev, "invalid irq mode\n");
-		return -EINVAL;
+	hw->cfg = SPRD_DMA_DONOT_WAIT_BDONE << SPRD_DMA_WAIT_BDONE_OFFSET;
+	hw->intc = SPRD_DMA_TRANS_INT | SPRD_DMA_CFG_ERR_INT_EN;
+	hw->src_addr = src & SPRD_DMA_LOW_ADDR_MASK;
+	hw->des_addr = dest & SPRD_DMA_LOW_ADDR_MASK;
+	hw->wrap_ptr = (src >> SPRD_DMA_HIGH_ADDR_OFFSET) &
+		SPRD_DMA_HIGH_ADDR_MASK;
+	hw->wrap_to = (dest >> SPRD_DMA_HIGH_ADDR_OFFSET) &
+		SPRD_DMA_HIGH_ADDR_MASK;
+
+	if (IS_ALIGNED(len, 8)) {
+		datawidth = SPRD_DMA_DATAWIDTH_8_BYTES;
+		step = SPRD_DMA_DWORD_STEP;
+	} else if (IS_ALIGNED(len, 4)) {
+		datawidth = SPRD_DMA_DATAWIDTH_4_BYTES;
+		step = SPRD_DMA_WORD_STEP;
+	} else if (IS_ALIGNED(len, 2)) {
+		datawidth = SPRD_DMA_DATAWIDTH_2_BYTES;
+		step = SPRD_DMA_SHORT_STEP;
+	} else {
+		datawidth = SPRD_DMA_DATAWIDTH_1_BYTE;
+		step = SPRD_DMA_BYTE_STEP;
 	}
 
-	if (transcation_len == 0)
-		hw->trsc_len = block_len & SPRD_DMA_TRSC_LEN_MASK;
-	else
-		hw->trsc_len = transcation_len & SPRD_DMA_TRSC_LEN_MASK;
+	temp = datawidth << SPRD_DMA_SRC_DATAWIDTH_OFFSET;
+	temp |= datawidth << SPRD_DMA_DES_DATAWIDTH_OFFSET;
+	temp |= SPRD_DMA_TRANS_REQ << SPRD_DMA_REQ_MODE_OFFSET;
+	temp |= len & SPRD_DMA_FRG_LEN_MASK;
+	hw->frg_len = temp;
 
-	hw->trsf_step = (des_step & SPRD_DMA_TRSF_STEP_MASK) <<
-			SPRD_DMA_DEST_TRSF_STEP_OFFSET |
-			(src_step & SPRD_DMA_TRSF_STEP_MASK) <<
-			SPRD_DMA_SRC_TRSF_STEP_OFFSET;
+	hw->blk_len = len & SPRD_DMA_BLK_LEN_MASK;
+	hw->trsc_len = len & SPRD_DMA_TRSC_LEN_MASK;
 
-	hw->frg_step = 0;
-	hw->src_blk_step = 0;
-	hw->des_blk_step = 0;
-	hw->src_blk_step = 0;
-	return 0;
+	temp = (step & SPRD_DMA_TRSF_STEP_MASK) << SPRD_DMA_DEST_TRSF_STEP_OFFSET;
+	temp |= (step & SPRD_DMA_TRSF_STEP_MASK) << SPRD_DMA_SRC_TRSF_STEP_OFFSET;
+	hw->trsf_step = temp;
+
+	return vchan_tx_prep(&schan->vc, &sdesc->vd, flags);
 }
 
 static struct dma_async_tx_descriptor *
-sprd_dma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
-			 size_t len, unsigned long flags)
+sprd_dma_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
+		       unsigned int sglen, enum dma_transfer_direction dir,
+		       unsigned long flags, void *context)
 {
 	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	struct dma_slave_config *slave_cfg = &schan->slave_cfg;
+	dma_addr_t src = 0, dst = 0;
 	struct sprd_dma_desc *sdesc;
-	int ret;
+	struct scatterlist *sg;
+	u32 len = 0;
+	int ret, i;
+
+	/* TODO: now we only support one sg for each DMA configuration. */
+	if (!is_slave_direction(dir) || sglen > 1)
+		return NULL;
 
 	sdesc = kzalloc(sizeof(*sdesc), GFP_NOWAIT);
 	if (!sdesc)
 		return NULL;
 
-	ret = sprd_dma_config(chan, sdesc, dest, src, len);
+	for_each_sg(sgl, sg, sglen, i) {
+		len = sg_dma_len(sg);
+
+		if (dir == DMA_MEM_TO_DEV) {
+			src = sg_dma_address(sg);
+			dst = slave_cfg->dst_addr;
+		} else {
+			src = slave_cfg->src_addr;
+			dst = sg_dma_address(sg);
+		}
+	}
+
+	ret = sprd_dma_fill_desc(chan, sdesc, src, dst, len, dir, flags,
+				 slave_cfg);
 	if (ret) {
 		kfree(sdesc);
 		return NULL;
@@ -731,6 +774,19 @@ sprd_dma_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	return vchan_tx_prep(&schan->vc, &sdesc->vd, flags);
 }
 
+static int sprd_dma_slave_config(struct dma_chan *chan,
+				 struct dma_slave_config *config)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	struct dma_slave_config *slave_cfg = &schan->slave_cfg;
+
+	if (!is_slave_direction(config->direction))
+		return -EINVAL;
+
+	memcpy(slave_cfg, config, sizeof(*config));
+	return 0;
+}
+
 static int sprd_dma_pause(struct dma_chan *chan)
 {
 	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
@@ -842,10 +898,9 @@ static int sprd_dma_probe(struct platform_device *pdev)
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	sdev->glb_base = devm_ioremap_nocache(&pdev->dev, res->start,
-					      resource_size(res));
-	if (!sdev->glb_base)
-		return -ENOMEM;
+	sdev->glb_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(sdev->glb_base))
+		return PTR_ERR(sdev->glb_base);
 
 	dma_cap_set(DMA_MEMCPY, sdev->dma_dev.cap_mask);
 	sdev->total_chns = chn_count;
@@ -858,6 +913,8 @@ static int sprd_dma_probe(struct platform_device *pdev)
 	sdev->dma_dev.device_tx_status = sprd_dma_tx_status;
 	sdev->dma_dev.device_issue_pending = sprd_dma_issue_pending;
 	sdev->dma_dev.device_prep_dma_memcpy = sprd_dma_prep_dma_memcpy;
+	sdev->dma_dev.device_prep_slave_sg = sprd_dma_prep_slave_sg;
+	sdev->dma_dev.device_config = sprd_dma_slave_config;
 	sdev->dma_dev.device_pause = sprd_dma_pause;
 	sdev->dma_dev.device_resume = sprd_dma_resume;
 	sdev->dma_dev.device_terminate_all = sprd_dma_terminate_all;
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index c2b089af0420..1bc149af990e 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -2889,8 +2889,7 @@ static int __init d40_dmaengine_init(struct d40_base *base,
 #ifdef CONFIG_PM_SLEEP
 static int dma40_suspend(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct d40_base *base = platform_get_drvdata(pdev);
+	struct d40_base *base = dev_get_drvdata(dev);
 	int ret;
 
 	ret = pm_runtime_force_suspend(dev);
@@ -2904,8 +2903,7 @@ static int dma40_suspend(struct device *dev)
 
 static int dma40_resume(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct d40_base *base = platform_get_drvdata(pdev);
+	struct d40_base *base = dev_get_drvdata(dev);
 	int ret = 0;
 
 	if (base->lcpa_regulator) {
@@ -2970,8 +2968,7 @@ static void d40_save_restore_registers(struct d40_base *base, bool save)
 
 static int dma40_runtime_suspend(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct d40_base *base = platform_get_drvdata(pdev);
+	struct d40_base *base = dev_get_drvdata(dev);
 
 	d40_save_restore_registers(base, true);
 
@@ -2985,8 +2982,7 @@ static int dma40_runtime_suspend(struct device *dev)
 
 static int dma40_runtime_resume(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct d40_base *base = platform_get_drvdata(pdev);
+	struct d40_base *base = dev_get_drvdata(dev);
 
 	d40_save_restore_registers(base, false);
 
diff --git a/drivers/dma/stm32-mdma.c b/drivers/dma/stm32-mdma.c
index daa1602eb9f5..9dc450b7ace6 100644
--- a/drivers/dma/stm32-mdma.c
+++ b/drivers/dma/stm32-mdma.c
@@ -252,13 +252,17 @@ struct stm32_mdma_hwdesc {
 	u32 cmdr;
 } __aligned(64);
 
+struct stm32_mdma_desc_node {
+	struct stm32_mdma_hwdesc *hwdesc;
+	dma_addr_t hwdesc_phys;
+};
+
 struct stm32_mdma_desc {
 	struct virt_dma_desc vdesc;
 	u32 ccr;
-	struct stm32_mdma_hwdesc *hwdesc;
-	dma_addr_t hwdesc_phys;
 	bool cyclic;
 	u32 count;
+	struct stm32_mdma_desc_node node[];
 };
 
 struct stm32_mdma_chan {
@@ -344,30 +348,42 @@ static struct stm32_mdma_desc *stm32_mdma_alloc_desc(
 		struct stm32_mdma_chan *chan, u32 count)
 {
 	struct stm32_mdma_desc *desc;
+	int i;
 
-	desc = kzalloc(sizeof(*desc), GFP_NOWAIT);
+	desc = kzalloc(offsetof(typeof(*desc), node[count]), GFP_NOWAIT);
 	if (!desc)
 		return NULL;
 
-	desc->hwdesc = dma_pool_alloc(chan->desc_pool, GFP_NOWAIT,
-				      &desc->hwdesc_phys);
-	if (!desc->hwdesc) {
-		dev_err(chan2dev(chan), "Failed to allocate descriptor\n");
-		kfree(desc);
-		return NULL;
+	for (i = 0; i < count; i++) {
+		desc->node[i].hwdesc =
+			dma_pool_alloc(chan->desc_pool, GFP_NOWAIT,
+				       &desc->node[i].hwdesc_phys);
+		if (!desc->node[i].hwdesc)
+			goto err;
 	}
 
 	desc->count = count;
 
 	return desc;
+
+err:
+	dev_err(chan2dev(chan), "Failed to allocate descriptor\n");
+	while (--i >= 0)
+		dma_pool_free(chan->desc_pool, desc->node[i].hwdesc,
+			      desc->node[i].hwdesc_phys);
+	kfree(desc);
+	return NULL;
 }
 
 static void stm32_mdma_desc_free(struct virt_dma_desc *vdesc)
 {
 	struct stm32_mdma_desc *desc = to_stm32_mdma_desc(vdesc);
 	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(vdesc->tx.chan);
+	int i;
 
-	dma_pool_free(chan->desc_pool, desc->hwdesc, desc->hwdesc_phys);
+	for (i = 0; i < desc->count; i++)
+		dma_pool_free(chan->desc_pool, desc->node[i].hwdesc,
+			      desc->node[i].hwdesc_phys);
 	kfree(desc);
 }
 
@@ -410,13 +426,10 @@ static enum dma_slave_buswidth stm32_mdma_get_max_width(dma_addr_t addr,
 static u32 stm32_mdma_get_best_burst(u32 buf_len, u32 tlen, u32 max_burst,
 				     enum dma_slave_buswidth width)
 {
-	u32 best_burst = max_burst;
-	u32 burst_len = best_burst * width;
+	u32 best_burst;
 
-	while ((burst_len > 0) && (tlen % burst_len)) {
-		best_burst = best_burst >> 1;
-		burst_len = best_burst * width;
-	}
+	best_burst = min((u32)1 << __ffs(tlen | buf_len),
+			 max_burst * width) / width;
 
 	return (best_burst > 0) ? best_burst : 1;
 }
@@ -669,18 +682,18 @@ static int stm32_mdma_set_xfer_param(struct stm32_mdma_chan *chan,
 }
 
 static void stm32_mdma_dump_hwdesc(struct stm32_mdma_chan *chan,
-				   struct stm32_mdma_hwdesc *hwdesc)
+				   struct stm32_mdma_desc_node *node)
 {
-	dev_dbg(chan2dev(chan), "hwdesc:  0x%p\n", hwdesc);
-	dev_dbg(chan2dev(chan), "CTCR:    0x%08x\n", hwdesc->ctcr);
-	dev_dbg(chan2dev(chan), "CBNDTR:  0x%08x\n", hwdesc->cbndtr);
-	dev_dbg(chan2dev(chan), "CSAR:    0x%08x\n", hwdesc->csar);
-	dev_dbg(chan2dev(chan), "CDAR:    0x%08x\n", hwdesc->cdar);
-	dev_dbg(chan2dev(chan), "CBRUR:   0x%08x\n", hwdesc->cbrur);
-	dev_dbg(chan2dev(chan), "CLAR:    0x%08x\n", hwdesc->clar);
-	dev_dbg(chan2dev(chan), "CTBR:    0x%08x\n", hwdesc->ctbr);
-	dev_dbg(chan2dev(chan), "CMAR:    0x%08x\n", hwdesc->cmar);
-	dev_dbg(chan2dev(chan), "CMDR:    0x%08x\n\n", hwdesc->cmdr);
+	dev_dbg(chan2dev(chan), "hwdesc:  %pad\n", &node->hwdesc_phys);
+	dev_dbg(chan2dev(chan), "CTCR:    0x%08x\n", node->hwdesc->ctcr);
+	dev_dbg(chan2dev(chan), "CBNDTR:  0x%08x\n", node->hwdesc->cbndtr);
+	dev_dbg(chan2dev(chan), "CSAR:    0x%08x\n", node->hwdesc->csar);
+	dev_dbg(chan2dev(chan), "CDAR:    0x%08x\n", node->hwdesc->cdar);
+	dev_dbg(chan2dev(chan), "CBRUR:   0x%08x\n", node->hwdesc->cbrur);
+	dev_dbg(chan2dev(chan), "CLAR:    0x%08x\n", node->hwdesc->clar);
+	dev_dbg(chan2dev(chan), "CTBR:    0x%08x\n", node->hwdesc->ctbr);
+	dev_dbg(chan2dev(chan), "CMAR:    0x%08x\n", node->hwdesc->cmar);
+	dev_dbg(chan2dev(chan), "CMDR:    0x%08x\n\n", node->hwdesc->cmdr);
 }
 
 static void stm32_mdma_setup_hwdesc(struct stm32_mdma_chan *chan,
@@ -694,7 +707,7 @@ static void stm32_mdma_setup_hwdesc(struct stm32_mdma_chan *chan,
 	struct stm32_mdma_hwdesc *hwdesc;
 	u32 next = count + 1;
 
-	hwdesc = &desc->hwdesc[count];
+	hwdesc = desc->node[count].hwdesc;
 	hwdesc->ctcr = ctcr;
 	hwdesc->cbndtr &= ~(STM32_MDMA_CBNDTR_BRC_MK |
 			STM32_MDMA_CBNDTR_BRDUM |
@@ -704,19 +717,20 @@ static void stm32_mdma_setup_hwdesc(struct stm32_mdma_chan *chan,
 	hwdesc->csar = src_addr;
 	hwdesc->cdar = dst_addr;
 	hwdesc->cbrur = 0;
-	hwdesc->clar = desc->hwdesc_phys + next * sizeof(*hwdesc);
 	hwdesc->ctbr = ctbr;
 	hwdesc->cmar = config->mask_addr;
 	hwdesc->cmdr = config->mask_data;
 
 	if (is_last) {
 		if (is_cyclic)
-			hwdesc->clar = desc->hwdesc_phys;
+			hwdesc->clar = desc->node[0].hwdesc_phys;
 		else
 			hwdesc->clar = 0;
+	} else {
+		hwdesc->clar = desc->node[next].hwdesc_phys;
 	}
 
-	stm32_mdma_dump_hwdesc(chan, hwdesc);
+	stm32_mdma_dump_hwdesc(chan, &desc->node[count]);
 }
 
 static int stm32_mdma_setup_xfer(struct stm32_mdma_chan *chan,
@@ -780,7 +794,7 @@ stm32_mdma_prep_slave_sg(struct dma_chan *c, struct scatterlist *sgl,
 {
 	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
 	struct stm32_mdma_desc *desc;
-	int ret;
+	int i, ret;
 
 	/*
 	 * Once DMA is in setup cyclic mode the channel we cannot assign this
@@ -806,7 +820,9 @@ stm32_mdma_prep_slave_sg(struct dma_chan *c, struct scatterlist *sgl,
 	return vchan_tx_prep(&chan->vchan, &desc->vdesc, flags);
 
 xfer_setup_err:
-	dma_pool_free(chan->desc_pool, &desc->hwdesc, desc->hwdesc_phys);
+	for (i = 0; i < desc->count; i++)
+		dma_pool_free(chan->desc_pool, desc->node[i].hwdesc,
+			      desc->node[i].hwdesc_phys);
 	kfree(desc);
 	return NULL;
 }
@@ -895,7 +911,9 @@ stm32_mdma_prep_dma_cyclic(struct dma_chan *c, dma_addr_t buf_addr,
 	return vchan_tx_prep(&chan->vchan, &desc->vdesc, flags);
 
 xfer_setup_err:
-	dma_pool_free(chan->desc_pool, &desc->hwdesc, desc->hwdesc_phys);
+	for (i = 0; i < desc->count; i++)
+		dma_pool_free(chan->desc_pool, desc->node[i].hwdesc,
+			      desc->node[i].hwdesc_phys);
 	kfree(desc);
 	return NULL;
 }
@@ -1009,7 +1027,7 @@ stm32_mdma_prep_dma_memcpy(struct dma_chan *c, dma_addr_t dest, dma_addr_t src,
 			ctcr |= STM32_MDMA_CTCR_PKE;
 
 		/* Prepare hardware descriptor */
-		hwdesc = desc->hwdesc;
+		hwdesc = desc->node[0].hwdesc;
 		hwdesc->ctcr = ctcr;
 		hwdesc->cbndtr = cbndtr;
 		hwdesc->csar = src;
@@ -1020,7 +1038,7 @@ stm32_mdma_prep_dma_memcpy(struct dma_chan *c, dma_addr_t dest, dma_addr_t src,
 		hwdesc->cmar = 0;
 		hwdesc->cmdr = 0;
 
-		stm32_mdma_dump_hwdesc(chan, hwdesc);
+		stm32_mdma_dump_hwdesc(chan, &desc->node[0]);
 	} else {
 		/* Setup a LLI transfer */
 		ctcr |= STM32_MDMA_CTCR_TRGM(STM32_MDMA_LINKED_LIST) |
@@ -1120,7 +1138,7 @@ static void stm32_mdma_start_transfer(struct stm32_mdma_chan *chan)
 	}
 
 	chan->desc = to_stm32_mdma_desc(vdesc);
-	hwdesc = chan->desc->hwdesc;
+	hwdesc = chan->desc->node[0].hwdesc;
 	chan->curr_hwdesc = 0;
 
 	stm32_mdma_write(dmadev, STM32_MDMA_CCR(id), chan->desc->ccr);
@@ -1198,7 +1216,7 @@ static int stm32_mdma_resume(struct dma_chan *c)
 	unsigned long flags;
 	u32 status, reg;
 
-	hwdesc = &chan->desc->hwdesc[chan->curr_hwdesc];
+	hwdesc = chan->desc->node[chan->curr_hwdesc].hwdesc;
 
 	spin_lock_irqsave(&chan->vchan.lock, flags);
 
@@ -1268,13 +1286,13 @@ static size_t stm32_mdma_desc_residue(struct stm32_mdma_chan *chan,
 				      u32 curr_hwdesc)
 {
 	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct stm32_mdma_hwdesc *hwdesc = desc->node[0].hwdesc;
 	u32 cbndtr, residue, modulo, burst_size;
 	int i;
 
 	residue = 0;
 	for (i = curr_hwdesc + 1; i < desc->count; i++) {
-		struct stm32_mdma_hwdesc *hwdesc = &desc->hwdesc[i];
-
+		hwdesc = desc->node[i].hwdesc;
 		residue += STM32_MDMA_CBNDTR_BNDT(hwdesc->cbndtr);
 	}
 	cbndtr = stm32_mdma_read(dmadev, STM32_MDMA_CBNDTR(chan->id));
@@ -1503,7 +1521,7 @@ static struct dma_chan *stm32_mdma_of_xlate(struct of_phandle_args *dma_spec,
 
 	c = dma_get_any_slave_channel(&dmadev->ddev);
 	if (!c) {
-		dev_err(mdma2dev(dmadev), "No more channel avalaible\n");
+		dev_err(mdma2dev(dmadev), "No more channels available\n");
 		return NULL;
 	}
 
diff --git a/drivers/dma/ti/Kconfig b/drivers/dma/ti/Kconfig
new file mode 100644
index 000000000000..e5e74e1361dc
--- /dev/null
+++ b/drivers/dma/ti/Kconfig
@@ -0,0 +1,37 @@
+#
+# Texas Instruments DMA drivers
+#
+
+config TI_CPPI41
+	tristate "Texas Instruments CPPI 4.1 DMA support"
+	depends on (ARCH_OMAP || ARCH_DAVINCI_DA8XX)
+	select DMA_ENGINE
+	help
+	  The Communications Port Programming Interface (CPPI) 4.1 DMA engine
+	  is currently used by the USB driver on AM335x and DA8xx platforms.
+
+config TI_EDMA
+	tristate "Texas Instruments EDMA support"
+	depends on ARCH_DAVINCI || ARCH_OMAP || ARCH_KEYSTONE || COMPILE_TEST
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	select TI_DMA_CROSSBAR if (ARCH_OMAP || COMPILE_TEST)
+	default y
+	help
+	  Enable support for the TI EDMA (Enhanced DMA) controller. This DMA
+	  engine is found on TI DaVinci, AM33xx, AM43xx, DRA7xx and Keystone 2
+	  parts.
+
+config DMA_OMAP
+	tristate "Texas Instruments sDMA (omap-dma) support"
+	depends on ARCH_OMAP || COMPILE_TEST
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	select TI_DMA_CROSSBAR if (SOC_DRA7XX || COMPILE_TEST)
+	default y
+	help
+	  Enable support for the TI sDMA (System DMA or DMA4) controller. This
+	  DMA engine is found on OMAP and DRA7xx parts.
+
+config TI_DMA_CROSSBAR
+	bool
diff --git a/drivers/dma/ti/Makefile b/drivers/dma/ti/Makefile
new file mode 100644
index 000000000000..113e59ec9c32
--- /dev/null
+++ b/drivers/dma/ti/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_TI_CPPI41) += cppi41.o
+obj-$(CONFIG_TI_EDMA) += edma.o
+obj-$(CONFIG_DMA_OMAP) += omap-dma.o
+obj-$(CONFIG_TI_DMA_CROSSBAR) += dma-crossbar.o
diff --git a/drivers/dma/cppi41.c b/drivers/dma/ti/cppi41.c
index d9bee65a18a4..1497da367710 100644
--- a/drivers/dma/cppi41.c
+++ b/drivers/dma/ti/cppi41.c
@@ -11,7 +11,7 @@
 #include <linux/interrupt.h>
 #include <linux/of_address.h>
 #include <linux/pm_runtime.h>
-#include "dmaengine.h"
+#include "../dmaengine.h"
 
 #define DESC_TYPE	27
 #define DESC_TYPE_HOST	0x10
diff --git a/drivers/dma/ti-dma-crossbar.c b/drivers/dma/ti/dma-crossbar.c
index 9272b173c746..9272b173c746 100644
--- a/drivers/dma/ti-dma-crossbar.c
+++ b/drivers/dma/ti/dma-crossbar.c
diff --git a/drivers/dma/edma.c b/drivers/dma/ti/edma.c
index 9bc722ca8329..ceabdea40ae0 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/ti/edma.c
@@ -33,8 +33,8 @@
 
 #include <linux/platform_data/edma.h>
 
-#include "dmaengine.h"
-#include "virt-dma.h"
+#include "../dmaengine.h"
+#include "../virt-dma.h"
 
 /* Offsets matching "struct edmacc_param" */
 #define PARM_OPT		0x00
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/ti/omap-dma.c
index 9483000fcf79..9b5ca8691f27 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/ti/omap-dma.c
@@ -21,7 +21,7 @@
 #include <linux/of_dma.h>
 #include <linux/of_device.h>
 
-#include "virt-dma.h"
+#include "../virt-dma.h"
 
 #define OMAP_SDMA_REQUESTS	127
 #define OMAP_SDMA_CHANNELS	32
diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 4d8c7b9078fd..eb45af71d3a3 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -1244,8 +1244,7 @@ static void txx9dmac_shutdown(struct platform_device *pdev)
 
 static int txx9dmac_suspend_noirq(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct txx9dmac_dev *ddev = platform_get_drvdata(pdev);
+	struct txx9dmac_dev *ddev = dev_get_drvdata(dev);
 
 	txx9dmac_off(ddev);
 	return 0;
@@ -1253,9 +1252,8 @@ static int txx9dmac_suspend_noirq(struct device *dev)
 
 static int txx9dmac_resume_noirq(struct device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev);
-	struct txx9dmac_dev *ddev = platform_get_drvdata(pdev);
-	struct txx9dmac_platform_data *pdata = dev_get_platdata(&pdev->dev);
+	struct txx9dmac_dev *ddev = dev_get_drvdata(dev);
+	struct txx9dmac_platform_data *pdata = dev_get_platdata(dev);
 	u32 mcr;
 
 	mcr = TXX9_DMA_MCR_MSTEN | MCR_LE;
diff --git a/include/linux/dma/sprd-dma.h b/include/linux/dma/sprd-dma.h
new file mode 100644
index 000000000000..b0115e340fbc
--- /dev/null
+++ b/include/linux/dma/sprd-dma.h
@@ -0,0 +1,61 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifndef _SPRD_DMA_H_
+#define _SPRD_DMA_H_
+
+#define SPRD_DMA_REQ_SHIFT 16
+#define SPRD_DMA_FLAGS(req_mode, int_type) \
+	((req_mode) << SPRD_DMA_REQ_SHIFT | (int_type))
+
+/*
+ * enum sprd_dma_req_mode: define the DMA request mode
+ * @SPRD_DMA_FRAG_REQ: fragment request mode
+ * @SPRD_DMA_BLK_REQ: block request mode
+ * @SPRD_DMA_TRANS_REQ: transaction request mode
+ * @SPRD_DMA_LIST_REQ: link-list request mode
+ *
+ * We have 4 types request mode: fragment mode, block mode, transaction mode
+ * and linklist mode. One transaction can contain several blocks, one block can
+ * contain several fragments. Link-list mode means we can save several DMA
+ * configuration into one reserved memory, then DMA can fetch each DMA
+ * configuration automatically to start transfer.
+ */
+enum sprd_dma_req_mode {
+	SPRD_DMA_FRAG_REQ,
+	SPRD_DMA_BLK_REQ,
+	SPRD_DMA_TRANS_REQ,
+	SPRD_DMA_LIST_REQ,
+};
+
+/*
+ * enum sprd_dma_int_type: define the DMA interrupt type
+ * @SPRD_DMA_NO_INT: do not need generate DMA interrupts.
+ * @SPRD_DMA_FRAG_INT: fragment done interrupt when one fragment request
+ * is done.
+ * @SPRD_DMA_BLK_INT: block done interrupt when one block request is done.
+ * @SPRD_DMA_BLK_FRAG_INT: block and fragment interrupt when one fragment
+ * or one block request is done.
+ * @SPRD_DMA_TRANS_INT: tansaction done interrupt when one transaction
+ * request is done.
+ * @SPRD_DMA_TRANS_FRAG_INT: transaction and fragment interrupt when one
+ * transaction request or fragment request is done.
+ * @SPRD_DMA_TRANS_BLK_INT: transaction and block interrupt when one
+ * transaction request or block request is done.
+ * @SPRD_DMA_LIST_INT: link-list done interrupt when one link-list request
+ * is done.
+ * @SPRD_DMA_CFGERR_INT: configure error interrupt when configuration is
+ * incorrect.
+ */
+enum sprd_dma_int_type {
+	SPRD_DMA_NO_INT,
+	SPRD_DMA_FRAG_INT,
+	SPRD_DMA_BLK_INT,
+	SPRD_DMA_BLK_FRAG_INT,
+	SPRD_DMA_TRANS_INT,
+	SPRD_DMA_TRANS_FRAG_INT,
+	SPRD_DMA_TRANS_BLK_INT,
+	SPRD_DMA_LIST_INT,
+	SPRD_DMA_CFGERR_INT,
+};
+
+#endif