Merge tag 'dmaengine-4.15-rc1' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine updates from Vinod Koul:
 "Updates for this cycle include:

   - new driver for Spreadtrum dma controller, ST MDMA and DMAMUX
     controllers

   - PM support for IMG MDC drivers

   - updates to bcm-sba-raid driver and improvements to sun6i driver

   - subsystem conversion for:
      - timers to use timer_setup()
      - remove usage of PCI pool API
      - usage of %p format specifier

   - minor updates to bunch of drivers"

* tag 'dmaengine-4.15-rc1' of git://git.infradead.org/users/vkoul/slave-dma: (49 commits)
  dmaengine: ti-dma-crossbar: Correct am335x/am43xx mux value type
  dmaengine: dmatest: warn user when dma test times out
  dmaengine: Revert "rcar-dmac: use TCRB instead of TCR for residue"
  dmaengine: stm32_mdma: activate pack/unpack feature
  dmaengine: at_hdmac: Remove unnecessary 0x prefixes before %pad
  dmaengine: coh901318: Remove unnecessary 0x prefixes before %pad
  MAINTAINERS: Step down from a co-maintaner of DW DMAC driver
  dmaengine: pch_dma: Replace PCI pool old API
  dmaengine: Convert timers to use timer_setup()
  dmaengine: sprd: Add Spreadtrum DMA driver
  dt-bindings: dmaengine: Add Spreadtrum SC9860 DMA controller
  dmaengine: sun6i: Retrieve channel count/max request from devicetree
  dmaengine: Build bcm-sba-raid driver as loadable module for iProc SoCs
  dmaengine: bcm-sba-raid: Use common GPL comment header
  dmaengine: bcm-sba-raid: Use only single mailbox channel
  dmaengine: bcm-sba-raid: serialize dma_cookie_complete() using reqs_lock
  dmaengine: pl330: fix descriptor allocation fail
  dmaengine: rcar-dmac: use TCRB instead of TCR for residue
  dmaengine: sun6i: Add support for Allwinner A64 and compatibles
  arm64: allwinner: a64: Add devicetree binding for DMA controller
  ...

26 files changed, 3597 insertions, 291 deletions

diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index fadc4d8783bd..27df3e2837fd 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -115,7 +115,7 @@ config BCM_SBA_RAID
 	select DMA_ENGINE_RAID
 	select ASYNC_TX_DISABLE_XOR_VAL_DMA
 	select ASYNC_TX_DISABLE_PQ_VAL_DMA
-	default ARCH_BCM_IPROC
+	default m if ARCH_BCM_IPROC
 	help
 	  Enable support for Broadcom SBA RAID Engine. The SBA RAID
 	  engine is available on most of the Broadcom iProc SoCs. It
@@ -483,6 +483,35 @@ config STM32_DMA
 	  If you have a board based on such a MCU and wish to use DMA say Y
 	  here.
 
+config STM32_DMAMUX
+	bool "STMicroelectronics STM32 dma multiplexer support"
+	depends on STM32_DMA || COMPILE_TEST
+	help
+	  Enable support for the on-chip DMA multiplexer on STMicroelectronics
+	  STM32 MCUs.
+	  If you have a board based on such a MCU and wish to use DMAMUX say Y
+	  here.
+
+config STM32_MDMA
+	bool "STMicroelectronics STM32 master dma support"
+	depends on ARCH_STM32 || COMPILE_TEST
+	depends on OF
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	help
+	  Enable support for the on-chip MDMA controller on STMicroelectronics
+	  STM32 platforms.
+	  If you have a board based on STM32 SoC and wish to use the master DMA
+	  say Y here.
+
+config SPRD_DMA
+	tristate "Spreadtrum DMA support"
+	depends on ARCH_SPRD || COMPILE_TEST
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	help
+	  Enable support for the on-chip DMA controller on Spreadtrum platform.
+
 config S3C24XX_DMAC
 	bool "Samsung S3C24XX DMA support"
 	depends on ARCH_S3C24XX || COMPILE_TEST
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 9d0156b50294..b9dca8a0e142 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -60,6 +60,9 @@ obj-$(CONFIG_RENESAS_DMA) += sh/
 obj-$(CONFIG_SIRF_DMA) += sirf-dma.o
 obj-$(CONFIG_STE_DMA40) += ste_dma40.o ste_dma40_ll.o
 obj-$(CONFIG_STM32_DMA) += stm32-dma.o
+obj-$(CONFIG_STM32_DMAMUX) += stm32-dmamux.o
+obj-$(CONFIG_STM32_MDMA) += stm32-mdma.o
+obj-$(CONFIG_SPRD_DMA) += sprd-dma.o
 obj-$(CONFIG_S3C24XX_DMAC) += s3c24xx-dma.o
 obj-$(CONFIG_TXX9_DMAC) += txx9dmac.o
 obj-$(CONFIG_TEGRA20_APB_DMA) += tegra20-apb-dma.o
diff --git a/drivers/dma/at_hdmac_regs.h b/drivers/dma/at_hdmac_regs.h
index 7f58f06157f6..ef3f227ce3e6 100644
--- a/drivers/dma/at_hdmac_regs.h
+++ b/drivers/dma/at_hdmac_regs.h
@@ -385,7 +385,7 @@ static void vdbg_dump_regs(struct at_dma_chan *atchan) {}
 static void atc_dump_lli(struct at_dma_chan *atchan, struct at_lli *lli)
 {
 	dev_crit(chan2dev(&atchan->chan_common),
-		 "  desc: s%pad d%pad ctrl0x%x:0x%x l0x%pad\n",
+		 "desc: s%pad d%pad ctrl0x%x:0x%x l%pad\n",
 		 &lli->saddr, &lli->daddr,
 		 lli->ctrla, lli->ctrlb, &lli->dscr);
 }
diff --git a/drivers/dma/bcm-sba-raid.c b/drivers/dma/bcm-sba-raid.c
index 6c2c44724637..3956a018bf5a 100644
--- a/drivers/dma/bcm-sba-raid.c
+++ b/drivers/dma/bcm-sba-raid.c
@@ -1,9 +1,14 @@
 /*
  * Copyright (C) 2017 Broadcom
  *
- * 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.
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
  */
 
 /*
@@ -25,11 +30,8 @@
  *
  * The Broadcom SBA RAID driver does not require any register programming
  * except submitting request to SBA hardware device via mailbox channels.
- * This driver implements a DMA device with one DMA channel using a set
- * of mailbox channels provided by Broadcom SoC specific ring manager
- * driver. To exploit parallelism (as described above), all DMA request
- * coming to SBA RAID DMA channel are broken down to smaller requests
- * and submitted to multiple mailbox channels in round-robin fashion.
+ * This driver implements a DMA device with one DMA channel using a single
+ * mailbox channel provided by Broadcom SoC specific ring manager driver.
  * For having more SBA DMA channels, we can create more SBA device nodes
  * in Broadcom SoC specific DTS based on number of hardware rings supported
  * by Broadcom SoC ring manager.
@@ -85,6 +87,7 @@
 #define SBA_CMD_GALOIS					0xe
 
 #define SBA_MAX_REQ_PER_MBOX_CHANNEL			8192
+#define SBA_MAX_MSG_SEND_PER_MBOX_CHANNEL		8
 
 /* Driver helper macros */
 #define to_sba_request(tx)		\
@@ -142,9 +145,7 @@ struct sba_device {
 	u32 max_cmds_pool_size;
 	/* Maibox client and Mailbox channels */
 	struct mbox_client client;
-	int mchans_count;
-	atomic_t mchans_current;
-	struct mbox_chan **mchans;
+	struct mbox_chan *mchan;
 	struct device *mbox_dev;
 	/* DMA device and DMA channel */
 	struct dma_device dma_dev;
@@ -200,14 +201,6 @@ static inline u32 __pure sba_cmd_pq_c_mdata(u32 d, u32 b1, u32 b0)
 
 /* ====== General helper routines ===== */
 
-static void sba_peek_mchans(struct sba_device *sba)
-{
-	int mchan_idx;
-
-	for (mchan_idx = 0; mchan_idx < sba->mchans_count; mchan_idx++)
-		mbox_client_peek_data(sba->mchans[mchan_idx]);
-}
-
 static struct sba_request *sba_alloc_request(struct sba_device *sba)
 {
 	bool found = false;
@@ -231,7 +224,7 @@ static struct sba_request *sba_alloc_request(struct sba_device *sba)
 		 * would have completed which will create more
 		 * room for new requests.
 		 */
-		sba_peek_mchans(sba);
+		mbox_client_peek_data(sba->mchan);
 		return NULL;
 	}
 
@@ -369,15 +362,11 @@ static void sba_cleanup_pending_requests(struct sba_device *sba)
 static int sba_send_mbox_request(struct sba_device *sba,
 				 struct sba_request *req)
 {
-	int mchans_idx, ret = 0;
-
-	/* Select mailbox channel in round-robin fashion */
-	mchans_idx = atomic_inc_return(&sba->mchans_current);
-	mchans_idx = mchans_idx % sba->mchans_count;
+	int ret = 0;
 
 	/* Send message for the request */
 	req->msg.error = 0;
-	ret = mbox_send_message(sba->mchans[mchans_idx], &req->msg);
+	ret = mbox_send_message(sba->mchan, &req->msg);
 	if (ret < 0) {
 		dev_err(sba->dev, "send message failed with error %d", ret);
 		return ret;
@@ -390,7 +379,7 @@ static int sba_send_mbox_request(struct sba_device *sba,
 	}
 
 	/* Signal txdone for mailbox channel */
-	mbox_client_txdone(sba->mchans[mchans_idx], ret);
+	mbox_client_txdone(sba->mchan, ret);
 
 	return ret;
 }
@@ -402,13 +391,8 @@ static void _sba_process_pending_requests(struct sba_device *sba)
 	u32 count;
 	struct sba_request *req;
 
-	/*
-	 * Process few pending requests
-	 *
-	 * For now, we process (<number_of_mailbox_channels> * 8)
-	 * number of requests at a time.
-	 */
-	count = sba->mchans_count * 8;
+	/* Process few pending requests */
+	count = SBA_MAX_MSG_SEND_PER_MBOX_CHANNEL;
 	while (!list_empty(&sba->reqs_pending_list) && count) {
 		/* Get the first pending request */
 		req = list_first_entry(&sba->reqs_pending_list,
@@ -442,7 +426,9 @@ static void sba_process_received_request(struct sba_device *sba,
 
 		WARN_ON(tx->cookie < 0);
 		if (tx->cookie > 0) {
+			spin_lock_irqsave(&sba->reqs_lock, flags);
 			dma_cookie_complete(tx);
+			spin_unlock_irqrestore(&sba->reqs_lock, flags);
 			dmaengine_desc_get_callback_invoke(tx, NULL);
 			dma_descriptor_unmap(tx);
 			tx->callback = NULL;
@@ -570,7 +556,7 @@ static enum dma_status sba_tx_status(struct dma_chan *dchan,
 	if (ret == DMA_COMPLETE)
 		return ret;
 
-	sba_peek_mchans(sba);
+	mbox_client_peek_data(sba->mchan);
 
 	return dma_cookie_status(dchan, cookie, txstate);
 }
@@ -1637,7 +1623,7 @@ static int sba_async_register(struct sba_device *sba)
 
 static int sba_probe(struct platform_device *pdev)
 {
-	int i, ret = 0, mchans_count;
+	int ret = 0;
 	struct sba_device *sba;
 	struct platform_device *mbox_pdev;
 	struct of_phandle_args args;
@@ -1650,12 +1636,11 @@ static int sba_probe(struct platform_device *pdev)
 	sba->dev = &pdev->dev;
 	platform_set_drvdata(pdev, sba);
 
-	/* Number of channels equals number of mailbox channels */
+	/* Number of mailbox channels should be atleast 1 */
 	ret = of_count_phandle_with_args(pdev->dev.of_node,
 					 "mboxes", "#mbox-cells");
 	if (ret <= 0)
 		return -ENODEV;
-	mchans_count = ret;
 
 	/* Determine SBA version from DT compatible string */
 	if (of_device_is_compatible(sba->dev->of_node, "brcm,iproc-sba"))
@@ -1688,7 +1673,7 @@ static int sba_probe(struct platform_device *pdev)
 	default:
 		return -EINVAL;
 	}
-	sba->max_req = SBA_MAX_REQ_PER_MBOX_CHANNEL * mchans_count;
+	sba->max_req = SBA_MAX_REQ_PER_MBOX_CHANNEL;
 	sba->max_cmd_per_req = sba->max_pq_srcs + 3;
 	sba->max_xor_srcs = sba->max_cmd_per_req - 1;
 	sba->max_resp_pool_size = sba->max_req * sba->hw_resp_size;
@@ -1702,55 +1687,30 @@ static int sba_probe(struct platform_device *pdev)
 	sba->client.knows_txdone	= true;
 	sba->client.tx_tout		= 0;
 
-	/* Allocate mailbox channel array */
-	sba->mchans = devm_kcalloc(&pdev->dev, mchans_count,
-				   sizeof(*sba->mchans), GFP_KERNEL);
-	if (!sba->mchans)
-		return -ENOMEM;
-
-	/* Request mailbox channels */
-	sba->mchans_count = 0;
-	for (i = 0; i < mchans_count; i++) {
-		sba->mchans[i] = mbox_request_channel(&sba->client, i);
-		if (IS_ERR(sba->mchans[i])) {
-			ret = PTR_ERR(sba->mchans[i]);
-			goto fail_free_mchans;
-		}
-		sba->mchans_count++;
+	/* Request mailbox channel */
+	sba->mchan = mbox_request_channel(&sba->client, 0);
+	if (IS_ERR(sba->mchan)) {
+		ret = PTR_ERR(sba->mchan);
+		goto fail_free_mchan;
 	}
-	atomic_set(&sba->mchans_current, 0);
 
 	/* Find-out underlying mailbox device */
 	ret = of_parse_phandle_with_args(pdev->dev.of_node,
 					 "mboxes", "#mbox-cells", 0, &args);
 	if (ret)
-		goto fail_free_mchans;
+		goto fail_free_mchan;
 	mbox_pdev = of_find_device_by_node(args.np);
 	of_node_put(args.np);
 	if (!mbox_pdev) {
 		ret = -ENODEV;
-		goto fail_free_mchans;
+		goto fail_free_mchan;
 	}
 	sba->mbox_dev = &mbox_pdev->dev;
 
-	/* All mailbox channels should be of same ring manager device */
-	for (i = 1; i < mchans_count; i++) {
-		ret = of_parse_phandle_with_args(pdev->dev.of_node,
-					 "mboxes", "#mbox-cells", i, &args);
-		if (ret)
-			goto fail_free_mchans;
-		mbox_pdev = of_find_device_by_node(args.np);
-		of_node_put(args.np);
-		if (sba->mbox_dev != &mbox_pdev->dev) {
-			ret = -EINVAL;
-			goto fail_free_mchans;
-		}
-	}
-
 	/* Prealloc channel resource */
 	ret = sba_prealloc_channel_resources(sba);
 	if (ret)
-		goto fail_free_mchans;
+		goto fail_free_mchan;
 
 	/* Check availability of debugfs */
 	if (!debugfs_initialized())
@@ -1777,24 +1737,22 @@ skip_debugfs:
 		goto fail_free_resources;
 
 	/* Print device info */
-	dev_info(sba->dev, "%s using SBAv%d and %d mailbox channels",
+	dev_info(sba->dev, "%s using SBAv%d mailbox channel from %s",
 		 dma_chan_name(&sba->dma_chan), sba->ver+1,
-		 sba->mchans_count);
+		 dev_name(sba->mbox_dev));
 
 	return 0;
 
 fail_free_resources:
 	debugfs_remove_recursive(sba->root);
 	sba_freeup_channel_resources(sba);
-fail_free_mchans:
-	for (i = 0; i < sba->mchans_count; i++)
-		mbox_free_channel(sba->mchans[i]);
+fail_free_mchan:
+	mbox_free_channel(sba->mchan);
 	return ret;
 }
 
 static int sba_remove(struct platform_device *pdev)
 {
-	int i;
 	struct sba_device *sba = platform_get_drvdata(pdev);
 
 	dma_async_device_unregister(&sba->dma_dev);
@@ -1803,8 +1761,7 @@ static int sba_remove(struct platform_device *pdev)
 
 	sba_freeup_channel_resources(sba);
 
-	for (i = 0; i < sba->mchans_count; i++)
-		mbox_free_channel(sba->mchans[i]);
+	mbox_free_channel(sba->mchan);
 
 	return 0;
 }
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index 74794c9859f6..da74fd74636b 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -1319,8 +1319,8 @@ static void coh901318_list_print(struct coh901318_chan *cohc,
 	int i = 0;
 
 	while (l) {
-		dev_vdbg(COHC_2_DEV(cohc), "i %d, lli %p, ctrl 0x%x, src 0x%pad"
-			 ", dst 0x%pad, link 0x%pad virt_link_addr 0x%p\n",
+		dev_vdbg(COHC_2_DEV(cohc), "i %d, lli %p, ctrl 0x%x, src %pad"
+			 ", dst %pad, link %pad virt_link_addr 0x%p\n",
 			 i, l, l->control, &l->src_addr, &l->dst_addr,
 			 &l->link_addr, l->virt_link_addr);
 		i++;
@@ -2231,7 +2231,7 @@ coh901318_prep_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 	spin_lock_irqsave(&cohc->lock, flg);
 
 	dev_vdbg(COHC_2_DEV(cohc),
-		 "[%s] channel %d src 0x%pad dest 0x%pad size %zu\n",
+		 "[%s] channel %d src %pad dest %pad size %zu\n",
 		 __func__, cohc->id, &src, &dest, size);
 
 	if (flags & DMA_PREP_INTERRUPT)
diff --git a/drivers/dma/dma-axi-dmac.c b/drivers/dma/dma-axi-dmac.c
index 7f0b9aa15867..2419fe524daa 100644
--- a/drivers/dma/dma-axi-dmac.c
+++ b/drivers/dma/dma-axi-dmac.c
@@ -72,6 +72,9 @@
 
 #define AXI_DMAC_FLAG_CYCLIC		BIT(0)
 
+/* The maximum ID allocated by the hardware is 31 */
+#define AXI_DMAC_SG_UNUSED 32U
+
 struct axi_dmac_sg {
 	dma_addr_t src_addr;
 	dma_addr_t dest_addr;
@@ -80,6 +83,7 @@ struct axi_dmac_sg {
 	unsigned int dest_stride;
 	unsigned int src_stride;
 	unsigned int id;
+	bool schedule_when_free;
 };
 
 struct axi_dmac_desc {
@@ -200,11 +204,21 @@ static void axi_dmac_start_transfer(struct axi_dmac_chan *chan)
 	}
 	sg = &desc->sg[desc->num_submitted];
 
+	/* Already queued in cyclic mode. Wait for it to finish */
+	if (sg->id != AXI_DMAC_SG_UNUSED) {
+		sg->schedule_when_free = true;
+		return;
+	}
+
 	desc->num_submitted++;
-	if (desc->num_submitted == desc->num_sgs)
-		chan->next_desc = NULL;
-	else
+	if (desc->num_submitted == desc->num_sgs) {
+		if (desc->cyclic)
+			desc->num_submitted = 0; /* Start again */
+		else
+			chan->next_desc = NULL;
+	} else {
 		chan->next_desc = desc;
+	}
 
 	sg->id = axi_dmac_read(dmac, AXI_DMAC_REG_TRANSFER_ID);
 
@@ -220,9 +234,11 @@ static void axi_dmac_start_transfer(struct axi_dmac_chan *chan)
 
 	/*
 	 * If the hardware supports cyclic transfers and there is no callback to
-	 * call, enable hw cyclic mode to avoid unnecessary interrupts.
+	 * call and only a single segment, enable hw cyclic mode to avoid
+	 * unnecessary interrupts.
 	 */
-	if (chan->hw_cyclic && desc->cyclic && !desc->vdesc.tx.callback)
+	if (chan->hw_cyclic && desc->cyclic && !desc->vdesc.tx.callback &&
+		desc->num_sgs == 1)
 		flags |= AXI_DMAC_FLAG_CYCLIC;
 
 	axi_dmac_write(dmac, AXI_DMAC_REG_X_LENGTH, sg->x_len - 1);
@@ -237,37 +253,52 @@ static struct axi_dmac_desc *axi_dmac_active_desc(struct axi_dmac_chan *chan)
 		struct axi_dmac_desc, vdesc.node);
 }
 
-static void axi_dmac_transfer_done(struct axi_dmac_chan *chan,
+static bool axi_dmac_transfer_done(struct axi_dmac_chan *chan,
 	unsigned int completed_transfers)
 {
 	struct axi_dmac_desc *active;
 	struct axi_dmac_sg *sg;
+	bool start_next = false;
 
 	active = axi_dmac_active_desc(chan);
 	if (!active)
-		return;
+		return false;
 
-	if (active->cyclic) {
-		vchan_cyclic_callback(&active->vdesc);
-	} else {
-		do {
-			sg = &active->sg[active->num_completed];
-			if (!(BIT(sg->id) & completed_transfers))
-				break;
-			active->num_completed++;
-			if (active->num_completed == active->num_sgs) {
+	do {
+		sg = &active->sg[active->num_completed];
+		if (sg->id == AXI_DMAC_SG_UNUSED) /* Not yet submitted */
+			break;
+		if (!(BIT(sg->id) & completed_transfers))
+			break;
+		active->num_completed++;
+		sg->id = AXI_DMAC_SG_UNUSED;
+		if (sg->schedule_when_free) {
+			sg->schedule_when_free = false;
+			start_next = true;
+		}
+
+		if (active->cyclic)
+			vchan_cyclic_callback(&active->vdesc);
+
+		if (active->num_completed == active->num_sgs) {
+			if (active->cyclic) {
+				active->num_completed = 0; /* wrap around */
+			} else {
 				list_del(&active->vdesc.node);
 				vchan_cookie_complete(&active->vdesc);
 				active = axi_dmac_active_desc(chan);
 			}
-		} while (active);
-	}
+		}
+	} while (active);
+
+	return start_next;
 }
 
 static irqreturn_t axi_dmac_interrupt_handler(int irq, void *devid)
 {
 	struct axi_dmac *dmac = devid;
 	unsigned int pending;
+	bool start_next = false;
 
 	pending = axi_dmac_read(dmac, AXI_DMAC_REG_IRQ_PENDING);
 	if (!pending)
@@ -281,10 +312,10 @@ static irqreturn_t axi_dmac_interrupt_handler(int irq, void *devid)
 		unsigned int completed;
 
 		completed = axi_dmac_read(dmac, AXI_DMAC_REG_TRANSFER_DONE);
-		axi_dmac_transfer_done(&dmac->chan, completed);
+		start_next = axi_dmac_transfer_done(&dmac->chan, completed);
 	}
 	/* Space has become available in the descriptor queue */
-	if (pending & AXI_DMAC_IRQ_SOT)
+	if ((pending & AXI_DMAC_IRQ_SOT) || start_next)
 		axi_dmac_start_transfer(&dmac->chan);
 	spin_unlock(&dmac->chan.vchan.lock);
 
@@ -334,12 +365,16 @@ static void axi_dmac_issue_pending(struct dma_chan *c)
 static struct axi_dmac_desc *axi_dmac_alloc_desc(unsigned int num_sgs)
 {
 	struct axi_dmac_desc *desc;
+	unsigned int i;
 
 	desc = kzalloc(sizeof(struct axi_dmac_desc) +
 		sizeof(struct axi_dmac_sg) * num_sgs, GFP_NOWAIT);
 	if (!desc)
 		return NULL;
 
+	for (i = 0; i < num_sgs; i++)
+		desc->sg[i].id = AXI_DMAC_SG_UNUSED;
+
 	desc->num_sgs = num_sgs;
 
 	return desc;
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index 34ff53290b03..47edc7fbf91f 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -702,6 +702,7 @@ static int dmatest_func(void *data)
 			 * free it this time?" dancing.  For now, just
 			 * leave it dangling.
 			 */
+			WARN(1, "dmatest: Kernel stack may be corrupted!!\n");
 			dmaengine_unmap_put(um);
 			result("test timed out", total_tests, src_off, dst_off,
 			       len, 0);
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index a7ea20e7b8e9..9364a3ed345a 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -891,6 +891,10 @@ static int edma_slave_config(struct dma_chan *chan,
 	    cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
 		return -EINVAL;
 
+	if (cfg->src_maxburst > chan->device->max_burst ||
+	    cfg->dst_maxburst > chan->device->max_burst)
+		return -EINVAL;
+
 	memcpy(&echan->cfg, cfg, sizeof(echan->cfg));
 
 	return 0;
@@ -1868,6 +1872,7 @@ static void edma_dma_init(struct edma_cc *ecc, bool legacy_mode)
 	s_ddev->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
 	s_ddev->directions |= (BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV));
 	s_ddev->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
+	s_ddev->max_burst = SZ_32K - 1; /* CIDX: 16bit signed */
 
 	s_ddev->dev = ecc->dev;
 	INIT_LIST_HEAD(&s_ddev->channels);
diff --git a/drivers/dma/img-mdc-dma.c b/drivers/dma/img-mdc-dma.c
index 54db1411ce73..0391f930aecc 100644
--- a/drivers/dma/img-mdc-dma.c
+++ b/drivers/dma/img-mdc-dma.c
@@ -23,6 +23,7 @@
 #include <linux/of_device.h>
 #include <linux/of_dma.h>
 #include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 #include <linux/regmap.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
@@ -730,14 +731,23 @@ static int mdc_slave_config(struct dma_chan *chan,
 	return 0;
 }
 
+static int mdc_alloc_chan_resources(struct dma_chan *chan)
+{
+	struct mdc_chan *mchan = to_mdc_chan(chan);
+	struct device *dev = mdma2dev(mchan->mdma);
+
+	return pm_runtime_get_sync(dev);
+}
+
 static void mdc_free_chan_resources(struct dma_chan *chan)
 {
 	struct mdc_chan *mchan = to_mdc_chan(chan);
 	struct mdc_dma *mdma = mchan->mdma;
+	struct device *dev = mdma2dev(mdma);
 
 	mdc_terminate_all(chan);
-
 	mdma->soc->disable_chan(mchan);
+	pm_runtime_put(dev);
 }
 
 static irqreturn_t mdc_chan_irq(int irq, void *dev_id)
@@ -854,6 +864,22 @@ static const struct of_device_id mdc_dma_of_match[] = {
 };
 MODULE_DEVICE_TABLE(of, mdc_dma_of_match);
 
+static int img_mdc_runtime_suspend(struct device *dev)
+{
+	struct mdc_dma *mdma = dev_get_drvdata(dev);
+
+	clk_disable_unprepare(mdma->clk);
+
+	return 0;
+}
+
+static int img_mdc_runtime_resume(struct device *dev)
+{
+	struct mdc_dma *mdma = dev_get_drvdata(dev);
+
+	return clk_prepare_enable(mdma->clk);
+}
+
 static int mdc_dma_probe(struct platform_device *pdev)
 {
 	struct mdc_dma *mdma;
@@ -883,10 +909,6 @@ static int mdc_dma_probe(struct platform_device *pdev)
 	if (IS_ERR(mdma->clk))
 		return PTR_ERR(mdma->clk);
 
-	ret = clk_prepare_enable(mdma->clk);
-	if (ret)
-		return ret;
-
 	dma_cap_zero(mdma->dma_dev.cap_mask);
 	dma_cap_set(DMA_SLAVE, mdma->dma_dev.cap_mask);
 	dma_cap_set(DMA_PRIVATE, mdma->dma_dev.cap_mask);
@@ -919,12 +941,13 @@ static int mdc_dma_probe(struct platform_device *pdev)
 				   "img,max-burst-multiplier",
 				   &mdma->max_burst_mult);
 	if (ret)
-		goto disable_clk;
+		return ret;
 
 	mdma->dma_dev.dev = &pdev->dev;
 	mdma->dma_dev.device_prep_slave_sg = mdc_prep_slave_sg;
 	mdma->dma_dev.device_prep_dma_cyclic = mdc_prep_dma_cyclic;
 	mdma->dma_dev.device_prep_dma_memcpy = mdc_prep_dma_memcpy;
+	mdma->dma_dev.device_alloc_chan_resources = mdc_alloc_chan_resources;
 	mdma->dma_dev.device_free_chan_resources = mdc_free_chan_resources;
 	mdma->dma_dev.device_tx_status = mdc_tx_status;
 	mdma->dma_dev.device_issue_pending = mdc_issue_pending;
@@ -945,15 +968,14 @@ static int mdc_dma_probe(struct platform_device *pdev)
 		mchan->mdma = mdma;
 		mchan->chan_nr = i;
 		mchan->irq = platform_get_irq(pdev, i);
-		if (mchan->irq < 0) {
-			ret = mchan->irq;
-			goto disable_clk;
-		}
+		if (mchan->irq < 0)
+			return mchan->irq;
+
 		ret = devm_request_irq(&pdev->dev, mchan->irq, mdc_chan_irq,
 				       IRQ_TYPE_LEVEL_HIGH,
 				       dev_name(&pdev->dev), mchan);
 		if (ret < 0)
-			goto disable_clk;
+			return ret;
 
 		mchan->vc.desc_free = mdc_desc_free;
 		vchan_init(&mchan->vc, &mdma->dma_dev);
@@ -962,14 +984,19 @@ static int mdc_dma_probe(struct platform_device *pdev)
 	mdma->desc_pool = dmam_pool_create(dev_name(&pdev->dev), &pdev->dev,
 					   sizeof(struct mdc_hw_list_desc),
 					   4, 0);
-	if (!mdma->desc_pool) {
-		ret = -ENOMEM;
-		goto disable_clk;
+	if (!mdma->desc_pool)
+		return -ENOMEM;
+
+	pm_runtime_enable(&pdev->dev);
+	if (!pm_runtime_enabled(&pdev->dev)) {
+		ret = img_mdc_runtime_resume(&pdev->dev);
+		if (ret)
+			return ret;
 	}
 
 	ret = dma_async_device_register(&mdma->dma_dev);
 	if (ret)
-		goto disable_clk;
+		goto suspend;
 
 	ret = of_dma_controller_register(pdev->dev.of_node, mdc_of_xlate, mdma);
 	if (ret)
@@ -982,8 +1009,10 @@ static int mdc_dma_probe(struct platform_device *pdev)
 
 unregister:
 	dma_async_device_unregister(&mdma->dma_dev);
-disable_clk:
-	clk_disable_unprepare(mdma->clk);
+suspend:
+	if (!pm_runtime_enabled(&pdev->dev))
+		img_mdc_runtime_suspend(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
 	return ret;
 }
 
@@ -1004,14 +1033,47 @@ static int mdc_dma_remove(struct platform_device *pdev)
 		tasklet_kill(&mchan->vc.task);
 	}
 
-	clk_disable_unprepare(mdma->clk);
+	pm_runtime_disable(&pdev->dev);
+	if (!pm_runtime_status_suspended(&pdev->dev))
+		img_mdc_runtime_suspend(&pdev->dev);
 
 	return 0;
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int img_mdc_suspend_late(struct device *dev)
+{
+	struct mdc_dma *mdma = dev_get_drvdata(dev);
+	int i;
+
+	/* Check that all channels are idle */
+	for (i = 0; i < mdma->nr_channels; i++) {
+		struct mdc_chan *mchan = &mdma->channels[i];
+
+		if (unlikely(mchan->desc))
+			return -EBUSY;
+	}
+
+	return pm_runtime_force_suspend(dev);
+}
+
+static int img_mdc_resume_early(struct device *dev)
+{
+	return pm_runtime_force_resume(dev);
+}
+#endif /* CONFIG_PM_SLEEP */
+
+static const struct dev_pm_ops img_mdc_pm_ops = {
+	SET_RUNTIME_PM_OPS(img_mdc_runtime_suspend,
+			   img_mdc_runtime_resume, NULL)
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(img_mdc_suspend_late,
+				     img_mdc_resume_early)
+};
+
 static struct platform_driver mdc_dma_driver = {
 	.driver = {
 		.name = "img-mdc-dma",
+		.pm = &img_mdc_pm_ops,
 		.of_match_table = of_match_ptr(mdc_dma_of_match),
 	},
 	.probe = mdc_dma_probe,
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index f681df8f0ed3..331f863c605e 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -364,9 +364,9 @@ static void imxdma_disable_hw(struct imxdma_channel *imxdmac)
 	local_irq_restore(flags);
 }
 
-static void imxdma_watchdog(unsigned long data)
+static void imxdma_watchdog(struct timer_list *t)
 {
-	struct imxdma_channel *imxdmac = (struct imxdma_channel *)data;
+	struct imxdma_channel *imxdmac = from_timer(imxdmac, t, watchdog);
 	struct imxdma_engine *imxdma = imxdmac->imxdma;
 	int channel = imxdmac->channel;
 
@@ -1153,9 +1153,7 @@ static int __init imxdma_probe(struct platform_device *pdev)
 			}
 
 			imxdmac->irq = irq + i;
-			init_timer(&imxdmac->watchdog);
-			imxdmac->watchdog.function = &imxdma_watchdog;
-			imxdmac->watchdog.data = (unsigned long)imxdmac;
+			timer_setup(&imxdmac->watchdog, imxdma_watchdog, 0);
 		}
 
 		imxdmac->imxdma = imxdma;
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index a67ec1bdc4e0..2184881afe76 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -178,6 +178,14 @@
 #define SDMA_WATERMARK_LEVEL_HWE	BIT(29)
 #define SDMA_WATERMARK_LEVEL_CONT	BIT(31)
 
+#define SDMA_DMA_BUSWIDTHS	(BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \
+				 BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
+				 BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
+
+#define SDMA_DMA_DIRECTIONS	(BIT(DMA_DEV_TO_MEM) | \
+				 BIT(DMA_MEM_TO_DEV) | \
+				 BIT(DMA_DEV_TO_DEV))
+
 /*
  * Mode/Count of data node descriptors - IPCv2
  */
@@ -1851,9 +1859,9 @@ static int sdma_probe(struct platform_device *pdev)
 	sdma->dma_device.device_prep_dma_cyclic = sdma_prep_dma_cyclic;
 	sdma->dma_device.device_config = sdma_config;
 	sdma->dma_device.device_terminate_all = sdma_disable_channel_with_delay;
-	sdma->dma_device.src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
-	sdma->dma_device.dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
-	sdma->dma_device.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+	sdma->dma_device.src_addr_widths = SDMA_DMA_BUSWIDTHS;
+	sdma->dma_device.dst_addr_widths = SDMA_DMA_BUSWIDTHS;
+	sdma->dma_device.directions = SDMA_DMA_DIRECTIONS;
 	sdma->dma_device.residue_granularity = DMA_RESIDUE_GRANULARITY_SEGMENT;
 	sdma->dma_device.device_issue_pending = sdma_issue_pending;
 	sdma->dma_device.dev->dma_parms = &sdma->dma_parms;
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index f70cc74032ea..58d4ccd33672 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -474,7 +474,7 @@ int ioat_check_space_lock(struct ioatdma_chan *ioat_chan, int num_descs)
 	if (time_is_before_jiffies(ioat_chan->timer.expires)
 	    && timer_pending(&ioat_chan->timer)) {
 		mod_timer(&ioat_chan->timer, jiffies + COMPLETION_TIMEOUT);
-		ioat_timer_event((unsigned long)ioat_chan);
+		ioat_timer_event(&ioat_chan->timer);
 	}
 
 	return -ENOMEM;
@@ -862,9 +862,9 @@ static void check_active(struct ioatdma_chan *ioat_chan)
 		mod_timer(&ioat_chan->timer, jiffies + IDLE_TIMEOUT);
 }
 
-void ioat_timer_event(unsigned long data)
+void ioat_timer_event(struct timer_list *t)
 {
-	struct ioatdma_chan *ioat_chan = to_ioat_chan((void *)data);
+	struct ioatdma_chan *ioat_chan = from_timer(ioat_chan, t, timer);
 	dma_addr_t phys_complete;
 	u64 status;
 
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h
index 56200eefcf5e..1ab42ec2b7ff 100644
--- a/drivers/dma/ioat/dma.h
+++ b/drivers/dma/ioat/dma.h
@@ -406,10 +406,9 @@ enum dma_status
 ioat_tx_status(struct dma_chan *c, dma_cookie_t cookie,
 		struct dma_tx_state *txstate);
 void ioat_cleanup_event(unsigned long data);
-void ioat_timer_event(unsigned long data);
+void ioat_timer_event(struct timer_list *t);
 int ioat_check_space_lock(struct ioatdma_chan *ioat_chan, int num_descs);
 void ioat_issue_pending(struct dma_chan *chan);
-void ioat_timer_event(unsigned long data);
 
 /* IOAT Init functions */
 bool is_bwd_ioat(struct pci_dev *pdev);
diff --git a/drivers/dma/ioat/init.c b/drivers/dma/ioat/init.c
index 93e006c3441d..2f31d3d0caa6 100644
--- a/drivers/dma/ioat/init.c
+++ b/drivers/dma/ioat/init.c
@@ -760,7 +760,7 @@ ioat_init_channel(struct ioatdma_device *ioat_dma,
 	dma_cookie_init(&ioat_chan->dma_chan);
 	list_add_tail(&ioat_chan->dma_chan.device_node, &dma->channels);
 	ioat_dma->idx[idx] = ioat_chan;
-	setup_timer(&ioat_chan->timer, ioat_timer_event, data);
+	timer_setup(&ioat_chan->timer, ioat_timer_event, 0);
 	tasklet_init(&ioat_chan->cleanup_task, ioat_cleanup_event, data);
 }
 
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index d3f918a9ee76..50559338239b 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1286,7 +1286,6 @@ MODULE_DEVICE_TABLE(of, nbpf_match);
 static int nbpf_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	const struct of_device_id *of_id = of_match_device(nbpf_match, dev);
 	struct device_node *np = dev->of_node;
 	struct nbpf_device *nbpf;
 	struct dma_device *dma_dev;
@@ -1300,10 +1299,10 @@ static int nbpf_probe(struct platform_device *pdev)
 	BUILD_BUG_ON(sizeof(struct nbpf_desc_page) > PAGE_SIZE);
 
 	/* DT only */
-	if (!np || !of_id || !of_id->data)
+	if (!np)
 		return -ENODEV;
 
-	cfg = of_id->data;
+	cfg = of_device_get_match_data(dev);
 	num_channels = cfg->num_channels;
 
 	nbpf = devm_kzalloc(dev, sizeof(*nbpf) + num_channels *
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 8c1665c8fe33..f6dd849159d8 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -1288,6 +1288,10 @@ static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config
 	    cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
 		return -EINVAL;
 
+	if (cfg->src_maxburst > chan->device->max_burst ||
+	    cfg->dst_maxburst > chan->device->max_burst)
+		return -EINVAL;
+
 	memcpy(&c->cfg, cfg, sizeof(c->cfg));
 
 	return 0;
@@ -1482,6 +1486,7 @@ static int omap_dma_probe(struct platform_device *pdev)
 	od->ddev.dst_addr_widths = OMAP_DMA_BUSWIDTHS;
 	od->ddev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
 	od->ddev.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
+	od->ddev.max_burst = SZ_16M - 1; /* CCEN: 24bit unsigned */
 	od->ddev.dev = &pdev->dev;
 	INIT_LIST_HEAD(&od->ddev.channels);
 	spin_lock_init(&od->lock);
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index f9028e9d0dfc..afd8f27bda96 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -123,7 +123,7 @@ struct pch_dma_chan {
 struct pch_dma {
 	struct dma_device	dma;
 	void __iomem *membase;
-	struct pci_pool		*pool;
+	struct dma_pool		*pool;
 	struct pch_dma_regs	regs;
 	struct pch_dma_desc_regs ch_regs[MAX_CHAN_NR];
 	struct pch_dma_chan	channels[MAX_CHAN_NR];
@@ -437,7 +437,7 @@ static struct pch_dma_desc *pdc_alloc_desc(struct dma_chan *chan, gfp_t flags)
 	struct pch_dma *pd = to_pd(chan->device);
 	dma_addr_t addr;
 
-	desc = pci_pool_zalloc(pd->pool, flags, &addr);
+	desc = dma_pool_zalloc(pd->pool, flags, &addr);
 	if (desc) {
 		INIT_LIST_HEAD(&desc->tx_list);
 		dma_async_tx_descriptor_init(&desc->txd, chan);
@@ -549,7 +549,7 @@ static void pd_free_chan_resources(struct dma_chan *chan)
 	spin_unlock_irq(&pd_chan->lock);
 
 	list_for_each_entry_safe(desc, _d, &tmp_list, desc_node)
-		pci_pool_free(pd->pool, desc, desc->txd.phys);
+		dma_pool_free(pd->pool, desc, desc->txd.phys);
 
 	pdc_enable_irq(chan, 0);
 }
@@ -880,7 +880,7 @@ static int pch_dma_probe(struct pci_dev *pdev,
 		goto err_iounmap;
 	}
 
-	pd->pool = pci_pool_create("pch_dma_desc_pool", pdev,
+	pd->pool = dma_pool_create("pch_dma_desc_pool", &pdev->dev,
 				   sizeof(struct pch_dma_desc), 4, 0);
 	if (!pd->pool) {
 		dev_err(&pdev->dev, "Failed to alloc DMA descriptors\n");
@@ -931,7 +931,7 @@ static int pch_dma_probe(struct pci_dev *pdev,
 	return 0;
 
 err_free_pool:
-	pci_pool_destroy(pd->pool);
+	dma_pool_destroy(pd->pool);
 err_free_irq:
 	free_irq(pdev->irq, pd);
 err_iounmap:
@@ -963,7 +963,7 @@ static void pch_dma_remove(struct pci_dev *pdev)
 			tasklet_kill(&pd_chan->tasklet);
 		}
 
-		pci_pool_destroy(pd->pool);
+		dma_pool_destroy(pd->pool);
 		pci_iounmap(pdev, pd->membase);
 		pci_release_regions(pdev);
 		pci_disable_device(pdev);
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index f122c2a7b9f0..d7327fd5f445 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2390,7 +2390,8 @@ static inline void _init_desc(struct dma_pl330_desc *desc)
 }
 
 /* Returns the number of descriptors added to the DMAC pool */
-static int add_desc(struct pl330_dmac *pl330, gfp_t flg, int count)
+static int add_desc(struct list_head *pool, spinlock_t *lock,
+		    gfp_t flg, int count)
 {
 	struct dma_pl330_desc *desc;
 	unsigned long flags;
@@ -2400,27 +2401,28 @@ static int add_desc(struct pl330_dmac *pl330, gfp_t flg, int count)
 	if (!desc)
 		return 0;
 
-	spin_lock_irqsave(&pl330->pool_lock, flags);
+	spin_lock_irqsave(lock, flags);
 
 	for (i = 0; i < count; i++) {
 		_init_desc(&desc[i]);
-		list_add_tail(&desc[i].node, &pl330->desc_pool);
+		list_add_tail(&desc[i].node, pool);
 	}
 
-	spin_unlock_irqrestore(&pl330->pool_lock, flags);
+	spin_unlock_irqrestore(lock, flags);
 
 	return count;
 }
 
-static struct dma_pl330_desc *pluck_desc(struct pl330_dmac *pl330)
+static struct dma_pl330_desc *pluck_desc(struct list_head *pool,
+					 spinlock_t *lock)
 {
 	struct dma_pl330_desc *desc = NULL;
 	unsigned long flags;
 
-	spin_lock_irqsave(&pl330->pool_lock, flags);
+	spin_lock_irqsave(lock, flags);
 
-	if (!list_empty(&pl330->desc_pool)) {
-		desc = list_entry(pl330->desc_pool.next,
+	if (!list_empty(pool)) {
+		desc = list_entry(pool->next,
 				struct dma_pl330_desc, node);
 
 		list_del_init(&desc->node);
@@ -2429,7 +2431,7 @@ static struct dma_pl330_desc *pluck_desc(struct pl330_dmac *pl330)
 		desc->txd.callback = NULL;
 	}
 
-	spin_unlock_irqrestore(&pl330->pool_lock, flags);
+	spin_unlock_irqrestore(lock, flags);
 
 	return desc;
 }
@@ -2441,20 +2443,18 @@ static struct dma_pl330_desc *pl330_get_desc(struct dma_pl330_chan *pch)
 	struct dma_pl330_desc *desc;
 
 	/* Pluck one desc from the pool of DMAC */
-	desc = pluck_desc(pl330);
+	desc = pluck_desc(&pl330->desc_pool, &pl330->pool_lock);
 
 	/* If the DMAC pool is empty, alloc new */
 	if (!desc) {
-		if (!add_desc(pl330, GFP_ATOMIC, 1))
-			return NULL;
+		DEFINE_SPINLOCK(lock);
+		LIST_HEAD(pool);
 
-		/* Try again */
-		desc = pluck_desc(pl330);
-		if (!desc) {
-			dev_err(pch->dmac->ddma.dev,
-				"%s:%d ALERT!\n", __func__, __LINE__);
+		if (!add_desc(&pool, &lock, GFP_ATOMIC, 1))
 			return NULL;
-		}
+
+		desc = pluck_desc(&pool, &lock);
+		WARN_ON(!desc || !list_empty(&pool));
 	}
 
 	/* Initialize the descriptor */
@@ -2868,7 +2868,8 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 	spin_lock_init(&pl330->pool_lock);
 
 	/* Create a descriptor pool of default size */
-	if (!add_desc(pl330, GFP_KERNEL, NR_DEFAULT_DESC))
+	if (!add_desc(&pl330->desc_pool, &pl330->pool_lock,
+		      GFP_KERNEL, NR_DEFAULT_DESC))
 		dev_warn(&adev->dev, "unable to allocate desc\n");
 
 	INIT_LIST_HEAD(&pd->channels);
diff --git a/drivers/dma/qcom/bam_dma.c b/drivers/dma/qcom/bam_dma.c
index 6d89fb6a6a92..d076940e0c69 100644
--- a/drivers/dma/qcom/bam_dma.c
+++ b/drivers/dma/qcom/bam_dma.c
@@ -46,6 +46,7 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/of_dma.h>
+#include <linux/circ_buf.h>
 #include <linux/clk.h>
 #include <linux/dmaengine.h>
 #include <linux/pm_runtime.h>
@@ -78,6 +79,8 @@ struct bam_async_desc {
 
 	struct bam_desc_hw *curr_desc;
 
+	/* list node for the desc in the bam_chan list of descriptors */
+	struct list_head desc_node;
 	enum dma_transfer_direction dir;
 	size_t length;
 	struct bam_desc_hw desc[0];
@@ -347,6 +350,8 @@ static const struct reg_offset_data bam_v1_7_reg_info[] = {
 #define BAM_DESC_FIFO_SIZE	SZ_32K
 #define MAX_DESCRIPTORS (BAM_DESC_FIFO_SIZE / sizeof(struct bam_desc_hw) - 1)
 #define BAM_FIFO_SIZE	(SZ_32K - 8)
+#define IS_BUSY(chan)	(CIRC_SPACE(bchan->tail, bchan->head,\
+			 MAX_DESCRIPTORS + 1) == 0)
 
 struct bam_chan {
 	struct virt_dma_chan vc;
@@ -356,8 +361,6 @@ struct bam_chan {
 	/* configuration from device tree */
 	u32 id;
 
-	struct bam_async_desc *curr_txd;	/* current running dma */
-
 	/* runtime configuration */
 	struct dma_slave_config slave;
 
@@ -372,6 +375,8 @@ struct bam_chan {
 	unsigned int initialized;	/* is the channel hw initialized? */
 	unsigned int paused;		/* is the channel paused? */
 	unsigned int reconfigure;	/* new slave config? */
+	/* list of descriptors currently processed */
+	struct list_head desc_list;
 
 	struct list_head node;
 };
@@ -539,7 +544,7 @@ static void bam_free_chan(struct dma_chan *chan)
 
 	vchan_free_chan_resources(to_virt_chan(chan));
 
-	if (bchan->curr_txd) {
+	if (!list_empty(&bchan->desc_list)) {
 		dev_err(bchan->bdev->dev, "Cannot free busy channel\n");
 		goto err;
 	}
@@ -632,8 +637,6 @@ static struct dma_async_tx_descriptor *bam_prep_slave_sg(struct dma_chan *chan,
 
 	if (flags & DMA_PREP_INTERRUPT)
 		async_desc->flags |= DESC_FLAG_EOT;
-	else
-		async_desc->flags |= DESC_FLAG_INT;
 
 	async_desc->num_desc = num_alloc;
 	async_desc->curr_desc = async_desc->desc;
@@ -684,14 +687,16 @@ err_out:
 static int bam_dma_terminate_all(struct dma_chan *chan)
 {
 	struct bam_chan *bchan = to_bam_chan(chan);
+	struct bam_async_desc *async_desc, *tmp;
 	unsigned long flag;
 	LIST_HEAD(head);
 
 	/* remove all transactions, including active transaction */
 	spin_lock_irqsave(&bchan->vc.lock, flag);
-	if (bchan->curr_txd) {
-		list_add(&bchan->curr_txd->vd.node, &bchan->vc.desc_issued);
-		bchan->curr_txd = NULL;
+	list_for_each_entry_safe(async_desc, tmp,
+				 &bchan->desc_list, desc_node) {
+		list_add(&async_desc->vd.node, &bchan->vc.desc_issued);
+		list_del(&async_desc->desc_node);
 	}
 
 	vchan_get_all_descriptors(&bchan->vc, &head);
@@ -763,9 +768,9 @@ static int bam_resume(struct dma_chan *chan)
  */
 static u32 process_channel_irqs(struct bam_device *bdev)
 {
-	u32 i, srcs, pipe_stts;
+	u32 i, srcs, pipe_stts, offset, avail;
 	unsigned long flags;
-	struct bam_async_desc *async_desc;
+	struct bam_async_desc *async_desc, *tmp;
 
 	srcs = readl_relaxed(bam_addr(bdev, 0, BAM_IRQ_SRCS_EE));
 
@@ -785,27 +790,40 @@ static u32 process_channel_irqs(struct bam_device *bdev)
 		writel_relaxed(pipe_stts, bam_addr(bdev, i, BAM_P_IRQ_CLR));
 
 		spin_lock_irqsave(&bchan->vc.lock, flags);
-		async_desc = bchan->curr_txd;
 
-		if (async_desc) {
-			async_desc->num_desc -= async_desc->xfer_len;
-			async_desc->curr_desc += async_desc->xfer_len;
-			bchan->curr_txd = NULL;
+		offset = readl_relaxed(bam_addr(bdev, i, BAM_P_SW_OFSTS)) &
+				       P_SW_OFSTS_MASK;
+		offset /= sizeof(struct bam_desc_hw);
+
+		/* Number of bytes available to read */
+		avail = CIRC_CNT(offset, bchan->head, MAX_DESCRIPTORS + 1);
+
+		list_for_each_entry_safe(async_desc, tmp,
+					 &bchan->desc_list, desc_node) {
+			/* Not enough data to read */
+			if (avail < async_desc->xfer_len)
+				break;
 
 			/* manage FIFO */
 			bchan->head += async_desc->xfer_len;
 			bchan->head %= MAX_DESCRIPTORS;
 
+			async_desc->num_desc -= async_desc->xfer_len;
+			async_desc->curr_desc += async_desc->xfer_len;
+			avail -= async_desc->xfer_len;
+
 			/*
-			 * if complete, process cookie.  Otherwise
+			 * if complete, process cookie. Otherwise
 			 * push back to front of desc_issued so that
 			 * it gets restarted by the tasklet
 			 */
-			if (!async_desc->num_desc)
+			if (!async_desc->num_desc) {
 				vchan_cookie_complete(&async_desc->vd);
-			else
+			} else {
 				list_add(&async_desc->vd.node,
-					&bchan->vc.desc_issued);
+					 &bchan->vc.desc_issued);
+			}
+			list_del(&async_desc->desc_node);
 		}
 
 		spin_unlock_irqrestore(&bchan->vc.lock, flags);
@@ -867,6 +885,7 @@ static enum dma_status bam_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 		struct dma_tx_state *txstate)
 {
 	struct bam_chan *bchan = to_bam_chan(chan);
+	struct bam_async_desc *async_desc;
 	struct virt_dma_desc *vd;
 	int ret;
 	size_t residue = 0;
@@ -882,11 +901,17 @@ static enum dma_status bam_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 
 	spin_lock_irqsave(&bchan->vc.lock, flags);
 	vd = vchan_find_desc(&bchan->vc, cookie);
-	if (vd)
+	if (vd) {
 		residue = container_of(vd, struct bam_async_desc, vd)->length;
-	else if (bchan->curr_txd && bchan->curr_txd->vd.tx.cookie == cookie)
-		for (i = 0; i < bchan->curr_txd->num_desc; i++)
-			residue += bchan->curr_txd->curr_desc[i].size;
+	} else {
+		list_for_each_entry(async_desc, &bchan->desc_list, desc_node) {
+			if (async_desc->vd.tx.cookie != cookie)
+				continue;
+
+			for (i = 0; i < async_desc->num_desc; i++)
+				residue += async_desc->curr_desc[i].size;
+		}
+	}
 
 	spin_unlock_irqrestore(&bchan->vc.lock, flags);
 
@@ -927,63 +952,86 @@ static void bam_start_dma(struct bam_chan *bchan)
 {
 	struct virt_dma_desc *vd = vchan_next_desc(&bchan->vc);
 	struct bam_device *bdev = bchan->bdev;
-	struct bam_async_desc *async_desc;
+	struct bam_async_desc *async_desc = NULL;
 	struct bam_desc_hw *desc;
 	struct bam_desc_hw *fifo = PTR_ALIGN(bchan->fifo_virt,
 					sizeof(struct bam_desc_hw));
 	int ret;
+	unsigned int avail;
+	struct dmaengine_desc_callback cb;
 
 	lockdep_assert_held(&bchan->vc.lock);
 
 	if (!vd)
 		return;
 
-	list_del(&vd->node);
-
-	async_desc = container_of(vd, struct bam_async_desc, vd);
-	bchan->curr_txd = async_desc;
-
 	ret = pm_runtime_get_sync(bdev->dev);
 	if (ret < 0)
 		return;
 
-	/* on first use, initialize the channel hardware */
-	if (!bchan->initialized)
-		bam_chan_init_hw(bchan, async_desc->dir);
+	while (vd && !IS_BUSY(bchan)) {
+		list_del(&vd->node);
 
-	/* apply new slave config changes, if necessary */
-	if (bchan->reconfigure)
-		bam_apply_new_config(bchan, async_desc->dir);
+		async_desc = container_of(vd, struct bam_async_desc, vd);
 
-	desc = bchan->curr_txd->curr_desc;
+		/* on first use, initialize the channel hardware */
+		if (!bchan->initialized)
+			bam_chan_init_hw(bchan, async_desc->dir);
 
-	if (async_desc->num_desc > MAX_DESCRIPTORS)
-		async_desc->xfer_len = MAX_DESCRIPTORS;
-	else
-		async_desc->xfer_len = async_desc->num_desc;
+		/* apply new slave config changes, if necessary */
+		if (bchan->reconfigure)
+			bam_apply_new_config(bchan, async_desc->dir);
 
-	/* set any special flags on the last descriptor */
-	if (async_desc->num_desc == async_desc->xfer_len)
-		desc[async_desc->xfer_len - 1].flags |=
-					cpu_to_le16(async_desc->flags);
-	else
-		desc[async_desc->xfer_len - 1].flags |=
-					cpu_to_le16(DESC_FLAG_INT);
+		desc = async_desc->curr_desc;
+		avail = CIRC_SPACE(bchan->tail, bchan->head,
+				   MAX_DESCRIPTORS + 1);
+
+		if (async_desc->num_desc > avail)
+			async_desc->xfer_len = avail;
+		else
+			async_desc->xfer_len = async_desc->num_desc;
+
+		/* set any special flags on the last descriptor */
+		if (async_desc->num_desc == async_desc->xfer_len)
+			desc[async_desc->xfer_len - 1].flags |=
+						cpu_to_le16(async_desc->flags);
 
-	if (bchan->tail + async_desc->xfer_len > MAX_DESCRIPTORS) {
-		u32 partial = MAX_DESCRIPTORS - bchan->tail;
+		vd = vchan_next_desc(&bchan->vc);
 
-		memcpy(&fifo[bchan->tail], desc,
-				partial * sizeof(struct bam_desc_hw));
-		memcpy(fifo, &desc[partial], (async_desc->xfer_len - partial) *
+		dmaengine_desc_get_callback(&async_desc->vd.tx, &cb);
+
+		/*
+		 * An interrupt is generated at this desc, if
+		 *  - FIFO is FULL.
+		 *  - No more descriptors to add.
+		 *  - If a callback completion was requested for this DESC,
+		 *     In this case, BAM will deliver the completion callback
+		 *     for this desc and continue processing the next desc.
+		 */
+		if (((avail <= async_desc->xfer_len) || !vd ||
+		     dmaengine_desc_callback_valid(&cb)) &&
+		    !(async_desc->flags & DESC_FLAG_EOT))
+			desc[async_desc->xfer_len - 1].flags |=
+				cpu_to_le16(DESC_FLAG_INT);
+
+		if (bchan->tail + async_desc->xfer_len > MAX_DESCRIPTORS) {
+			u32 partial = MAX_DESCRIPTORS - bchan->tail;
+
+			memcpy(&fifo[bchan->tail], desc,
+			       partial * sizeof(struct bam_desc_hw));
+			memcpy(fifo, &desc[partial],
+			       (async_desc->xfer_len - partial) *
 				sizeof(struct bam_desc_hw));
-	} else {
-		memcpy(&fifo[bchan->tail], desc,
-			async_desc->xfer_len * sizeof(struct bam_desc_hw));
-	}
+		} else {
+			memcpy(&fifo[bchan->tail], desc,
+			       async_desc->xfer_len *
+			       sizeof(struct bam_desc_hw));
+		}
 
-	bchan->tail += async_desc->xfer_len;
-	bchan->tail %= MAX_DESCRIPTORS;
+		bchan->tail += async_desc->xfer_len;
+		bchan->tail %= MAX_DESCRIPTORS;
+		list_add_tail(&async_desc->desc_node, &bchan->desc_list);
+	}
 
 	/* ensure descriptor writes and dma start not reordered */
 	wmb();
@@ -1012,7 +1060,7 @@ static void dma_tasklet(unsigned long data)
 		bchan = &bdev->channels[i];
 		spin_lock_irqsave(&bchan->vc.lock, flags);
 
-		if (!list_empty(&bchan->vc.desc_issued) && !bchan->curr_txd)
+		if (!list_empty(&bchan->vc.desc_issued) && !IS_BUSY(bchan))
 			bam_start_dma(bchan);
 		spin_unlock_irqrestore(&bchan->vc.lock, flags);
 	}
@@ -1033,7 +1081,7 @@ static void bam_issue_pending(struct dma_chan *chan)
 	spin_lock_irqsave(&bchan->vc.lock, flags);
 
 	/* if work pending and idle, start a transaction */
-	if (vchan_issue_pending(&bchan->vc) && !bchan->curr_txd)
+	if (vchan_issue_pending(&bchan->vc) && !IS_BUSY(bchan))
 		bam_start_dma(bchan);
 
 	spin_unlock_irqrestore(&bchan->vc.lock, flags);
@@ -1133,6 +1181,7 @@ static void bam_channel_init(struct bam_device *bdev, struct bam_chan *bchan,
 
 	vchan_init(&bchan->vc, &bdev->common);
 	bchan->vc.desc_free = bam_dma_free_desc;
+	INIT_LIST_HEAD(&bchan->desc_list);
 }
 
 static const struct of_device_id bam_of_match[] = {
diff --git a/drivers/dma/sa11x0-dma.c b/drivers/dma/sa11x0-dma.c
index 1adeb3265085..c7a89c22890e 100644
--- a/drivers/dma/sa11x0-dma.c
+++ b/drivers/dma/sa11x0-dma.c
@@ -823,6 +823,13 @@ static const struct sa11x0_dma_channel_desc chan_desc[] = {
 	CD(Ser4SSPRc, DDAR_RW),
 };
 
+static const struct dma_slave_map sa11x0_dma_map[] = {
+	{ "sa11x0-ir", "tx", "Ser2ICPTr" },
+	{ "sa11x0-ir", "rx", "Ser2ICPRc" },
+	{ "sa11x0-ssp", "tx", "Ser4SSPTr" },
+	{ "sa11x0-ssp", "rx", "Ser4SSPRc" },
+};
+
 static int sa11x0_dma_init_dmadev(struct dma_device *dmadev,
 	struct device *dev)
 {
@@ -909,6 +916,10 @@ static int sa11x0_dma_probe(struct platform_device *pdev)
 	spin_lock_init(&d->lock);
 	INIT_LIST_HEAD(&d->chan_pending);
 
+	d->slave.filter.fn = sa11x0_dma_filter_fn;
+	d->slave.filter.mapcnt = ARRAY_SIZE(sa11x0_dma_map);
+	d->slave.filter.map = sa11x0_dma_map;
+
 	d->base = ioremap(res->start, resource_size(res));
 	if (!d->base) {
 		ret = -ENOMEM;
diff --git a/drivers/dma/sprd-dma.c b/drivers/dma/sprd-dma.c
new file mode 100644
index 000000000000..b652071a2096
--- /dev/null
+++ b/drivers/dma/sprd-dma.c
@@ -0,0 +1,988 @@
+/*
+ * Copyright (C) 2017 Spreadtrum Communications Inc.
+ *
+ * SPDX-License-Identifier: GPL-2.0
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_dma.h>
+#include <linux/of_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/slab.h>
+
+#include "virt-dma.h"
+
+#define SPRD_DMA_CHN_REG_OFFSET		0x1000
+#define SPRD_DMA_CHN_REG_LENGTH		0x40
+#define SPRD_DMA_MEMCPY_MIN_SIZE	64
+
+/* DMA global registers definition */
+#define SPRD_DMA_GLB_PAUSE		0x0
+#define SPRD_DMA_GLB_FRAG_WAIT		0x4
+#define SPRD_DMA_GLB_REQ_PEND0_EN	0x8
+#define SPRD_DMA_GLB_REQ_PEND1_EN	0xc
+#define SPRD_DMA_GLB_INT_RAW_STS	0x10
+#define SPRD_DMA_GLB_INT_MSK_STS	0x14
+#define SPRD_DMA_GLB_REQ_STS		0x18
+#define SPRD_DMA_GLB_CHN_EN_STS		0x1c
+#define SPRD_DMA_GLB_DEBUG_STS		0x20
+#define SPRD_DMA_GLB_ARB_SEL_STS	0x24
+#define SPRD_DMA_GLB_REQ_UID(uid)	(0x4 * ((uid) - 1))
+#define SPRD_DMA_GLB_REQ_UID_OFFSET	0x2000
+
+/* DMA channel registers definition */
+#define SPRD_DMA_CHN_PAUSE		0x0
+#define SPRD_DMA_CHN_REQ		0x4
+#define SPRD_DMA_CHN_CFG		0x8
+#define SPRD_DMA_CHN_INTC		0xc
+#define SPRD_DMA_CHN_SRC_ADDR		0x10
+#define SPRD_DMA_CHN_DES_ADDR		0x14
+#define SPRD_DMA_CHN_FRG_LEN		0x18
+#define SPRD_DMA_CHN_BLK_LEN		0x1c
+#define SPRD_DMA_CHN_TRSC_LEN		0x20
+#define SPRD_DMA_CHN_TRSF_STEP		0x24
+#define SPRD_DMA_CHN_WARP_PTR		0x28
+#define SPRD_DMA_CHN_WARP_TO		0x2c
+#define SPRD_DMA_CHN_LLIST_PTR		0x30
+#define SPRD_DMA_CHN_FRAG_STEP		0x34
+#define SPRD_DMA_CHN_SRC_BLK_STEP	0x38
+#define SPRD_DMA_CHN_DES_BLK_STEP	0x3c
+
+/* SPRD_DMA_CHN_INTC register definition */
+#define SPRD_DMA_INT_MASK		GENMASK(4, 0)
+#define SPRD_DMA_INT_CLR_OFFSET		24
+#define SPRD_DMA_FRAG_INT_EN		BIT(0)
+#define SPRD_DMA_BLK_INT_EN		BIT(1)
+#define SPRD_DMA_TRANS_INT_EN		BIT(2)
+#define SPRD_DMA_LIST_INT_EN		BIT(3)
+#define SPRD_DMA_CFG_ERR_INT_EN		BIT(4)
+
+/* SPRD_DMA_CHN_CFG register definition */
+#define SPRD_DMA_CHN_EN			BIT(0)
+#define SPRD_DMA_WAIT_BDONE_OFFSET	24
+#define SPRD_DMA_DONOT_WAIT_BDONE	1
+
+/* SPRD_DMA_CHN_REQ register definition */
+#define SPRD_DMA_REQ_EN			BIT(0)
+
+/* SPRD_DMA_CHN_PAUSE register definition */
+#define SPRD_DMA_PAUSE_EN		BIT(0)
+#define SPRD_DMA_PAUSE_STS		BIT(2)
+#define SPRD_DMA_PAUSE_CNT		0x2000
+
+/* DMA_CHN_WARP_* register definition */
+#define SPRD_DMA_HIGH_ADDR_MASK		GENMASK(31, 28)
+#define SPRD_DMA_LOW_ADDR_MASK		GENMASK(31, 0)
+#define SPRD_DMA_HIGH_ADDR_OFFSET	4
+
+/* SPRD_DMA_CHN_INTC register definition */
+#define SPRD_DMA_FRAG_INT_STS		BIT(16)
+#define SPRD_DMA_BLK_INT_STS		BIT(17)
+#define SPRD_DMA_TRSC_INT_STS		BIT(18)
+#define SPRD_DMA_LIST_INT_STS		BIT(19)
+#define SPRD_DMA_CFGERR_INT_STS		BIT(20)
+#define SPRD_DMA_CHN_INT_STS					\
+	(SPRD_DMA_FRAG_INT_STS | SPRD_DMA_BLK_INT_STS |		\
+	 SPRD_DMA_TRSC_INT_STS | SPRD_DMA_LIST_INT_STS |	\
+	 SPRD_DMA_CFGERR_INT_STS)
+
+/* SPRD_DMA_CHN_FRG_LEN register definition */
+#define SPRD_DMA_SRC_DATAWIDTH_OFFSET	30
+#define SPRD_DMA_DES_DATAWIDTH_OFFSET	28
+#define SPRD_DMA_SWT_MODE_OFFSET	26
+#define SPRD_DMA_REQ_MODE_OFFSET	24
+#define SPRD_DMA_REQ_MODE_MASK		GENMASK(1, 0)
+#define SPRD_DMA_FIX_SEL_OFFSET		21
+#define SPRD_DMA_FIX_EN_OFFSET		20
+#define SPRD_DMA_LLIST_END_OFFSET	19
+#define SPRD_DMA_FRG_LEN_MASK		GENMASK(16, 0)
+
+/* SPRD_DMA_CHN_BLK_LEN register definition */
+#define SPRD_DMA_BLK_LEN_MASK		GENMASK(16, 0)
+
+/* SPRD_DMA_CHN_TRSC_LEN register definition */
+#define SPRD_DMA_TRSC_LEN_MASK		GENMASK(27, 0)
+
+/* SPRD_DMA_CHN_TRSF_STEP register definition */
+#define SPRD_DMA_DEST_TRSF_STEP_OFFSET	16
+#define SPRD_DMA_SRC_TRSF_STEP_OFFSET	0
+#define SPRD_DMA_TRSF_STEP_MASK		GENMASK(15, 0)
+
+#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 channel hardware configuration */
+struct sprd_dma_chn_hw {
+	u32 pause;
+	u32 req;
+	u32 cfg;
+	u32 intc;
+	u32 src_addr;
+	u32 des_addr;
+	u32 frg_len;
+	u32 blk_len;
+	u32 trsc_len;
+	u32 trsf_step;
+	u32 wrap_ptr;
+	u32 wrap_to;
+	u32 llist_ptr;
+	u32 frg_step;
+	u32 src_blk_step;
+	u32 des_blk_step;
+};
+
+/* dma request description */
+struct sprd_dma_desc {
+	struct virt_dma_desc	vd;
+	struct sprd_dma_chn_hw	chn_hw;
+};
+
+/* dma channel description */
+struct sprd_dma_chn {
+	struct virt_dma_chan	vc;
+	void __iomem		*chn_base;
+	u32			chn_num;
+	u32			dev_id;
+	struct sprd_dma_desc	*cur_desc;
+};
+
+/* SPRD dma device */
+struct sprd_dma_dev {
+	struct dma_device	dma_dev;
+	void __iomem		*glb_base;
+	struct clk		*clk;
+	struct clk		*ashb_clk;
+	int			irq;
+	u32			total_chns;
+	struct sprd_dma_chn	channels[0];
+};
+
+static bool sprd_dma_filter_fn(struct dma_chan *chan, void *param);
+static struct of_dma_filter_info sprd_dma_info = {
+	.filter_fn = sprd_dma_filter_fn,
+};
+
+static inline struct sprd_dma_chn *to_sprd_dma_chan(struct dma_chan *c)
+{
+	return container_of(c, struct sprd_dma_chn, vc.chan);
+}
+
+static inline struct sprd_dma_dev *to_sprd_dma_dev(struct dma_chan *c)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(c);
+
+	return container_of(schan, struct sprd_dma_dev, channels[c->chan_id]);
+}
+
+static inline struct sprd_dma_desc *to_sprd_dma_desc(struct virt_dma_desc *vd)
+{
+	return container_of(vd, struct sprd_dma_desc, vd);
+}
+
+static void sprd_dma_chn_update(struct sprd_dma_chn *schan, u32 reg,
+				u32 mask, u32 val)
+{
+	u32 orig = readl(schan->chn_base + reg);
+	u32 tmp;
+
+	tmp = (orig & ~mask) | val;
+	writel(tmp, schan->chn_base + reg);
+}
+
+static int sprd_dma_enable(struct sprd_dma_dev *sdev)
+{
+	int ret;
+
+	ret = clk_prepare_enable(sdev->clk);
+	if (ret)
+		return ret;
+
+	/*
+	 * The ashb_clk is optional and only for AGCP DMA controller, so we
+	 * need add one condition to check if the ashb_clk need enable.
+	 */
+	if (!IS_ERR(sdev->ashb_clk))
+		ret = clk_prepare_enable(sdev->ashb_clk);
+
+	return ret;
+}
+
+static void sprd_dma_disable(struct sprd_dma_dev *sdev)
+{
+	clk_disable_unprepare(sdev->clk);
+
+	/*
+	 * Need to check if we need disable the optional ashb_clk for AGCP DMA.
+	 */
+	if (!IS_ERR(sdev->ashb_clk))
+		clk_disable_unprepare(sdev->ashb_clk);
+}
+
+static void sprd_dma_set_uid(struct sprd_dma_chn *schan)
+{
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(&schan->vc.chan);
+	u32 dev_id = schan->dev_id;
+
+	if (dev_id != SPRD_DMA_SOFTWARE_UID) {
+		u32 uid_offset = SPRD_DMA_GLB_REQ_UID_OFFSET +
+				 SPRD_DMA_GLB_REQ_UID(dev_id);
+
+		writel(schan->chn_num + 1, sdev->glb_base + uid_offset);
+	}
+}
+
+static void sprd_dma_unset_uid(struct sprd_dma_chn *schan)
+{
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(&schan->vc.chan);
+	u32 dev_id = schan->dev_id;
+
+	if (dev_id != SPRD_DMA_SOFTWARE_UID) {
+		u32 uid_offset = SPRD_DMA_GLB_REQ_UID_OFFSET +
+				 SPRD_DMA_GLB_REQ_UID(dev_id);
+
+		writel(0, sdev->glb_base + uid_offset);
+	}
+}
+
+static void sprd_dma_clear_int(struct sprd_dma_chn *schan)
+{
+	sprd_dma_chn_update(schan, SPRD_DMA_CHN_INTC,
+			    SPRD_DMA_INT_MASK << SPRD_DMA_INT_CLR_OFFSET,
+			    SPRD_DMA_INT_MASK << SPRD_DMA_INT_CLR_OFFSET);
+}
+
+static void sprd_dma_enable_chn(struct sprd_dma_chn *schan)
+{
+	sprd_dma_chn_update(schan, SPRD_DMA_CHN_CFG, SPRD_DMA_CHN_EN,
+			    SPRD_DMA_CHN_EN);
+}
+
+static void sprd_dma_disable_chn(struct sprd_dma_chn *schan)
+{
+	sprd_dma_chn_update(schan, SPRD_DMA_CHN_CFG, SPRD_DMA_CHN_EN, 0);
+}
+
+static void sprd_dma_soft_request(struct sprd_dma_chn *schan)
+{
+	sprd_dma_chn_update(schan, SPRD_DMA_CHN_REQ, SPRD_DMA_REQ_EN,
+			    SPRD_DMA_REQ_EN);
+}
+
+static void sprd_dma_pause_resume(struct sprd_dma_chn *schan, bool enable)
+{
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(&schan->vc.chan);
+	u32 pause, timeout = SPRD_DMA_PAUSE_CNT;
+
+	if (enable) {
+		sprd_dma_chn_update(schan, SPRD_DMA_CHN_PAUSE,
+				    SPRD_DMA_PAUSE_EN, SPRD_DMA_PAUSE_EN);
+
+		do {
+			pause = readl(schan->chn_base + SPRD_DMA_CHN_PAUSE);
+			if (pause & SPRD_DMA_PAUSE_STS)
+				break;
+
+			cpu_relax();
+		} while (--timeout > 0);
+
+		if (!timeout)
+			dev_warn(sdev->dma_dev.dev,
+				 "pause dma controller timeout\n");
+	} else {
+		sprd_dma_chn_update(schan, SPRD_DMA_CHN_PAUSE,
+				    SPRD_DMA_PAUSE_EN, 0);
+	}
+}
+
+static void sprd_dma_stop_and_disable(struct sprd_dma_chn *schan)
+{
+	u32 cfg = readl(schan->chn_base + SPRD_DMA_CHN_CFG);
+
+	if (!(cfg & SPRD_DMA_CHN_EN))
+		return;
+
+	sprd_dma_pause_resume(schan, true);
+	sprd_dma_disable_chn(schan);
+}
+
+static unsigned long sprd_dma_get_dst_addr(struct sprd_dma_chn *schan)
+{
+	unsigned long addr, addr_high;
+
+	addr = readl(schan->chn_base + SPRD_DMA_CHN_DES_ADDR);
+	addr_high = readl(schan->chn_base + SPRD_DMA_CHN_WARP_TO) &
+		    SPRD_DMA_HIGH_ADDR_MASK;
+
+	return addr | (addr_high << SPRD_DMA_HIGH_ADDR_OFFSET);
+}
+
+static enum sprd_dma_int_type sprd_dma_get_int_type(struct sprd_dma_chn *schan)
+{
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(&schan->vc.chan);
+	u32 intc_sts = readl(schan->chn_base + SPRD_DMA_CHN_INTC) &
+		       SPRD_DMA_CHN_INT_STS;
+
+	switch (intc_sts) {
+	case SPRD_DMA_CFGERR_INT_STS:
+		return SPRD_DMA_CFGERR_INT;
+
+	case SPRD_DMA_LIST_INT_STS:
+		return SPRD_DMA_LIST_INT;
+
+	case SPRD_DMA_TRSC_INT_STS:
+		return SPRD_DMA_TRANS_INT;
+
+	case SPRD_DMA_BLK_INT_STS:
+		return SPRD_DMA_BLK_INT;
+
+	case SPRD_DMA_FRAG_INT_STS:
+		return SPRD_DMA_FRAG_INT;
+
+	default:
+		dev_warn(sdev->dma_dev.dev, "incorrect dma interrupt type\n");
+		return SPRD_DMA_NO_INT;
+	}
+}
+
+static enum sprd_dma_req_mode sprd_dma_get_req_type(struct sprd_dma_chn *schan)
+{
+	u32 frag_reg = readl(schan->chn_base + SPRD_DMA_CHN_FRG_LEN);
+
+	return (frag_reg >> SPRD_DMA_REQ_MODE_OFFSET) & SPRD_DMA_REQ_MODE_MASK;
+}
+
+static void sprd_dma_set_chn_config(struct sprd_dma_chn *schan,
+				    struct sprd_dma_desc *sdesc)
+{
+	struct sprd_dma_chn_hw *cfg = &sdesc->chn_hw;
+
+	writel(cfg->pause, schan->chn_base + SPRD_DMA_CHN_PAUSE);
+	writel(cfg->cfg, schan->chn_base + SPRD_DMA_CHN_CFG);
+	writel(cfg->intc, schan->chn_base + SPRD_DMA_CHN_INTC);
+	writel(cfg->src_addr, schan->chn_base + SPRD_DMA_CHN_SRC_ADDR);
+	writel(cfg->des_addr, schan->chn_base + SPRD_DMA_CHN_DES_ADDR);
+	writel(cfg->frg_len, schan->chn_base + SPRD_DMA_CHN_FRG_LEN);
+	writel(cfg->blk_len, schan->chn_base + SPRD_DMA_CHN_BLK_LEN);
+	writel(cfg->trsc_len, schan->chn_base + SPRD_DMA_CHN_TRSC_LEN);
+	writel(cfg->trsf_step, schan->chn_base + SPRD_DMA_CHN_TRSF_STEP);
+	writel(cfg->wrap_ptr, schan->chn_base + SPRD_DMA_CHN_WARP_PTR);
+	writel(cfg->wrap_to, schan->chn_base + SPRD_DMA_CHN_WARP_TO);
+	writel(cfg->llist_ptr, schan->chn_base + SPRD_DMA_CHN_LLIST_PTR);
+	writel(cfg->frg_step, schan->chn_base + SPRD_DMA_CHN_FRAG_STEP);
+	writel(cfg->src_blk_step, schan->chn_base + SPRD_DMA_CHN_SRC_BLK_STEP);
+	writel(cfg->des_blk_step, schan->chn_base + SPRD_DMA_CHN_DES_BLK_STEP);
+	writel(cfg->req, schan->chn_base + SPRD_DMA_CHN_REQ);
+}
+
+static void sprd_dma_start(struct sprd_dma_chn *schan)
+{
+	struct virt_dma_desc *vd = vchan_next_desc(&schan->vc);
+
+	if (!vd)
+		return;
+
+	list_del(&vd->node);
+	schan->cur_desc = to_sprd_dma_desc(vd);
+
+	/*
+	 * Copy the DMA configuration from DMA descriptor to this hardware
+	 * channel.
+	 */
+	sprd_dma_set_chn_config(schan, schan->cur_desc);
+	sprd_dma_set_uid(schan);
+	sprd_dma_enable_chn(schan);
+
+	if (schan->dev_id == SPRD_DMA_SOFTWARE_UID)
+		sprd_dma_soft_request(schan);
+}
+
+static void sprd_dma_stop(struct sprd_dma_chn *schan)
+{
+	sprd_dma_stop_and_disable(schan);
+	sprd_dma_unset_uid(schan);
+	sprd_dma_clear_int(schan);
+}
+
+static bool sprd_dma_check_trans_done(struct sprd_dma_desc *sdesc,
+				      enum sprd_dma_int_type int_type,
+				      enum sprd_dma_req_mode req_mode)
+{
+	if (int_type == SPRD_DMA_NO_INT)
+		return false;
+
+	if (int_type >= req_mode + 1)
+		return true;
+	else
+		return false;
+}
+
+static irqreturn_t dma_irq_handle(int irq, void *dev_id)
+{
+	struct sprd_dma_dev *sdev = (struct sprd_dma_dev *)dev_id;
+	u32 irq_status = readl(sdev->glb_base + SPRD_DMA_GLB_INT_MSK_STS);
+	struct sprd_dma_chn *schan;
+	struct sprd_dma_desc *sdesc;
+	enum sprd_dma_req_mode req_type;
+	enum sprd_dma_int_type int_type;
+	bool trans_done = false;
+	u32 i;
+
+	while (irq_status) {
+		i = __ffs(irq_status);
+		irq_status &= (irq_status - 1);
+		schan = &sdev->channels[i];
+
+		spin_lock(&schan->vc.lock);
+		int_type = sprd_dma_get_int_type(schan);
+		req_type = sprd_dma_get_req_type(schan);
+		sprd_dma_clear_int(schan);
+
+		sdesc = schan->cur_desc;
+
+		/* Check if the dma request descriptor is done. */
+		trans_done = sprd_dma_check_trans_done(sdesc, int_type,
+						       req_type);
+		if (trans_done == true) {
+			vchan_cookie_complete(&sdesc->vd);
+			schan->cur_desc = NULL;
+			sprd_dma_start(schan);
+		}
+		spin_unlock(&schan->vc.lock);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int sprd_dma_alloc_chan_resources(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	int ret;
+
+	ret = pm_runtime_get_sync(chan->device->dev);
+	if (ret < 0)
+		return ret;
+
+	schan->dev_id = SPRD_DMA_SOFTWARE_UID;
+	return 0;
+}
+
+static void sprd_dma_free_chan_resources(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	sprd_dma_stop(schan);
+	spin_unlock_irqrestore(&schan->vc.lock, flags);
+
+	vchan_free_chan_resources(&schan->vc);
+	pm_runtime_put(chan->device->dev);
+}
+
+static enum dma_status sprd_dma_tx_status(struct dma_chan *chan,
+					  dma_cookie_t cookie,
+					  struct dma_tx_state *txstate)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	struct virt_dma_desc *vd;
+	unsigned long flags;
+	enum dma_status ret;
+	u32 pos;
+
+	ret = dma_cookie_status(chan, cookie, txstate);
+	if (ret == DMA_COMPLETE || !txstate)
+		return ret;
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	vd = vchan_find_desc(&schan->vc, cookie);
+	if (vd) {
+		struct sprd_dma_desc *sdesc = to_sprd_dma_desc(vd);
+		struct sprd_dma_chn_hw *hw = &sdesc->chn_hw;
+
+		if (hw->trsc_len > 0)
+			pos = hw->trsc_len;
+		else if (hw->blk_len > 0)
+			pos = hw->blk_len;
+		else if (hw->frg_len > 0)
+			pos = hw->frg_len;
+		else
+			pos = 0;
+	} else if (schan->cur_desc && schan->cur_desc->vd.tx.cookie == cookie) {
+		pos = sprd_dma_get_dst_addr(schan);
+	} else {
+		pos = 0;
+	}
+	spin_unlock_irqrestore(&schan->vc.lock, flags);
+
+	dma_set_residue(txstate, pos);
+	return ret;
+}
+
+static void sprd_dma_issue_pending(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	if (vchan_issue_pending(&schan->vc) && !schan->cur_desc)
+		sprd_dma_start(schan);
+	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)
+{
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(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;
+	} else {
+		datawidth = 0;
+		src_step = 1;
+		des_step = 1;
+	}
+
+	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;
+	}
+
+	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);
+
+	if ((src_step != 0 && des_step != 0) || (src_step | des_step) == 0) {
+		fix_en = 0;
+	} else {
+		fix_en = 1;
+		if (src_step)
+			fix_mode = 1;
+		else
+			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;
+
+	case SPRD_DMA_BLK_INT:
+		hw->intc |= SPRD_DMA_BLK_INT_EN;
+		break;
+
+	case SPRD_DMA_BLK_FRAG_INT:
+		hw->intc |= SPRD_DMA_BLK_INT_EN | SPRD_DMA_FRAG_INT_EN;
+		break;
+
+	case SPRD_DMA_TRANS_INT:
+		hw->intc |= SPRD_DMA_TRANS_INT_EN;
+		break;
+
+	case SPRD_DMA_TRANS_FRAG_INT:
+		hw->intc |= SPRD_DMA_TRANS_INT_EN | SPRD_DMA_FRAG_INT_EN;
+		break;
+
+	case SPRD_DMA_TRANS_BLK_INT:
+		hw->intc |= SPRD_DMA_TRANS_INT_EN | SPRD_DMA_BLK_INT_EN;
+		break;
+
+	case SPRD_DMA_LIST_INT:
+		hw->intc |= SPRD_DMA_LIST_INT_EN;
+		break;
+
+	case SPRD_DMA_CFGERR_INT:
+		hw->intc |= SPRD_DMA_CFG_ERR_INT_EN;
+		break;
+
+	default:
+		dev_err(sdev->dma_dev.dev, "invalid irq mode\n");
+		return -EINVAL;
+	}
+
+	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;
+
+	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->frg_step = 0;
+	hw->src_blk_step = 0;
+	hw->des_blk_step = 0;
+	hw->src_blk_step = 0;
+	return 0;
+}
+
+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;
+	int ret;
+
+	sdesc = kzalloc(sizeof(*sdesc), GFP_NOWAIT);
+	if (!sdesc)
+		return NULL;
+
+	ret = sprd_dma_config(chan, sdesc, dest, src, len);
+	if (ret) {
+		kfree(sdesc);
+		return NULL;
+	}
+
+	return vchan_tx_prep(&schan->vc, &sdesc->vd, flags);
+}
+
+static int sprd_dma_pause(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	sprd_dma_pause_resume(schan, true);
+	spin_unlock_irqrestore(&schan->vc.lock, flags);
+
+	return 0;
+}
+
+static int sprd_dma_resume(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	unsigned long flags;
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	sprd_dma_pause_resume(schan, false);
+	spin_unlock_irqrestore(&schan->vc.lock, flags);
+
+	return 0;
+}
+
+static int sprd_dma_terminate_all(struct dma_chan *chan)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	unsigned long flags;
+	LIST_HEAD(head);
+
+	spin_lock_irqsave(&schan->vc.lock, flags);
+	sprd_dma_stop(schan);
+
+	vchan_get_all_descriptors(&schan->vc, &head);
+	spin_unlock_irqrestore(&schan->vc.lock, flags);
+
+	vchan_dma_desc_free_list(&schan->vc, &head);
+	return 0;
+}
+
+static void sprd_dma_free_desc(struct virt_dma_desc *vd)
+{
+	struct sprd_dma_desc *sdesc = to_sprd_dma_desc(vd);
+
+	kfree(sdesc);
+}
+
+static bool sprd_dma_filter_fn(struct dma_chan *chan, void *param)
+{
+	struct sprd_dma_chn *schan = to_sprd_dma_chan(chan);
+	struct sprd_dma_dev *sdev = to_sprd_dma_dev(&schan->vc.chan);
+	u32 req = *(u32 *)param;
+
+	if (req < sdev->total_chns)
+		return req == schan->chn_num + 1;
+	else
+		return false;
+}
+
+static int sprd_dma_probe(struct platform_device *pdev)
+{
+	struct device_node *np = pdev->dev.of_node;
+	struct sprd_dma_dev *sdev;
+	struct sprd_dma_chn *dma_chn;
+	struct resource *res;
+	u32 chn_count;
+	int ret, i;
+
+	ret = device_property_read_u32(&pdev->dev, "#dma-channels", &chn_count);
+	if (ret) {
+		dev_err(&pdev->dev, "get dma channels count failed\n");
+		return ret;
+	}
+
+	sdev = devm_kzalloc(&pdev->dev, sizeof(*sdev) +
+			    sizeof(*dma_chn) * chn_count,
+			    GFP_KERNEL);
+	if (!sdev)
+		return -ENOMEM;
+
+	sdev->clk = devm_clk_get(&pdev->dev, "enable");
+	if (IS_ERR(sdev->clk)) {
+		dev_err(&pdev->dev, "get enable clock failed\n");
+		return PTR_ERR(sdev->clk);
+	}
+
+	/* ashb clock is optional for AGCP DMA */
+	sdev->ashb_clk = devm_clk_get(&pdev->dev, "ashb_eb");
+	if (IS_ERR(sdev->ashb_clk))
+		dev_warn(&pdev->dev, "no optional ashb eb clock\n");
+
+	/*
+	 * We have three DMA controllers: AP DMA, AON DMA and AGCP DMA. For AGCP
+	 * DMA controller, it can or do not request the irq, which will save
+	 * system power without resuming system by DMA interrupts if AGCP DMA
+	 * does not request the irq. Thus the DMA interrupts property should
+	 * be optional.
+	 */
+	sdev->irq = platform_get_irq(pdev, 0);
+	if (sdev->irq > 0) {
+		ret = devm_request_irq(&pdev->dev, sdev->irq, dma_irq_handle,
+				       0, "sprd_dma", (void *)sdev);
+		if (ret < 0) {
+			dev_err(&pdev->dev, "request dma irq failed\n");
+			return ret;
+		}
+	} else {
+		dev_warn(&pdev->dev, "no interrupts for the dma controller\n");
+	}
+
+	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;
+
+	dma_cap_set(DMA_MEMCPY, sdev->dma_dev.cap_mask);
+	sdev->total_chns = chn_count;
+	sdev->dma_dev.chancnt = chn_count;
+	INIT_LIST_HEAD(&sdev->dma_dev.channels);
+	INIT_LIST_HEAD(&sdev->dma_dev.global_node);
+	sdev->dma_dev.dev = &pdev->dev;
+	sdev->dma_dev.device_alloc_chan_resources = sprd_dma_alloc_chan_resources;
+	sdev->dma_dev.device_free_chan_resources = sprd_dma_free_chan_resources;
+	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_pause = sprd_dma_pause;
+	sdev->dma_dev.device_resume = sprd_dma_resume;
+	sdev->dma_dev.device_terminate_all = sprd_dma_terminate_all;
+
+	for (i = 0; i < chn_count; i++) {
+		dma_chn = &sdev->channels[i];
+		dma_chn->chn_num = i;
+		dma_chn->cur_desc = NULL;
+		/* get each channel's registers base address. */
+		dma_chn->chn_base = sdev->glb_base + SPRD_DMA_CHN_REG_OFFSET +
+				    SPRD_DMA_CHN_REG_LENGTH * i;
+
+		dma_chn->vc.desc_free = sprd_dma_free_desc;
+		vchan_init(&dma_chn->vc, &sdev->dma_dev);
+	}
+
+	platform_set_drvdata(pdev, sdev);
+	ret = sprd_dma_enable(sdev);
+	if (ret)
+		return ret;
+
+	pm_runtime_set_active(&pdev->dev);
+	pm_runtime_enable(&pdev->dev);
+
+	ret = pm_runtime_get_sync(&pdev->dev);
+	if (ret < 0)
+		goto err_rpm;
+
+	ret = dma_async_device_register(&sdev->dma_dev);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "register dma device failed:%d\n", ret);
+		goto err_register;
+	}
+
+	sprd_dma_info.dma_cap = sdev->dma_dev.cap_mask;
+	ret = of_dma_controller_register(np, of_dma_simple_xlate,
+					 &sprd_dma_info);
+	if (ret)
+		goto err_of_register;
+
+	pm_runtime_put(&pdev->dev);
+	return 0;
+
+err_of_register:
+	dma_async_device_unregister(&sdev->dma_dev);
+err_register:
+	pm_runtime_put_noidle(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+err_rpm:
+	sprd_dma_disable(sdev);
+	return ret;
+}
+
+static int sprd_dma_remove(struct platform_device *pdev)
+{
+	struct sprd_dma_dev *sdev = platform_get_drvdata(pdev);
+	struct sprd_dma_chn *c, *cn;
+	int ret;
+
+	ret = pm_runtime_get_sync(&pdev->dev);
+	if (ret < 0)
+		return ret;
+
+	/* explicitly free the irq */
+	if (sdev->irq > 0)
+		devm_free_irq(&pdev->dev, sdev->irq, sdev);
+
+	list_for_each_entry_safe(c, cn, &sdev->dma_dev.channels,
+				 vc.chan.device_node) {
+		list_del(&c->vc.chan.device_node);
+		tasklet_kill(&c->vc.task);
+	}
+
+	of_dma_controller_free(pdev->dev.of_node);
+	dma_async_device_unregister(&sdev->dma_dev);
+	sprd_dma_disable(sdev);
+
+	pm_runtime_put_noidle(&pdev->dev);
+	pm_runtime_disable(&pdev->dev);
+	return 0;
+}
+
+static const struct of_device_id sprd_dma_match[] = {
+	{ .compatible = "sprd,sc9860-dma", },
+	{},
+};
+
+static int __maybe_unused sprd_dma_runtime_suspend(struct device *dev)
+{
+	struct sprd_dma_dev *sdev = dev_get_drvdata(dev);
+
+	sprd_dma_disable(sdev);
+	return 0;
+}
+
+static int __maybe_unused sprd_dma_runtime_resume(struct device *dev)
+{
+	struct sprd_dma_dev *sdev = dev_get_drvdata(dev);
+	int ret;
+
+	ret = sprd_dma_enable(sdev);
+	if (ret)
+		dev_err(sdev->dma_dev.dev, "enable dma failed\n");
+
+	return ret;
+}
+
+static const struct dev_pm_ops sprd_dma_pm_ops = {
+	SET_RUNTIME_PM_OPS(sprd_dma_runtime_suspend,
+			   sprd_dma_runtime_resume,
+			   NULL)
+};
+
+static struct platform_driver sprd_dma_driver = {
+	.probe = sprd_dma_probe,
+	.remove = sprd_dma_remove,
+	.driver = {
+		.name = "sprd-dma",
+		.of_match_table = sprd_dma_match,
+		.pm = &sprd_dma_pm_ops,
+	},
+};
+module_platform_driver(sprd_dma_driver);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("DMA driver for Spreadtrum");
+MODULE_AUTHOR("Baolin Wang <baolin.wang@spreadtrum.com>");
+MODULE_ALIAS("platform:sprd-dma");
diff --git a/drivers/dma/stm32-dmamux.c b/drivers/dma/stm32-dmamux.c
new file mode 100644
index 000000000000..d5db0f6e1ff8
--- /dev/null
+++ b/drivers/dma/stm32-dmamux.c
@@ -0,0 +1,327 @@
+/*
+ *
+ * Copyright (C) STMicroelectronics SA 2017
+ * Author(s): M'boumba Cedric Madianga <cedric.madianga@gmail.com>
+ *            Pierre-Yves Mordret <pierre-yves.mordret@st.com>
+ *
+ * License terms: GPL V2.0.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
+ * details.
+ *
+ * DMA Router driver for STM32 DMA MUX
+ *
+ * Based on TI DMA Crossbar driver
+ *
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/of_dma.h>
+#include <linux/reset.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+#define STM32_DMAMUX_CCR(x)		(0x4 * (x))
+#define STM32_DMAMUX_MAX_DMA_REQUESTS	32
+#define STM32_DMAMUX_MAX_REQUESTS	255
+
+struct stm32_dmamux {
+	u32 master;
+	u32 request;
+	u32 chan_id;
+};
+
+struct stm32_dmamux_data {
+	struct dma_router dmarouter;
+	struct clk *clk;
+	struct reset_control *rst;
+	void __iomem *iomem;
+	u32 dma_requests; /* Number of DMA requests connected to DMAMUX */
+	u32 dmamux_requests; /* Number of DMA requests routed toward DMAs */
+	spinlock_t lock; /* Protects register access */
+	unsigned long *dma_inuse; /* Used DMA channel */
+	u32 dma_reqs[]; /* Number of DMA Request per DMA masters.
+			 *  [0] holds number of DMA Masters.
+			 *  To be kept at very end end of this structure
+			 */
+};
+
+static inline u32 stm32_dmamux_read(void __iomem *iomem, u32 reg)
+{
+	return readl_relaxed(iomem + reg);
+}
+
+static inline void stm32_dmamux_write(void __iomem *iomem, u32 reg, u32 val)
+{
+	writel_relaxed(val, iomem + reg);
+}
+
+static void stm32_dmamux_free(struct device *dev, void *route_data)
+{
+	struct stm32_dmamux_data *dmamux = dev_get_drvdata(dev);
+	struct stm32_dmamux *mux = route_data;
+	unsigned long flags;
+
+	/* Clear dma request */
+	spin_lock_irqsave(&dmamux->lock, flags);
+
+	stm32_dmamux_write(dmamux->iomem, STM32_DMAMUX_CCR(mux->chan_id), 0);
+	clear_bit(mux->chan_id, dmamux->dma_inuse);
+
+	if (!IS_ERR(dmamux->clk))
+		clk_disable(dmamux->clk);
+
+	spin_unlock_irqrestore(&dmamux->lock, flags);
+
+	dev_dbg(dev, "Unmapping DMAMUX(%u) to DMA%u(%u)\n",
+		mux->request, mux->master, mux->chan_id);
+
+	kfree(mux);
+}
+
+static void *stm32_dmamux_route_allocate(struct of_phandle_args *dma_spec,
+					 struct of_dma *ofdma)
+{
+	struct platform_device *pdev = of_find_device_by_node(ofdma->of_node);
+	struct stm32_dmamux_data *dmamux = platform_get_drvdata(pdev);
+	struct stm32_dmamux *mux;
+	u32 i, min, max;
+	int ret;
+	unsigned long flags;
+
+	if (dma_spec->args_count != 3) {
+		dev_err(&pdev->dev, "invalid number of dma mux args\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (dma_spec->args[0] > dmamux->dmamux_requests) {
+		dev_err(&pdev->dev, "invalid mux request number: %d\n",
+			dma_spec->args[0]);
+		return ERR_PTR(-EINVAL);
+	}
+
+	mux = kzalloc(sizeof(*mux), GFP_KERNEL);
+	if (!mux)
+		return ERR_PTR(-ENOMEM);
+
+	spin_lock_irqsave(&dmamux->lock, flags);
+	mux->chan_id = find_first_zero_bit(dmamux->dma_inuse,
+					   dmamux->dma_requests);
+	set_bit(mux->chan_id, dmamux->dma_inuse);
+	spin_unlock_irqrestore(&dmamux->lock, flags);
+
+	if (mux->chan_id == dmamux->dma_requests) {
+		dev_err(&pdev->dev, "Run out of free DMA requests\n");
+		ret = -ENOMEM;
+		goto error;
+	}
+
+	/* Look for DMA Master */
+	for (i = 1, min = 0, max = dmamux->dma_reqs[i];
+	     i <= dmamux->dma_reqs[0];
+	     min += dmamux->dma_reqs[i], max += dmamux->dma_reqs[++i])
+		if (mux->chan_id < max)
+			break;
+	mux->master = i - 1;
+
+	/* The of_node_put() will be done in of_dma_router_xlate function */
+	dma_spec->np = of_parse_phandle(ofdma->of_node, "dma-masters", i - 1);
+	if (!dma_spec->np) {
+		dev_err(&pdev->dev, "can't get dma master\n");
+		ret = -EINVAL;
+		goto error;
+	}
+
+	/* Set dma request */
+	spin_lock_irqsave(&dmamux->lock, flags);
+	if (!IS_ERR(dmamux->clk)) {
+		ret = clk_enable(dmamux->clk);
+		if (ret < 0) {
+			spin_unlock_irqrestore(&dmamux->lock, flags);
+			dev_err(&pdev->dev, "clk_prep_enable issue: %d\n", ret);
+			goto error;
+		}
+	}
+	spin_unlock_irqrestore(&dmamux->lock, flags);
+
+	mux->request = dma_spec->args[0];
+
+	/*  craft DMA spec */
+	dma_spec->args[3] = dma_spec->args[2];
+	dma_spec->args[2] = dma_spec->args[1];
+	dma_spec->args[1] = 0;
+	dma_spec->args[0] = mux->chan_id - min;
+	dma_spec->args_count = 4;
+
+	stm32_dmamux_write(dmamux->iomem, STM32_DMAMUX_CCR(mux->chan_id),
+			   mux->request);
+	dev_dbg(&pdev->dev, "Mapping DMAMUX(%u) to DMA%u(%u)\n",
+		mux->request, mux->master, mux->chan_id);
+
+	return mux;
+
+error:
+	clear_bit(mux->chan_id, dmamux->dma_inuse);
+	kfree(mux);
+	return ERR_PTR(ret);
+}
+
+static const struct of_device_id stm32_stm32dma_master_match[] = {
+	{ .compatible = "st,stm32-dma", },
+	{},
+};
+
+static int stm32_dmamux_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	const struct of_device_id *match;
+	struct device_node *dma_node;
+	struct stm32_dmamux_data *stm32_dmamux;
+	struct resource *res;
+	void __iomem *iomem;
+	int i, count, ret;
+	u32 dma_req;
+
+	if (!node)
+		return -ENODEV;
+
+	count = device_property_read_u32_array(&pdev->dev, "dma-masters",
+					       NULL, 0);
+	if (count < 0) {
+		dev_err(&pdev->dev, "Can't get DMA master(s) node\n");
+		return -ENODEV;
+	}
+
+	stm32_dmamux = devm_kzalloc(&pdev->dev, sizeof(*stm32_dmamux) +
+				    sizeof(u32) * (count + 1), GFP_KERNEL);
+	if (!stm32_dmamux)
+		return -ENOMEM;
+
+	dma_req = 0;
+	for (i = 1; i <= count; i++) {
+		dma_node = of_parse_phandle(node, "dma-masters", i - 1);
+
+		match = of_match_node(stm32_stm32dma_master_match, dma_node);
+		if (!match) {
+			dev_err(&pdev->dev, "DMA master is not supported\n");
+			of_node_put(dma_node);
+			return -EINVAL;
+		}
+
+		if (of_property_read_u32(dma_node, "dma-requests",
+					 &stm32_dmamux->dma_reqs[i])) {
+			dev_info(&pdev->dev,
+				 "Missing MUX output information, using %u.\n",
+				 STM32_DMAMUX_MAX_DMA_REQUESTS);
+			stm32_dmamux->dma_reqs[i] =
+				STM32_DMAMUX_MAX_DMA_REQUESTS;
+		}
+		dma_req += stm32_dmamux->dma_reqs[i];
+		of_node_put(dma_node);
+	}
+
+	if (dma_req > STM32_DMAMUX_MAX_DMA_REQUESTS) {
+		dev_err(&pdev->dev, "Too many DMA Master Requests to manage\n");
+		return -ENODEV;
+	}
+
+	stm32_dmamux->dma_requests = dma_req;
+	stm32_dmamux->dma_reqs[0] = count;
+	stm32_dmamux->dma_inuse = devm_kcalloc(&pdev->dev,
+					       BITS_TO_LONGS(dma_req),
+					       sizeof(unsigned long),
+					       GFP_KERNEL);
+	if (!stm32_dmamux->dma_inuse)
+		return -ENOMEM;
+
+	if (device_property_read_u32(&pdev->dev, "dma-requests",
+				     &stm32_dmamux->dmamux_requests)) {
+		stm32_dmamux->dmamux_requests = STM32_DMAMUX_MAX_REQUESTS;
+		dev_warn(&pdev->dev, "DMAMUX defaulting on %u requests\n",
+			 stm32_dmamux->dmamux_requests);
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENODEV;
+
+	iomem = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(iomem))
+		return PTR_ERR(iomem);
+
+	spin_lock_init(&stm32_dmamux->lock);
+
+	stm32_dmamux->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(stm32_dmamux->clk)) {
+		ret = PTR_ERR(stm32_dmamux->clk);
+		if (ret == -EPROBE_DEFER)
+			dev_info(&pdev->dev, "Missing controller clock\n");
+		return ret;
+	}
+
+	stm32_dmamux->rst = devm_reset_control_get(&pdev->dev, NULL);
+	if (!IS_ERR(stm32_dmamux->rst)) {
+		reset_control_assert(stm32_dmamux->rst);
+		udelay(2);
+		reset_control_deassert(stm32_dmamux->rst);
+	}
+
+	stm32_dmamux->iomem = iomem;
+	stm32_dmamux->dmarouter.dev = &pdev->dev;
+	stm32_dmamux->dmarouter.route_free = stm32_dmamux_free;
+
+	platform_set_drvdata(pdev, stm32_dmamux);
+
+	if (!IS_ERR(stm32_dmamux->clk)) {
+		ret = clk_prepare_enable(stm32_dmamux->clk);
+		if (ret < 0) {
+			dev_err(&pdev->dev, "clk_prep_enable error: %d\n", ret);
+			return ret;
+		}
+	}
+
+	/* Reset the dmamux */
+	for (i = 0; i < stm32_dmamux->dma_requests; i++)
+		stm32_dmamux_write(stm32_dmamux->iomem, STM32_DMAMUX_CCR(i), 0);
+
+	if (!IS_ERR(stm32_dmamux->clk))
+		clk_disable(stm32_dmamux->clk);
+
+	return of_dma_router_register(node, stm32_dmamux_route_allocate,
+				     &stm32_dmamux->dmarouter);
+}
+
+static const struct of_device_id stm32_dmamux_match[] = {
+	{ .compatible = "st,stm32h7-dmamux" },
+	{},
+};
+
+static struct platform_driver stm32_dmamux_driver = {
+	.probe	= stm32_dmamux_probe,
+	.driver = {
+		.name = "stm32-dmamux",
+		.of_match_table = stm32_dmamux_match,
+	},
+};
+
+static int __init stm32_dmamux_init(void)
+{
+	return platform_driver_register(&stm32_dmamux_driver);
+}
+arch_initcall(stm32_dmamux_init);
+
+MODULE_DESCRIPTION("DMA Router driver for STM32 DMA MUX");
+MODULE_AUTHOR("M'boumba Cedric Madianga <cedric.madianga@gmail.com>");
+MODULE_AUTHOR("Pierre-Yves Mordret <pierre-yves.mordret@st.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/dma/stm32-mdma.c b/drivers/dma/stm32-mdma.c
new file mode 100644
index 000000000000..daa1602eb9f5
--- /dev/null
+++ b/drivers/dma/stm32-mdma.c
@@ -0,0 +1,1682 @@
+/*
+ *
+ * Copyright (C) STMicroelectronics SA 2017
+ * Author(s): M'boumba Cedric Madianga <cedric.madianga@gmail.com>
+ *            Pierre-Yves Mordret <pierre-yves.mordret@st.com>
+ *
+ * License terms: GPL V2.0.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
+ * details.
+ *
+ * Driver for STM32 MDMA controller
+ *
+ * Inspired by stm32-dma.c and dma-jz4780.c
+ *
+ */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/iopoll.h>
+#include <linux/jiffies.h>
+#include <linux/list.h>
+#include <linux/log2.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/of_dma.h>
+#include <linux/platform_device.h>
+#include <linux/reset.h>
+#include <linux/slab.h>
+
+#include "virt-dma.h"
+
+/*  MDMA Generic getter/setter */
+#define STM32_MDMA_SHIFT(n)		(ffs(n) - 1)
+#define STM32_MDMA_SET(n, mask)		(((n) << STM32_MDMA_SHIFT(mask)) & \
+					 (mask))
+#define STM32_MDMA_GET(n, mask)		(((n) & (mask)) >> \
+					 STM32_MDMA_SHIFT(mask))
+
+#define STM32_MDMA_GISR0		0x0000 /* MDMA Int Status Reg 1 */
+#define STM32_MDMA_GISR1		0x0004 /* MDMA Int Status Reg 2 */
+
+/* MDMA Channel x interrupt/status register */
+#define STM32_MDMA_CISR(x)		(0x40 + 0x40 * (x)) /* x = 0..62 */
+#define STM32_MDMA_CISR_CRQA		BIT(16)
+#define STM32_MDMA_CISR_TCIF		BIT(4)
+#define STM32_MDMA_CISR_BTIF		BIT(3)
+#define STM32_MDMA_CISR_BRTIF		BIT(2)
+#define STM32_MDMA_CISR_CTCIF		BIT(1)
+#define STM32_MDMA_CISR_TEIF		BIT(0)
+
+/* MDMA Channel x interrupt flag clear register */
+#define STM32_MDMA_CIFCR(x)		(0x44 + 0x40 * (x))
+#define STM32_MDMA_CIFCR_CLTCIF		BIT(4)
+#define STM32_MDMA_CIFCR_CBTIF		BIT(3)
+#define STM32_MDMA_CIFCR_CBRTIF		BIT(2)
+#define STM32_MDMA_CIFCR_CCTCIF		BIT(1)
+#define STM32_MDMA_CIFCR_CTEIF		BIT(0)
+#define STM32_MDMA_CIFCR_CLEAR_ALL	(STM32_MDMA_CIFCR_CLTCIF \
+					| STM32_MDMA_CIFCR_CBTIF \
+					| STM32_MDMA_CIFCR_CBRTIF \
+					| STM32_MDMA_CIFCR_CCTCIF \
+					| STM32_MDMA_CIFCR_CTEIF)
+
+/* MDMA Channel x error status register */
+#define STM32_MDMA_CESR(x)		(0x48 + 0x40 * (x))
+#define STM32_MDMA_CESR_BSE		BIT(11)
+#define STM32_MDMA_CESR_ASR		BIT(10)
+#define STM32_MDMA_CESR_TEMD		BIT(9)
+#define STM32_MDMA_CESR_TELD		BIT(8)
+#define STM32_MDMA_CESR_TED		BIT(7)
+#define STM32_MDMA_CESR_TEA_MASK	GENMASK(6, 0)
+
+/* MDMA Channel x control register */
+#define STM32_MDMA_CCR(x)		(0x4C + 0x40 * (x))
+#define STM32_MDMA_CCR_SWRQ		BIT(16)
+#define STM32_MDMA_CCR_WEX		BIT(14)
+#define STM32_MDMA_CCR_HEX		BIT(13)
+#define STM32_MDMA_CCR_BEX		BIT(12)
+#define STM32_MDMA_CCR_PL_MASK		GENMASK(7, 6)
+#define STM32_MDMA_CCR_PL(n)		STM32_MDMA_SET(n, \
+						       STM32_MDMA_CCR_PL_MASK)
+#define STM32_MDMA_CCR_TCIE		BIT(5)
+#define STM32_MDMA_CCR_BTIE		BIT(4)
+#define STM32_MDMA_CCR_BRTIE		BIT(3)
+#define STM32_MDMA_CCR_CTCIE		BIT(2)
+#define STM32_MDMA_CCR_TEIE		BIT(1)
+#define STM32_MDMA_CCR_EN		BIT(0)
+#define STM32_MDMA_CCR_IRQ_MASK		(STM32_MDMA_CCR_TCIE \
+					| STM32_MDMA_CCR_BTIE \
+					| STM32_MDMA_CCR_BRTIE \
+					| STM32_MDMA_CCR_CTCIE \
+					| STM32_MDMA_CCR_TEIE)
+
+/* MDMA Channel x transfer configuration register */
+#define STM32_MDMA_CTCR(x)		(0x50 + 0x40 * (x))
+#define STM32_MDMA_CTCR_BWM		BIT(31)
+#define STM32_MDMA_CTCR_SWRM		BIT(30)
+#define STM32_MDMA_CTCR_TRGM_MSK	GENMASK(29, 28)
+#define STM32_MDMA_CTCR_TRGM(n)		STM32_MDMA_SET((n), \
+						       STM32_MDMA_CTCR_TRGM_MSK)
+#define STM32_MDMA_CTCR_TRGM_GET(n)	STM32_MDMA_GET((n), \
+						       STM32_MDMA_CTCR_TRGM_MSK)
+#define STM32_MDMA_CTCR_PAM_MASK	GENMASK(27, 26)
+#define STM32_MDMA_CTCR_PAM(n)		STM32_MDMA_SET(n, \
+						       STM32_MDMA_CTCR_PAM_MASK)
+#define STM32_MDMA_CTCR_PKE		BIT(25)
+#define STM32_MDMA_CTCR_TLEN_MSK	GENMASK(24, 18)
+#define STM32_MDMA_CTCR_TLEN(n)		STM32_MDMA_SET((n), \
+						       STM32_MDMA_CTCR_TLEN_MSK)
+#define STM32_MDMA_CTCR_TLEN_GET(n)	STM32_MDMA_GET((n), \
+						       STM32_MDMA_CTCR_TLEN_MSK)
+#define STM32_MDMA_CTCR_LEN2_MSK	GENMASK(25, 18)
+#define STM32_MDMA_CTCR_LEN2(n)		STM32_MDMA_SET((n), \
+						       STM32_MDMA_CTCR_LEN2_MSK)
+#define STM32_MDMA_CTCR_LEN2_GET(n)	STM32_MDMA_GET((n), \
+						       STM32_MDMA_CTCR_LEN2_MSK)
+#define STM32_MDMA_CTCR_DBURST_MASK	GENMASK(17, 15)
+#define STM32_MDMA_CTCR_DBURST(n)	STM32_MDMA_SET(n, \
+						    STM32_MDMA_CTCR_DBURST_MASK)
+#define STM32_MDMA_CTCR_SBURST_MASK	GENMASK(14, 12)
+#define STM32_MDMA_CTCR_SBURST(n)	STM32_MDMA_SET(n, \
+						    STM32_MDMA_CTCR_SBURST_MASK)
+#define STM32_MDMA_CTCR_DINCOS_MASK	GENMASK(11, 10)
+#define STM32_MDMA_CTCR_DINCOS(n)	STM32_MDMA_SET((n), \
+						    STM32_MDMA_CTCR_DINCOS_MASK)
+#define STM32_MDMA_CTCR_SINCOS_MASK	GENMASK(9, 8)
+#define STM32_MDMA_CTCR_SINCOS(n)	STM32_MDMA_SET((n), \
+						    STM32_MDMA_CTCR_SINCOS_MASK)
+#define STM32_MDMA_CTCR_DSIZE_MASK	GENMASK(7, 6)
+#define STM32_MDMA_CTCR_DSIZE(n)	STM32_MDMA_SET(n, \
+						     STM32_MDMA_CTCR_DSIZE_MASK)
+#define STM32_MDMA_CTCR_SSIZE_MASK	GENMASK(5, 4)
+#define STM32_MDMA_CTCR_SSIZE(n)	STM32_MDMA_SET(n, \
+						     STM32_MDMA_CTCR_SSIZE_MASK)
+#define STM32_MDMA_CTCR_DINC_MASK	GENMASK(3, 2)
+#define STM32_MDMA_CTCR_DINC(n)		STM32_MDMA_SET((n), \
+						      STM32_MDMA_CTCR_DINC_MASK)
+#define STM32_MDMA_CTCR_SINC_MASK	GENMASK(1, 0)
+#define STM32_MDMA_CTCR_SINC(n)		STM32_MDMA_SET((n), \
+						      STM32_MDMA_CTCR_SINC_MASK)
+#define STM32_MDMA_CTCR_CFG_MASK	(STM32_MDMA_CTCR_SINC_MASK \
+					| STM32_MDMA_CTCR_DINC_MASK \
+					| STM32_MDMA_CTCR_SINCOS_MASK \
+					| STM32_MDMA_CTCR_DINCOS_MASK \
+					| STM32_MDMA_CTCR_LEN2_MSK \
+					| STM32_MDMA_CTCR_TRGM_MSK)
+
+/* MDMA Channel x block number of data register */
+#define STM32_MDMA_CBNDTR(x)		(0x54 + 0x40 * (x))
+#define STM32_MDMA_CBNDTR_BRC_MK	GENMASK(31, 20)
+#define STM32_MDMA_CBNDTR_BRC(n)	STM32_MDMA_SET(n, \
+						       STM32_MDMA_CBNDTR_BRC_MK)
+#define STM32_MDMA_CBNDTR_BRC_GET(n)	STM32_MDMA_GET((n), \
+						       STM32_MDMA_CBNDTR_BRC_MK)
+
+#define STM32_MDMA_CBNDTR_BRDUM		BIT(19)
+#define STM32_MDMA_CBNDTR_BRSUM		BIT(18)
+#define STM32_MDMA_CBNDTR_BNDT_MASK	GENMASK(16, 0)
+#define STM32_MDMA_CBNDTR_BNDT(n)	STM32_MDMA_SET(n, \
+						    STM32_MDMA_CBNDTR_BNDT_MASK)
+
+/* MDMA Channel x source address register */
+#define STM32_MDMA_CSAR(x)		(0x58 + 0x40 * (x))
+
+/* MDMA Channel x destination address register */
+#define STM32_MDMA_CDAR(x)		(0x5C + 0x40 * (x))
+
+/* MDMA Channel x block repeat address update register */
+#define STM32_MDMA_CBRUR(x)		(0x60 + 0x40 * (x))
+#define STM32_MDMA_CBRUR_DUV_MASK	GENMASK(31, 16)
+#define STM32_MDMA_CBRUR_DUV(n)		STM32_MDMA_SET(n, \
+						      STM32_MDMA_CBRUR_DUV_MASK)
+#define STM32_MDMA_CBRUR_SUV_MASK	GENMASK(15, 0)
+#define STM32_MDMA_CBRUR_SUV(n)		STM32_MDMA_SET(n, \
+						      STM32_MDMA_CBRUR_SUV_MASK)
+
+/* MDMA Channel x link address register */
+#define STM32_MDMA_CLAR(x)		(0x64 + 0x40 * (x))
+
+/* MDMA Channel x trigger and bus selection register */
+#define STM32_MDMA_CTBR(x)		(0x68 + 0x40 * (x))
+#define STM32_MDMA_CTBR_DBUS		BIT(17)
+#define STM32_MDMA_CTBR_SBUS		BIT(16)
+#define STM32_MDMA_CTBR_TSEL_MASK	GENMASK(7, 0)
+#define STM32_MDMA_CTBR_TSEL(n)		STM32_MDMA_SET(n, \
+						      STM32_MDMA_CTBR_TSEL_MASK)
+
+/* MDMA Channel x mask address register */
+#define STM32_MDMA_CMAR(x)		(0x70 + 0x40 * (x))
+
+/* MDMA Channel x mask data register */
+#define STM32_MDMA_CMDR(x)		(0x74 + 0x40 * (x))
+
+#define STM32_MDMA_MAX_BUF_LEN		128
+#define STM32_MDMA_MAX_BLOCK_LEN	65536
+#define STM32_MDMA_MAX_CHANNELS		63
+#define STM32_MDMA_MAX_REQUESTS		256
+#define STM32_MDMA_MAX_BURST		128
+#define STM32_MDMA_VERY_HIGH_PRIORITY	0x11
+
+enum stm32_mdma_trigger_mode {
+	STM32_MDMA_BUFFER,
+	STM32_MDMA_BLOCK,
+	STM32_MDMA_BLOCK_REP,
+	STM32_MDMA_LINKED_LIST,
+};
+
+enum stm32_mdma_width {
+	STM32_MDMA_BYTE,
+	STM32_MDMA_HALF_WORD,
+	STM32_MDMA_WORD,
+	STM32_MDMA_DOUBLE_WORD,
+};
+
+enum stm32_mdma_inc_mode {
+	STM32_MDMA_FIXED = 0,
+	STM32_MDMA_INC = 2,
+	STM32_MDMA_DEC = 3,
+};
+
+struct stm32_mdma_chan_config {
+	u32 request;
+	u32 priority_level;
+	u32 transfer_config;
+	u32 mask_addr;
+	u32 mask_data;
+};
+
+struct stm32_mdma_hwdesc {
+	u32 ctcr;
+	u32 cbndtr;
+	u32 csar;
+	u32 cdar;
+	u32 cbrur;
+	u32 clar;
+	u32 ctbr;
+	u32 dummy;
+	u32 cmar;
+	u32 cmdr;
+} __aligned(64);
+
+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_chan {
+	struct virt_dma_chan vchan;
+	struct dma_pool *desc_pool;
+	u32 id;
+	struct stm32_mdma_desc *desc;
+	u32 curr_hwdesc;
+	struct dma_slave_config dma_config;
+	struct stm32_mdma_chan_config chan_config;
+	bool busy;
+	u32 mem_burst;
+	u32 mem_width;
+};
+
+struct stm32_mdma_device {
+	struct dma_device ddev;
+	void __iomem *base;
+	struct clk *clk;
+	int irq;
+	struct reset_control *rst;
+	u32 nr_channels;
+	u32 nr_requests;
+	u32 nr_ahb_addr_masks;
+	struct stm32_mdma_chan chan[STM32_MDMA_MAX_CHANNELS];
+	u32 ahb_addr_masks[];
+};
+
+static struct stm32_mdma_device *stm32_mdma_get_dev(
+	struct stm32_mdma_chan *chan)
+{
+	return container_of(chan->vchan.chan.device, struct stm32_mdma_device,
+			    ddev);
+}
+
+static struct stm32_mdma_chan *to_stm32_mdma_chan(struct dma_chan *c)
+{
+	return container_of(c, struct stm32_mdma_chan, vchan.chan);
+}
+
+static struct stm32_mdma_desc *to_stm32_mdma_desc(struct virt_dma_desc *vdesc)
+{
+	return container_of(vdesc, struct stm32_mdma_desc, vdesc);
+}
+
+static struct device *chan2dev(struct stm32_mdma_chan *chan)
+{
+	return &chan->vchan.chan.dev->device;
+}
+
+static struct device *mdma2dev(struct stm32_mdma_device *mdma_dev)
+{
+	return mdma_dev->ddev.dev;
+}
+
+static u32 stm32_mdma_read(struct stm32_mdma_device *dmadev, u32 reg)
+{
+	return readl_relaxed(dmadev->base + reg);
+}
+
+static void stm32_mdma_write(struct stm32_mdma_device *dmadev, u32 reg, u32 val)
+{
+	writel_relaxed(val, dmadev->base + reg);
+}
+
+static void stm32_mdma_set_bits(struct stm32_mdma_device *dmadev, u32 reg,
+				u32 mask)
+{
+	void __iomem *addr = dmadev->base + reg;
+
+	writel_relaxed(readl_relaxed(addr) | mask, addr);
+}
+
+static void stm32_mdma_clr_bits(struct stm32_mdma_device *dmadev, u32 reg,
+				u32 mask)
+{
+	void __iomem *addr = dmadev->base + reg;
+
+	writel_relaxed(readl_relaxed(addr) & ~mask, addr);
+}
+
+static struct stm32_mdma_desc *stm32_mdma_alloc_desc(
+		struct stm32_mdma_chan *chan, u32 count)
+{
+	struct stm32_mdma_desc *desc;
+
+	desc = kzalloc(sizeof(*desc), 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;
+	}
+
+	desc->count = count;
+
+	return desc;
+}
+
+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);
+
+	dma_pool_free(chan->desc_pool, desc->hwdesc, desc->hwdesc_phys);
+	kfree(desc);
+}
+
+static int stm32_mdma_get_width(struct stm32_mdma_chan *chan,
+				enum dma_slave_buswidth width)
+{
+	switch (width) {
+	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(width) - 1;
+	default:
+		dev_err(chan2dev(chan), "Dma bus width %i not supported\n",
+			width);
+		return -EINVAL;
+	}
+}
+
+static enum dma_slave_buswidth stm32_mdma_get_max_width(dma_addr_t addr,
+							u32 buf_len, u32 tlen)
+{
+	enum dma_slave_buswidth max_width = DMA_SLAVE_BUSWIDTH_8_BYTES;
+
+	for (max_width = DMA_SLAVE_BUSWIDTH_8_BYTES;
+	     max_width > DMA_SLAVE_BUSWIDTH_1_BYTE;
+	     max_width >>= 1) {
+		/*
+		 * Address and buffer length both have to be aligned on
+		 * bus width
+		 */
+		if ((((buf_len | addr) & (max_width - 1)) == 0) &&
+		    tlen >= max_width)
+			break;
+	}
+
+	return max_width;
+}
+
+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;
+
+	while ((burst_len > 0) && (tlen % burst_len)) {
+		best_burst = best_burst >> 1;
+		burst_len = best_burst * width;
+	}
+
+	return (best_burst > 0) ? best_burst : 1;
+}
+
+static int stm32_mdma_disable_chan(struct stm32_mdma_chan *chan)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	u32 ccr, cisr, id, reg;
+	int ret;
+
+	id = chan->id;
+	reg = STM32_MDMA_CCR(id);
+
+	/* Disable interrupts */
+	stm32_mdma_clr_bits(dmadev, reg, STM32_MDMA_CCR_IRQ_MASK);
+
+	ccr = stm32_mdma_read(dmadev, reg);
+	if (ccr & STM32_MDMA_CCR_EN) {
+		stm32_mdma_clr_bits(dmadev, reg, STM32_MDMA_CCR_EN);
+
+		/* Ensure that any ongoing transfer has been completed */
+		ret = readl_relaxed_poll_timeout_atomic(
+				dmadev->base + STM32_MDMA_CISR(id), cisr,
+				(cisr & STM32_MDMA_CISR_CTCIF), 10, 1000);
+		if (ret) {
+			dev_err(chan2dev(chan), "%s: timeout!\n", __func__);
+			return -EBUSY;
+		}
+	}
+
+	return 0;
+}
+
+static void stm32_mdma_stop(struct stm32_mdma_chan *chan)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	u32 status;
+	int ret;
+
+	/* Disable DMA */
+	ret = stm32_mdma_disable_chan(chan);
+	if (ret < 0)
+		return;
+
+	/* Clear interrupt status if it is there */
+	status = stm32_mdma_read(dmadev, STM32_MDMA_CISR(chan->id));
+	if (status) {
+		dev_dbg(chan2dev(chan), "%s(): clearing interrupt: 0x%08x\n",
+			__func__, status);
+		stm32_mdma_set_bits(dmadev, STM32_MDMA_CIFCR(chan->id), status);
+	}
+
+	chan->busy = false;
+}
+
+static void stm32_mdma_set_bus(struct stm32_mdma_device *dmadev, u32 *ctbr,
+			       u32 ctbr_mask, u32 src_addr)
+{
+	u32 mask;
+	int i;
+
+	/* Check if memory device is on AHB or AXI */
+	*ctbr &= ~ctbr_mask;
+	mask = src_addr & 0xF0000000;
+	for (i = 0; i < dmadev->nr_ahb_addr_masks; i++) {
+		if (mask == dmadev->ahb_addr_masks[i]) {
+			*ctbr |= ctbr_mask;
+			break;
+		}
+	}
+}
+
+static int stm32_mdma_set_xfer_param(struct stm32_mdma_chan *chan,
+				     enum dma_transfer_direction direction,
+				     u32 *mdma_ccr, u32 *mdma_ctcr,
+				     u32 *mdma_ctbr, dma_addr_t addr,
+				     u32 buf_len)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct stm32_mdma_chan_config *chan_config = &chan->chan_config;
+	enum dma_slave_buswidth src_addr_width, dst_addr_width;
+	phys_addr_t src_addr, dst_addr;
+	int src_bus_width, dst_bus_width;
+	u32 src_maxburst, dst_maxburst, src_best_burst, dst_best_burst;
+	u32 ccr, ctcr, ctbr, tlen;
+
+	src_addr_width = chan->dma_config.src_addr_width;
+	dst_addr_width = chan->dma_config.dst_addr_width;
+	src_maxburst = chan->dma_config.src_maxburst;
+	dst_maxburst = chan->dma_config.dst_maxburst;
+
+	ccr = stm32_mdma_read(dmadev, STM32_MDMA_CCR(chan->id));
+	ctcr = stm32_mdma_read(dmadev, STM32_MDMA_CTCR(chan->id));
+	ctbr = stm32_mdma_read(dmadev, STM32_MDMA_CTBR(chan->id));
+
+	/* Enable HW request mode */
+	ctcr &= ~STM32_MDMA_CTCR_SWRM;
+
+	/* Set DINC, SINC, DINCOS, SINCOS, TRGM and TLEN retrieve from DT */
+	ctcr &= ~STM32_MDMA_CTCR_CFG_MASK;
+	ctcr |= chan_config->transfer_config & STM32_MDMA_CTCR_CFG_MASK;
+
+	/*
+	 * For buffer transfer length (TLEN) we have to set
+	 * the number of bytes - 1 in CTCR register
+	 */
+	tlen = STM32_MDMA_CTCR_LEN2_GET(ctcr);
+	ctcr &= ~STM32_MDMA_CTCR_LEN2_MSK;
+	ctcr |= STM32_MDMA_CTCR_TLEN((tlen - 1));
+
+	/* Disable Pack Enable */
+	ctcr &= ~STM32_MDMA_CTCR_PKE;
+
+	/* Check burst size constraints */
+	if (src_maxburst * src_addr_width > STM32_MDMA_MAX_BURST ||
+	    dst_maxburst * dst_addr_width > STM32_MDMA_MAX_BURST) {
+		dev_err(chan2dev(chan),
+			"burst size * bus width higher than %d bytes\n",
+			STM32_MDMA_MAX_BURST);
+		return -EINVAL;
+	}
+
+	if ((!is_power_of_2(src_maxburst) && src_maxburst > 0) ||
+	    (!is_power_of_2(dst_maxburst) && dst_maxburst > 0)) {
+		dev_err(chan2dev(chan), "burst size must be a power of 2\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * Configure channel control:
+	 * - Clear SW request as in this case this is a HW one
+	 * - Clear WEX, HEX and BEX bits
+	 * - Set priority level
+	 */
+	ccr &= ~(STM32_MDMA_CCR_SWRQ | STM32_MDMA_CCR_WEX | STM32_MDMA_CCR_HEX |
+		 STM32_MDMA_CCR_BEX | STM32_MDMA_CCR_PL_MASK);
+	ccr |= STM32_MDMA_CCR_PL(chan_config->priority_level);
+
+	/* Configure Trigger selection */
+	ctbr &= ~STM32_MDMA_CTBR_TSEL_MASK;
+	ctbr |= STM32_MDMA_CTBR_TSEL(chan_config->request);
+
+	switch (direction) {
+	case DMA_MEM_TO_DEV:
+		dst_addr = chan->dma_config.dst_addr;
+
+		/* Set device data size */
+		dst_bus_width = stm32_mdma_get_width(chan, dst_addr_width);
+		if (dst_bus_width < 0)
+			return dst_bus_width;
+		ctcr &= ~STM32_MDMA_CTCR_DSIZE_MASK;
+		ctcr |= STM32_MDMA_CTCR_DSIZE(dst_bus_width);
+
+		/* Set device burst value */
+		dst_best_burst = stm32_mdma_get_best_burst(buf_len, tlen,
+							   dst_maxburst,
+							   dst_addr_width);
+		chan->mem_burst = dst_best_burst;
+		ctcr &= ~STM32_MDMA_CTCR_DBURST_MASK;
+		ctcr |= STM32_MDMA_CTCR_DBURST((ilog2(dst_best_burst)));
+
+		/* Set memory data size */
+		src_addr_width = stm32_mdma_get_max_width(addr, buf_len, tlen);
+		chan->mem_width = src_addr_width;
+		src_bus_width = stm32_mdma_get_width(chan, src_addr_width);
+		if (src_bus_width < 0)
+			return src_bus_width;
+		ctcr &= ~STM32_MDMA_CTCR_SSIZE_MASK |
+			STM32_MDMA_CTCR_SINCOS_MASK;
+		ctcr |= STM32_MDMA_CTCR_SSIZE(src_bus_width) |
+			STM32_MDMA_CTCR_SINCOS(src_bus_width);
+
+		/* Set memory burst value */
+		src_maxburst = STM32_MDMA_MAX_BUF_LEN / src_addr_width;
+		src_best_burst = stm32_mdma_get_best_burst(buf_len, tlen,
+							   src_maxburst,
+							   src_addr_width);
+		chan->mem_burst = src_best_burst;
+		ctcr &= ~STM32_MDMA_CTCR_SBURST_MASK;
+		ctcr |= STM32_MDMA_CTCR_SBURST((ilog2(src_best_burst)));
+
+		/* Select bus */
+		stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_DBUS,
+				   dst_addr);
+
+		if (dst_bus_width != src_bus_width)
+			ctcr |= STM32_MDMA_CTCR_PKE;
+
+		/* Set destination address */
+		stm32_mdma_write(dmadev, STM32_MDMA_CDAR(chan->id), dst_addr);
+		break;
+
+	case DMA_DEV_TO_MEM:
+		src_addr = chan->dma_config.src_addr;
+
+		/* Set device data size */
+		src_bus_width = stm32_mdma_get_width(chan, src_addr_width);
+		if (src_bus_width < 0)
+			return src_bus_width;
+		ctcr &= ~STM32_MDMA_CTCR_SSIZE_MASK;
+		ctcr |= STM32_MDMA_CTCR_SSIZE(src_bus_width);
+
+		/* Set device burst value */
+		src_best_burst = stm32_mdma_get_best_burst(buf_len, tlen,
+							   src_maxburst,
+							   src_addr_width);
+		ctcr &= ~STM32_MDMA_CTCR_SBURST_MASK;
+		ctcr |= STM32_MDMA_CTCR_SBURST((ilog2(src_best_burst)));
+
+		/* Set memory data size */
+		dst_addr_width = stm32_mdma_get_max_width(addr, buf_len, tlen);
+		chan->mem_width = dst_addr_width;
+		dst_bus_width = stm32_mdma_get_width(chan, dst_addr_width);
+		if (dst_bus_width < 0)
+			return dst_bus_width;
+		ctcr &= ~(STM32_MDMA_CTCR_DSIZE_MASK |
+			STM32_MDMA_CTCR_DINCOS_MASK);
+		ctcr |= STM32_MDMA_CTCR_DSIZE(dst_bus_width) |
+			STM32_MDMA_CTCR_DINCOS(dst_bus_width);
+
+		/* Set memory burst value */
+		dst_maxburst = STM32_MDMA_MAX_BUF_LEN / dst_addr_width;
+		dst_best_burst = stm32_mdma_get_best_burst(buf_len, tlen,
+							   dst_maxburst,
+							   dst_addr_width);
+		ctcr &= ~STM32_MDMA_CTCR_DBURST_MASK;
+		ctcr |= STM32_MDMA_CTCR_DBURST((ilog2(dst_best_burst)));
+
+		/* Select bus */
+		stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_SBUS,
+				   src_addr);
+
+		if (dst_bus_width != src_bus_width)
+			ctcr |= STM32_MDMA_CTCR_PKE;
+
+		/* Set source address */
+		stm32_mdma_write(dmadev, STM32_MDMA_CSAR(chan->id), src_addr);
+		break;
+
+	default:
+		dev_err(chan2dev(chan), "Dma direction is not supported\n");
+		return -EINVAL;
+	}
+
+	*mdma_ccr = ccr;
+	*mdma_ctcr = ctcr;
+	*mdma_ctbr = ctbr;
+
+	return 0;
+}
+
+static void stm32_mdma_dump_hwdesc(struct stm32_mdma_chan *chan,
+				   struct stm32_mdma_hwdesc *hwdesc)
+{
+	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);
+}
+
+static void stm32_mdma_setup_hwdesc(struct stm32_mdma_chan *chan,
+				    struct stm32_mdma_desc *desc,
+				    enum dma_transfer_direction dir, u32 count,
+				    dma_addr_t src_addr, dma_addr_t dst_addr,
+				    u32 len, u32 ctcr, u32 ctbr, bool is_last,
+				    bool is_first, bool is_cyclic)
+{
+	struct stm32_mdma_chan_config *config = &chan->chan_config;
+	struct stm32_mdma_hwdesc *hwdesc;
+	u32 next = count + 1;
+
+	hwdesc = &desc->hwdesc[count];
+	hwdesc->ctcr = ctcr;
+	hwdesc->cbndtr &= ~(STM32_MDMA_CBNDTR_BRC_MK |
+			STM32_MDMA_CBNDTR_BRDUM |
+			STM32_MDMA_CBNDTR_BRSUM |
+			STM32_MDMA_CBNDTR_BNDT_MASK);
+	hwdesc->cbndtr |= STM32_MDMA_CBNDTR_BNDT(len);
+	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;
+		else
+			hwdesc->clar = 0;
+	}
+
+	stm32_mdma_dump_hwdesc(chan, hwdesc);
+}
+
+static int stm32_mdma_setup_xfer(struct stm32_mdma_chan *chan,
+				 struct stm32_mdma_desc *desc,
+				 struct scatterlist *sgl, u32 sg_len,
+				 enum dma_transfer_direction direction)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct dma_slave_config *dma_config = &chan->dma_config;
+	struct scatterlist *sg;
+	dma_addr_t src_addr, dst_addr;
+	u32 ccr, ctcr, ctbr;
+	int i, ret = 0;
+
+	for_each_sg(sgl, sg, sg_len, i) {
+		if (sg_dma_len(sg) > STM32_MDMA_MAX_BLOCK_LEN) {
+			dev_err(chan2dev(chan), "Invalid block len\n");
+			return -EINVAL;
+		}
+
+		if (direction == DMA_MEM_TO_DEV) {
+			src_addr = sg_dma_address(sg);
+			dst_addr = dma_config->dst_addr;
+			ret = stm32_mdma_set_xfer_param(chan, direction, &ccr,
+							&ctcr, &ctbr, src_addr,
+							sg_dma_len(sg));
+			stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_SBUS,
+					   src_addr);
+		} else {
+			src_addr = dma_config->src_addr;
+			dst_addr = sg_dma_address(sg);
+			ret = stm32_mdma_set_xfer_param(chan, direction, &ccr,
+							&ctcr, &ctbr, dst_addr,
+							sg_dma_len(sg));
+			stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_DBUS,
+					   dst_addr);
+		}
+
+		if (ret < 0)
+			return ret;
+
+		stm32_mdma_setup_hwdesc(chan, desc, direction, i, src_addr,
+					dst_addr, sg_dma_len(sg), ctcr, ctbr,
+					i == sg_len - 1, i == 0, false);
+	}
+
+	/* Enable interrupts */
+	ccr &= ~STM32_MDMA_CCR_IRQ_MASK;
+	ccr |= STM32_MDMA_CCR_TEIE | STM32_MDMA_CCR_CTCIE;
+	if (sg_len > 1)
+		ccr |= STM32_MDMA_CCR_BTIE;
+	desc->ccr = ccr;
+
+	return 0;
+}
+
+static struct dma_async_tx_descriptor *
+stm32_mdma_prep_slave_sg(struct dma_chan *c, struct scatterlist *sgl,
+			 u32 sg_len, enum dma_transfer_direction direction,
+			 unsigned long flags, void *context)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_desc *desc;
+	int ret;
+
+	/*
+	 * Once DMA is in setup cyclic mode the channel we cannot assign this
+	 * channel anymore. The DMA channel needs to be aborted or terminated
+	 * for allowing another request.
+	 */
+	if (chan->desc && chan->desc->cyclic) {
+		dev_err(chan2dev(chan),
+			"Request not allowed when dma in cyclic mode\n");
+		return NULL;
+	}
+
+	desc = stm32_mdma_alloc_desc(chan, sg_len);
+	if (!desc)
+		return NULL;
+
+	ret = stm32_mdma_setup_xfer(chan, desc, sgl, sg_len, direction);
+	if (ret < 0)
+		goto xfer_setup_err;
+
+	desc->cyclic = false;
+
+	return vchan_tx_prep(&chan->vchan, &desc->vdesc, flags);
+
+xfer_setup_err:
+	dma_pool_free(chan->desc_pool, &desc->hwdesc, desc->hwdesc_phys);
+	kfree(desc);
+	return NULL;
+}
+
+static struct dma_async_tx_descriptor *
+stm32_mdma_prep_dma_cyclic(struct dma_chan *c, dma_addr_t buf_addr,
+			   size_t buf_len, size_t period_len,
+			   enum dma_transfer_direction direction,
+			   unsigned long flags)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct dma_slave_config *dma_config = &chan->dma_config;
+	struct stm32_mdma_desc *desc;
+	dma_addr_t src_addr, dst_addr;
+	u32 ccr, ctcr, ctbr, count;
+	int i, ret;
+
+	/*
+	 * Once DMA is in setup cyclic mode the channel we cannot assign this
+	 * channel anymore. The DMA channel needs to be aborted or terminated
+	 * for allowing another request.
+	 */
+	if (chan->desc && chan->desc->cyclic) {
+		dev_err(chan2dev(chan),
+			"Request not allowed when dma in cyclic mode\n");
+		return NULL;
+	}
+
+	if (!buf_len || !period_len || period_len > STM32_MDMA_MAX_BLOCK_LEN) {
+		dev_err(chan2dev(chan), "Invalid buffer/period len\n");
+		return NULL;
+	}
+
+	if (buf_len % period_len) {
+		dev_err(chan2dev(chan), "buf_len not multiple of period_len\n");
+		return NULL;
+	}
+
+	count = buf_len / period_len;
+
+	desc = stm32_mdma_alloc_desc(chan, count);
+	if (!desc)
+		return NULL;
+
+	/* Select bus */
+	if (direction == DMA_MEM_TO_DEV) {
+		src_addr = buf_addr;
+		ret = stm32_mdma_set_xfer_param(chan, direction, &ccr, &ctcr,
+						&ctbr, src_addr, period_len);
+		stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_SBUS,
+				   src_addr);
+	} else {
+		dst_addr = buf_addr;
+		ret = stm32_mdma_set_xfer_param(chan, direction, &ccr, &ctcr,
+						&ctbr, dst_addr, period_len);
+		stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_DBUS,
+				   dst_addr);
+	}
+
+	if (ret < 0)
+		goto xfer_setup_err;
+
+	/* Enable interrupts */
+	ccr &= ~STM32_MDMA_CCR_IRQ_MASK;
+	ccr |= STM32_MDMA_CCR_TEIE | STM32_MDMA_CCR_CTCIE | STM32_MDMA_CCR_BTIE;
+	desc->ccr = ccr;
+
+	/* Configure hwdesc list */
+	for (i = 0; i < count; i++) {
+		if (direction == DMA_MEM_TO_DEV) {
+			src_addr = buf_addr + i * period_len;
+			dst_addr = dma_config->dst_addr;
+		} else {
+			src_addr = dma_config->src_addr;
+			dst_addr = buf_addr + i * period_len;
+		}
+
+		stm32_mdma_setup_hwdesc(chan, desc, direction, i, src_addr,
+					dst_addr, period_len, ctcr, ctbr,
+					i == count - 1, i == 0, true);
+	}
+
+	desc->cyclic = true;
+
+	return vchan_tx_prep(&chan->vchan, &desc->vdesc, flags);
+
+xfer_setup_err:
+	dma_pool_free(chan->desc_pool, &desc->hwdesc, desc->hwdesc_phys);
+	kfree(desc);
+	return NULL;
+}
+
+static struct dma_async_tx_descriptor *
+stm32_mdma_prep_dma_memcpy(struct dma_chan *c, dma_addr_t dest, dma_addr_t src,
+			   size_t len, unsigned long flags)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	enum dma_slave_buswidth max_width;
+	struct stm32_mdma_desc *desc;
+	struct stm32_mdma_hwdesc *hwdesc;
+	u32 ccr, ctcr, ctbr, cbndtr, count, max_burst, mdma_burst;
+	u32 best_burst, tlen;
+	size_t xfer_count, offset;
+	int src_bus_width, dst_bus_width;
+	int i;
+
+	/*
+	 * Once DMA is in setup cyclic mode the channel we cannot assign this
+	 * channel anymore. The DMA channel needs to be aborted or terminated
+	 * to allow another request
+	 */
+	if (chan->desc && chan->desc->cyclic) {
+		dev_err(chan2dev(chan),
+			"Request not allowed when dma in cyclic mode\n");
+		return NULL;
+	}
+
+	count = DIV_ROUND_UP(len, STM32_MDMA_MAX_BLOCK_LEN);
+	desc = stm32_mdma_alloc_desc(chan, count);
+	if (!desc)
+		return NULL;
+
+	ccr = stm32_mdma_read(dmadev, STM32_MDMA_CCR(chan->id));
+	ctcr = stm32_mdma_read(dmadev, STM32_MDMA_CTCR(chan->id));
+	ctbr = stm32_mdma_read(dmadev, STM32_MDMA_CTBR(chan->id));
+	cbndtr = stm32_mdma_read(dmadev, STM32_MDMA_CBNDTR(chan->id));
+
+	/* Enable sw req, some interrupts and clear other bits */
+	ccr &= ~(STM32_MDMA_CCR_WEX | STM32_MDMA_CCR_HEX |
+		 STM32_MDMA_CCR_BEX | STM32_MDMA_CCR_PL_MASK |
+		 STM32_MDMA_CCR_IRQ_MASK);
+	ccr |= STM32_MDMA_CCR_TEIE;
+
+	/* Enable SW request mode, dest/src inc and clear other bits */
+	ctcr &= ~(STM32_MDMA_CTCR_BWM | STM32_MDMA_CTCR_TRGM_MSK |
+		  STM32_MDMA_CTCR_PAM_MASK | STM32_MDMA_CTCR_PKE |
+		  STM32_MDMA_CTCR_TLEN_MSK | STM32_MDMA_CTCR_DBURST_MASK |
+		  STM32_MDMA_CTCR_SBURST_MASK | STM32_MDMA_CTCR_DINCOS_MASK |
+		  STM32_MDMA_CTCR_SINCOS_MASK | STM32_MDMA_CTCR_DSIZE_MASK |
+		  STM32_MDMA_CTCR_SSIZE_MASK | STM32_MDMA_CTCR_DINC_MASK |
+		  STM32_MDMA_CTCR_SINC_MASK);
+	ctcr |= STM32_MDMA_CTCR_SWRM | STM32_MDMA_CTCR_SINC(STM32_MDMA_INC) |
+		STM32_MDMA_CTCR_DINC(STM32_MDMA_INC);
+
+	/* Reset HW request */
+	ctbr &= ~STM32_MDMA_CTBR_TSEL_MASK;
+
+	/* Select bus */
+	stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_SBUS, src);
+	stm32_mdma_set_bus(dmadev, &ctbr, STM32_MDMA_CTBR_DBUS, dest);
+
+	/* Clear CBNDTR registers */
+	cbndtr &= ~(STM32_MDMA_CBNDTR_BRC_MK | STM32_MDMA_CBNDTR_BRDUM |
+			STM32_MDMA_CBNDTR_BRSUM | STM32_MDMA_CBNDTR_BNDT_MASK);
+
+	if (len <= STM32_MDMA_MAX_BLOCK_LEN) {
+		cbndtr |= STM32_MDMA_CBNDTR_BNDT(len);
+		if (len <= STM32_MDMA_MAX_BUF_LEN) {
+			/* Setup a buffer transfer */
+			ccr |= STM32_MDMA_CCR_TCIE | STM32_MDMA_CCR_CTCIE;
+			ctcr |= STM32_MDMA_CTCR_TRGM(STM32_MDMA_BUFFER);
+		} else {
+			/* Setup a block transfer */
+			ccr |= STM32_MDMA_CCR_BTIE | STM32_MDMA_CCR_CTCIE;
+			ctcr |= STM32_MDMA_CTCR_TRGM(STM32_MDMA_BLOCK);
+		}
+
+		tlen = STM32_MDMA_MAX_BUF_LEN;
+		ctcr |= STM32_MDMA_CTCR_TLEN((tlen - 1));
+
+		/* Set source best burst size */
+		max_width = stm32_mdma_get_max_width(src, len, tlen);
+		src_bus_width = stm32_mdma_get_width(chan, max_width);
+
+		max_burst = tlen / max_width;
+		best_burst = stm32_mdma_get_best_burst(len, tlen, max_burst,
+						       max_width);
+		mdma_burst = ilog2(best_burst);
+
+		ctcr |= STM32_MDMA_CTCR_SBURST(mdma_burst) |
+			STM32_MDMA_CTCR_SSIZE(src_bus_width) |
+			STM32_MDMA_CTCR_SINCOS(src_bus_width);
+
+		/* Set destination best burst size */
+		max_width = stm32_mdma_get_max_width(dest, len, tlen);
+		dst_bus_width = stm32_mdma_get_width(chan, max_width);
+
+		max_burst = tlen / max_width;
+		best_burst = stm32_mdma_get_best_burst(len, tlen, max_burst,
+						       max_width);
+		mdma_burst = ilog2(best_burst);
+
+		ctcr |= STM32_MDMA_CTCR_DBURST(mdma_burst) |
+			STM32_MDMA_CTCR_DSIZE(dst_bus_width) |
+			STM32_MDMA_CTCR_DINCOS(dst_bus_width);
+
+		if (dst_bus_width != src_bus_width)
+			ctcr |= STM32_MDMA_CTCR_PKE;
+
+		/* Prepare hardware descriptor */
+		hwdesc = desc->hwdesc;
+		hwdesc->ctcr = ctcr;
+		hwdesc->cbndtr = cbndtr;
+		hwdesc->csar = src;
+		hwdesc->cdar = dest;
+		hwdesc->cbrur = 0;
+		hwdesc->clar = 0;
+		hwdesc->ctbr = ctbr;
+		hwdesc->cmar = 0;
+		hwdesc->cmdr = 0;
+
+		stm32_mdma_dump_hwdesc(chan, hwdesc);
+	} else {
+		/* Setup a LLI transfer */
+		ctcr |= STM32_MDMA_CTCR_TRGM(STM32_MDMA_LINKED_LIST) |
+			STM32_MDMA_CTCR_TLEN((STM32_MDMA_MAX_BUF_LEN - 1));
+		ccr |= STM32_MDMA_CCR_BTIE | STM32_MDMA_CCR_CTCIE;
+		tlen = STM32_MDMA_MAX_BUF_LEN;
+
+		for (i = 0, offset = 0; offset < len;
+		     i++, offset += xfer_count) {
+			xfer_count = min_t(size_t, len - offset,
+					   STM32_MDMA_MAX_BLOCK_LEN);
+
+			/* Set source best burst size */
+			max_width = stm32_mdma_get_max_width(src, len, tlen);
+			src_bus_width = stm32_mdma_get_width(chan, max_width);
+
+			max_burst = tlen / max_width;
+			best_burst = stm32_mdma_get_best_burst(len, tlen,
+							       max_burst,
+							       max_width);
+			mdma_burst = ilog2(best_burst);
+
+			ctcr |= STM32_MDMA_CTCR_SBURST(mdma_burst) |
+				STM32_MDMA_CTCR_SSIZE(src_bus_width) |
+				STM32_MDMA_CTCR_SINCOS(src_bus_width);
+
+			/* Set destination best burst size */
+			max_width = stm32_mdma_get_max_width(dest, len, tlen);
+			dst_bus_width = stm32_mdma_get_width(chan, max_width);
+
+			max_burst = tlen / max_width;
+			best_burst = stm32_mdma_get_best_burst(len, tlen,
+							       max_burst,
+							       max_width);
+			mdma_burst = ilog2(best_burst);
+
+			ctcr |= STM32_MDMA_CTCR_DBURST(mdma_burst) |
+				STM32_MDMA_CTCR_DSIZE(dst_bus_width) |
+				STM32_MDMA_CTCR_DINCOS(dst_bus_width);
+
+			if (dst_bus_width != src_bus_width)
+				ctcr |= STM32_MDMA_CTCR_PKE;
+
+			/* Prepare hardware descriptor */
+			stm32_mdma_setup_hwdesc(chan, desc, DMA_MEM_TO_MEM, i,
+						src + offset, dest + offset,
+						xfer_count, ctcr, ctbr,
+						i == count - 1, i == 0, false);
+		}
+	}
+
+	desc->ccr = ccr;
+
+	desc->cyclic = false;
+
+	return vchan_tx_prep(&chan->vchan, &desc->vdesc, flags);
+}
+
+static void stm32_mdma_dump_reg(struct stm32_mdma_chan *chan)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+
+	dev_dbg(chan2dev(chan), "CCR:     0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CCR(chan->id)));
+	dev_dbg(chan2dev(chan), "CTCR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CTCR(chan->id)));
+	dev_dbg(chan2dev(chan), "CBNDTR:  0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CBNDTR(chan->id)));
+	dev_dbg(chan2dev(chan), "CSAR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CSAR(chan->id)));
+	dev_dbg(chan2dev(chan), "CDAR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CDAR(chan->id)));
+	dev_dbg(chan2dev(chan), "CBRUR:   0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CBRUR(chan->id)));
+	dev_dbg(chan2dev(chan), "CLAR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CLAR(chan->id)));
+	dev_dbg(chan2dev(chan), "CTBR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CTBR(chan->id)));
+	dev_dbg(chan2dev(chan), "CMAR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CMAR(chan->id)));
+	dev_dbg(chan2dev(chan), "CMDR:    0x%08x\n",
+		stm32_mdma_read(dmadev, STM32_MDMA_CMDR(chan->id)));
+}
+
+static void stm32_mdma_start_transfer(struct stm32_mdma_chan *chan)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct virt_dma_desc *vdesc;
+	struct stm32_mdma_hwdesc *hwdesc;
+	u32 id = chan->id;
+	u32 status, reg;
+
+	vdesc = vchan_next_desc(&chan->vchan);
+	if (!vdesc) {
+		chan->desc = NULL;
+		return;
+	}
+
+	chan->desc = to_stm32_mdma_desc(vdesc);
+	hwdesc = chan->desc->hwdesc;
+	chan->curr_hwdesc = 0;
+
+	stm32_mdma_write(dmadev, STM32_MDMA_CCR(id), chan->desc->ccr);
+	stm32_mdma_write(dmadev, STM32_MDMA_CTCR(id), hwdesc->ctcr);
+	stm32_mdma_write(dmadev, STM32_MDMA_CBNDTR(id), hwdesc->cbndtr);
+	stm32_mdma_write(dmadev, STM32_MDMA_CSAR(id), hwdesc->csar);
+	stm32_mdma_write(dmadev, STM32_MDMA_CDAR(id), hwdesc->cdar);
+	stm32_mdma_write(dmadev, STM32_MDMA_CBRUR(id), hwdesc->cbrur);
+	stm32_mdma_write(dmadev, STM32_MDMA_CLAR(id), hwdesc->clar);
+	stm32_mdma_write(dmadev, STM32_MDMA_CTBR(id), hwdesc->ctbr);
+	stm32_mdma_write(dmadev, STM32_MDMA_CMAR(id), hwdesc->cmar);
+	stm32_mdma_write(dmadev, STM32_MDMA_CMDR(id), hwdesc->cmdr);
+
+	/* Clear interrupt status if it is there */
+	status = stm32_mdma_read(dmadev, STM32_MDMA_CISR(id));
+	if (status)
+		stm32_mdma_set_bits(dmadev, STM32_MDMA_CIFCR(id), status);
+
+	stm32_mdma_dump_reg(chan);
+
+	/* Start DMA */
+	stm32_mdma_set_bits(dmadev, STM32_MDMA_CCR(id), STM32_MDMA_CCR_EN);
+
+	/* Set SW request in case of MEM2MEM transfer */
+	if (hwdesc->ctcr & STM32_MDMA_CTCR_SWRM) {
+		reg = STM32_MDMA_CCR(id);
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CCR_SWRQ);
+	}
+
+	chan->busy = true;
+
+	dev_dbg(chan2dev(chan), "vchan %p: started\n", &chan->vchan);
+}
+
+static void stm32_mdma_issue_pending(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	unsigned long flags;
+
+	spin_lock_irqsave(&chan->vchan.lock, flags);
+
+	if (!vchan_issue_pending(&chan->vchan))
+		goto end;
+
+	dev_dbg(chan2dev(chan), "vchan %p: issued\n", &chan->vchan);
+
+	if (!chan->desc && !chan->busy)
+		stm32_mdma_start_transfer(chan);
+
+end:
+	spin_unlock_irqrestore(&chan->vchan.lock, flags);
+}
+
+static int stm32_mdma_pause(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&chan->vchan.lock, flags);
+	ret = stm32_mdma_disable_chan(chan);
+	spin_unlock_irqrestore(&chan->vchan.lock, flags);
+
+	if (!ret)
+		dev_dbg(chan2dev(chan), "vchan %p: pause\n", &chan->vchan);
+
+	return ret;
+}
+
+static int stm32_mdma_resume(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	struct stm32_mdma_hwdesc *hwdesc;
+	unsigned long flags;
+	u32 status, reg;
+
+	hwdesc = &chan->desc->hwdesc[chan->curr_hwdesc];
+
+	spin_lock_irqsave(&chan->vchan.lock, flags);
+
+	/* Re-configure control register */
+	stm32_mdma_write(dmadev, STM32_MDMA_CCR(chan->id), chan->desc->ccr);
+
+	/* Clear interrupt status if it is there */
+	status = stm32_mdma_read(dmadev, STM32_MDMA_CISR(chan->id));
+	if (status)
+		stm32_mdma_set_bits(dmadev, STM32_MDMA_CIFCR(chan->id), status);
+
+	stm32_mdma_dump_reg(chan);
+
+	/* Re-start DMA */
+	reg = STM32_MDMA_CCR(chan->id);
+	stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CCR_EN);
+
+	/* Set SW request in case of MEM2MEM transfer */
+	if (hwdesc->ctcr & STM32_MDMA_CTCR_SWRM)
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CCR_SWRQ);
+
+	spin_unlock_irqrestore(&chan->vchan.lock, flags);
+
+	dev_dbg(chan2dev(chan), "vchan %p: resume\n", &chan->vchan);
+
+	return 0;
+}
+
+static int stm32_mdma_terminate_all(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	unsigned long flags;
+	LIST_HEAD(head);
+
+	spin_lock_irqsave(&chan->vchan.lock, flags);
+	if (chan->busy) {
+		stm32_mdma_stop(chan);
+		chan->desc = NULL;
+	}
+	vchan_get_all_descriptors(&chan->vchan, &head);
+	spin_unlock_irqrestore(&chan->vchan.lock, flags);
+
+	vchan_dma_desc_free_list(&chan->vchan, &head);
+
+	return 0;
+}
+
+static void stm32_mdma_synchronize(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+
+	vchan_synchronize(&chan->vchan);
+}
+
+static int stm32_mdma_slave_config(struct dma_chan *c,
+				   struct dma_slave_config *config)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+
+	memcpy(&chan->dma_config, config, sizeof(*config));
+
+	return 0;
+}
+
+static size_t stm32_mdma_desc_residue(struct stm32_mdma_chan *chan,
+				      struct stm32_mdma_desc *desc,
+				      u32 curr_hwdesc)
+{
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	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];
+
+		residue += STM32_MDMA_CBNDTR_BNDT(hwdesc->cbndtr);
+	}
+	cbndtr = stm32_mdma_read(dmadev, STM32_MDMA_CBNDTR(chan->id));
+	residue += cbndtr & STM32_MDMA_CBNDTR_BNDT_MASK;
+
+	if (!chan->mem_burst)
+		return residue;
+
+	burst_size = chan->mem_burst * chan->mem_width;
+	modulo = residue % burst_size;
+	if (modulo)
+		residue = residue - modulo + burst_size;
+
+	return residue;
+}
+
+static enum dma_status stm32_mdma_tx_status(struct dma_chan *c,
+					    dma_cookie_t cookie,
+					    struct dma_tx_state *state)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct virt_dma_desc *vdesc;
+	enum dma_status status;
+	unsigned long flags;
+	u32 residue = 0;
+
+	status = dma_cookie_status(c, cookie, state);
+	if ((status == DMA_COMPLETE) || (!state))
+		return status;
+
+	spin_lock_irqsave(&chan->vchan.lock, flags);
+
+	vdesc = vchan_find_desc(&chan->vchan, cookie);
+	if (chan->desc && cookie == chan->desc->vdesc.tx.cookie)
+		residue = stm32_mdma_desc_residue(chan, chan->desc,
+						  chan->curr_hwdesc);
+	else if (vdesc)
+		residue = stm32_mdma_desc_residue(chan,
+						  to_stm32_mdma_desc(vdesc), 0);
+	dma_set_residue(state, residue);
+
+	spin_unlock_irqrestore(&chan->vchan.lock, flags);
+
+	return status;
+}
+
+static void stm32_mdma_xfer_end(struct stm32_mdma_chan *chan)
+{
+	list_del(&chan->desc->vdesc.node);
+	vchan_cookie_complete(&chan->desc->vdesc);
+	chan->desc = NULL;
+	chan->busy = false;
+
+	/* Start the next transfer if this driver has a next desc */
+	stm32_mdma_start_transfer(chan);
+}
+
+static irqreturn_t stm32_mdma_irq_handler(int irq, void *devid)
+{
+	struct stm32_mdma_device *dmadev = devid;
+	struct stm32_mdma_chan *chan = devid;
+	u32 reg, id, ien, status, flag;
+
+	/* Find out which channel generates the interrupt */
+	status = readl_relaxed(dmadev->base + STM32_MDMA_GISR0);
+	if (status) {
+		id = __ffs(status);
+	} else {
+		status = readl_relaxed(dmadev->base + STM32_MDMA_GISR1);
+		if (!status) {
+			dev_dbg(mdma2dev(dmadev), "spurious it\n");
+			return IRQ_NONE;
+		}
+		id = __ffs(status);
+		/*
+		 * As GISR0 provides status for channel id from 0 to 31,
+		 * so GISR1 provides status for channel id from 32 to 62
+		 */
+		id += 32;
+	}
+
+	chan = &dmadev->chan[id];
+	if (!chan) {
+		dev_err(chan2dev(chan), "MDMA channel not initialized\n");
+		goto exit;
+	}
+
+	/* Handle interrupt for the channel */
+	spin_lock(&chan->vchan.lock);
+	status = stm32_mdma_read(dmadev, STM32_MDMA_CISR(chan->id));
+	ien = stm32_mdma_read(dmadev, STM32_MDMA_CCR(chan->id));
+	ien &= STM32_MDMA_CCR_IRQ_MASK;
+	ien >>= 1;
+
+	if (!(status & ien)) {
+		spin_unlock(&chan->vchan.lock);
+		dev_dbg(chan2dev(chan),
+			"spurious it (status=0x%04x, ien=0x%04x)\n",
+			status, ien);
+		return IRQ_NONE;
+	}
+
+	flag = __ffs(status & ien);
+	reg = STM32_MDMA_CIFCR(chan->id);
+
+	switch (1 << flag) {
+	case STM32_MDMA_CISR_TEIF:
+		id = chan->id;
+		status = readl_relaxed(dmadev->base + STM32_MDMA_CESR(id));
+		dev_err(chan2dev(chan), "Transfer Err: stat=0x%08x\n", status);
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CIFCR_CTEIF);
+		break;
+
+	case STM32_MDMA_CISR_CTCIF:
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CIFCR_CCTCIF);
+		stm32_mdma_xfer_end(chan);
+		break;
+
+	case STM32_MDMA_CISR_BRTIF:
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CIFCR_CBRTIF);
+		break;
+
+	case STM32_MDMA_CISR_BTIF:
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CIFCR_CBTIF);
+		chan->curr_hwdesc++;
+		if (chan->desc && chan->desc->cyclic) {
+			if (chan->curr_hwdesc == chan->desc->count)
+				chan->curr_hwdesc = 0;
+			vchan_cyclic_callback(&chan->desc->vdesc);
+		}
+		break;
+
+	case STM32_MDMA_CISR_TCIF:
+		stm32_mdma_set_bits(dmadev, reg, STM32_MDMA_CIFCR_CLTCIF);
+		break;
+
+	default:
+		dev_err(chan2dev(chan), "it %d unhandled (status=0x%04x)\n",
+			1 << flag, status);
+	}
+
+	spin_unlock(&chan->vchan.lock);
+
+exit:
+	return IRQ_HANDLED;
+}
+
+static int stm32_mdma_alloc_chan_resources(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	int ret;
+
+	chan->desc_pool = dmam_pool_create(dev_name(&c->dev->device),
+					   c->device->dev,
+					   sizeof(struct stm32_mdma_hwdesc),
+					  __alignof__(struct stm32_mdma_hwdesc),
+					   0);
+	if (!chan->desc_pool) {
+		dev_err(chan2dev(chan), "failed to allocate descriptor pool\n");
+		return -ENOMEM;
+	}
+
+	ret = clk_prepare_enable(dmadev->clk);
+	if (ret < 0) {
+		dev_err(chan2dev(chan), "clk_prepare_enable failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = stm32_mdma_disable_chan(chan);
+	if (ret < 0)
+		clk_disable_unprepare(dmadev->clk);
+
+	return ret;
+}
+
+static void stm32_mdma_free_chan_resources(struct dma_chan *c)
+{
+	struct stm32_mdma_chan *chan = to_stm32_mdma_chan(c);
+	struct stm32_mdma_device *dmadev = stm32_mdma_get_dev(chan);
+	unsigned long flags;
+
+	dev_dbg(chan2dev(chan), "Freeing channel %d\n", chan->id);
+
+	if (chan->busy) {
+		spin_lock_irqsave(&chan->vchan.lock, flags);
+		stm32_mdma_stop(chan);
+		chan->desc = NULL;
+		spin_unlock_irqrestore(&chan->vchan.lock, flags);
+	}
+
+	clk_disable_unprepare(dmadev->clk);
+	vchan_free_chan_resources(to_virt_chan(c));
+	dmam_pool_destroy(chan->desc_pool);
+	chan->desc_pool = NULL;
+}
+
+static struct dma_chan *stm32_mdma_of_xlate(struct of_phandle_args *dma_spec,
+					    struct of_dma *ofdma)
+{
+	struct stm32_mdma_device *dmadev = ofdma->of_dma_data;
+	struct stm32_mdma_chan *chan;
+	struct dma_chan *c;
+	struct stm32_mdma_chan_config config;
+
+	if (dma_spec->args_count < 5) {
+		dev_err(mdma2dev(dmadev), "Bad number of args\n");
+		return NULL;
+	}
+
+	config.request = dma_spec->args[0];
+	config.priority_level = dma_spec->args[1];
+	config.transfer_config = dma_spec->args[2];
+	config.mask_addr = dma_spec->args[3];
+	config.mask_data = dma_spec->args[4];
+
+	if (config.request >= dmadev->nr_requests) {
+		dev_err(mdma2dev(dmadev), "Bad request line\n");
+		return NULL;
+	}
+
+	if (config.priority_level > STM32_MDMA_VERY_HIGH_PRIORITY) {
+		dev_err(mdma2dev(dmadev), "Priority level not supported\n");
+		return NULL;
+	}
+
+	c = dma_get_any_slave_channel(&dmadev->ddev);
+	if (!c) {
+		dev_err(mdma2dev(dmadev), "No more channel avalaible\n");
+		return NULL;
+	}
+
+	chan = to_stm32_mdma_chan(c);
+	chan->chan_config = config;
+
+	return c;
+}
+
+static const struct of_device_id stm32_mdma_of_match[] = {
+	{ .compatible = "st,stm32h7-mdma", },
+	{ /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, stm32_mdma_of_match);
+
+static int stm32_mdma_probe(struct platform_device *pdev)
+{
+	struct stm32_mdma_chan *chan;
+	struct stm32_mdma_device *dmadev;
+	struct dma_device *dd;
+	struct device_node *of_node;
+	struct resource *res;
+	u32 nr_channels, nr_requests;
+	int i, count, ret;
+
+	of_node = pdev->dev.of_node;
+	if (!of_node)
+		return -ENODEV;
+
+	ret = device_property_read_u32(&pdev->dev, "dma-channels",
+				       &nr_channels);
+	if (ret) {
+		nr_channels = STM32_MDMA_MAX_CHANNELS;
+		dev_warn(&pdev->dev, "MDMA defaulting on %i channels\n",
+			 nr_channels);
+	}
+
+	ret = device_property_read_u32(&pdev->dev, "dma-requests",
+				       &nr_requests);
+	if (ret) {
+		nr_requests = STM32_MDMA_MAX_REQUESTS;
+		dev_warn(&pdev->dev, "MDMA defaulting on %i request lines\n",
+			 nr_requests);
+	}
+
+	count = device_property_read_u32_array(&pdev->dev, "st,ahb-addr-masks",
+					       NULL, 0);
+	if (count < 0)
+		count = 0;
+
+	dmadev = devm_kzalloc(&pdev->dev, sizeof(*dmadev) + sizeof(u32) * count,
+			      GFP_KERNEL);
+	if (!dmadev)
+		return -ENOMEM;
+
+	dmadev->nr_channels = nr_channels;
+	dmadev->nr_requests = nr_requests;
+	device_property_read_u32_array(&pdev->dev, "st,ahb-addr-masks",
+				       dmadev->ahb_addr_masks,
+				       count);
+	dmadev->nr_ahb_addr_masks = count;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	dmadev->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(dmadev->base))
+		return PTR_ERR(dmadev->base);
+
+	dmadev->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(dmadev->clk)) {
+		ret = PTR_ERR(dmadev->clk);
+		if (ret == -EPROBE_DEFER)
+			dev_info(&pdev->dev, "Missing controller clock\n");
+		return ret;
+	}
+
+	dmadev->rst = devm_reset_control_get(&pdev->dev, NULL);
+	if (!IS_ERR(dmadev->rst)) {
+		reset_control_assert(dmadev->rst);
+		udelay(2);
+		reset_control_deassert(dmadev->rst);
+	}
+
+	dd = &dmadev->ddev;
+	dma_cap_set(DMA_SLAVE, dd->cap_mask);
+	dma_cap_set(DMA_PRIVATE, dd->cap_mask);
+	dma_cap_set(DMA_CYCLIC, dd->cap_mask);
+	dma_cap_set(DMA_MEMCPY, dd->cap_mask);
+	dd->device_alloc_chan_resources = stm32_mdma_alloc_chan_resources;
+	dd->device_free_chan_resources = stm32_mdma_free_chan_resources;
+	dd->device_tx_status = stm32_mdma_tx_status;
+	dd->device_issue_pending = stm32_mdma_issue_pending;
+	dd->device_prep_slave_sg = stm32_mdma_prep_slave_sg;
+	dd->device_prep_dma_cyclic = stm32_mdma_prep_dma_cyclic;
+	dd->device_prep_dma_memcpy = stm32_mdma_prep_dma_memcpy;
+	dd->device_config = stm32_mdma_slave_config;
+	dd->device_pause = stm32_mdma_pause;
+	dd->device_resume = stm32_mdma_resume;
+	dd->device_terminate_all = stm32_mdma_terminate_all;
+	dd->device_synchronize = stm32_mdma_synchronize;
+	dd->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+		BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+		BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+		BIT(DMA_SLAVE_BUSWIDTH_8_BYTES);
+	dd->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+		BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+		BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+		BIT(DMA_SLAVE_BUSWIDTH_8_BYTES);
+	dd->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV) |
+		BIT(DMA_MEM_TO_MEM);
+	dd->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
+	dd->max_burst = STM32_MDMA_MAX_BURST;
+	dd->dev = &pdev->dev;
+	INIT_LIST_HEAD(&dd->channels);
+
+	for (i = 0; i < dmadev->nr_channels; i++) {
+		chan = &dmadev->chan[i];
+		chan->id = i;
+		chan->vchan.desc_free = stm32_mdma_desc_free;
+		vchan_init(&chan->vchan, dd);
+	}
+
+	dmadev->irq = platform_get_irq(pdev, 0);
+	if (dmadev->irq < 0) {
+		dev_err(&pdev->dev, "failed to get IRQ\n");
+		return dmadev->irq;
+	}
+
+	ret = devm_request_irq(&pdev->dev, dmadev->irq, stm32_mdma_irq_handler,
+			       0, dev_name(&pdev->dev), dmadev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to request IRQ\n");
+		return ret;
+	}
+
+	ret = dma_async_device_register(dd);
+	if (ret)
+		return ret;
+
+	ret = of_dma_controller_register(of_node, stm32_mdma_of_xlate, dmadev);
+	if (ret < 0) {
+		dev_err(&pdev->dev,
+			"STM32 MDMA DMA OF registration failed %d\n", ret);
+		goto err_unregister;
+	}
+
+	platform_set_drvdata(pdev, dmadev);
+
+	dev_info(&pdev->dev, "STM32 MDMA driver registered\n");
+
+	return 0;
+
+err_unregister:
+	dma_async_device_unregister(dd);
+
+	return ret;
+}
+
+static struct platform_driver stm32_mdma_driver = {
+	.probe = stm32_mdma_probe,
+	.driver = {
+		.name = "stm32-mdma",
+		.of_match_table = stm32_mdma_of_match,
+	},
+};
+
+static int __init stm32_mdma_init(void)
+{
+	return platform_driver_register(&stm32_mdma_driver);
+}
+
+subsys_initcall(stm32_mdma_init);
+
+MODULE_DESCRIPTION("Driver for STM32 MDMA controller");
+MODULE_AUTHOR("M'boumba Cedric Madianga <cedric.madianga@gmail.com>");
+MODULE_AUTHOR("Pierre-Yves Mordret <pierre-yves.mordret@st.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index bcd496edc70f..0cd13f17fc11 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -42,12 +42,18 @@
 
 #define DMA_STAT		0x30
 
+/* Offset between DMA_IRQ_EN and DMA_IRQ_STAT limits number of channels */
+#define DMA_MAX_CHANNELS	(DMA_IRQ_CHAN_NR * 0x10 / 4)
+
 /*
  * sun8i specific registers
  */
 #define SUN8I_DMA_GATE		0x20
 #define SUN8I_DMA_GATE_ENABLE	0x4
 
+#define SUNXI_H3_SECURE_REG		0x20
+#define SUNXI_H3_DMA_GATE		0x28
+#define SUNXI_H3_DMA_GATE_ENABLE	0x4
 /*
  * Channels specific registers
  */
@@ -62,16 +68,19 @@
 #define DMA_CHAN_LLI_ADDR	0x08
 
 #define DMA_CHAN_CUR_CFG	0x0c
-#define DMA_CHAN_CFG_SRC_DRQ(x)		((x) & 0x1f)
+#define DMA_CHAN_MAX_DRQ		0x1f
+#define DMA_CHAN_CFG_SRC_DRQ(x)		((x) & DMA_CHAN_MAX_DRQ)
 #define DMA_CHAN_CFG_SRC_IO_MODE	BIT(5)
 #define DMA_CHAN_CFG_SRC_LINEAR_MODE	(0 << 5)
-#define DMA_CHAN_CFG_SRC_BURST(x)	(((x) & 0x3) << 7)
+#define DMA_CHAN_CFG_SRC_BURST_A31(x)	(((x) & 0x3) << 7)
+#define DMA_CHAN_CFG_SRC_BURST_H3(x)	(((x) & 0x3) << 6)
 #define DMA_CHAN_CFG_SRC_WIDTH(x)	(((x) & 0x3) << 9)
 
 #define DMA_CHAN_CFG_DST_DRQ(x)		(DMA_CHAN_CFG_SRC_DRQ(x) << 16)
 #define DMA_CHAN_CFG_DST_IO_MODE	(DMA_CHAN_CFG_SRC_IO_MODE << 16)
 #define DMA_CHAN_CFG_DST_LINEAR_MODE	(DMA_CHAN_CFG_SRC_LINEAR_MODE << 16)
-#define DMA_CHAN_CFG_DST_BURST(x)	(DMA_CHAN_CFG_SRC_BURST(x) << 16)
+#define DMA_CHAN_CFG_DST_BURST_A31(x)	(DMA_CHAN_CFG_SRC_BURST_A31(x) << 16)
+#define DMA_CHAN_CFG_DST_BURST_H3(x)	(DMA_CHAN_CFG_SRC_BURST_H3(x) << 16)
 #define DMA_CHAN_CFG_DST_WIDTH(x)	(DMA_CHAN_CFG_SRC_WIDTH(x) << 16)
 
 #define DMA_CHAN_CUR_SRC	0x10
@@ -90,6 +99,9 @@
 #define NORMAL_WAIT	8
 #define DRQ_SDRAM	1
 
+/* forward declaration */
+struct sun6i_dma_dev;
+
 /*
  * Hardware channels / ports representation
  *
@@ -111,7 +123,12 @@ struct sun6i_dma_config {
 	 * however these SoCs really have and need this bit, as seen in the
 	 * BSP kernel source code.
 	 */
-	bool gate_needed;
+	void (*clock_autogate_enable)(struct sun6i_dma_dev *);
+	void (*set_burst_length)(u32 *p_cfg, s8 src_burst, s8 dst_burst);
+	u32 src_burst_lengths;
+	u32 dst_burst_lengths;
+	u32 src_addr_widths;
+	u32 dst_addr_widths;
 };
 
 /*
@@ -175,6 +192,9 @@ struct sun6i_dma_dev {
 	struct sun6i_pchan	*pchans;
 	struct sun6i_vchan	*vchans;
 	const struct sun6i_dma_config *cfg;
+	u32			num_pchans;
+	u32			num_vchans;
+	u32			max_request;
 };
 
 static struct device *chan2dev(struct dma_chan *chan)
@@ -251,8 +271,12 @@ static inline s8 convert_burst(u32 maxburst)
 	switch (maxburst) {
 	case 1:
 		return 0;
+	case 4:
+		return 1;
 	case 8:
 		return 2;
+	case 16:
+		return 3;
 	default:
 		return -EINVAL;
 	}
@@ -260,11 +284,29 @@ static inline s8 convert_burst(u32 maxburst)
 
 static inline s8 convert_buswidth(enum dma_slave_buswidth addr_width)
 {
-	if ((addr_width < DMA_SLAVE_BUSWIDTH_1_BYTE) ||
-	    (addr_width > DMA_SLAVE_BUSWIDTH_4_BYTES))
-		return -EINVAL;
+	return ilog2(addr_width);
+}
+
+static void sun6i_enable_clock_autogate_a23(struct sun6i_dma_dev *sdev)
+{
+	writel(SUN8I_DMA_GATE_ENABLE, sdev->base + SUN8I_DMA_GATE);
+}
+
+static void sun6i_enable_clock_autogate_h3(struct sun6i_dma_dev *sdev)
+{
+	writel(SUNXI_H3_DMA_GATE_ENABLE, sdev->base + SUNXI_H3_DMA_GATE);
+}
 
-	return addr_width >> 1;
+static void sun6i_set_burst_length_a31(u32 *p_cfg, s8 src_burst, s8 dst_burst)
+{
+	*p_cfg |= DMA_CHAN_CFG_SRC_BURST_A31(src_burst) |
+		  DMA_CHAN_CFG_DST_BURST_A31(dst_burst);
+}
+
+static void sun6i_set_burst_length_h3(u32 *p_cfg, s8 src_burst, s8 dst_burst)
+{
+	*p_cfg |= DMA_CHAN_CFG_SRC_BURST_H3(src_burst) |
+		  DMA_CHAN_CFG_DST_BURST_H3(dst_burst);
 }
 
 static size_t sun6i_get_chan_size(struct sun6i_pchan *pchan)
@@ -399,7 +441,6 @@ static int sun6i_dma_start_desc(struct sun6i_vchan *vchan)
 static void sun6i_dma_tasklet(unsigned long data)
 {
 	struct sun6i_dma_dev *sdev = (struct sun6i_dma_dev *)data;
-	const struct sun6i_dma_config *cfg = sdev->cfg;
 	struct sun6i_vchan *vchan;
 	struct sun6i_pchan *pchan;
 	unsigned int pchan_alloc = 0;
@@ -427,7 +468,7 @@ static void sun6i_dma_tasklet(unsigned long data)
 	}
 
 	spin_lock_irq(&sdev->lock);
-	for (pchan_idx = 0; pchan_idx < cfg->nr_max_channels; pchan_idx++) {
+	for (pchan_idx = 0; pchan_idx < sdev->num_pchans; pchan_idx++) {
 		pchan = &sdev->pchans[pchan_idx];
 
 		if (pchan->vchan || list_empty(&sdev->pending))
@@ -448,7 +489,7 @@ static void sun6i_dma_tasklet(unsigned long data)
 	}
 	spin_unlock_irq(&sdev->lock);
 
-	for (pchan_idx = 0; pchan_idx < cfg->nr_max_channels; pchan_idx++) {
+	for (pchan_idx = 0; pchan_idx < sdev->num_pchans; pchan_idx++) {
 		if (!(pchan_alloc & BIT(pchan_idx)))
 			continue;
 
@@ -470,7 +511,7 @@ static irqreturn_t sun6i_dma_interrupt(int irq, void *dev_id)
 	int i, j, ret = IRQ_NONE;
 	u32 status;
 
-	for (i = 0; i < sdev->cfg->nr_max_channels / DMA_IRQ_CHAN_NR; i++) {
+	for (i = 0; i < sdev->num_pchans / DMA_IRQ_CHAN_NR; i++) {
 		status = readl(sdev->base + DMA_IRQ_STAT(i));
 		if (!status)
 			continue;
@@ -510,47 +551,49 @@ static int set_config(struct sun6i_dma_dev *sdev,
 			enum dma_transfer_direction direction,
 			u32 *p_cfg)
 {
+	enum dma_slave_buswidth src_addr_width, dst_addr_width;
+	u32 src_maxburst, dst_maxburst;
 	s8 src_width, dst_width, src_burst, dst_burst;
 
+	src_addr_width = sconfig->src_addr_width;
+	dst_addr_width = sconfig->dst_addr_width;
+	src_maxburst = sconfig->src_maxburst;
+	dst_maxburst = sconfig->dst_maxburst;
+
 	switch (direction) {
 	case DMA_MEM_TO_DEV:
-		src_burst = convert_burst(sconfig->src_maxburst ?
-					sconfig->src_maxburst : 8);
-		src_width = convert_buswidth(sconfig->src_addr_width !=
-						DMA_SLAVE_BUSWIDTH_UNDEFINED ?
-				sconfig->src_addr_width :
-				DMA_SLAVE_BUSWIDTH_4_BYTES);
-		dst_burst = convert_burst(sconfig->dst_maxburst);
-		dst_width = convert_buswidth(sconfig->dst_addr_width);
+		if (src_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+			src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		src_maxburst = src_maxburst ? src_maxburst : 8;
 		break;
 	case DMA_DEV_TO_MEM:
-		src_burst = convert_burst(sconfig->src_maxburst);
-		src_width = convert_buswidth(sconfig->src_addr_width);
-		dst_burst = convert_burst(sconfig->dst_maxburst ?
-					sconfig->dst_maxburst : 8);
-		dst_width = convert_buswidth(sconfig->dst_addr_width !=
-						DMA_SLAVE_BUSWIDTH_UNDEFINED ?
-				sconfig->dst_addr_width :
-				DMA_SLAVE_BUSWIDTH_4_BYTES);
+		if (dst_addr_width == DMA_SLAVE_BUSWIDTH_UNDEFINED)
+			dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+		dst_maxburst = dst_maxburst ? dst_maxburst : 8;
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	if (src_burst < 0)
-		return src_burst;
-	if (src_width < 0)
-		return src_width;
-	if (dst_burst < 0)
-		return dst_burst;
-	if (dst_width < 0)
-		return dst_width;
-
-	*p_cfg = DMA_CHAN_CFG_SRC_BURST(src_burst) |
-		DMA_CHAN_CFG_SRC_WIDTH(src_width) |
-		DMA_CHAN_CFG_DST_BURST(dst_burst) |
+	if (!(BIT(src_addr_width) & sdev->slave.src_addr_widths))
+		return -EINVAL;
+	if (!(BIT(dst_addr_width) & sdev->slave.dst_addr_widths))
+		return -EINVAL;
+	if (!(BIT(src_maxburst) & sdev->cfg->src_burst_lengths))
+		return -EINVAL;
+	if (!(BIT(dst_maxburst) & sdev->cfg->dst_burst_lengths))
+		return -EINVAL;
+
+	src_width = convert_buswidth(src_addr_width);
+	dst_width = convert_buswidth(dst_addr_width);
+	dst_burst = convert_burst(dst_maxburst);
+	src_burst = convert_burst(src_maxburst);
+
+	*p_cfg = DMA_CHAN_CFG_SRC_WIDTH(src_width) |
 		DMA_CHAN_CFG_DST_WIDTH(dst_width);
 
+	sdev->cfg->set_burst_length(p_cfg, src_burst, dst_burst);
+
 	return 0;
 }
 
@@ -593,11 +636,11 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 		DMA_CHAN_CFG_DST_DRQ(DRQ_SDRAM) |
 		DMA_CHAN_CFG_DST_LINEAR_MODE |
 		DMA_CHAN_CFG_SRC_LINEAR_MODE |
-		DMA_CHAN_CFG_SRC_BURST(burst) |
 		DMA_CHAN_CFG_SRC_WIDTH(width) |
-		DMA_CHAN_CFG_DST_BURST(burst) |
 		DMA_CHAN_CFG_DST_WIDTH(width);
 
+	sdev->cfg->set_burst_length(&v_lli->cfg, burst, burst);
+
 	sun6i_dma_lli_add(NULL, v_lli, p_lli, txd);
 
 	sun6i_dma_dump_lli(vchan, v_lli);
@@ -948,7 +991,7 @@ static struct dma_chan *sun6i_dma_of_xlate(struct of_phandle_args *dma_spec,
 	struct dma_chan *chan;
 	u8 port = dma_spec->args[0];
 
-	if (port > sdev->cfg->nr_max_requests)
+	if (port > sdev->max_request)
 		return NULL;
 
 	chan = dma_get_any_slave_channel(&sdev->slave);
@@ -981,7 +1024,7 @@ static inline void sun6i_dma_free(struct sun6i_dma_dev *sdev)
 {
 	int i;
 
-	for (i = 0; i < sdev->cfg->nr_max_vchans; i++) {
+	for (i = 0; i < sdev->num_vchans; i++) {
 		struct sun6i_vchan *vchan = &sdev->vchans[i];
 
 		list_del(&vchan->vc.chan.device_node);
@@ -1009,6 +1052,15 @@ static struct sun6i_dma_config sun6i_a31_dma_cfg = {
 	.nr_max_channels = 16,
 	.nr_max_requests = 30,
 	.nr_max_vchans   = 53,
+	.set_burst_length = sun6i_set_burst_length_a31,
+	.src_burst_lengths = BIT(1) | BIT(8),
+	.dst_burst_lengths = BIT(1) | BIT(8),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
 };
 
 /*
@@ -1020,24 +1072,76 @@ static struct sun6i_dma_config sun8i_a23_dma_cfg = {
 	.nr_max_channels = 8,
 	.nr_max_requests = 24,
 	.nr_max_vchans   = 37,
-	.gate_needed	 = true,
+	.clock_autogate_enable = sun6i_enable_clock_autogate_a23,
+	.set_burst_length = sun6i_set_burst_length_a31,
+	.src_burst_lengths = BIT(1) | BIT(8),
+	.dst_burst_lengths = BIT(1) | BIT(8),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
 };
 
 static struct sun6i_dma_config sun8i_a83t_dma_cfg = {
 	.nr_max_channels = 8,
 	.nr_max_requests = 28,
 	.nr_max_vchans   = 39,
+	.clock_autogate_enable = sun6i_enable_clock_autogate_a23,
+	.set_burst_length = sun6i_set_burst_length_a31,
+	.src_burst_lengths = BIT(1) | BIT(8),
+	.dst_burst_lengths = BIT(1) | BIT(8),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
 };
 
 /*
  * The H3 has 12 physical channels, a maximum DRQ port id of 27,
  * and a total of 34 usable source and destination endpoints.
+ * It also supports additional burst lengths and bus widths,
+ * and the burst length fields have different offsets.
  */
 
 static struct sun6i_dma_config sun8i_h3_dma_cfg = {
 	.nr_max_channels = 12,
 	.nr_max_requests = 27,
 	.nr_max_vchans   = 34,
+	.clock_autogate_enable = sun6i_enable_clock_autogate_h3,
+	.set_burst_length = sun6i_set_burst_length_h3,
+	.src_burst_lengths = BIT(1) | BIT(4) | BIT(8) | BIT(16),
+	.dst_burst_lengths = BIT(1) | BIT(4) | BIT(8) | BIT(16),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_8_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_8_BYTES),
+};
+
+/*
+ * The A64 binding uses the number of dma channels from the
+ * device tree node.
+ */
+static struct sun6i_dma_config sun50i_a64_dma_cfg = {
+	.clock_autogate_enable = sun6i_enable_clock_autogate_h3,
+	.set_burst_length = sun6i_set_burst_length_h3,
+	.src_burst_lengths = BIT(1) | BIT(4) | BIT(8) | BIT(16),
+	.dst_burst_lengths = BIT(1) | BIT(4) | BIT(8) | BIT(16),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_8_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_8_BYTES),
 };
 
 /*
@@ -1049,7 +1153,16 @@ static struct sun6i_dma_config sun8i_v3s_dma_cfg = {
 	.nr_max_channels = 8,
 	.nr_max_requests = 23,
 	.nr_max_vchans   = 24,
-	.gate_needed	 = true,
+	.clock_autogate_enable = sun6i_enable_clock_autogate_a23,
+	.set_burst_length = sun6i_set_burst_length_a31,
+	.src_burst_lengths = BIT(1) | BIT(8),
+	.dst_burst_lengths = BIT(1) | BIT(8),
+	.src_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
+	.dst_addr_widths   = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+			     BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+			     BIT(DMA_SLAVE_BUSWIDTH_4_BYTES),
 };
 
 static const struct of_device_id sun6i_dma_match[] = {
@@ -1058,13 +1171,14 @@ static const struct of_device_id sun6i_dma_match[] = {
 	{ .compatible = "allwinner,sun8i-a83t-dma", .data = &sun8i_a83t_dma_cfg },
 	{ .compatible = "allwinner,sun8i-h3-dma", .data = &sun8i_h3_dma_cfg },
 	{ .compatible = "allwinner,sun8i-v3s-dma", .data = &sun8i_v3s_dma_cfg },
+	{ .compatible = "allwinner,sun50i-a64-dma", .data = &sun50i_a64_dma_cfg },
 	{ /* sentinel */ }
 };
 MODULE_DEVICE_TABLE(of, sun6i_dma_match);
 
 static int sun6i_dma_probe(struct platform_device *pdev)
 {
-	const struct of_device_id *device;
+	struct device_node *np = pdev->dev.of_node;
 	struct sun6i_dma_dev *sdc;
 	struct resource *res;
 	int ret, i;
@@ -1073,10 +1187,9 @@ static int sun6i_dma_probe(struct platform_device *pdev)
 	if (!sdc)
 		return -ENOMEM;
 
-	device = of_match_device(sun6i_dma_match, &pdev->dev);
-	if (!device)
+	sdc->cfg = of_device_get_match_data(&pdev->dev);
+	if (!sdc->cfg)
 		return -ENODEV;
-	sdc->cfg = device->data;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	sdc->base = devm_ioremap_resource(&pdev->dev, res);
@@ -1129,37 +1242,57 @@ static int sun6i_dma_probe(struct platform_device *pdev)
 	sdc->slave.device_pause			= sun6i_dma_pause;
 	sdc->slave.device_resume		= sun6i_dma_resume;
 	sdc->slave.device_terminate_all		= sun6i_dma_terminate_all;
-	sdc->slave.src_addr_widths		= BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
-						  BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
-						  BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
-	sdc->slave.dst_addr_widths		= BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
-						  BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
-						  BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+	sdc->slave.src_addr_widths		= sdc->cfg->src_addr_widths;
+	sdc->slave.dst_addr_widths		= sdc->cfg->dst_addr_widths;
 	sdc->slave.directions			= BIT(DMA_DEV_TO_MEM) |
 						  BIT(DMA_MEM_TO_DEV);
 	sdc->slave.residue_granularity		= DMA_RESIDUE_GRANULARITY_BURST;
 	sdc->slave.dev = &pdev->dev;
 
-	sdc->pchans = devm_kcalloc(&pdev->dev, sdc->cfg->nr_max_channels,
+	sdc->num_pchans = sdc->cfg->nr_max_channels;
+	sdc->num_vchans = sdc->cfg->nr_max_vchans;
+	sdc->max_request = sdc->cfg->nr_max_requests;
+
+	ret = of_property_read_u32(np, "dma-channels", &sdc->num_pchans);
+	if (ret && !sdc->num_pchans) {
+		dev_err(&pdev->dev, "Can't get dma-channels.\n");
+		return ret;
+	}
+
+	ret = of_property_read_u32(np, "dma-requests", &sdc->max_request);
+	if (ret && !sdc->max_request) {
+		dev_info(&pdev->dev, "Missing dma-requests, using %u.\n",
+			 DMA_CHAN_MAX_DRQ);
+		sdc->max_request = DMA_CHAN_MAX_DRQ;
+	}
+
+	/*
+	 * If the number of vchans is not specified, derive it from the
+	 * highest port number, at most one channel per port and direction.
+	 */
+	if (!sdc->num_vchans)
+		sdc->num_vchans = 2 * (sdc->max_request + 1);
+
+	sdc->pchans = devm_kcalloc(&pdev->dev, sdc->num_pchans,
 				   sizeof(struct sun6i_pchan), GFP_KERNEL);
 	if (!sdc->pchans)
 		return -ENOMEM;
 
-	sdc->vchans = devm_kcalloc(&pdev->dev, sdc->cfg->nr_max_vchans,
+	sdc->vchans = devm_kcalloc(&pdev->dev, sdc->num_vchans,
 				   sizeof(struct sun6i_vchan), GFP_KERNEL);
 	if (!sdc->vchans)
 		return -ENOMEM;
 
 	tasklet_init(&sdc->task, sun6i_dma_tasklet, (unsigned long)sdc);
 
-	for (i = 0; i < sdc->cfg->nr_max_channels; i++) {
+	for (i = 0; i < sdc->num_pchans; i++) {
 		struct sun6i_pchan *pchan = &sdc->pchans[i];
 
 		pchan->idx = i;
 		pchan->base = sdc->base + 0x100 + i * 0x40;
 	}
 
-	for (i = 0; i < sdc->cfg->nr_max_vchans; i++) {
+	for (i = 0; i < sdc->num_vchans; i++) {
 		struct sun6i_vchan *vchan = &sdc->vchans[i];
 
 		INIT_LIST_HEAD(&vchan->node);
@@ -1199,8 +1332,8 @@ static int sun6i_dma_probe(struct platform_device *pdev)
 		goto err_dma_unregister;
 	}
 
-	if (sdc->cfg->gate_needed)
-		writel(SUN8I_DMA_GATE_ENABLE, sdc->base + SUN8I_DMA_GATE);
+	if (sdc->cfg->clock_autogate_enable)
+		sdc->cfg->clock_autogate_enable(sdc);
 
 	return 0;
 
diff --git a/drivers/dma/ti-dma-crossbar.c b/drivers/dma/ti-dma-crossbar.c
index f1d04b70ee67..7df910e7c348 100644
--- a/drivers/dma/ti-dma-crossbar.c
+++ b/drivers/dma/ti-dma-crossbar.c
@@ -49,12 +49,12 @@ struct ti_am335x_xbar_data {
 
 struct ti_am335x_xbar_map {
 	u16 dma_line;
-	u16 mux_val;
+	u8 mux_val;
 };
 
-static inline void ti_am335x_xbar_write(void __iomem *iomem, int event, u16 val)
+static inline void ti_am335x_xbar_write(void __iomem *iomem, int event, u8 val)
 {
-	writeb_relaxed(val & 0x1f, iomem + event);
+	writeb_relaxed(val, iomem + event);
 }
 
 static void ti_am335x_xbar_free(struct device *dev, void *route_data)
@@ -105,7 +105,7 @@ static void *ti_am335x_xbar_route_allocate(struct of_phandle_args *dma_spec,
 	}
 
 	map->dma_line = (u16)dma_spec->args[0];
-	map->mux_val = (u16)dma_spec->args[2];
+	map->mux_val = (u8)dma_spec->args[2];
 
 	dma_spec->args[2] = 0;
 	dma_spec->args_count = 2;
diff --git a/drivers/dma/xilinx/xilinx_dma.c b/drivers/dma/xilinx/xilinx_dma.c
index 8722bcba489d..5eef13380ca8 100644
--- a/drivers/dma/xilinx/xilinx_dma.c
+++ b/drivers/dma/xilinx/xilinx_dma.c
@@ -366,6 +366,20 @@ struct xilinx_dma_chan {
 	u16 tdest;
 };
 
+/**
+ * enum xdma_ip_type: DMA IP type.
+ *
+ * XDMA_TYPE_AXIDMA: Axi dma ip.
+ * XDMA_TYPE_CDMA: Axi cdma ip.
+ * XDMA_TYPE_VDMA: Axi vdma ip.
+ *
+ */
+enum xdma_ip_type {
+	XDMA_TYPE_AXIDMA = 0,
+	XDMA_TYPE_CDMA,
+	XDMA_TYPE_VDMA,
+};
+
 struct xilinx_dma_config {
 	enum xdma_ip_type dmatype;
 	int (*clk_init)(struct platform_device *pdev, struct clk **axi_clk,