Merge branch 'topic/xilinx' into for-linus

2 files changed, 313 insertions, 168 deletions

diff --git a/drivers/dma/xilinx/xilinx_dma.c b/drivers/dma/xilinx/xilinx_dma.c
index 5eef133..27b5235 100644
--- a/drivers/dma/xilinx/xilinx_dma.c
+++ b/drivers/dma/xilinx/xilinx_dma.c
@@ -99,7 +99,9 @@
 #define XILINX_DMA_REG_FRMPTR_STS		0x0024
 #define XILINX_DMA_REG_PARK_PTR		0x0028
 #define XILINX_DMA_PARK_PTR_WR_REF_SHIFT	8
+#define XILINX_DMA_PARK_PTR_WR_REF_MASK		GENMASK(12, 8)
 #define XILINX_DMA_PARK_PTR_RD_REF_SHIFT	0
+#define XILINX_DMA_PARK_PTR_RD_REF_MASK		GENMASK(4, 0)
 #define XILINX_DMA_REG_VDMA_VERSION		0x002c
 
 /* Register Direct Mode Registers */
@@ -163,6 +165,7 @@
 #define XILINX_DMA_BD_SOP		BIT(27)
 #define XILINX_DMA_BD_EOP		BIT(26)
 #define XILINX_DMA_COALESCE_MAX		255
+#define XILINX_DMA_NUM_DESCS		255
 #define XILINX_DMA_NUM_APP_WORDS	5
 
 /* Multi-Channel DMA Descriptor offsets*/
@@ -211,8 +214,8 @@ struct xilinx_vdma_desc_hw {
  * @next_desc_msb: MSB of Next Descriptor Pointer @0x04
  * @buf_addr: Buffer address @0x08
  * @buf_addr_msb: MSB of Buffer address @0x0C
- * @pad1: Reserved @0x10
- * @pad2: Reserved @0x14
+ * @mcdma_control: Control field for mcdma @0x10
+ * @vsize_stride: Vsize and Stride field for mcdma @0x14
  * @control: Control field @0x18
  * @status: Status field @0x1C
  * @app: APP Fields @0x20 - 0x30
@@ -232,11 +235,11 @@ struct xilinx_axidma_desc_hw {
 /**
  * struct xilinx_cdma_desc_hw - Hardware Descriptor
  * @next_desc: Next Descriptor Pointer @0x00
- * @next_descmsb: Next Descriptor Pointer MSB @0x04
+ * @next_desc_msb: Next Descriptor Pointer MSB @0x04
  * @src_addr: Source address @0x08
- * @src_addrmsb: Source address MSB @0x0C
+ * @src_addr_msb: Source address MSB @0x0C
  * @dest_addr: Destination address @0x10
- * @dest_addrmsb: Destination address MSB @0x14
+ * @dest_addr_msb: Destination address MSB @0x14
  * @control: Control field @0x18
  * @status: Status field @0x1C
  */
@@ -310,6 +313,7 @@ struct xilinx_dma_tx_descriptor {
  * @pending_list: Descriptors waiting
  * @active_list: Descriptors ready to submit
  * @done_list: Complete descriptors
+ * @free_seg_list: Free descriptors
  * @common: DMA common channel
  * @desc_pool: Descriptors pool
  * @dev: The dma device
@@ -321,6 +325,7 @@ struct xilinx_dma_tx_descriptor {
  * @cyclic: Check for cyclic transfers.
  * @genlock: Support genlock mode
  * @err: Channel has errors
+ * @idle: Check for channel idle
  * @tasklet: Cleanup work after irq
  * @config: Device configuration info
  * @flush_on_fsync: Flush on Frame sync
@@ -329,9 +334,12 @@ struct xilinx_dma_tx_descriptor {
  * @desc_submitcount: Descriptor h/w submitted count
  * @residue: Residue for AXI DMA
  * @seg_v: Statically allocated segments base
+ * @seg_p: Physical allocated segments base
  * @cyclic_seg_v: Statically allocated segment base for cyclic transfers
+ * @cyclic_seg_p: Physical allocated segments base for cyclic dma
  * @start_transfer: Differentiate b/w DMA IP's transfer
  * @stop_transfer: Differentiate b/w DMA IP's quiesce
+ * @tdest: TDEST value for mcdma
  */
 struct xilinx_dma_chan {
 	struct xilinx_dma_device *xdev;
@@ -341,6 +349,7 @@ struct xilinx_dma_chan {
 	struct list_head pending_list;
 	struct list_head active_list;
 	struct list_head done_list;
+	struct list_head free_seg_list;
 	struct dma_chan common;
 	struct dma_pool *desc_pool;
 	struct device *dev;
@@ -352,6 +361,7 @@ struct xilinx_dma_chan {
 	bool cyclic;
 	bool genlock;
 	bool err;
+	bool idle;
 	struct tasklet_struct tasklet;
 	struct xilinx_vdma_config config;
 	bool flush_on_fsync;
@@ -360,18 +370,20 @@ struct xilinx_dma_chan {
 	u32 desc_submitcount;
 	u32 residue;
 	struct xilinx_axidma_tx_segment *seg_v;
+	dma_addr_t seg_p;
 	struct xilinx_axidma_tx_segment *cyclic_seg_v;
+	dma_addr_t cyclic_seg_p;
 	void (*start_transfer)(struct xilinx_dma_chan *chan);
 	int (*stop_transfer)(struct xilinx_dma_chan *chan);
 	u16 tdest;
 };
 
 /**
- * enum xdma_ip_type: DMA IP type.
+ * 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.
+ * @XDMA_TYPE_AXIDMA: Axi dma ip.
+ * @XDMA_TYPE_CDMA: Axi cdma ip.
+ * @XDMA_TYPE_VDMA: Axi vdma ip.
  *
  */
 enum xdma_ip_type {
@@ -580,18 +592,32 @@ xilinx_cdma_alloc_tx_segment(struct xilinx_dma_chan *chan)
 static struct xilinx_axidma_tx_segment *
 xilinx_axidma_alloc_tx_segment(struct xilinx_dma_chan *chan)
 {
-	struct xilinx_axidma_tx_segment *segment;
-	dma_addr_t phys;
-
-	segment = dma_pool_zalloc(chan->desc_pool, GFP_ATOMIC, &phys);
-	if (!segment)
-		return NULL;
+	struct xilinx_axidma_tx_segment *segment = NULL;
+	unsigned long flags;
 
-	segment->phys = phys;
+	spin_lock_irqsave(&chan->lock, flags);
+	if (!list_empty(&chan->free_seg_list)) {
+		segment = list_first_entry(&chan->free_seg_list,
+					   struct xilinx_axidma_tx_segment,
+					   node);
+		list_del(&segment->node);
+	}
+	spin_unlock_irqrestore(&chan->lock, flags);
 
 	return segment;
 }
 
+static void xilinx_dma_clean_hw_desc(struct xilinx_axidma_desc_hw *hw)
+{
+	u32 next_desc = hw->next_desc;
+	u32 next_desc_msb = hw->next_desc_msb;
+
+	memset(hw, 0, sizeof(struct xilinx_axidma_desc_hw));
+
+	hw->next_desc = next_desc;
+	hw->next_desc_msb = next_desc_msb;
+}
+
 /**
  * xilinx_dma_free_tx_segment - Free transaction segment
  * @chan: Driver specific DMA channel
@@ -600,7 +626,9 @@ xilinx_axidma_alloc_tx_segment(struct xilinx_dma_chan *chan)
 static void xilinx_dma_free_tx_segment(struct xilinx_dma_chan *chan,
 				struct xilinx_axidma_tx_segment *segment)
 {
-	dma_pool_free(chan->desc_pool, segment, segment->phys);
+	xilinx_dma_clean_hw_desc(&segment->hw);
+
+	list_add_tail(&segment->node, &chan->free_seg_list);
 }
 
 /**
@@ -725,16 +753,31 @@ static void xilinx_dma_free_descriptors(struct xilinx_dma_chan *chan)
 static void xilinx_dma_free_chan_resources(struct dma_chan *dchan)
 {
 	struct xilinx_dma_chan *chan = to_xilinx_chan(dchan);
+	unsigned long flags;
 
 	dev_dbg(chan->dev, "Free all channel resources.\n");
 
 	xilinx_dma_free_descriptors(chan);
+
 	if (chan->xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) {
-		xilinx_dma_free_tx_segment(chan, chan->cyclic_seg_v);
-		xilinx_dma_free_tx_segment(chan, chan->seg_v);
+		spin_lock_irqsave(&chan->lock, flags);
+		INIT_LIST_HEAD(&chan->free_seg_list);
+		spin_unlock_irqrestore(&chan->lock, flags);
+
+		/* Free memory that is allocated for BD */
+		dma_free_coherent(chan->dev, sizeof(*chan->seg_v) *
+				  XILINX_DMA_NUM_DESCS, chan->seg_v,
+				  chan->seg_p);
+
+		/* Free Memory that is allocated for cyclic DMA Mode */
+		dma_free_coherent(chan->dev, sizeof(*chan->cyclic_seg_v),
+				  chan->cyclic_seg_v, chan->cyclic_seg_p);
+	}
+
+	if (chan->xdev->dma_config->dmatype != XDMA_TYPE_AXIDMA) {
+		dma_pool_destroy(chan->desc_pool);
+		chan->desc_pool = NULL;
 	}
-	dma_pool_destroy(chan->desc_pool);
-	chan->desc_pool = NULL;
 }
 
 /**
@@ -817,6 +860,7 @@ static void xilinx_dma_do_tasklet(unsigned long data)
 static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 {
 	struct xilinx_dma_chan *chan = to_xilinx_chan(dchan);
+	int i;
 
 	/* Has this channel already been allocated? */
 	if (chan->desc_pool)
@@ -827,11 +871,30 @@ static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 	 * for meeting Xilinx VDMA specification requirement.
 	 */
 	if (chan->xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) {
-		chan->desc_pool = dma_pool_create("xilinx_dma_desc_pool",
-				   chan->dev,
-				   sizeof(struct xilinx_axidma_tx_segment),
-				   __alignof__(struct xilinx_axidma_tx_segment),
-				   0);
+		/* Allocate the buffer descriptors. */
+		chan->seg_v = dma_zalloc_coherent(chan->dev,
+						  sizeof(*chan->seg_v) *
+						  XILINX_DMA_NUM_DESCS,
+						  &chan->seg_p, GFP_KERNEL);
+		if (!chan->seg_v) {
+			dev_err(chan->dev,
+				"unable to allocate channel %d descriptors\n",
+				chan->id);
+			return -ENOMEM;
+		}
+
+		for (i = 0; i < XILINX_DMA_NUM_DESCS; i++) {
+			chan->seg_v[i].hw.next_desc =
+			lower_32_bits(chan->seg_p + sizeof(*chan->seg_v) *
+				((i + 1) % XILINX_DMA_NUM_DESCS));
+			chan->seg_v[i].hw.next_desc_msb =
+			upper_32_bits(chan->seg_p + sizeof(*chan->seg_v) *
+				((i + 1) % XILINX_DMA_NUM_DESCS));
+			chan->seg_v[i].phys = chan->seg_p +
+				sizeof(*chan->seg_v) * i;
+			list_add_tail(&chan->seg_v[i].node,
+				      &chan->free_seg_list);
+		}
 	} else if (chan->xdev->dma_config->dmatype == XDMA_TYPE_CDMA) {
 		chan->desc_pool = dma_pool_create("xilinx_cdma_desc_pool",
 				   chan->dev,
@@ -846,7 +909,8 @@ static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 				     0);
 	}
 
-	if (!chan->desc_pool) {
+	if (!chan->desc_pool &&
+	    (chan->xdev->dma_config->dmatype != XDMA_TYPE_AXIDMA)) {
 		dev_err(chan->dev,
 			"unable to allocate channel %d descriptor pool\n",
 			chan->id);
@@ -855,22 +919,20 @@ static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 
 	if (chan->xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) {
 		/*
-		 * For AXI DMA case after submitting a pending_list, keep
-		 * an extra segment allocated so that the "next descriptor"
-		 * pointer on the tail descriptor always points to a
-		 * valid descriptor, even when paused after reaching taildesc.
-		 * This way, it is possible to issue additional
-		 * transfers without halting and restarting the channel.
-		 */
-		chan->seg_v = xilinx_axidma_alloc_tx_segment(chan);
-
-		/*
 		 * For cyclic DMA mode we need to program the tail Descriptor
 		 * register with a value which is not a part of the BD chain
 		 * so allocating a desc segment during channel allocation for
 		 * programming tail descriptor.
 		 */
-		chan->cyclic_seg_v = xilinx_axidma_alloc_tx_segment(chan);
+		chan->cyclic_seg_v = dma_zalloc_coherent(chan->dev,
+					sizeof(*chan->cyclic_seg_v),
+					&chan->cyclic_seg_p, GFP_KERNEL);
+		if (!chan->cyclic_seg_v) {
+			dev_err(chan->dev,
+				"unable to allocate desc segment for cyclic DMA\n");
+			return -ENOMEM;
+		}
+		chan->cyclic_seg_v->phys = chan->cyclic_seg_p;
 	}
 
 	dma_cookie_init(dchan);
@@ -936,34 +998,10 @@ static enum dma_status xilinx_dma_tx_status(struct dma_chan *dchan,
 }
 
 /**
- * xilinx_dma_is_running - Check if DMA channel is running
- * @chan: Driver specific DMA channel
- *
- * Return: '1' if running, '0' if not.
- */
-static bool xilinx_dma_is_running(struct xilinx_dma_chan *chan)
-{
-	return !(dma_ctrl_read(chan, XILINX_DMA_REG_DMASR) &
-		 XILINX_DMA_DMASR_HALTED) &&
-		(dma_ctrl_read(chan, XILINX_DMA_REG_DMACR) &
-		 XILINX_DMA_DMACR_RUNSTOP);
-}
-
-/**
- * xilinx_dma_is_idle - Check if DMA channel is idle
- * @chan: Driver specific DMA channel
- *
- * Return: '1' if idle, '0' if not.
- */
-static bool xilinx_dma_is_idle(struct xilinx_dma_chan *chan)
-{
-	return dma_ctrl_read(chan, XILINX_DMA_REG_DMASR) &
-		XILINX_DMA_DMASR_IDLE;
-}
-
-/**
  * xilinx_dma_stop_transfer - Halt DMA channel
  * @chan: Driver specific DMA channel
+ *
+ * Return: '0' on success and failure value on error
  */
 static int xilinx_dma_stop_transfer(struct xilinx_dma_chan *chan)
 {
@@ -980,6 +1018,8 @@ static int xilinx_dma_stop_transfer(struct xilinx_dma_chan *chan)
 /**
  * xilinx_cdma_stop_transfer - Wait for the current transfer to complete
  * @chan: Driver specific DMA channel
+ *
+ * Return: '0' on success and failure value on error
  */
 static int xilinx_cdma_stop_transfer(struct xilinx_dma_chan *chan)
 {
@@ -1022,13 +1062,16 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 {
 	struct xilinx_vdma_config *config = &chan->config;
 	struct xilinx_dma_tx_descriptor *desc, *tail_desc;
-	u32 reg;
+	u32 reg, j;
 	struct xilinx_vdma_tx_segment *tail_segment;
 
 	/* This function was invoked with lock held */
 	if (chan->err)
 		return;
 
+	if (!chan->idle)
+		return;
+
 	if (list_empty(&chan->pending_list))
 		return;
 
@@ -1040,13 +1083,6 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 	tail_segment = list_last_entry(&tail_desc->segments,
 				       struct xilinx_vdma_tx_segment, node);
 
-	/* If it is SG mode and hardware is busy, cannot submit */
-	if (chan->has_sg && xilinx_dma_is_running(chan) &&
-	    !xilinx_dma_is_idle(chan)) {
-		dev_dbg(chan->dev, "DMA controller still busy\n");
-		return;
-	}
-
 	/*
 	 * If hardware is idle, then all descriptors on the running lists are
 	 * done, start new transfers
@@ -1063,10 +1099,6 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 	else
 		reg &= ~XILINX_DMA_DMACR_FRAMECNT_EN;
 
-	/* Configure channel to allow number frame buffers */
-	dma_ctrl_write(chan, XILINX_DMA_REG_FRMSTORE,
-			chan->desc_pendingcount);
-
 	/*
 	 * With SG, start with circular mode, so that BDs can be fetched.
 	 * In direct register mode, if not parking, enable circular mode
@@ -1079,17 +1111,16 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 
 	dma_ctrl_write(chan, XILINX_DMA_REG_DMACR, reg);
 
-	if (config->park && (config->park_frm >= 0) &&
-			(config->park_frm < chan->num_frms)) {
-		if (chan->direction == DMA_MEM_TO_DEV)
-			dma_write(chan, XILINX_DMA_REG_PARK_PTR,
-				config->park_frm <<
-					XILINX_DMA_PARK_PTR_RD_REF_SHIFT);
-		else
-			dma_write(chan, XILINX_DMA_REG_PARK_PTR,
-				config->park_frm <<
-					XILINX_DMA_PARK_PTR_WR_REF_SHIFT);
+	j = chan->desc_submitcount;
+	reg = dma_read(chan, XILINX_DMA_REG_PARK_PTR);
+	if (chan->direction == DMA_MEM_TO_DEV) {
+		reg &= ~XILINX_DMA_PARK_PTR_RD_REF_MASK;
+		reg |= j << XILINX_DMA_PARK_PTR_RD_REF_SHIFT;
+	} else {
+		reg &= ~XILINX_DMA_PARK_PTR_WR_REF_MASK;
+		reg |= j << XILINX_DMA_PARK_PTR_WR_REF_SHIFT;
 	}
+	dma_write(chan, XILINX_DMA_REG_PARK_PTR, reg);
 
 	/* Start the hardware */
 	xilinx_dma_start(chan);
@@ -1101,6 +1132,8 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->has_sg) {
 		dma_ctrl_write(chan, XILINX_DMA_REG_TAILDESC,
 				tail_segment->phys);
+		list_splice_tail_init(&chan->pending_list, &chan->active_list);
+		chan->desc_pendingcount = 0;
 	} else {
 		struct xilinx_vdma_tx_segment *segment, *last = NULL;
 		int i = 0;
@@ -1130,19 +1163,16 @@ static void xilinx_vdma_start_transfer(struct xilinx_dma_chan *chan)
 		vdma_desc_write(chan, XILINX_DMA_REG_FRMDLY_STRIDE,
 				last->hw.stride);
 		vdma_desc_write(chan, XILINX_DMA_REG_VSIZE, last->hw.vsize);
-	}
 
-	if (!chan->has_sg) {
-		list_del(&desc->node);
-		list_add_tail(&desc->node, &chan->active_list);
 		chan->desc_submitcount++;
 		chan->desc_pendingcount--;
+		list_del(&desc->node);
+		list_add_tail(&desc->node, &chan->active_list);
 		if (chan->desc_submitcount == chan->num_frms)
 			chan->desc_submitcount = 0;
-	} else {
-		list_splice_tail_init(&chan->pending_list, &chan->active_list);
-		chan->desc_pendingcount = 0;
 	}
+
+	chan->idle = false;
 }
 
 /**
@@ -1158,6 +1188,9 @@ static void xilinx_cdma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->err)
 		return;
 
+	if (!chan->idle)
+		return;
+
 	if (list_empty(&chan->pending_list))
 		return;
 
@@ -1176,6 +1209,12 @@ static void xilinx_cdma_start_transfer(struct xilinx_dma_chan *chan)
 	}
 
 	if (chan->has_sg) {
+		dma_ctrl_clr(chan, XILINX_DMA_REG_DMACR,
+			     XILINX_CDMA_CR_SGMODE);
+
+		dma_ctrl_set(chan, XILINX_DMA_REG_DMACR,
+			     XILINX_CDMA_CR_SGMODE);
+
 		xilinx_write(chan, XILINX_DMA_REG_CURDESC,
 			     head_desc->async_tx.phys);
 
@@ -1203,6 +1242,7 @@ static void xilinx_cdma_start_transfer(struct xilinx_dma_chan *chan)
 
 	list_splice_tail_init(&chan->pending_list, &chan->active_list);
 	chan->desc_pendingcount = 0;
+	chan->idle = false;
 }
 
 /**
@@ -1212,7 +1252,7 @@ static void xilinx_cdma_start_transfer(struct xilinx_dma_chan *chan)
 static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 {
 	struct xilinx_dma_tx_descriptor *head_desc, *tail_desc;
-	struct xilinx_axidma_tx_segment *tail_segment, *old_head, *new_head;
+	struct xilinx_axidma_tx_segment *tail_segment;
 	u32 reg;
 
 	if (chan->err)
@@ -1221,12 +1261,8 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	if (list_empty(&chan->pending_list))
 		return;
 
-	/* If it is SG mode and hardware is busy, cannot submit */
-	if (chan->has_sg && xilinx_dma_is_running(chan) &&
-	    !xilinx_dma_is_idle(chan)) {
-		dev_dbg(chan->dev, "DMA controller still busy\n");
+	if (!chan->idle)
 		return;
-	}
 
 	head_desc = list_first_entry(&chan->pending_list,
 				     struct xilinx_dma_tx_descriptor, node);
@@ -1235,21 +1271,6 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	tail_segment = list_last_entry(&tail_desc->segments,
 				       struct xilinx_axidma_tx_segment, node);
 
-	if (chan->has_sg && !chan->xdev->mcdma) {
-		old_head = list_first_entry(&head_desc->segments,
-					struct xilinx_axidma_tx_segment, node);
-		new_head = chan->seg_v;
-		/* Copy Buffer Descriptor fields. */
-		new_head->hw = old_head->hw;
-
-		/* Swap and save new reserve */
-		list_replace_init(&old_head->node, &new_head->node);
-		chan->seg_v = old_head;
-
-		tail_segment->hw.next_desc = chan->seg_v->phys;
-		head_desc->async_tx.phys = new_head->phys;
-	}
-
 	reg = dma_ctrl_read(chan, XILINX_DMA_REG_DMACR);
 
 	if (chan->desc_pendingcount <= XILINX_DMA_COALESCE_MAX) {
@@ -1324,6 +1345,7 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 
 	list_splice_tail_init(&chan->pending_list, &chan->active_list);
 	chan->desc_pendingcount = 0;
+	chan->idle = false;
 }
 
 /**
@@ -1388,6 +1410,8 @@ static int xilinx_dma_reset(struct xilinx_dma_chan *chan)
 	}
 
 	chan->err = false;
+	chan->idle = true;
+	chan->desc_submitcount = 0;
 
 	return err;
 }
@@ -1469,6 +1493,7 @@ static irqreturn_t xilinx_dma_irq_handler(int irq, void *data)
 	if (status & XILINX_DMA_DMASR_FRM_CNT_IRQ) {
 		spin_lock(&chan->lock);
 		xilinx_dma_complete_descriptor(chan);
+		chan->idle = true;
 		chan->start_transfer(chan);
 		spin_unlock(&chan->lock);
 	}
@@ -1591,7 +1616,7 @@ xilinx_vdma_dma_prep_interleaved(struct dma_chan *dchan,
 {
 	struct xilinx_dma_chan *chan = to_xilinx_chan(dchan);
 	struct xilinx_dma_tx_descriptor *desc;
-	struct xilinx_vdma_tx_segment *segment, *prev = NULL;
+	struct xilinx_vdma_tx_segment *segment;
 	struct xilinx_vdma_desc_hw *hw;
 
 	if (!is_slave_direction(xt->dir))
@@ -1645,8 +1670,6 @@ xilinx_vdma_dma_prep_interleaved(struct dma_chan *dchan,
 	/* Insert the segment into the descriptor segments list. */
 	list_add_tail(&segment->node, &desc->segments);
 
-	prev = segment;
-
 	/* Link the last hardware descriptor with the first. */
 	segment = list_first_entry(&desc->segments,
 				   struct xilinx_vdma_tx_segment, node);
@@ -1733,7 +1756,7 @@ static struct dma_async_tx_descriptor *xilinx_dma_prep_slave_sg(
 {
 	struct xilinx_dma_chan *chan = to_xilinx_chan(dchan);
 	struct xilinx_dma_tx_descriptor *desc;
-	struct xilinx_axidma_tx_segment *segment = NULL, *prev = NULL;
+	struct xilinx_axidma_tx_segment *segment = NULL;
 	u32 *app_w = (u32 *)context;
 	struct scatterlist *sg;
 	size_t copy;
@@ -1784,10 +1807,6 @@ static struct dma_async_tx_descriptor *xilinx_dma_prep_slave_sg(
 					       XILINX_DMA_NUM_APP_WORDS);
 			}
 
-			if (prev)
-				prev->hw.next_desc = segment->phys;
-
-			prev = segment;
 			sg_used += copy;
 
 			/*
@@ -1801,7 +1820,6 @@ static struct dma_async_tx_descriptor *xilinx_dma_prep_slave_sg(
 	segment = list_first_entry(&desc->segments,
 				   struct xilinx_axidma_tx_segment, node);
 	desc->async_tx.phys = segment->phys;
-	prev->hw.next_desc = segment->phys;
 
 	/* For the last DMA_MEM_TO_DEV transfer, set EOP */
 	if (chan->direction == DMA_MEM_TO_DEV) {
@@ -1821,11 +1839,14 @@ error:
 
 /**
  * xilinx_dma_prep_dma_cyclic - prepare descriptors for a DMA_SLAVE transaction
- * @chan: DMA channel
- * @sgl: scatterlist to transfer to/from
- * @sg_len: number of entries in @scatterlist
+ * @dchan: DMA channel
+ * @buf_addr: Physical address of the buffer
+ * @buf_len: Total length of the cyclic buffers
+ * @period_len: length of individual cyclic buffer
  * @direction: DMA direction
  * @flags: transfer ack flags
+ *
+ * Return: Async transaction descriptor on success and NULL on failure
  */
 static struct dma_async_tx_descriptor *xilinx_dma_prep_dma_cyclic(
 	struct dma_chan *dchan, dma_addr_t buf_addr, size_t buf_len,
@@ -2009,7 +2030,9 @@ error:
 
 /**
  * xilinx_dma_terminate_all - Halt the channel and free descriptors
- * @chan: Driver specific DMA Channel pointer
+ * @dchan: Driver specific DMA Channel pointer
+ *
+ * Return: '0' always.
  */
 static int xilinx_dma_terminate_all(struct dma_chan *dchan)
 {
@@ -2029,6 +2052,7 @@ static int xilinx_dma_terminate_all(struct dma_chan *dchan)
 
 	/* Remove and free all of the descriptors in the lists */
 	xilinx_dma_free_descriptors(chan);
+	chan->idle = true;
 
 	if (chan->cyclic) {
 		reg = dma_ctrl_read(chan, XILINX_DMA_REG_DMACR);
@@ -2037,6 +2061,10 @@ static int xilinx_dma_terminate_all(struct dma_chan *dchan)
 		chan->cyclic = false;
 	}
 
+	if ((chan->xdev->dma_config->dmatype == XDMA_TYPE_CDMA) && chan->has_sg)
+		dma_ctrl_clr(chan, XILINX_DMA_REG_DMACR,
+			     XILINX_CDMA_CR_SGMODE);
+
 	return 0;
 }
 
@@ -2323,6 +2351,7 @@ static void xdma_disable_allclks(struct xilinx_dma_device *xdev)
  *
  * @xdev: Driver specific device structure
  * @node: Device node
+ * @chan_id: DMA Channel id
  *
  * Return: '0' on success and failure value on error
  */
@@ -2344,11 +2373,18 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 	chan->has_sg = xdev->has_sg;
 	chan->desc_pendingcount = 0x0;
 	chan->ext_addr = xdev->ext_addr;
+	/* This variable ensures that descriptors are not
+	 * Submitted when dma engine is in progress. This variable is
+	 * Added to avoid polling for a bit in the status register to
+	 * Know dma state in the driver hot path.
+	 */
+	chan->idle = true;
 
 	spin_lock_init(&chan->lock);
 	INIT_LIST_HEAD(&chan->pending_list);
 	INIT_LIST_HEAD(&chan->done_list);
 	INIT_LIST_HEAD(&chan->active_list);
+	INIT_LIST_HEAD(&chan->free_seg_list);
 
 	/* Retrieve the channel properties from the device tree */
 	has_dre = of_property_read_bool(node, "xlnx,include-dre");
@@ -2379,6 +2415,7 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 		chan->ctrl_offset = XILINX_DMA_MM2S_CTRL_OFFSET;
 		if (xdev->dma_config->dmatype == XDMA_TYPE_VDMA) {
 			chan->desc_offset = XILINX_VDMA_MM2S_DESC_OFFSET;
+			chan->config.park = 1;
 
 			if (xdev->flush_on_fsync == XILINX_DMA_FLUSH_BOTH ||
 			    xdev->flush_on_fsync == XILINX_DMA_FLUSH_MM2S)
@@ -2395,6 +2432,7 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 		chan->ctrl_offset = XILINX_DMA_S2MM_CTRL_OFFSET;
 		if (xdev->dma_config->dmatype == XDMA_TYPE_VDMA) {
 			chan->desc_offset = XILINX_VDMA_S2MM_DESC_OFFSET;
+			chan->config.park = 1;
 
 			if (xdev->flush_on_fsync == XILINX_DMA_FLUSH_BOTH ||
 			    xdev->flush_on_fsync == XILINX_DMA_FLUSH_S2MM)
@@ -2459,7 +2497,8 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
  * Return: 0 always.
  */
 static int xilinx_dma_child_probe(struct xilinx_dma_device *xdev,
-				    struct device_node *node) {
+				    struct device_node *node)
+{
 	int ret, i, nr_channels = 1;
 
 	ret = of_property_read_u32(node, "dma-channels", &nr_channels);
@@ -2654,7 +2693,12 @@ static int xilinx_dma_probe(struct platform_device *pdev)
 		goto error;
 	}
 
-	dev_info(&pdev->dev, "Xilinx AXI VDMA Engine Driver Probed!!\n");
+	if (xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA)
+		dev_info(&pdev->dev, "Xilinx AXI DMA Engine Driver Probed!!\n");
+	else if (xdev->dma_config->dmatype == XDMA_TYPE_CDMA)
+		dev_info(&pdev->dev, "Xilinx AXI CDMA Engine Driver Probed!!\n");
+	else
+		dev_info(&pdev->dev, "Xilinx AXI VDMA Engine Driver Probed!!\n");
 
 	return 0;
 
diff --git a/drivers/dma/xilinx/zynqmp_dma.c b/drivers/dma/xilinx/zynqmp_dma.c
index 1ee1241..f146458 100644
--- a/drivers/dma/xilinx/zynqmp_dma.c
+++ b/drivers/dma/xilinx/zynqmp_dma.c
@@ -23,6 +23,7 @@
 #include <linux/slab.h>
 #include <linux/clk.h>
 #include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/pm_runtime.h>
 
 #include "../dmaengine.h"
 
@@ -47,6 +48,7 @@
 #define ZYNQMP_DMA_SRC_START_MSB	0x15C
 #define ZYNQMP_DMA_DST_START_LSB	0x160
 #define ZYNQMP_DMA_DST_START_MSB	0x164
+#define ZYNQMP_DMA_TOTAL_BYTE		0x188
 #define ZYNQMP_DMA_RATE_CTRL		0x18C
 #define ZYNQMP_DMA_IRQ_SRC_ACCT		0x190
 #define ZYNQMP_DMA_IRQ_DST_ACCT		0x194
@@ -138,6 +140,8 @@
 #define ZYNQMP_DMA_BUS_WIDTH_64		64
 #define ZYNQMP_DMA_BUS_WIDTH_128	128
 
+#define ZDMA_PM_TIMEOUT			100
+
 #define ZYNQMP_DMA_DESC_SIZE(chan)	(chan->desc_size)
 
 #define to_chan(chan)		container_of(chan, struct zynqmp_dma_chan, \
@@ -211,8 +215,6 @@ struct zynqmp_dma_desc_sw {
  * @bus_width: Bus width
  * @src_burst_len: Source burst length
  * @dst_burst_len: Dest burst length
- * @clk_main: Pointer to main clock
- * @clk_apb: Pointer to apb clock
  */
 struct zynqmp_dma_chan {
 	struct zynqmp_dma_device *zdev;
@@ -237,8 +239,6 @@ struct zynqmp_dma_chan {
 	u32 bus_width;
 	u32 src_burst_len;
 	u32 dst_burst_len;
-	struct clk *clk_main;
-	struct clk *clk_apb;
 };
 
 /**
@@ -246,11 +246,15 @@ struct zynqmp_dma_chan {
  * @dev: Device Structure
  * @common: DMA device structure
  * @chan: Driver specific DMA channel
+ * @clk_main: Pointer to main clock
+ * @clk_apb: Pointer to apb clock
  */
 struct zynqmp_dma_device {
 	struct device *dev;
 	struct dma_device common;
 	struct zynqmp_dma_chan *chan;
+	struct clk *clk_main;
+	struct clk *clk_apb;
 };
 
 static inline void zynqmp_dma_writeq(struct zynqmp_dma_chan *chan, u32 reg,
@@ -461,7 +465,11 @@ static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan)
 {
 	struct zynqmp_dma_chan *chan = to_chan(dchan);
 	struct zynqmp_dma_desc_sw *desc;
-	int i;
+	int i, ret;
+
+	ret = pm_runtime_get_sync(chan->dev);
+	if (ret < 0)
+		return ret;
 
 	chan->sw_desc_pool = kzalloc(sizeof(*desc) * ZYNQMP_DMA_NUM_DESCS,
 				     GFP_KERNEL);
@@ -506,6 +514,7 @@ static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan)
 static void zynqmp_dma_start(struct zynqmp_dma_chan *chan)
 {
 	writel(ZYNQMP_DMA_INT_EN_DEFAULT_MASK, chan->regs + ZYNQMP_DMA_IER);
+	writel(0, chan->regs + ZYNQMP_DMA_TOTAL_BYTE);
 	chan->idle = false;
 	writel(ZYNQMP_DMA_ENABLE, chan->regs + ZYNQMP_DMA_CTRL2);
 }
@@ -517,12 +526,12 @@ static void zynqmp_dma_start(struct zynqmp_dma_chan *chan)
  */
 static void zynqmp_dma_handle_ovfl_int(struct zynqmp_dma_chan *chan, u32 status)
 {
-	u32 val;
-
+	if (status & ZYNQMP_DMA_BYTE_CNT_OVRFL)
+		writel(0, chan->regs + ZYNQMP_DMA_TOTAL_BYTE);
 	if (status & ZYNQMP_DMA_IRQ_DST_ACCT_ERR)
-		val = readl(chan->regs + ZYNQMP_DMA_IRQ_DST_ACCT);
+		readl(chan->regs + ZYNQMP_DMA_IRQ_DST_ACCT);
 	if (status & ZYNQMP_DMA_IRQ_SRC_ACCT_ERR)
-		val = readl(chan->regs + ZYNQMP_DMA_IRQ_SRC_ACCT);
+		readl(chan->regs + ZYNQMP_DMA_IRQ_SRC_ACCT);
 }
 
 static void zynqmp_dma_config(struct zynqmp_dma_chan *chan)
@@ -545,6 +554,8 @@ static void zynqmp_dma_config(struct zynqmp_dma_chan *chan)
  * zynqmp_dma_device_config - Zynqmp dma device configuration
  * @dchan: DMA channel
  * @config: DMA device config
+ *
+ * Return: 0 always
  */
 static int zynqmp_dma_device_config(struct dma_chan *dchan,
 				    struct dma_slave_config *config)
@@ -640,7 +651,7 @@ static void zynqmp_dma_issue_pending(struct dma_chan *dchan)
 
 /**
  * zynqmp_dma_free_descriptors - Free channel descriptors
- * @dchan: DMA channel pointer
+ * @chan: ZynqMP DMA channel pointer
  */
 static void zynqmp_dma_free_descriptors(struct zynqmp_dma_chan *chan)
 {
@@ -664,6 +675,8 @@ static void zynqmp_dma_free_chan_resources(struct dma_chan *dchan)
 		(2 * ZYNQMP_DMA_DESC_SIZE(chan) * ZYNQMP_DMA_NUM_DESCS),
 		chan->desc_pool_v, chan->desc_pool_p);
 	kfree(chan->sw_desc_pool);
+	pm_runtime_mark_last_busy(chan->dev);
+	pm_runtime_put_autosuspend(chan->dev);
 }
 
 /**
@@ -715,7 +728,7 @@ static irqreturn_t zynqmp_dma_irq_handler(int irq, void *data)
 
 	if (status & ZYNQMP_DMA_INT_OVRFL) {
 		zynqmp_dma_handle_ovfl_int(chan, status);
-		dev_info(chan->dev, "Channel %p overflow interrupt\n", chan);
+		dev_dbg(chan->dev, "Channel %p overflow interrupt\n", chan);
 		ret = IRQ_HANDLED;
 	}
 
@@ -838,11 +851,10 @@ static void zynqmp_dma_chan_remove(struct zynqmp_dma_chan *chan)
 	if (!chan)
 		return;
 
-	devm_free_irq(chan->zdev->dev, chan->irq, chan);
+	if (chan->irq)
+		devm_free_irq(chan->zdev->dev, chan->irq, chan);
 	tasklet_kill(&chan->tasklet);
 	list_del(&chan->common.device_node);
-	clk_disable_unprepare(chan->clk_apb);
-	clk_disable_unprepare(chan->clk_main);
 }
 
 /**
@@ -907,30 +919,6 @@ static int zynqmp_dma_chan_probe(struct zynqmp_dma_device *zdev,
 			       "zynqmp-dma", chan);
 	if (err)
 		return err;
-	chan->clk_main = devm_clk_get(&pdev->dev, "clk_main");
-	if (IS_ERR(chan->clk_main)) {
-		dev_err(&pdev->dev, "main clock not found.\n");
-		return PTR_ERR(chan->clk_main);
-	}
-
-	chan->clk_apb = devm_clk_get(&pdev->dev, "clk_apb");
-	if (IS_ERR(chan->clk_apb)) {
-		dev_err(&pdev->dev, "apb clock not found.\n");
-		return PTR_ERR(chan->clk_apb);
-	}
-
-	err = clk_prepare_enable(chan->clk_main);
-	if (err) {
-		dev_err(&pdev->dev, "Unable to enable main clock.\n");
-		return err;
-	}
-
-	err = clk_prepare_enable(chan->clk_apb);
-	if (err) {
-		clk_disable_unprepare(chan->clk_main);
-		dev_err(&pdev->dev, "Unable to enable apb clock.\n");
-		return err;
-	}
 
 	chan->desc_size = sizeof(struct zynqmp_dma_desc_ll);
 	chan->idle = true;
@@ -953,6 +941,87 @@ static struct dma_chan *of_zynqmp_dma_xlate(struct of_phandle_args *dma_spec,
 }
 
 /**
+ * zynqmp_dma_suspend - Suspend method for the driver
+ * @dev:	Address of the device structure
+ *
+ * Put the driver into low power mode.
+ * Return: 0 on success and failure value on error
+ */
+static int __maybe_unused zynqmp_dma_suspend(struct device *dev)
+{
+	if (!device_may_wakeup(dev))
+		return pm_runtime_force_suspend(dev);
+
+	return 0;
+}
+
+/**
+ * zynqmp_dma_resume - Resume from suspend
+ * @dev:	Address of the device structure
+ *
+ * Resume operation after suspend.
+ * Return: 0 on success and failure value on error
+ */
+static int __maybe_unused zynqmp_dma_resume(struct device *dev)
+{
+	if (!device_may_wakeup(dev))
+		return pm_runtime_force_resume(dev);
+
+	return 0;
+}
+
+/**
+ * zynqmp_dma_runtime_suspend - Runtime suspend method for the driver
+ * @dev:	Address of the device structure
+ *
+ * Put the driver into low power mode.
+ * Return: 0 always
+ */
+static int __maybe_unused zynqmp_dma_runtime_suspend(struct device *dev)
+{
+	struct zynqmp_dma_device *zdev = dev_get_drvdata(dev);
+
+	clk_disable_unprepare(zdev->clk_main);
+	clk_disable_unprepare(zdev->clk_apb);
+
+	return 0;
+}
+
+/**
+ * zynqmp_dma_runtime_resume - Runtime suspend method for the driver
+ * @dev:	Address of the device structure
+ *
+ * Put the driver into low power mode.
+ * Return: 0 always
+ */
+static int __maybe_unused zynqmp_dma_runtime_resume(struct device *dev)
+{
+	struct zynqmp_dma_device *zdev = dev_get_drvdata(dev);
+	int err;
+
+	err = clk_prepare_enable(zdev->clk_main);
+	if (err) {
+		dev_err(dev, "Unable to enable main clock.\n");
+		return err;
+	}
+
+	err = clk_prepare_enable(zdev->clk_apb);
+	if (err) {
+		dev_err(dev, "Unable to enable apb clock.\n");
+		clk_disable_unprepare(zdev->clk_main);
+		return err;
+	}
+
+	return 0;
+}
+
+static const struct dev_pm_ops zynqmp_dma_dev_pm_ops = {
+	SET_SYSTEM_SLEEP_PM_OPS(zynqmp_dma_suspend, zynqmp_dma_resume)
+	SET_RUNTIME_PM_OPS(zynqmp_dma_runtime_suspend,
+			   zynqmp_dma_runtime_resume, NULL)
+};
+
+/**
  * zynqmp_dma_probe - Driver probe function
  * @pdev: Pointer to the platform_device structure
  *
@@ -984,12 +1053,33 @@ static int zynqmp_dma_probe(struct platform_device *pdev)
 	p->device_config = zynqmp_dma_device_config;
 	p->dev = &pdev->dev;
 
+	zdev->clk_main = devm_clk_get(&pdev->dev, "clk_main");
+	if (IS_ERR(zdev->clk_main)) {
+		dev_err(&pdev->dev, "main clock not found.\n");
+		return PTR_ERR(zdev->clk_main);
+	}
+
+	zdev->clk_apb = devm_clk_get(&pdev->dev, "clk_apb");
+	if (IS_ERR(zdev->clk_apb)) {
+		dev_err(&pdev->dev, "apb clock not found.\n");
+		return PTR_ERR(zdev->clk_apb);
+	}
+
 	platform_set_drvdata(pdev, zdev);
+	pm_runtime_set_autosuspend_delay(zdev->dev, ZDMA_PM_TIMEOUT);
+	pm_runtime_use_autosuspend(zdev->dev);
+	pm_runtime_enable(zdev->dev);
+	pm_runtime_get_sync(zdev->dev);
+	if (!pm_runtime_enabled(zdev->dev)) {
+		ret = zynqmp_dma_runtime_resume(zdev->dev);
+		if (ret)
+			return ret;
+	}
 
 	ret = zynqmp_dma_chan_probe(zdev, pdev);
 	if (ret) {
 		dev_err(&pdev->dev, "Probing channel failed\n");
-		goto free_chan_resources;
+		goto err_disable_pm;
 	}
 
 	p->dst_addr_widths = BIT(zdev->chan->bus_width / 8);
@@ -1005,12 +1095,19 @@ static int zynqmp_dma_probe(struct platform_device *pdev)
 		goto free_chan_resources;
 	}
 
+	pm_runtime_mark_last_busy(zdev->dev);
+	pm_runtime_put_sync_autosuspend(zdev->dev);
+
 	dev_info(&pdev->dev, "ZynqMP DMA driver Probe success\n");
 
 	return 0;
 
 free_chan_resources:
 	zynqmp_dma_chan_remove(zdev->chan);
+err_disable_pm:
+	if (!pm_runtime_enabled(zdev->dev))
+		zynqmp_dma_runtime_suspend(zdev->dev);
+	pm_runtime_disable(zdev->dev);
 	return ret;
 }
 
@@ -1028,6 +1125,9 @@ static int zynqmp_dma_remove(struct platform_device *pdev)
 	dma_async_device_unregister(&zdev->common);
 
 	zynqmp_dma_chan_remove(zdev->chan);
+	pm_runtime_disable(zdev->dev);
+	if (!pm_runtime_enabled(zdev->dev))
+		zynqmp_dma_runtime_suspend(zdev->dev);
 
 	return 0;
 }
@@ -1042,6 +1142,7 @@ static struct platform_driver zynqmp_dma_driver = {
 	.driver = {
 		.name = "xilinx-zynqmp-dma",
 		.of_match_table = zynqmp_dma_of_match,
+		.pm = &zynqmp_dma_dev_pm_ops,
 	},
 	.probe = zynqmp_dma_probe,
 	.remove = zynqmp_dma_remove,