Merge tag 'dmaengine-fix-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine

Pull dmaengine updates from Vinod Koul:
 "New support:
   - Loongson LS2X APB DMA controller
   - sf-pdma: mpfs-pdma support
   - Qualcomm X1E80100 GPI dma controller support

  Updates:
   - Xilinx XDMA updates to support interleaved DMA transfers
   - TI PSIL threads for AM62P and J722S and cfg register regions
     description
   - axi-dmac Improving the cyclic DMA transfers
   - Tegra Support dma-channel-mask property
   - Remaining platform remove callback returning void conversions

 Driver fixes for:
   - Xilinx xdma driver operator precedence and initialization fix
   - Excess kernel-doc warning fix in imx-sdma xilinx xdma drivers
   - format-overflow warning fix for rz-dmac, sh usb dmac drivers
   - 'output may be truncated' fix for shdma, fsl-qdma and dw-edma
     drivers"

* tag 'dmaengine-fix-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (58 commits)
  dmaengine: dw-edma: increase size of 'name' in debugfs code
  dmaengine: fsl-qdma: increase size of 'irq_name'
  dmaengine: shdma: increase size of 'dev_id'
  dmaengine: xilinx: xdma: Fix kernel-doc warnings
  dmaengine: usb-dmac: Avoid format-overflow warning
  dmaengine: sh: rz-dmac: Avoid format-overflow warning
  dmaengine: imx-sdma: fix Excess kernel-doc warnings
  dmaengine: xilinx: xdma: Fix initialization location of desc in xdma_channel_isr()
  dmaengine: xilinx: xdma: Fix operator precedence in xdma_prep_interleaved_dma()
  dmaengine: xilinx: xdma: statify xdma_prep_interleaved_dma
  dmaengine: xilinx: xdma: Workaround truncation compilation error
  dmaengine: pl330: issue_pending waits until WFP state
  dmaengine: xilinx: xdma: Implement interleaved DMA transfers
  dmaengine: xilinx: xdma: Prepare the introduction of interleaved DMA transfers
  dmaengine: xilinx: xdma: Add transfer error reporting
  dmaengine: xilinx: xdma: Add error checking in xdma_channel_isr()
  dmaengine: xilinx: xdma: Rework xdma_terminate_all()
  dmaengine: xilinx: xdma: Ease dma_pool alignment requirements
  dmaengine: xilinx: xdma: Add necessary macro definitions
  dmaengine: xilinx: xdma: Get rid of unused code
  ...

34 files changed, 1712 insertions, 261 deletions

diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 70ba506dabab..e928f2ca0f1e 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -378,6 +378,20 @@ config LPC18XX_DMAMUX
 	  Enable support for DMA on NXP LPC18xx/43xx platforms
 	  with PL080 and multiplexed DMA request lines.
 
+config LS2X_APB_DMA
+	tristate "Loongson LS2X APB DMA support"
+	depends on LOONGARCH || COMPILE_TEST
+	select DMA_ENGINE
+	select DMA_VIRTUAL_CHANNELS
+	help
+	  Support for the Loongson LS2X APB DMA controller driver. The
+	  DMA controller is having single DMA channel which can be
+	  configured for different peripherals like audio, nand, sdio
+	  etc which is in APB bus.
+
+	  This DMA controller transfers data from memory to peripheral fifo.
+	  It does not support memory to memory data transfer.
+
 config MCF_EDMA
 	tristate "Freescale eDMA engine support, ColdFire mcf5441x SoCs"
 	depends on M5441x || COMPILE_TEST
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 83553a97a010..dfd40d14e408 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_INTEL_IOATDMA) += ioat/
 obj-y += idxd/
 obj-$(CONFIG_K3_DMA) += k3dma.o
 obj-$(CONFIG_LPC18XX_DMAMUX) += lpc18xx-dmamux.o
+obj-$(CONFIG_LS2X_APB_DMA) += ls2x-apb-dma.o
 obj-$(CONFIG_MILBEAUT_HDMAC) += milbeaut-hdmac.o
 obj-$(CONFIG_MILBEAUT_XDMAC) += milbeaut-xdmac.o
 obj-$(CONFIG_MMP_PDMA) += mmp_pdma.o
diff --git a/drivers/dma/apple-admac.c b/drivers/dma/apple-admac.c
index 5b63996640d9..9588773dd2eb 100644
--- a/drivers/dma/apple-admac.c
+++ b/drivers/dma/apple-admac.c
@@ -57,6 +57,8 @@
 
 #define REG_BUS_WIDTH(ch)	(0x8040 + (ch) * 0x200)
 
+#define BUS_WIDTH_WORD_SIZE	GENMASK(3, 0)
+#define BUS_WIDTH_FRAME_SIZE	GENMASK(7, 4)
 #define BUS_WIDTH_8BIT		0x00
 #define BUS_WIDTH_16BIT		0x01
 #define BUS_WIDTH_32BIT		0x02
@@ -740,7 +742,8 @@ static int admac_device_config(struct dma_chan *chan,
 	struct admac_data *ad = adchan->host;
 	bool is_tx = admac_chan_direction(adchan->no) == DMA_MEM_TO_DEV;
 	int wordsize = 0;
-	u32 bus_width = 0;
+	u32 bus_width = readl_relaxed(ad->base + REG_BUS_WIDTH(adchan->no)) &
+		~(BUS_WIDTH_WORD_SIZE | BUS_WIDTH_FRAME_SIZE);
 
 	switch (is_tx ? config->dst_addr_width : config->src_addr_width) {
 	case DMA_SLAVE_BUSWIDTH_1_BYTE:
diff --git a/drivers/dma/dma-axi-dmac.c b/drivers/dma/dma-axi-dmac.c
index 2457a420c13d..4e339c04fc1e 100644
--- a/drivers/dma/dma-axi-dmac.c
+++ b/drivers/dma/dma-axi-dmac.c
@@ -81,9 +81,13 @@
 #define AXI_DMAC_REG_CURRENT_DEST_ADDR	0x438
 #define AXI_DMAC_REG_PARTIAL_XFER_LEN	0x44c
 #define AXI_DMAC_REG_PARTIAL_XFER_ID	0x450
+#define AXI_DMAC_REG_CURRENT_SG_ID	0x454
+#define AXI_DMAC_REG_SG_ADDRESS		0x47c
+#define AXI_DMAC_REG_SG_ADDRESS_HIGH	0x4bc
 
 #define AXI_DMAC_CTRL_ENABLE		BIT(0)
 #define AXI_DMAC_CTRL_PAUSE		BIT(1)
+#define AXI_DMAC_CTRL_ENABLE_SG		BIT(2)
 
 #define AXI_DMAC_IRQ_SOT		BIT(0)
 #define AXI_DMAC_IRQ_EOT		BIT(1)
@@ -97,20 +101,35 @@
 /* The maximum ID allocated by the hardware is 31 */
 #define AXI_DMAC_SG_UNUSED 32U
 
+/* Flags for axi_dmac_hw_desc.flags */
+#define AXI_DMAC_HW_FLAG_LAST		BIT(0)
+#define AXI_DMAC_HW_FLAG_IRQ		BIT(1)
+
+struct axi_dmac_hw_desc {
+	u32 flags;
+	u32 id;
+	u64 dest_addr;
+	u64 src_addr;
+	u64 next_sg_addr;
+	u32 y_len;
+	u32 x_len;
+	u32 src_stride;
+	u32 dst_stride;
+	u64 __pad[2];
+};
+
 struct axi_dmac_sg {
-	dma_addr_t src_addr;
-	dma_addr_t dest_addr;
-	unsigned int x_len;
-	unsigned int y_len;
-	unsigned int dest_stride;
-	unsigned int src_stride;
-	unsigned int id;
 	unsigned int partial_len;
 	bool schedule_when_free;
+
+	struct axi_dmac_hw_desc *hw;
+	dma_addr_t hw_phys;
 };
 
 struct axi_dmac_desc {
 	struct virt_dma_desc vdesc;
+	struct axi_dmac_chan *chan;
+
 	bool cyclic;
 	bool have_partial_xfer;
 
@@ -139,6 +158,7 @@ struct axi_dmac_chan {
 	bool hw_partial_xfer;
 	bool hw_cyclic;
 	bool hw_2d;
+	bool hw_sg;
 };
 
 struct axi_dmac {
@@ -213,9 +233,11 @@ static void axi_dmac_start_transfer(struct axi_dmac_chan *chan)
 	unsigned int flags = 0;
 	unsigned int val;
 
-	val = axi_dmac_read(dmac, AXI_DMAC_REG_START_TRANSFER);
-	if (val) /* Queue is full, wait for the next SOT IRQ */
-		return;
+	if (!chan->hw_sg) {
+		val = axi_dmac_read(dmac, AXI_DMAC_REG_START_TRANSFER);
+		if (val) /* Queue is full, wait for the next SOT IRQ */
+			return;
+	}
 
 	desc = chan->next_desc;
 
@@ -229,14 +251,15 @@ 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) {
+	if (sg->hw->id != AXI_DMAC_SG_UNUSED) {
 		sg->schedule_when_free = true;
 		return;
 	}
 
-	desc->num_submitted++;
-	if (desc->num_submitted == desc->num_sgs ||
-	    desc->have_partial_xfer) {
+	if (chan->hw_sg) {
+		chan->next_desc = NULL;
+	} else if (++desc->num_submitted == desc->num_sgs ||
+		   desc->have_partial_xfer) {
 		if (desc->cyclic)
 			desc->num_submitted = 0; /* Start again */
 		else
@@ -246,32 +269,42 @@ static void axi_dmac_start_transfer(struct axi_dmac_chan *chan)
 		chan->next_desc = desc;
 	}
 
-	sg->id = axi_dmac_read(dmac, AXI_DMAC_REG_TRANSFER_ID);
+	sg->hw->id = axi_dmac_read(dmac, AXI_DMAC_REG_TRANSFER_ID);
 
-	if (axi_dmac_dest_is_mem(chan)) {
-		axi_dmac_write(dmac, AXI_DMAC_REG_DEST_ADDRESS, sg->dest_addr);
-		axi_dmac_write(dmac, AXI_DMAC_REG_DEST_STRIDE, sg->dest_stride);
-	}
+	if (!chan->hw_sg) {
+		if (axi_dmac_dest_is_mem(chan)) {
+			axi_dmac_write(dmac, AXI_DMAC_REG_DEST_ADDRESS, sg->hw->dest_addr);
+			axi_dmac_write(dmac, AXI_DMAC_REG_DEST_STRIDE, sg->hw->dst_stride);
+		}
 
-	if (axi_dmac_src_is_mem(chan)) {
-		axi_dmac_write(dmac, AXI_DMAC_REG_SRC_ADDRESS, sg->src_addr);
-		axi_dmac_write(dmac, AXI_DMAC_REG_SRC_STRIDE, sg->src_stride);
+		if (axi_dmac_src_is_mem(chan)) {
+			axi_dmac_write(dmac, AXI_DMAC_REG_SRC_ADDRESS, sg->hw->src_addr);
+			axi_dmac_write(dmac, AXI_DMAC_REG_SRC_STRIDE, sg->hw->src_stride);
+		}
 	}
 
 	/*
 	 * If the hardware supports cyclic transfers and there is no callback to
-	 * call and only a single segment, enable hw cyclic mode to avoid
-	 * unnecessary interrupts.
+	 * call, enable hw cyclic mode to avoid unnecessary interrupts.
 	 */
-	if (chan->hw_cyclic && desc->cyclic && !desc->vdesc.tx.callback &&
-		desc->num_sgs == 1)
-		flags |= AXI_DMAC_FLAG_CYCLIC;
+	if (chan->hw_cyclic && desc->cyclic && !desc->vdesc.tx.callback) {
+		if (chan->hw_sg)
+			desc->sg[desc->num_sgs - 1].hw->flags &= ~AXI_DMAC_HW_FLAG_IRQ;
+		else if (desc->num_sgs == 1)
+			flags |= AXI_DMAC_FLAG_CYCLIC;
+	}
 
 	if (chan->hw_partial_xfer)
 		flags |= AXI_DMAC_FLAG_PARTIAL_REPORT;
 
-	axi_dmac_write(dmac, AXI_DMAC_REG_X_LENGTH, sg->x_len - 1);
-	axi_dmac_write(dmac, AXI_DMAC_REG_Y_LENGTH, sg->y_len - 1);
+	if (chan->hw_sg) {
+		axi_dmac_write(dmac, AXI_DMAC_REG_SG_ADDRESS, (u32)sg->hw_phys);
+		axi_dmac_write(dmac, AXI_DMAC_REG_SG_ADDRESS_HIGH,
+			       (u64)sg->hw_phys >> 32);
+	} else {
+		axi_dmac_write(dmac, AXI_DMAC_REG_X_LENGTH, sg->hw->x_len);
+		axi_dmac_write(dmac, AXI_DMAC_REG_Y_LENGTH, sg->hw->y_len);
+	}
 	axi_dmac_write(dmac, AXI_DMAC_REG_FLAGS, flags);
 	axi_dmac_write(dmac, AXI_DMAC_REG_START_TRANSFER, 1);
 }
@@ -286,9 +319,9 @@ static inline unsigned int axi_dmac_total_sg_bytes(struct axi_dmac_chan *chan,
 	struct axi_dmac_sg *sg)
 {
 	if (chan->hw_2d)
-		return sg->x_len * sg->y_len;
+		return (sg->hw->x_len + 1) * (sg->hw->y_len + 1);
 	else
-		return sg->x_len;
+		return (sg->hw->x_len + 1);
 }
 
 static void axi_dmac_dequeue_partial_xfers(struct axi_dmac_chan *chan)
@@ -307,9 +340,9 @@ static void axi_dmac_dequeue_partial_xfers(struct axi_dmac_chan *chan)
 		list_for_each_entry(desc, &chan->active_descs, vdesc.node) {
 			for (i = 0; i < desc->num_sgs; i++) {
 				sg = &desc->sg[i];
-				if (sg->id == AXI_DMAC_SG_UNUSED)
+				if (sg->hw->id == AXI_DMAC_SG_UNUSED)
 					continue;
-				if (sg->id == id) {
+				if (sg->hw->id == id) {
 					desc->have_partial_xfer = true;
 					sg->partial_len = len;
 					found_sg = true;
@@ -348,6 +381,9 @@ static void axi_dmac_compute_residue(struct axi_dmac_chan *chan,
 	rslt->result = DMA_TRANS_NOERROR;
 	rslt->residue = 0;
 
+	if (chan->hw_sg)
+		return;
+
 	/*
 	 * We get here if the last completed segment is partial, which
 	 * means we can compute the residue from that segment onwards
@@ -374,36 +410,47 @@ static bool axi_dmac_transfer_done(struct axi_dmac_chan *chan,
 	    (completed_transfers & AXI_DMAC_FLAG_PARTIAL_XFER_DONE))
 		axi_dmac_dequeue_partial_xfers(chan);
 
-	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 (chan->hw_sg) {
+		if (active->cyclic) {
+			vchan_cyclic_callback(&active->vdesc);
+		} else {
+			list_del(&active->vdesc.node);
+			vchan_cookie_complete(&active->vdesc);
+			active = axi_dmac_active_desc(chan);
+			start_next = !!active;
 		}
+	} else {
+		do {
+			sg = &active->sg[active->num_completed];
+			if (sg->hw->id == AXI_DMAC_SG_UNUSED) /* Not yet submitted */
+				break;
+			if (!(BIT(sg->hw->id) & completed_transfers))
+				break;
+			active->num_completed++;
+			sg->hw->id = AXI_DMAC_SG_UNUSED;
+			if (sg->schedule_when_free) {
+				sg->schedule_when_free = false;
+				start_next = true;
+			}
 
-		if (sg->partial_len)
-			axi_dmac_compute_residue(chan, active);
+			if (sg->partial_len)
+				axi_dmac_compute_residue(chan, active);
 
-		if (active->cyclic)
-			vchan_cyclic_callback(&active->vdesc);
+			if (active->cyclic)
+				vchan_cyclic_callback(&active->vdesc);
 
-		if (active->num_completed == active->num_sgs ||
-		    sg->partial_len) {
-			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);
+			if (active->num_completed == active->num_sgs ||
+			    sg->partial_len) {
+				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;
 }
@@ -467,8 +514,12 @@ static void axi_dmac_issue_pending(struct dma_chan *c)
 	struct axi_dmac_chan *chan = to_axi_dmac_chan(c);
 	struct axi_dmac *dmac = chan_to_axi_dmac(chan);
 	unsigned long flags;
+	u32 ctrl = AXI_DMAC_CTRL_ENABLE;
 
-	axi_dmac_write(dmac, AXI_DMAC_REG_CTRL, AXI_DMAC_CTRL_ENABLE);
+	if (chan->hw_sg)
+		ctrl |= AXI_DMAC_CTRL_ENABLE_SG;
+
+	axi_dmac_write(dmac, AXI_DMAC_REG_CTRL, ctrl);
 
 	spin_lock_irqsave(&chan->vchan.lock, flags);
 	if (vchan_issue_pending(&chan->vchan))
@@ -476,22 +527,58 @@ static void axi_dmac_issue_pending(struct dma_chan *c)
 	spin_unlock_irqrestore(&chan->vchan.lock, flags);
 }
 
-static struct axi_dmac_desc *axi_dmac_alloc_desc(unsigned int num_sgs)
+static struct axi_dmac_desc *
+axi_dmac_alloc_desc(struct axi_dmac_chan *chan, unsigned int num_sgs)
 {
+	struct axi_dmac *dmac = chan_to_axi_dmac(chan);
+	struct device *dev = dmac->dma_dev.dev;
+	struct axi_dmac_hw_desc *hws;
 	struct axi_dmac_desc *desc;
+	dma_addr_t hw_phys;
 	unsigned int i;
 
 	desc = kzalloc(struct_size(desc, sg, num_sgs), GFP_NOWAIT);
 	if (!desc)
 		return NULL;
 	desc->num_sgs = num_sgs;
+	desc->chan = chan;
+
+	hws = dma_alloc_coherent(dev, PAGE_ALIGN(num_sgs * sizeof(*hws)),
+				&hw_phys, GFP_ATOMIC);
+	if (!hws) {
+		kfree(desc);
+		return NULL;
+	}
 
-	for (i = 0; i < num_sgs; i++)
-		desc->sg[i].id = AXI_DMAC_SG_UNUSED;
+	for (i = 0; i < num_sgs; i++) {
+		desc->sg[i].hw = &hws[i];
+		desc->sg[i].hw_phys = hw_phys + i * sizeof(*hws);
+
+		hws[i].id = AXI_DMAC_SG_UNUSED;
+		hws[i].flags = 0;
+
+		/* Link hardware descriptors */
+		hws[i].next_sg_addr = hw_phys + (i + 1) * sizeof(*hws);
+	}
+
+	/* The last hardware descriptor will trigger an interrupt */
+	desc->sg[num_sgs - 1].hw->flags = AXI_DMAC_HW_FLAG_LAST | AXI_DMAC_HW_FLAG_IRQ;
 
 	return desc;
 }
 
+static void axi_dmac_free_desc(struct axi_dmac_desc *desc)
+{
+	struct axi_dmac *dmac = chan_to_axi_dmac(desc->chan);
+	struct device *dev = dmac->dma_dev.dev;
+	struct axi_dmac_hw_desc *hw = desc->sg[0].hw;
+	dma_addr_t hw_phys = desc->sg[0].hw_phys;
+
+	dma_free_coherent(dev, PAGE_ALIGN(desc->num_sgs * sizeof(*hw)),
+			  hw, hw_phys);
+	kfree(desc);
+}
+
 static struct axi_dmac_sg *axi_dmac_fill_linear_sg(struct axi_dmac_chan *chan,
 	enum dma_transfer_direction direction, dma_addr_t addr,
 	unsigned int num_periods, unsigned int period_len,
@@ -508,26 +595,24 @@ static struct axi_dmac_sg *axi_dmac_fill_linear_sg(struct axi_dmac_chan *chan,
 	segment_size = ((segment_size - 1) | chan->length_align_mask) + 1;
 
 	for (i = 0; i < num_periods; i++) {
-		len = period_len;
-
-		while (len > segment_size) {
+		for (len = period_len; len > segment_size; sg++) {
 			if (direction == DMA_DEV_TO_MEM)
-				sg->dest_addr = addr;
+				sg->hw->dest_addr = addr;
 			else
-				sg->src_addr = addr;
-			sg->x_len = segment_size;
-			sg->y_len = 1;
-			sg++;
+				sg->hw->src_addr = addr;
+			sg->hw->x_len = segment_size - 1;
+			sg->hw->y_len = 0;
+			sg->hw->flags = 0;
 			addr += segment_size;
 			len -= segment_size;
 		}
 
 		if (direction == DMA_DEV_TO_MEM)
-			sg->dest_addr = addr;
+			sg->hw->dest_addr = addr;
 		else
-			sg->src_addr = addr;
-		sg->x_len = len;
-		sg->y_len = 1;
+			sg->hw->src_addr = addr;
+		sg->hw->x_len = len - 1;
+		sg->hw->y_len = 0;
 		sg++;
 		addr += len;
 	}
@@ -554,7 +639,7 @@ static struct dma_async_tx_descriptor *axi_dmac_prep_slave_sg(
 	for_each_sg(sgl, sg, sg_len, i)
 		num_sgs += DIV_ROUND_UP(sg_dma_len(sg), chan->max_length);
 
-	desc = axi_dmac_alloc_desc(num_sgs);
+	desc = axi_dmac_alloc_desc(chan, num_sgs);
 	if (!desc)
 		return NULL;
 
@@ -563,7 +648,7 @@ static struct dma_async_tx_descriptor *axi_dmac_prep_slave_sg(
 	for_each_sg(sgl, sg, sg_len, i) {
 		if (!axi_dmac_check_addr(chan, sg_dma_address(sg)) ||
 		    !axi_dmac_check_len(chan, sg_dma_len(sg))) {
-			kfree(desc);
+			axi_dmac_free_desc(desc);
 			return NULL;
 		}
 
@@ -583,7 +668,7 @@ static struct dma_async_tx_descriptor *axi_dmac_prep_dma_cyclic(
 {
 	struct axi_dmac_chan *chan = to_axi_dmac_chan(c);
 	struct axi_dmac_desc *desc;
-	unsigned int num_periods, num_segments;
+	unsigned int num_periods, num_segments, num_sgs;
 
 	if (direction != chan->direction)
 		return NULL;
@@ -597,11 +682,16 @@ static struct dma_async_tx_descriptor *axi_dmac_prep_dma_cyclic(
 
 	num_periods = buf_len / period_len;
 	num_segments = DIV_ROUND_UP(period_len, chan->max_length);
+	num_sgs = num_periods * num_segments;
 
-	desc = axi_dmac_alloc_desc(num_periods * num_segments);
+	desc = axi_dmac_alloc_desc(chan, num_sgs);
 	if (!desc)
 		return NULL;
 
+	/* Chain the last descriptor to the first, and remove its "last" flag */
+	desc->sg[num_sgs - 1].hw->next_sg_addr = desc->sg[0].hw_phys;
+	desc->sg[num_sgs - 1].hw->flags &= ~AXI_DMAC_HW_FLAG_LAST;
+
 	axi_dmac_fill_linear_sg(chan, direction, buf_addr, num_periods,
 		period_len, desc->sg);
 
@@ -653,26 +743,26 @@ static struct dma_async_tx_descriptor *axi_dmac_prep_interleaved(
 			return NULL;
 	}
 
-	desc = axi_dmac_alloc_desc(1);
+	desc = axi_dmac_alloc_desc(chan, 1);
 	if (!desc)
 		return NULL;
 
 	if (axi_dmac_src_is_mem(chan)) {
-		desc->sg[0].src_addr = xt->src_start;
-		desc->sg[0].src_stride = xt->sgl[0].size + src_icg;
+		desc->sg[0].hw->src_addr = xt->src_start;
+		desc->sg[0].hw->src_stride = xt->sgl[0].size + src_icg;
 	}
 
 	if (axi_dmac_dest_is_mem(chan)) {
-		desc->sg[0].dest_addr = xt->dst_start;
-		desc->sg[0].dest_stride = xt->sgl[0].size + dst_icg;
+		desc->sg[0].hw->dest_addr = xt->dst_start;
+		desc->sg[0].hw->dst_stride = xt->sgl[0].size + dst_icg;
 	}
 
 	if (chan->hw_2d) {
-		desc->sg[0].x_len = xt->sgl[0].size;
-		desc->sg[0].y_len = xt->numf;
+		desc->sg[0].hw->x_len = xt->sgl[0].size - 1;
+		desc->sg[0].hw->y_len = xt->numf - 1;
 	} else {
-		desc->sg[0].x_len = xt->sgl[0].size * xt->numf;
-		desc->sg[0].y_len = 1;
+		desc->sg[0].hw->x_len = xt->sgl[0].size * xt->numf - 1;
+		desc->sg[0].hw->y_len = 0;
 	}
 
 	if (flags & DMA_CYCLIC)
@@ -688,7 +778,7 @@ static void axi_dmac_free_chan_resources(struct dma_chan *c)
 
 static void axi_dmac_desc_free(struct virt_dma_desc *vdesc)
 {
-	kfree(container_of(vdesc, struct axi_dmac_desc, vdesc));
+	axi_dmac_free_desc(to_axi_dmac_desc(vdesc));
 }
 
 static bool axi_dmac_regmap_rdwr(struct device *dev, unsigned int reg)
@@ -714,6 +804,9 @@ static bool axi_dmac_regmap_rdwr(struct device *dev, unsigned int reg)
 	case AXI_DMAC_REG_CURRENT_DEST_ADDR:
 	case AXI_DMAC_REG_PARTIAL_XFER_LEN:
 	case AXI_DMAC_REG_PARTIAL_XFER_ID:
+	case AXI_DMAC_REG_CURRENT_SG_ID:
+	case AXI_DMAC_REG_SG_ADDRESS:
+	case AXI_DMAC_REG_SG_ADDRESS_HIGH:
 		return true;
 	default:
 		return false;
@@ -866,6 +959,10 @@ static int axi_dmac_detect_caps(struct axi_dmac *dmac, unsigned int version)
 	if (axi_dmac_read(dmac, AXI_DMAC_REG_FLAGS) == AXI_DMAC_FLAG_CYCLIC)
 		chan->hw_cyclic = true;
 
+	axi_dmac_write(dmac, AXI_DMAC_REG_SG_ADDRESS, 0xffffffff);
+	if (axi_dmac_read(dmac, AXI_DMAC_REG_SG_ADDRESS))
+		chan->hw_sg = true;
+
 	axi_dmac_write(dmac, AXI_DMAC_REG_Y_LENGTH, 1);
 	if (axi_dmac_read(dmac, AXI_DMAC_REG_Y_LENGTH) == 1)
 		chan->hw_2d = true;
@@ -911,6 +1008,7 @@ static int axi_dmac_probe(struct platform_device *pdev)
 	struct axi_dmac *dmac;
 	struct regmap *regmap;
 	unsigned int version;
+	u32 irq_mask = 0;
 	int ret;
 
 	dmac = devm_kzalloc(&pdev->dev, sizeof(*dmac), GFP_KERNEL);
@@ -966,6 +1064,7 @@ static int axi_dmac_probe(struct platform_device *pdev)
 	dma_dev->dst_addr_widths = BIT(dmac->chan.dest_width);
 	dma_dev->directions = BIT(dmac->chan.direction);
 	dma_dev->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
+	dma_dev->max_sg_burst = 31; /* 31 SGs maximum in one burst */
 	INIT_LIST_HEAD(&dma_dev->channels);
 
 	dmac->chan.vchan.desc_free = axi_dmac_desc_free;
@@ -977,7 +1076,10 @@ static int axi_dmac_probe(struct platform_device *pdev)
 
 	dma_dev->copy_align = (dmac->chan.address_align_mask + 1);
 
-	axi_dmac_write(dmac, AXI_DMAC_REG_IRQ_MASK, 0x00);
+	if (dmac->chan.hw_sg)
+		irq_mask |= AXI_DMAC_IRQ_SOT;
+
+	axi_dmac_write(dmac, AXI_DMAC_REG_IRQ_MASK, irq_mask);
 
 	if (of_dma_is_coherent(pdev->dev.of_node)) {
 		ret = axi_dmac_read(dmac, AXI_DMAC_REG_COHERENCY_DESC);
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index b7388ae62d7f..491b22240221 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -1103,6 +1103,9 @@ EXPORT_SYMBOL_GPL(dma_async_device_channel_register);
 static void __dma_async_device_channel_unregister(struct dma_device *device,
 						  struct dma_chan *chan)
 {
+	if (chan->local == NULL)
+		return;
+
 	WARN_ONCE(!device->device_release && chan->client_count,
 		  "%s called while %d clients hold a reference\n",
 		  __func__, chan->client_count);
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index ffe621695e47..a4f608837849 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -21,6 +21,10 @@
 #include <linux/slab.h>
 #include <linux/wait.h>
 
+static bool nobounce;
+module_param(nobounce, bool, 0644);
+MODULE_PARM_DESC(nobounce, "Prevent using swiotlb buffer (default: use swiotlb buffer)");
+
 static unsigned int test_buf_size = 16384;
 module_param(test_buf_size, uint, 0644);
 MODULE_PARM_DESC(test_buf_size, "Size of the memcpy test buffer");
@@ -90,6 +94,7 @@ MODULE_PARM_DESC(polled, "Use polling for completion instead of interrupts");
 
 /**
  * struct dmatest_params - test parameters.
+ * @nobounce:		prevent using swiotlb buffer
  * @buf_size:		size of the memcpy test buffer
  * @channel:		bus ID of the channel to test
  * @device:		bus ID of the DMA Engine to test
@@ -106,6 +111,7 @@ MODULE_PARM_DESC(polled, "Use polling for completion instead of interrupts");
  * @polled:		use polling for completion instead of interrupts
  */
 struct dmatest_params {
+	bool		nobounce;
 	unsigned int	buf_size;
 	char		channel[20];
 	char		device[32];
@@ -215,6 +221,7 @@ struct dmatest_done {
 struct dmatest_data {
 	u8		**raw;
 	u8		**aligned;
+	gfp_t		gfp_flags;
 	unsigned int	cnt;
 	unsigned int	off;
 };
@@ -533,7 +540,7 @@ static int dmatest_alloc_test_data(struct dmatest_data *d,
 		goto err;
 
 	for (i = 0; i < d->cnt; i++) {
-		d->raw[i] = kmalloc(buf_size + align, GFP_KERNEL);
+		d->raw[i] = kmalloc(buf_size + align, d->gfp_flags);
 		if (!d->raw[i])
 			goto err;
 
@@ -655,6 +662,13 @@ static int dmatest_func(void *data)
 		goto err_free_coefs;
 	}
 
+	src->gfp_flags = GFP_KERNEL;
+	dst->gfp_flags = GFP_KERNEL;
+	if (params->nobounce) {
+		src->gfp_flags = GFP_DMA;
+		dst->gfp_flags = GFP_DMA;
+	}
+
 	if (dmatest_alloc_test_data(src, buf_size, align) < 0)
 		goto err_free_coefs;
 
@@ -1093,6 +1107,7 @@ static void add_threaded_test(struct dmatest_info *info)
 	struct dmatest_params *params = &info->params;
 
 	/* Copy test parameters */
+	params->nobounce = nobounce;
 	params->buf_size = test_buf_size;
 	strscpy(params->channel, strim(test_channel), sizeof(params->channel));
 	strscpy(params->device, strim(test_device), sizeof(params->device));
diff --git a/drivers/dma/dw-edma/dw-edma-v0-debugfs.c b/drivers/dma/dw-edma/dw-edma-v0-debugfs.c
index 0745d9e7d259..406f169b09a7 100644
--- a/drivers/dma/dw-edma/dw-edma-v0-debugfs.c
+++ b/drivers/dma/dw-edma/dw-edma-v0-debugfs.c
@@ -176,7 +176,7 @@ dw_edma_debugfs_regs_wr(struct dw_edma *dw, struct dentry *dent)
 	};
 	struct dentry *regs_dent, *ch_dent;
 	int nr_entries, i;
-	char name[16];
+	char name[32];
 
 	regs_dent = debugfs_create_dir(WRITE_STR, dent);
 
@@ -239,7 +239,7 @@ static noinline_for_stack void dw_edma_debugfs_regs_rd(struct dw_edma *dw,
 	};
 	struct dentry *regs_dent, *ch_dent;
 	int nr_entries, i;
-	char name[16];
+	char name[32];
 
 	regs_dent = debugfs_create_dir(READ_STR, dent);
 
diff --git a/drivers/dma/dw-edma/dw-hdma-v0-debugfs.c b/drivers/dma/dw-edma/dw-hdma-v0-debugfs.c
index 520c81978b08..dcdc57fe976c 100644
--- a/drivers/dma/dw-edma/dw-hdma-v0-debugfs.c
+++ b/drivers/dma/dw-edma/dw-hdma-v0-debugfs.c
@@ -116,7 +116,7 @@ static void dw_hdma_debugfs_regs_ch(struct dw_edma *dw, enum dw_edma_dir dir,
 static void dw_hdma_debugfs_regs_wr(struct dw_edma *dw, struct dentry *dent)
 {
 	struct dentry *regs_dent, *ch_dent;
-	char name[16];
+	char name[32];
 	int i;
 
 	regs_dent = debugfs_create_dir(WRITE_STR, dent);
@@ -133,7 +133,7 @@ static void dw_hdma_debugfs_regs_wr(struct dw_edma *dw, struct dentry *dent)
 static void dw_hdma_debugfs_regs_rd(struct dw_edma *dw, struct dentry *dent)
 {
 	struct dentry *regs_dent, *ch_dent;
-	char name[16];
+	char name[32];
 	int i;
 
 	regs_dent = debugfs_create_dir(READ_STR, dent);
diff --git a/drivers/dma/fsl-edma-main.c b/drivers/dma/fsl-edma-main.c
index 238a69bd0d6f..45cc419b1b4a 100644
--- a/drivers/dma/fsl-edma-main.c
+++ b/drivers/dma/fsl-edma-main.c
@@ -9,6 +9,7 @@
  * Vybrid and Layerscape SoCs.
  */
 
+#include <dt-bindings/dma/fsl-edma.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
 #include <linux/clk.h>
@@ -21,10 +22,6 @@
 
 #include "fsl-edma-common.h"
 
-#define ARGS_RX                         BIT(0)