Merge branch 'next' of git://git.infradead.org/users/vkoul/slave-dma

* 'next' of git://git.infradead.org/users/vkoul/slave-dma: (63 commits)
  dmaengine: mid_dma: mask_peripheral_interrupt only when dmac is idle
  dmaengine/ep93xx_dma: add module.h include
  pch_dma: Reduce wasting memory
  pch_dma: Fix suspend issue
  dma/timberdale: free_irq() on an error path
  dma: shdma: transfer based runtime PM
  dmaengine: shdma: protect against the IRQ handler
  dmaengine i.MX DMA/SDMA: add missing include of linux/module.h
  dmaengine: delete redundant chan_id and chancnt initialization in dma drivers
  dmaengine/amba-pl08x: Check txd->llis_va before freeing dma_pool
  dmaengine/amba-pl08x: Add support for sg len greater than one for slave transfers
  serial: sh-sci: don't filter on DMA device, use only channel ID
  ARM: SAMSUNG: Remove Samsung specific enum type for dma direction
  ASoC: Samsung: Update DMA interface
  spi/s3c64xx: Merge dma control code
  spi/s3c64xx: Add support DMA engine API
  ARM: SAMSUNG: Remove S3C-PL330-DMA driver
  ARM: S5P64X0: Use generic DMA PL330 driver
  ARM: S5PC100: Use generic DMA PL330 driver
  ARM: S5PV210: Use generic DMA PL330 driver
  ...

Fix up fairly trivial conflicts in
 - arch/arm/mach-exynos4/{Kconfig,clock.c}
 - arch/arm/mach-s5p64x0/dma.c

17 files changed, 864 insertions, 479 deletions

diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 2e3b3d38c465..ab8f469f5cf8 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -193,7 +193,8 @@ config ARCH_HAS_ASYNC_TX_FIND_CHANNEL
 config PL330_DMA
 	tristate "DMA API Driver for PL330"
 	select DMA_ENGINE
-	depends on PL330
+	depends on ARM_AMBA
+	select PL330
 	help
 	  Select if your platform has one or more PL330 DMACs.
 	  You need to provide platform specific settings via
diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c
index be21e3f138a8..b7cbd1ab1db1 100644
--- a/drivers/dma/amba-pl08x.c
+++ b/drivers/dma/amba-pl08x.c
@@ -66,32 +66,29 @@
  *    after the final transfer signalled by LBREQ or LSREQ.  The DMAC
  *    will then move to the next LLI entry.
  *
- * Only the former works sanely with scatter lists, so we only implement
- * the DMAC flow control method.  However, peripherals which use the LBREQ
- * and LSREQ signals (eg, MMCI) are unable to use this mode, which through
- * these hardware restrictions prevents them from using scatter DMA.
- *
  * Global TODO:
  * - Break out common code from arch/arm/mach-s3c64xx and share
  */
-#include <linux/device.h>
-#include <linux/init.h>
-#include <linux/module.h>
-#include <linux/interrupt.h>
-#include <linux/slab.h>
-#include <linux/delay.h>
-#include <linux/dma-mapping.h>
-#include <linux/dmapool.h>
-#include <linux/dmaengine.h>
 #include <linux/amba/bus.h>
 #include <linux/amba/pl08x.h>
 #include <linux/debugfs.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/dmaengine.h>
+#include <linux/dmapool.h>
+#include <linux/dma-mapping.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/pm_runtime.h>
 #include <linux/seq_file.h>
-
+#include <linux/slab.h>
 #include <asm/hardware/pl080.h>
 
 #define DRIVER_NAME	"pl08xdmac"
 
+static struct amba_driver pl08x_amba_driver;
+
 /**
  * struct vendor_data - vendor-specific config parameters for PL08x derivatives
  * @channels: the number of channels available in this variant
@@ -126,7 +123,8 @@ struct pl08x_lli {
  * @phy_chans: array of data for the physical channels
  * @pool: a pool for the LLI descriptors
  * @pool_ctr: counter of LLIs in the pool
- * @lli_buses: bitmask to or in to LLI pointer selecting AHB port for LLI fetches
+ * @lli_buses: bitmask to or in to LLI pointer selecting AHB port for LLI
+ * fetches
  * @mem_buses: set to indicate memory transfers on AHB2.
  * @lock: a spinlock for this struct
  */
@@ -149,14 +147,6 @@ struct pl08x_driver_data {
  * PL08X specific defines
  */
 
-/*
- * Memory boundaries: the manual for PL08x says that the controller
- * cannot read past a 1KiB boundary, so these defines are used to
- * create transfer LLIs that do not cross such boundaries.
- */
-#define PL08X_BOUNDARY_SHIFT		(10)	/* 1KB 0x400 */
-#define PL08X_BOUNDARY_SIZE		(1 << PL08X_BOUNDARY_SHIFT)
-
 /* Size (bytes) of each LLI buffer allocated for one transfer */
 # define PL08X_LLI_TSFR_SIZE	0x2000
 
@@ -272,7 +262,6 @@ static void pl08x_resume_phy_chan(struct pl08x_phy_chan *ch)
 	writel(val, ch->base + PL080_CH_CONFIG);
 }
 
-
 /*
  * pl08x_terminate_phy_chan() stops the channel, clears the FIFO and
  * clears any pending interrupt status.  This should not be used for
@@ -363,7 +352,9 @@ static u32 pl08x_getbytes_chan(struct pl08x_dma_chan *plchan)
 	if (!list_empty(&plchan->pend_list)) {
 		struct pl08x_txd *txdi;
 		list_for_each_entry(txdi, &plchan->pend_list, node) {
-			bytes += txdi->len;
+			struct pl08x_sg *dsg;
+			list_for_each_entry(dsg, &txd->dsg_list, node)
+				bytes += dsg->len;
 		}
 	}
 
@@ -407,6 +398,7 @@ pl08x_get_phy_channel(struct pl08x_driver_data *pl08x,
 		return NULL;
 	}
 
+	pm_runtime_get_sync(&pl08x->adev->dev);
 	return ch;
 }
 
@@ -420,6 +412,8 @@ static inline void pl08x_put_phy_channel(struct pl08x_driver_data *pl08x,
 	/* Stop the channel and clear its interrupts */
 	pl08x_terminate_phy_chan(pl08x, ch);
 
+	pm_runtime_put(&pl08x->adev->dev);
+
 	/* Mark it as free */
 	ch->serving = NULL;
 	spin_unlock_irqrestore(&ch->lock, flags);
@@ -499,36 +493,30 @@ struct pl08x_lli_build_data {
 };
 
 /*
- * Autoselect a master bus to use for the transfer this prefers the
- * destination bus if both available if fixed address on one bus the
- * other will be chosen
+ * Autoselect a master bus to use for the transfer. Slave will be the chosen as
+ * victim in case src & dest are not similarly aligned. i.e. If after aligning
+ * masters address with width requirements of transfer (by sending few byte by
+ * byte data), slave is still not aligned, then its width will be reduced to
+ * BYTE.
+ * - prefers the destination bus if both available
+ * - prefers bus with fixed address (i.e. peripheral)
  */
 static void pl08x_choose_master_bus(struct pl08x_lli_build_data *bd,
 	struct pl08x_bus_data **mbus, struct pl08x_bus_data **sbus, u32 cctl)
 {
 	if (!(cctl & PL080_CONTROL_DST_INCR)) {
-		*mbus = &bd->srcbus;
-		*sbus = &bd->dstbus;
-	} else if (!(cctl & PL080_CONTROL_SRC_INCR)) {
 		*mbus = &bd->dstbus;
 		*sbus = &bd->srcbus;
+	} else if (!(cctl & PL080_CONTROL_SRC_INCR)) {
+		*mbus = &bd->srcbus;
+		*sbus = &bd->dstbus;
 	} else {
-		if (bd->dstbus.buswidth == 4) {
+		if (bd->dstbus.buswidth >= bd->srcbus.buswidth) {
 			*mbus = &bd->dstbus;
 			*sbus = &bd->srcbus;
-		} else if (bd->srcbus.buswidth == 4) {
-			*mbus = &bd->srcbus;
-			*sbus = &bd->dstbus;
-		} else if (bd->dstbus.buswidth == 2) {
-			*mbus = &bd->dstbus;
-			*sbus = &bd->srcbus;
-		} else if (bd->srcbus.buswidth == 2) {
+		} else {
 			*mbus = &bd->srcbus;
 			*sbus = &bd->dstbus;
-		} else {
-			/* bd->srcbus.buswidth == 1 */
-			*mbus = &bd->dstbus;
-			*sbus = &bd->srcbus;
 		}
 	}
 }
@@ -547,7 +535,8 @@ static void pl08x_fill_lli_for_desc(struct pl08x_lli_build_data *bd,
 	llis_va[num_llis].cctl = cctl;
 	llis_va[num_llis].src = bd->srcbus.addr;
 	llis_va[num_llis].dst = bd->dstbus.addr;
-	llis_va[num_llis].lli = llis_bus + (num_llis + 1) * sizeof(struct pl08x_lli);
+	llis_va[num_llis].lli = llis_bus + (num_llis + 1) *
+		sizeof(struct pl08x_lli);
 	llis_va[num_llis].lli |= bd->lli_bus;
 
 	if (cctl & PL080_CONTROL_SRC_INCR)
@@ -560,16 +549,12 @@ static void pl08x_fill_lli_for_desc(struct pl08x_lli_build_data *bd,
 	bd->remainder -= len;
 }
 
-/*
- * Return number of bytes to fill to boundary, or len.
- * This calculation works for any value of addr.
- */
-static inline size_t pl08x_pre_boundary(u32 addr, size_t len)
+static inline void prep_byte_width_lli(struct pl08x_lli_build_data *bd,
+		u32 *cctl, u32 len, int num_llis, size_t *total_bytes)
 {
-	size_t boundary_len = PL08X_BOUNDARY_SIZE -
-			(addr & (PL08X_BOUNDARY_SIZE - 1));
-
-	return min(boundary_len, len);
+	*cctl = pl08x_cctl_bits(*cctl, 1, 1, len);
+	pl08x_fill_lli_for_desc(bd, num_llis, len, *cctl);
+	(*total_bytes) += len;
 }
 
 /*
@@ -583,13 +568,12 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x,
 	struct pl08x_bus_data *mbus, *sbus;
 	struct pl08x_lli_build_data bd;
 	int num_llis = 0;
-	u32 cctl;
-	size_t max_bytes_per_lli;
-	size_t total_bytes = 0;
+	u32 cctl, early_bytes = 0;
+	size_t max_bytes_per_lli, total_bytes;
 	struct pl08x_lli *llis_va;
+	struct pl08x_sg *dsg;
 
-	txd->llis_va = dma_pool_alloc(pl08x->pool, GFP_NOWAIT,
-				      &txd->llis_bus);
+	txd->llis_va = dma_pool_alloc(pl08x->pool, GFP_NOWAIT, &txd->llis_bus);
 	if (!txd->llis_va) {
 		dev_err(&pl08x->adev->dev, "%s no memory for llis\n", __func__);
 		return 0;
@@ -597,13 +581,9 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x,
 
 	pl08x->pool_ctr++;
 
-	/* Get the default CCTL */
-	cctl = txd->cctl;
-
 	bd.txd = txd;
-	bd.srcbus.addr = txd->src_addr;
-	bd.dstbus.addr = txd->dst_addr;
 	bd.lli_bus = (pl08x->lli_buses & PL08X_AHB2) ? PL080_LLI_LM_AHB2 : 0;
+	cctl = txd->cctl;
 
 	/* Find maximum width of the source bus */
 	bd.srcbus.maxwidth =
@@ -615,215 +595,179 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x,
 		pl08x_get_bytes_for_cctl((cctl & PL080_CONTROL_DWIDTH_MASK) >>
 				       PL080_CONTROL_DWIDTH_SHIFT);
 
-	/* Set up the bus widths to the maximum */
-	bd.srcbus.buswidth = bd.srcbus.maxwidth;
-	bd.dstbus.buswidth = bd.dstbus.maxwidth;
+	list_for_each_entry(dsg, &txd->dsg_list, node) {
+		total_bytes = 0;
+		cctl = txd->cctl;
 
-	/*
-	 * Bytes transferred == tsize * MIN(buswidths), not max(buswidths)
-	 */
-	max_bytes_per_lli = min(bd.srcbus.buswidth, bd.dstbus.buswidth) *
-		PL080_CONTROL_TRANSFER_SIZE_MASK;
+		bd.srcbus.addr = dsg->src_addr;
+		bd.dstbus.addr = dsg->dst_addr;
+		bd.remainder = dsg->len;
+		bd.srcbus.buswidth = bd.srcbus.maxwidth;
+		bd.dstbus.buswidth = bd.dstbus.maxwidth;
 
-	/* We need to count this down to zero */
-	bd.remainder = txd->len;
+		pl08x_choose_master_bus(&bd, &mbus, &sbus, cctl);
 
-	/*
-	 * Choose bus to align to
-	 * - prefers destination bus if both available
-	 * - if fixed address on one bus chooses other
-	 */
-	pl08x_choose_master_bus(&bd, &mbus, &sbus, cctl);
-
-	dev_vdbg(&pl08x->adev->dev, "src=0x%08x%s/%u dst=0x%08x%s/%u len=%zu llimax=%zu\n",
-		 bd.srcbus.addr, cctl & PL080_CONTROL_SRC_INCR ? "+" : "",
-		 bd.srcbus.buswidth,
-		 bd.dstbus.addr, cctl & PL080_CONTROL_DST_INCR ? "+" : "",
-		 bd.dstbus.buswidth,
-		 bd.remainder, max_bytes_per_lli);
-	dev_vdbg(&pl08x->adev->dev, "mbus=%s sbus=%s\n",
-		 mbus == &bd.srcbus ? "src" : "dst",
-		 sbus == &bd.srcbus ? "src" : "dst");
-
-	if (txd->len < mbus->buswidth) {
-		/* Less than a bus width available - send as single bytes */
-		while (bd.remainder) {
-			dev_vdbg(&pl08x->adev->dev,
-				 "%s single byte LLIs for a transfer of "
-				 "less than a bus width (remain 0x%08x)\n",
-				 __func__, bd.remainder);
-			cctl = pl08x_cctl_bits(cctl, 1, 1, 1);
-			pl08x_fill_lli_for_desc(&bd, num_llis++, 1, cctl);
-			total_bytes++;
-		}
-	} else {
-		/* Make one byte LLIs until master bus is aligned */
-		while ((mbus->addr) % (mbus->buswidth)) {
-			dev_vdbg(&pl08x->adev->dev,
-				"%s adjustment lli for less than bus width "
-				 "(remain 0x%08x)\n",
-				 __func__, bd.remainder);
-			cctl = pl08x_cctl_bits(cctl, 1, 1, 1);
-			pl08x_fill_lli_for_desc(&bd, num_llis++, 1, cctl);
-			total_bytes++;
-		}
+		dev_vdbg(&pl08x->adev->dev, "src=0x%08x%s/%u dst=0x%08x%s/%u len=%zu\n",
+			bd.srcbus.addr, cctl & PL080_CONTROL_SRC_INCR ? "+" : "",
+			bd.srcbus.buswidth,
+			bd.dstbus.addr, cctl & PL080_CONTROL_DST_INCR ? "+" : "",
+			bd.dstbus.buswidth,
+			bd.remainder);
+		dev_vdbg(&pl08x->adev->dev, "mbus=%s sbus=%s\n",
+			mbus == &bd.srcbus ? "src" : "dst",
+			sbus == &bd.srcbus ? "src" : "dst");
 
 		/*
-		 * Master now aligned
-		 * - if slave is not then we must set its width down
+		 * Zero length is only allowed if all these requirements are
+		 * met:
+		 * - flow controller is peripheral.
+		 * - src.addr is aligned to src.width
+		 * - dst.addr is aligned to dst.width
+		 *
+		 * sg_len == 1 should be true, as there can be two cases here:
+		 *
+		 * - Memory addresses are contiguous and are not scattered.
+		 *   Here, Only one sg will be passed by user driver, with
+		 *   memory address and zero length. We pass this to controller
+		 *   and after the transfer it will receive the last burst
+		 *   request from peripheral and so transfer finishes.
+		 *
+		 * - Memory addresses are scattered and are not contiguous.
+		 *   Here, Obviously as DMA controller doesn't know when a lli's
+		 *   transfer gets over, it can't load next lli. So in this
+		 *   case, there has to be an assumption that only one lli is
+		 *   supported. Thus, we can't have scattered addresses.
 		 */
-		if (sbus->addr % sbus->buswidth) {
-			dev_dbg(&pl08x->adev->dev,
-				"%s set down bus width to one byte\n",
-				 __func__);
+		if (!bd.remainder) {
+			u32 fc = (txd->ccfg & PL080_CONFIG_FLOW_CONTROL_MASK) >>
+				PL080_CONFIG_FLOW_CONTROL_SHIFT;
+			if (!((fc >= PL080_FLOW_SRC2DST_DST) &&
+					(fc <= PL080_FLOW_SRC2DST_SRC))) {
+				dev_err(&pl08x->adev->dev, "%s sg len can't be zero",
+					__func__);
+				return 0;
+			}
+
+			if ((bd.srcbus.addr % bd.srcbus.buswidth) ||
+					(bd.srcbus.addr % bd.srcbus.buswidth)) {
+				dev_err(&pl08x->adev->dev,
+					"%s src & dst address must be aligned to src"
+					" & dst width if peripheral is flow controller",
+					__func__);
+				return 0;
+			}
 
-			sbus->buswidth = 1;
+			cctl = pl08x_cctl_bits(cctl, bd.srcbus.buswidth,
+					bd.dstbus.buswidth, 0);
+			pl08x_fill_lli_for_desc(&bd, num_llis++, 0, cctl);
+			break;
 		}
 
 		/*
-		 * Make largest possible LLIs until less than one bus
-		 * width left
+		 * Send byte by byte for following cases
+		 * - Less than a bus width available
+		 * - until master bus is aligned
 		 */
-		while (bd.remainder > (mbus->buswidth - 1)) {
-			size_t lli_len, target_len, tsize, odd_bytes;
+		if (bd.remainder < mbus->buswidth)
+			early_bytes = bd.remainder;
+		else if ((mbus->addr) % (mbus->buswidth)) {
+			early_bytes = mbus->buswidth - (mbus->addr) %
+				(mbus->buswidth);
+			if ((bd.remainder - early_bytes) < mbus->buswidth)
+				early_bytes = bd.remainder;
+		}
 
+		if (early_bytes) {
+			dev_vdbg(&pl08x->adev->dev,
+				"%s byte width LLIs (remain 0x%08x)\n",
+				__func__, bd.remainder);
+			prep_byte_width_lli(&bd, &cctl, early_bytes, num_llis++,
+				&total_bytes);
+		}
+
+		if (bd.remainder) {
 			/*
-			 * If enough left try to send max possible,
-			 * otherwise try to send the remainder
+			 * Master now aligned
+			 * - if slave is not then we must set its width down
 			 */
-			target_len = min(bd.remainder, max_bytes_per_lli);
+			if (sbus->addr % sbus->buswidth) {
+				dev_dbg(&pl08x->adev->dev,
+					"%s set down bus width to one byte\n",
+					__func__);
+
+				sbus->buswidth = 1;
+			}
 
 			/*
-			 * Set bus lengths for incrementing buses to the
-			 * number of bytes which fill to next memory boundary,
-			 * limiting on the target length calculated above.
+			 * Bytes transferred = tsize * src width, not
+			 * MIN(buswidths)
 			 */
-			if (cctl & PL080_CONTROL_SRC_INCR)
-				bd.srcbus.fill_bytes =
-					pl08x_pre_boundary(bd.srcbus.addr,
-						target_len);
-			else
-				bd.srcbus.fill_bytes = target_len;
-
-			if (cctl & PL080_CONTROL_DST_INCR)
-				bd.dstbus.fill_bytes =
-					pl08x_pre_boundary(bd.dstbus.addr,
-						target_len);
-			else
-				bd.dstbus.fill_bytes = target_len;
-
-			/* Find the nearest */
-			lli_len	= min(bd.srcbus.fill_bytes,
-				      bd.dstbus.fill_bytes);
-
-			BUG_ON(lli_len > bd.remainder);
-
-			if (lli_len <= 0) {
-				dev_err(&pl08x->adev->dev,
-					"%s lli_len is %zu, <= 0\n",
-						__func__, lli_len);
-				return 0;
-			}
+			max_bytes_per_lli = bd.srcbus.buswidth *
+				PL080_CONTROL_TRANSFER_SIZE_MASK;
+			dev_vdbg(&pl08x->adev->dev,
+				"%s max bytes per lli = %zu\n",
+				__func__, max_bytes_per_lli);
+
+			/*
+			 * Make largest possible LLIs until less than one bus
+			 * width left
+			 */
+			while (bd.remainder > (mbus->buswidth - 1)) {
+				size_t lli_len, tsize, width;
 
-			if (lli_len == target_len) {
-				/*
-				 * Can send what we wanted.
-				 * Maintain alignment
-				 */
-				lli_len	= (lli_len/mbus->buswidth) *
-							mbus->buswidth;
-				odd_bytes = 0;
-			} else {
 				/*
-				 * So now we know how many bytes to transfer
-				 * to get to the nearest boundary.  The next
-				 * LLI will past the boundary.  However, we
-				 * may be working to a boundary on the slave
-				 * bus.  We need to ensure the master stays
-				 * aligned, and that we are working in
-				 * multiples of the bus widths.
+				 * If enough left try to send max possible,
+				 * otherwise try to send the remainder
 				 */
-				odd_bytes = lli_len % mbus->buswidth;
-				lli_len -= odd_bytes;
-
-			}
+				lli_len = min(bd.remainder, max_bytes_per_lli);
 
-			if (lli_len) {
 				/*
-				 * Check against minimum bus alignment:
-				 * Calculate actual transfer size in relation
-				 * to bus width an get a maximum remainder of
-				 * the smallest bus width - 1
+				 * Check against maximum bus alignment:
+				 * Calculate actual transfer size in relation to
+				 * bus width an get a maximum remainder of the
+				 * highest bus width - 1
 				 */
-				/* FIXME: use round_down()? */
-				tsize = lli_len / min(mbus->buswidth,
-						      sbus->buswidth);
-				lli_len	= tsize * min(mbus->buswidth,
-						      sbus->buswidth);
-
-				if (target_len != lli_len) {
-					dev_vdbg(&pl08x->adev->dev,
-					"%s can't send what we want. Desired 0x%08zx, lli of 0x%08zx bytes in txd of 0x%08zx\n",
-					__func__, target_len, lli_len, txd->len);
-				}
-
-				cctl = pl08x_cctl_bits(cctl,
-						       bd.srcbus.buswidth,
-						       bd.dstbus.buswidth,
-						       tsize);
+				width = max(mbus->buswidth, sbus->buswidth);
+				lli_len = (lli_len / width) * width;
+				tsize = lli_len / bd.srcbus.buswidth;
 
 				dev_vdbg(&pl08x->adev->dev,
-					"%s fill lli with single lli chunk of size 0x%08zx (remainder 0x%08zx)\n",
+					"%s fill lli with single lli chunk of "
+					"size 0x%08zx (remainder 0x%08zx)\n",
 					__func__, lli_len, bd.remainder);
+
+				cctl = pl08x_cctl_bits(cctl, bd.srcbus.buswidth,
+					bd.dstbus.buswidth, tsize);
 				pl08x_fill_lli_for_desc(&bd, num_llis++,
-					lli_len, cctl);
+						lli_len, cctl);
 				total_bytes += lli_len;
 			}
 
-
-			if (odd_bytes) {
-				/*
-				 * Creep past the boundary, maintaining
-				 * master alignment
-				 */
-				int j;
-				for (j = 0; (j < mbus->buswidth)
-						&& (bd.remainder); j++) {
-					cctl = pl08x_cctl_bits(cctl, 1, 1, 1);
-					dev_vdbg(&pl08x->adev->dev,
-						"%s align with boundary, single byte (remain 0x%08zx)\n",
-						__func__, bd.remainder);
-					pl08x_fill_lli_for_desc(&bd,
-						num_llis++, 1, cctl);
-					total_bytes++;
-				}
+			/*
+			 * Send any odd bytes
+			 */
+			if (bd.remainder) {
+				dev_vdbg(&pl08x->adev->dev,
+					"%s align with boundary, send odd bytes (remain %zu)\n",
+					__func__, bd.remainder);
+				prep_byte_width_lli(&bd, &cctl, bd.remainder,
+						num_llis++, &total_bytes);
 			}
 		}
 
-		/*
-		 * Send any odd bytes
-		 */
-		while (bd.remainder) {
-			cctl = pl08x_cctl_bits(cctl, 1, 1, 1);
-			dev_vdbg(&pl08x->adev->dev,
-				"%s align with boundary, single odd byte (remain %zu)\n",
-				__func__, bd.remainder);
-			pl08x_fill_lli_for_desc(&bd, num_llis++, 1, cctl);
-			total_bytes++;
+		if (total_bytes != dsg->len) {
+			dev_err(&pl08x->adev->dev,
+				"%s size of encoded lli:s don't match total txd, transferred 0x%08zx from size 0x%08zx\n",
+				__func__, total_bytes, dsg->len);
+			return 0;
 		}
-	}
-	if (total_bytes != txd->len) {
-		dev_err(&pl08x->adev->dev,
-			"%s size of encoded lli:s don't match total txd, transferred 0x%08zx from size 0x%08zx\n",
-			__func__, total_bytes, txd->len);
-		return 0;
-	}
 
-	if (num_llis >= MAX_NUM_TSFR_LLIS) {
-		dev_err(&pl08x->adev->dev,
-			"%s need to increase MAX_NUM_TSFR_LLIS from 0x%08x\n",
-			__func__, (u32) MAX_NUM_TSFR_LLIS);
-		return 0;
+		if (num_llis >= MAX_NUM_TSFR_LLIS) {
+			dev_err(&pl08x->adev->dev,
+				"%s need to increase MAX_NUM_TSFR_LLIS from 0x%08x\n",
+				__func__, (u32) MAX_NUM_TSFR_LLIS);
+			return 0;
+		}
 	}
 
 	llis_va = txd->llis_va;
@@ -856,11 +800,19 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x,
 static void pl08x_free_txd(struct pl08x_driver_data *pl08x,
 			   struct pl08x_txd *txd)
 {
+	struct pl08x_sg *dsg, *_dsg;
+
 	/* Free the LLI */
-	dma_pool_free(pl08x->pool, txd->llis_va, txd->llis_bus);
+	if (txd->llis_va)
+		dma_pool_free(pl08x->pool, txd->llis_va, txd->llis_bus);
 
 	pl08x->pool_ctr--;
 
+	list_for_each_entry_safe(dsg, _dsg, &txd->dsg_list, node) {
+		list_del(&dsg->node);
+		kfree(dsg);
+	}
+
 	kfree(txd);
 }
 
@@ -917,9 +869,7 @@ static int prep_phy_channel(struct pl08x_dma_chan *plchan,
 	 * need, but for slaves the physical signals may be muxed!
 	 * Can the platform allow us to use this channel?
 	 */
-	if (plchan->slave &&
-	    ch->signal < 0 &&
-	    pl08x->pd->get_signal) {
+	if (plchan->slave && pl08x->pd->get_signal) {
 		ret = pl08x->pd->get_signal(plchan);
 		if (ret < 0) {
 			dev_dbg(&pl08x->adev->dev,
@@ -1008,10 +958,8 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_interrupt(
  * If slaves are relying on interrupts to signal completion this function
  * must not be called with interrupts disabled.
  */
-static enum dma_status
-pl08x_dma_tx_status(struct dma_chan *chan,
-		    dma_cookie_t cookie,
-		    struct dma_tx_state *txstate)
+static enum dma_status pl08x_dma_tx_status(struct dma_chan *chan,
+		dma_cookie_t cookie, struct dma_tx_state *txstate)
 {
 	struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
 	dma_cookie_t last_used;
@@ -1253,7 +1201,9 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan,
 
 	num_llis = pl08x_fill_llis_for_desc(pl08x, txd);
 	if (!num_llis) {
-		kfree(txd);
+		spin_lock_irqsave(&plchan->lock, flags);
+		pl08x_free_txd(pl08x, txd);
+		spin_unlock_irqrestore(&plchan->lock, flags);
 		return -EINVAL;
 	}
 
@@ -1301,13 +1251,14 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan,
 static struct pl08x_txd *pl08x_get_txd(struct pl08x_dma_chan *plchan,
 	unsigned long flags)
 {
-	struct pl08x_txd *txd = kzalloc(sizeof(struct pl08x_txd), GFP_NOWAIT);
+	struct pl08x_txd *txd = kzalloc(sizeof(*txd), GFP_NOWAIT);
 
 	if (txd) {
 		dma_async_tx_descriptor_init(&txd->tx, &plchan->chan);
 		txd->tx.flags = flags;
 		txd->tx.tx_submit = pl08x_tx_submit;
 		INIT_LIST_HEAD(&txd->node);
+		INIT_LIST_HEAD(&txd->dsg_list);
 
 		/* Always enable error and terminal interrupts */
 		txd->ccfg = PL080_CONFIG_ERR_IRQ_MASK |
@@ -1326,6 +1277,7 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy(
 	struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
 	struct pl08x_driver_data *pl08x = plchan->host;
 	struct pl08x_txd *txd;
+	struct pl08x_sg *dsg;
 	int ret;
 
 	txd = pl08x_get_txd(plchan, flags);
@@ -1335,10 +1287,19 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy(
 		return NULL;
 	}
 
+	dsg = kzalloc(sizeof(struct pl08x_sg), GFP_NOWAIT);
+	if (!dsg) {
+		pl08x_free_txd(pl08x, txd);
+		dev_err(&pl08x->adev->dev, "%s no memory for pl080 sg\n",
+				__func__);
+		return NULL;
+	}
+	list_add_tail(&dsg->node, &txd->dsg_list);
+
 	txd->direction = DMA_NONE;
-	txd->src_addr = src;
-	txd->dst_addr = dest;
-	txd->len = len;
+	dsg->src_addr = src;
+	dsg->dst_addr = dest;
+	dsg->len = len;
 
 	/* Set platform data for m2m */
 	txd->ccfg |= PL080_FLOW_MEM2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT;
@@ -1367,19 +1328,13 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg(
 	struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
 	struct pl08x_driver_data *pl08x = plchan->host;
 	struct pl08x_txd *txd;
-	int ret;
-
-	/*
-	 * Current implementation ASSUMES only one sg
-	 */
-	if (sg_len != 1) {
-		dev_err(&pl08x->adev->dev, "%s prepared too long sglist\n",
-			__func__);
-		BUG();
-	}
+	struct pl08x_sg *dsg;
+	struct scatterlist *sg;
+	dma_addr_t slave_addr;
+	int ret, tmp;
 
 	dev_dbg(&pl08x->adev->dev, "%s prepare transaction of %d bytes from %s\n",
-		__func__, sgl->length, plchan->name);
+			__func__, sgl->length, plchan->name);
 
 	txd = pl08x_get_txd(plchan, flags);
 	if (!txd) {
@@ -1398,24 +1353,49 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg(
 	 * channel target address dynamically at runtime.
 	 */
 	txd->direction = direction;
-	txd->len = sgl->length;
 
 	if (direction == DMA_TO_DEVICE) {
-		txd->ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT;
 		txd->cctl = plchan->dst_cctl;
-		txd->src_addr = sgl->dma_address;
-		txd->dst_addr = plchan->dst_addr;
+		slave_addr = plchan->dst_addr;
 	} else if (direction == DMA_FROM_DEVICE) {
-		txd->ccfg |= PL080_FLOW_PER2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT;
 		txd->cctl = plchan->src_cctl;
-		txd->src_addr = plchan->src_addr;
-		txd->dst_addr = sgl->dma_address;
+		slave_addr = plchan->src_addr;
 	} else {
+		pl08x_free_txd(pl08x, txd);
 		dev_err(&pl08x->adev->dev,
 			"%s direction unsupported\n", __func__);
 		return NULL;
 	}
 
+	if (plchan->cd->device_fc)
+		tmp = (direction == DMA_TO_DEVICE) ? PL080_FLOW_MEM2PER_PER :
+			PL080_FLOW_PER2MEM_PER;
+	else
+		tmp = (direction == DMA_TO_DEVICE) ? PL080_FLOW_MEM2PER :
+			PL080_FLOW_PER2MEM;
+
+	txd->ccfg |= tmp << PL080_CONFIG_FLOW_CONTROL_SHIFT;
+
+	for_each_sg(sgl, sg, sg_len, tmp) {
+		dsg = kzalloc(sizeof(struct pl08x_sg), GFP_NOWAIT);
+		if (!dsg) {
+			pl08x_free_txd(pl08x, txd);
+			dev_err(&pl08x->adev->dev, "%s no mem for pl080 sg\n",
+					__func__);
+			return NULL;
+		}
+		list_add_tail(&dsg->node, &txd->dsg_list);
+
+		dsg->len = sg_dma_len(sg);
+		if (direction == DMA_TO_DEVICE) {
+			dsg->src_addr = sg_phys(sg);
+			dsg->dst_addr = slave_addr;
+		} else {
+			dsg->src_addr = slave_addr;
+			dsg->dst_addr = sg_phys(sg);
+		}
+	}
+
 	ret = pl08x_prep_channel_resources(plchan, txd);
 	if (ret)
 		return NULL;
@@ -1489,9 +1469,15 @@ static int pl08x_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 
 bool pl08x_filter_id(struct dma_chan *chan, void *chan_id)
 {
-	struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+	struct pl08x_dma_chan *plchan;
 	char *name = chan_id;
 
+	/* Reject channels for devices not bound to this driver */
+	if (chan->device->dev->driver != &pl08x_amba_driver.drv)
+		return false;
+
+	plchan = to_pl08x_chan(chan);
+
 	/* Check that the channel is not taken! */
 	if (!strcmp(plchan->name, name))
 		return true;
@@ -1507,34 +1493,34 @@ bool pl08x_filter_id(struct dma_chan *chan, void *chan_id)
  */
 static void pl08x_ensure_on(struct pl08x_driver_data *pl08x)
 {
-	u32 val;
-
-	val = readl(pl08x->base + PL080_CONFIG);
-	val &= ~(PL080_CONFIG_M2_BE | PL080_CONFIG_M1_BE | PL080_CONFIG_ENABLE);
-	/* We implicitly clear bit 1 and that means little-endian mode */
-	val |= PL080_CONFIG_ENABLE;
-	writel(val, pl08x->base + PL080_CONFIG);
+	writel(PL080_CONFIG_ENABLE, pl08x->base + PL080_CONFIG);
 }
 
 static void pl08x_unmap_buffers(struct pl08x_txd *txd)
 {
 	struct device *dev = txd->tx.chan->device->dev;
+	struct pl08x_sg *dsg;
 
 	if (!(txd->tx.flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
 		if (txd->tx.flags & DMA_COMPL_SRC_UNMAP_SINGLE)
-			dma_unmap_single(dev, txd->src_addr, txd->len,
-				DMA_TO_DEVICE);
-		else
-			dma_unmap_page(dev, txd->src_addr, txd->len,
-				DMA_TO_DEVICE);
+			list_for_each_entry(dsg, &txd->dsg_list, node)
+				dma_unmap_single(dev, dsg->src_addr, dsg->len,
+						DMA_TO_DEVICE);
+		else {
+			list_for_each_entry(dsg, &txd->dsg_list, node)
+				dma_unmap_page(dev, dsg->src_addr, dsg->len,
+						DMA_TO_DEVICE);
+		}
 	}
 	if (!(txd->tx.flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
 		if (txd->tx.flags & DMA_COMPL_DEST_UNMAP_SINGLE)
-			dma_unmap_single(dev, txd->dst_addr, txd->len,
-				DMA_FROM_DEVICE);
+			list_for_each_entry(dsg, &txd->dsg_list, node)
+				dma_unmap_single(dev, dsg->dst_addr, dsg->len,
+						DMA_FROM_DEVICE);
 		else
-			dma_unmap_page(dev, txd->dst_addr, txd->len,
-				DMA_FROM_DEVICE);
+			list_for_each_entry(dsg, &txd->dsg_list, node)
+				dma_unmap_page(dev, dsg->dst_addr, dsg->len,
+						DMA_FROM_DEVICE);
 	}
 }
 
@@ -1589,8 +1575,8 @@ static void pl08x_tasklet(unsigned long data)
 		 */
 		list_for_each_entry(waiting, &pl08x->memcpy.channels,
 				    chan.device_node) {
-		  if (waiting->state == PL08X_CHAN_WAITING &&
-			    waiting->waiting != NULL) {
+			if (waiting->state == PL08X_CHAN_WAITING &&
+				waiting->waiting != NULL) {
 				int ret;
 
 				/* This should REALLY not fail now */
@@ -1630,38 +1616,40 @@ static void pl08x_tasklet(unsigned long data)
 static irqreturn_t pl08x_irq(int irq, void *dev)
 {
 	struct pl08x_driver_data *pl08x = dev;
-	u32 mask = 0;
-	u32 val;
-	int i;
-
-	val = readl(pl08x->base + PL080_ERR_STATUS);
-	if (val) {
-		/* An error interrupt (on one or more channels) */
-		dev_err(&pl08x->adev->dev,
-			"%s error interrupt, register value 0x%08x\n",
-				__func__, val);
-		/*
-		 * Simply clear ALL PL08X error interrupts,
-		 * regardless of channel and cause
-		 * FIXME: should be 0x00000003 on PL081 really.
-		 */
-		writel(0x000000FF, pl08x->base + PL080_ERR_CLEAR);
+	u32 mask = 0, err, tc, i;
+
+	/* check & clear - ERR & TC interrupts */
+	err = readl(pl08x->base + PL080_ERR_STATUS);
+	if (err) {
+		dev_err(&pl08x->adev->dev, "%s error interrupt, register value 0x%08x\n",
+			__func__, err);
+		writel(err, pl08x->base + PL080_ERR_CLEAR);
 	}
-	val = readl(pl08x->base + PL080_INT_STATUS);
+	tc = readl(pl08x->base + PL080_INT_STATUS);
+	if (tc)
+		writel(tc, pl08x->base + PL080_TC_CLEAR);
+
+	if (!err && !tc)
+		return IRQ_NONE;
+
 	for (i = 0; i < pl08x->vd->channels; i++) {
-		if ((1 << i) & val) {
+		if (((1 << i) & err) || ((1 << i) & tc)) {
 			/* Locate physical channel */
 			struct pl08x_phy_chan *phychan = &pl08x->phy_chans[i];
 			struct pl08x_dma_chan *plchan = phychan->serving;
 
+			if (!plchan) {
+				dev_err(&pl08x->adev->dev,
+					"%s Error TC interrupt on unused channel: 0x%08x\n",
+					__func__, i);
+				continue;
+			}
+
 			/* Schedule tasklet on this channel */
 			tasklet_schedule(&plchan->tasklet);
-
 			mask |= (1 << i);
 		}
 	}
-	/* Clear only the terminal interrupts on channels we processed */
-	writel(mask, pl08x->base + PL080_TC_CLEAR);
 
 	return mask ? IRQ_HANDLED : IRQ_NONE;
 }
@@ -1685,9 +1673,7 @@ static void pl08x_dma_slave_init(struct pl08x_dma_chan *chan)
  * Make a local wrapper to hold required data
  */
 static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x,
-					   struct dma_device *dmadev,
-					   unsigned int channels,
-					   bool slave)
+		struct dma_device *dmadev, unsigned int channels, bool slave)
 {
 	struct pl08x_dma_chan *chan;
 	int i;
@@ -1700,7 +1686,7 @@ static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x,
 	 * to cope with that situation.
 	 */
 	for (i = 0; i < channels; i++) {
-		chan = kzalloc(sizeof(struct pl08x_dma_chan), GFP_KERNEL);
+		chan = kzalloc(sizeof(*chan), GFP_KERNEL);
 		if (!chan) {
 			dev_err(&pl08x->adev->dev,
 				"%s no memory for channel\n", __func__);
@@ -1728,7 +1714,7 @@ static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x,
 			kfree(chan);
 			continue;
 		}
-		dev_info(&pl08x->adev->dev,
+		dev_dbg(&pl08x->adev->dev,
 			 "initialize virtual channel \"%s\"\n",
 			 chan->name);
 
@@ -1837,9 +1823,9 @@ static const struct file_operations pl08x_debugfs_operations = {
 static void init_pl08x_debugfs(struct pl08x_driver_data *pl08x)
 {
 	/* Expose a simple debugfs interface to view all clocks */
-	(void) debugfs_create_file(dev_name(&pl08x->adev->dev), S_IFREG | S_IRUGO,
-				   NULL, pl08x,
-				   &pl08x_debugfs_operations);
+	(void) debugfs_create_file(dev_name(&pl08x->adev->dev),
+			S_IFREG | S_IRUGO, NULL, pl08x,
+			&pl08x_debugfs_operations);
 }
 
 #else
@@ -1860,12 +1846,15 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 		return ret;
 
 	/* Create the driver state holder */
-	pl08x = kzalloc(sizeof(struct pl08x_driver_data), GFP_KERNEL);
+	pl08x = kzalloc(sizeof(*pl08x), GFP_KERNEL);
 	if (!pl08x) {
 		ret = -ENOMEM;
 		goto out_no_pl08x;
 	}
 
+	pm_runtime_set_active(&adev->dev);
+	pm_runtime_enable(&adev->dev);
+
 	/* Initialize memcpy engine */
 	dma_cap_set(DMA_MEMCPY, pl08x->memcpy.cap_mask);
 	pl08x->memcpy.dev = &adev->dev;
@@ -1939,7 +1928,7 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 	}
 
 	/* Initialize physical channels */
-	pl08x->phy_chans = kmalloc((vd->channels * sizeof(struct pl08x_phy_chan)),
+	pl08x->phy_chans = kmalloc((vd->channels * sizeof(*pl08x->phy_chans)),
 			GFP_KERNEL);
 	if (!pl08x->phy_chans) {
 		dev_err(&adev->dev, "%s failed to allocate "
@@ -1956,9 +1945,8 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 		spin_lock_init(&ch->lock);
 		ch->serving = NULL;
 		ch->signal = -1;
-		dev_info(&adev->dev,
-			 "physical channel %d is %s\n", i,
-			 pl08x_phy_channel_busy(ch) ? "BUSY" : "FREE");
+		dev_dbg(&adev->dev, "physical channel %d is %s\n",
+			i, pl08x_phy_channel_busy(ch) ? "BUSY" : "FREE");
 	}
 
 	/* Register as many memcpy channels as there are physical channels */
@@ -1974,8 +1962,7 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 
 	/* Register slave channels */
 	ret = pl08x_dma_init_virtual_channels(pl08x, &pl08x->slave,
-					      pl08x->pd->num_slave_channels,
-					      true);
+			pl08x->pd->num_slave_channels, true);
 	if (ret <= 0) {
 		dev_warn(&pl08x->adev->dev,
 			"%s failed to enumerate slave channels - %d\n",
@@ -2005,6 +1992,8 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
 	dev_info(&pl08x->adev->dev, "DMA: PL%03x rev%u at 0x%08llx irq %d\n",
 		 amba_part(adev), amba_rev(adev),
 		 (unsigned long long)adev->res.start, adev->irq[0]);
+
+	pm_runtime_put(&adev->dev);
 	return 0;
 
 out_no_slave_reg:
@@ -2023,6 +2012,9 @@ out_no_ioremap:
 	dma_pool_destroy(pl08x->pool);
 out_no_lli_pool:
 out_no_platdata:
+	pm_runtime_put(&adev->dev);
+	pm_runtime_disable(&adev->dev);
+
 	kfree(pl08x);
 out_no_pl08x:
 	amba_release_regions(adev);
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index 6a483eac7b3f..fcfa0a8b5c59 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -107,10 +107,11 @@ static struct at_desc *atc_desc_get(struct at_dma_chan *atchan)
 {
 	struct at_desc *desc, *_desc;
 	struct at_desc *ret = NULL;
+	unsigned long flags;
 	unsigned int i = 0;
 	LIST_HEAD(tmp_list);
 
-	spin_lock_bh(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 	list_for_each_entry_safe(desc, _desc, &atchan->free_list, desc_node) {
 		i++;
 		if (async_tx_test_ack(&desc->txd)) {
@@ -121,7 +122,7 @@ static struct at_desc *atc_desc_get(struct at_dma_chan *atchan)
 		dev_dbg(chan2dev(&atchan->chan_common),
 				"desc %p not ACKed\n", desc);
 	}
-	spin_unlock_bh(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 	dev_vdbg(chan2dev(&atchan->chan_common),
 		"scanned %u descriptors on freelist\n", i);
 
@@ -129,9 +130,9 @@ static struct at_desc *atc_desc_get(struct at_dma_chan *atchan)
 	if (!ret) {
 		ret = atc_alloc_descriptor(&atchan->chan_common, GFP_ATOMIC);
 		if (ret) {
-			spin_lock_bh(&atchan->lock);
+			spin_lock_irqsave(&atchan->lock, flags);
 			atchan->descs_allocated++;
-			spin_unlock_bh(&atchan->lock);
+			spin_unlock_irqrestore(&atchan->lock, flags);
 		} else {
 			dev_err(chan2dev(&atchan->chan_common),
 					"not enough descriptors available\n");
@@ -150,8 +151,9 @@ static void atc_desc_put(struct at_dma_chan *atchan, struct at_desc *desc)
 {
 	if (desc) {
 		struct at_desc *child;
+		unsigned long flags;
 
-		spin_lock_bh(&atchan->lock);
+		spin_lock_irqsave(&atchan->lock, flags);
 		list_for_each_entry(child, &desc->tx_list, desc_node)
 			dev_vdbg(chan2dev(&atchan->chan_common),
 					"moving child desc %p to freelist\n",
@@ -160,7 +162,7 @@ static void atc_desc_put(struct at_dma_chan *atchan, struct at_desc *desc)
 		dev_vdbg(chan2dev(&atchan->chan_common),
 			 "moving desc %p to freelist\n", desc);
 		list_add(&desc->desc_node, &atchan->free_list);
-		spin_unlock_bh(&atchan->lock);
+		spin_unlock_irqrestore(&atchan->lock, flags);
 	}
 }
 
@@ -299,7 +301,7 @@ atc_chain_complete(struct at_dma_chan *atchan, struct at_desc *desc)
 
 	/* for cyclic transfers,
 	 * no need to replay callback function while stopping */
-	if (!test_bit(ATC_IS_CYCLIC, &atchan->status)) {
+	if (!atc_chan_is_cyclic(atchan)) {
 		dma_async_tx_callback	callback = txd->callback;
 		void			*param = txd->callback_param;
 
@@ -471,16 +473,17 @@ static void atc_handle_cyclic(struct at_dma_chan *atchan)
 static void atc_tasklet(unsigned long data)
 {
 	struct at_dma_chan *atchan = (struct at_dma_chan *)data;
+	unsigned long flags;
 
-	spin_lock(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 	if (test_and_clear_bit(ATC_IS_ERROR, &atchan->status))
 		atc_handle_error(atchan);
-	else if (test_bit(ATC_IS_CYCLIC, &atchan->status))
+	else if (atc_chan_is_cyclic(atchan))
 		atc_handle_cyclic(atchan);
 	else
 		atc_advance_work(atchan);
 
-	spin_unlock(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 }
 
 static irqreturn_t at_dma_interrupt(int irq, void *dev_id)
@@ -539,8 +542,9 @@ static dma_cookie_t atc_tx_submit(struct dma_async_tx_descriptor *tx)
 	struct at_desc		*desc = txd_to_at_desc(tx);
 	struct at_dma_chan	*atchan = to_at_dma_chan(tx->chan);
 	dma_cookie_t		cookie;
+	unsigned long		flags;
 
-	spin_lock_bh(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 	cookie = atc_assign_cookie(atchan, desc);
 
 	if (list_empty(&atchan->active_list)) {
@@ -554,7 +558,7 @@ static dma_cookie_t atc_tx_submit(struct dma_async_tx_descriptor *tx)
 		list_add_tail(&desc->desc_node, &atchan->queue);
 	}
 
-	spin_unlock_bh(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 
 	return cookie;
 }
@@ -927,28 +931,29 @@ static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 	struct at_dma_chan	*atchan = to_at_dma_chan(chan);
 	struct at_dma		*atdma = to_at_dma(chan->device);
 	int			chan_id = atchan->chan_common.chan_id;
+	unsigned long		flags;
 
 	LIST_HEAD(list);
 
 	dev_vdbg(chan2dev(chan), "atc_control (%d)\n", cmd);
 
 	if (cmd == DMA_PAUSE) {
-		spin_lock_bh(&atchan->lock);
+		spin_lock_irqsave(&atchan->lock, flags);
 
 		dma_writel(atdma, CHER, AT_DMA_SUSP(chan_id));
 		set_bit(ATC_IS_PAUSED, &atchan->status);
 
-		spin_unlock_bh(&atchan->lock);
+		spin_unlock_irqrestore(&atchan->lock, flags);
 	} else if (cmd == DMA_RESUME) {
-		if (!test_bit(ATC_IS_PAUSED, &atchan->status))
+		if (!atc_chan_is_paused(atchan))
 			return 0;
 
-		spin_lock_bh(&atchan->lock);
+		spin_lock_irqsave(&atchan->lock, flags);
 
 		dma_writel(atdma, CHDR, AT_DMA_RES(chan_id));
 		clear_bit(ATC_IS_PAUSED, &atchan->status);
 
-		spin_unlock_bh(&atchan->lock);
+		spin_unlock_irqrestore(&atchan->lock, flags);
 	} else if (cmd == DMA_TERMINATE_ALL) {
 		struct at_desc	*desc, *_desc;
 		/*
@@ -957,7 +962,7 @@ static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 		 * channel. We still have to poll the channel enable bit due
 		 * to AHB/HSB limitations.
 		 */
-		spin_lock_bh(&atchan->lock);
+		spin_lock_irqsave(&atchan->lock, flags);
 
 		/* disabling channel: must also remove suspend state */
 		dma_writel(atdma, CHDR, AT_DMA_RES(chan_id) | atchan->mask);
@@ -978,7 +983,7 @@ static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 		/* if channel dedicated to cyclic operations, free it */
 		clear_bit(ATC_IS_CYCLIC, &atchan->status);
 
-		spin_unlock_bh(&atchan->lock);
+		spin_unlock_irqrestore(&atchan->lock, flags);
 	} else {
 		return -ENXIO;
 	}
@@ -1004,9 +1009,10 @@ atc_tx_status(struct dma_chan *chan,
 	struct at_dma_chan	*atchan = to_at_dma_chan(chan);
 	dma_cookie_t		last_used;
 	dma_cookie_t		last_complete;
+	unsigned long		flags;
 	enum dma_status		ret;
 
-	spin_lock_bh(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 
 	last_complete = atchan->completed_cookie;
 	last_used = chan->cookie;
@@ -1021,7 +1027,7 @@ atc_tx_status(struct dma_chan *chan,
 		ret = dma_async_is_complete(cookie, last_complete, last_used);
 	}
 
-	spin_unlock_bh(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 
 	if (ret != DMA_SUCCESS)
 		dma_set_tx_state(txstate, last_complete, last_used,
@@ -1029,7 +1035,7 @@ atc_tx_status(struct dma_chan *chan,
 	else
 		dma_set_tx_state(txstate, last_complete, last_used, 0);
 
-	if (test_bit(ATC_IS_PAUSED, &atchan->status))
+	if (atc_chan_is_paused(atchan))
 		ret = DMA_PAUSED;
 
 	dev_vdbg(chan2dev(chan), "tx_status %d: cookie = %d (d%d, u%d)\n",
@@ -1046,18 +1052,19 @@ atc_tx_status(struct dma_chan *chan,
 static void atc_issue_pending(struct dma_chan *chan)
 {
 	struct at_dma_chan	*atchan = to_at_dma_chan(chan);
+	unsigned long		flags;
 
 	dev_vdbg(chan2dev(chan), "issue_pending\n");
 
 	/* Not needed for cyclic transfers */
-	if (test_bit(ATC_IS_CYCLIC, &atchan->status))
+	if (atc_chan_is_cyclic(atchan))
 		return;
 
-	spin_lock_bh(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 	if (!atc_chan_is_enabled(atchan)) {
 		atc_advance_work(atchan);
 	}
-	spin_unlock_bh(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 }
 
 /**
@@ -1073,6 +1080,7 @@ static int atc_alloc_chan_resources(struct dma_chan *chan)
 	struct at_dma		*atdma = to_at_dma(chan->device);
 	struct at_desc		*desc;
 	struct at_dma_slave	*atslave;
+	unsigned long		flags;
 	int			i;
 	u32			cfg;
 	LIST_HEAD(tmp_list);
@@ -1116,11 +1124,11 @@ static int atc_alloc_chan_resources(struct dma_chan *chan)
 		list_add_tail(&desc->desc_node, &tmp_list);
 	}
 
-	spin_lock_bh(&atchan->lock);
+	spin_lock_irqsave(&atchan->lock, flags);
 	atchan->descs_allocated = i;
 	list_splice(&tmp_list, &atchan->free_list);
 	atchan->completed_cookie = chan->cookie = 1;
-	spin_unlock_bh(&atchan->lock);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 
 	/* channel parameters */
 	channel_writel(atchan, CFG, cfg);
@@ -1260,12 +1268,11 @@ static int __init at_dma_probe(struct platform_device *pdev)
 
 	/* initialize channels related values */
 	INIT_LIST_HEAD(&atdma->dma_common.channels);
-	for (i = 0; i < pdata->nr_channels; i++, atdma->dma_common.chancnt++) {
+	for (i = 0; i < pdata->nr_channels; i++) {
 		struct at_dma_chan	*atchan = &atdma->chan[i];
 
 		atchan->chan_common.device = &atdma->dma_common;
 		atchan->chan_common.cookie = atchan->completed_cookie = 1;
-		atchan->chan_common.chan_id = i;
 		list_add_tail(&atchan->chan_common.device_node,
 				&atdma->dma_common.channels);
 
@@ -1293,22 +1300,20 @@ static int __init at_dma_probe(struct platform_device *pdev)
 	if (dma_has_cap(DMA_MEMCPY, atdma->dma_common.cap_mask))
 		atdma->dma_common.device_prep_dma_memcpy = atc_prep_dma_memcpy;
 
-	if (dma_has_cap(DMA_SLAVE, atdma->dma_common.cap_mask))
+	if (dma_has_cap(DMA_SLAVE, atdma->dma_common.cap_mask)) {
 		atdma->dma_common.device_prep_slave_sg = atc_prep_slave_sg;
-
-	if (dma_has_cap(DMA_CYCLIC, atdma->dma_common.cap_mask))
+		/* controller can do slave DMA: can trigger cyclic transfers */
+		dma_cap_set(DMA_CYCLIC, atdma->dma_common.cap_mask);
 		atdma->dma_common.device_prep_dma_cyclic = atc_prep_dma_cyclic;
-
-	if (dma_has_cap(DMA_SLAVE, atdma->dma_common.cap_mask) ||
-	    dma_has_cap(DMA_CYCLIC, atdma->dma_common.cap_mask))
 		atdma->dma_common.device_control = atc_control;
+	}
 
 	dma_writel(atdma, EN, AT_DMA_ENABLE);
 
 	dev_info(&pdev->dev, "Atmel AHB DMA Controller ( %s%s), %d channels\n",
 	  dma_has_cap(DMA_MEMCPY, atdma->dma_common.cap_mask) ? "cpy " : "",
 	  dma_has_cap(DMA_SLAVE, atdma->dma_common.cap_mask)  ? "slave " : "",
-	  atdma->dma_common.chancnt);
+	  pdata->nr_channels);
 
 	dma_async_device_register(&atdma->dma_common);
 
@@ -1377,27 +1382,112 @@ static void at_dma_shutdown(struct platform_device *pdev)
 	clk_disable(atdma->clk);
 }
 
+static int at_dma_prepare(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct dma_chan *chan, *_chan;
+
+	list_for_each_entry_safe(chan, _chan, &atdma->dma_common.channels,
+			device_node) {
+		struct at_dma_chan *atchan = to_at_dma_chan(chan);
+		/* wait for transaction completion (except in cyclic case) */
+		if (atc_chan_is_enabled(atchan) && !atc_chan_is_cyclic(atchan))
+			return -EAGAIN;
+	}
+	return 0;
+}
+
+static void atc_suspend_cyclic(struct at_dma_chan *atchan)
+{
+	struct dma_chan	*chan = &atchan->chan_common;
+
+	/* Channel should be paused by user
+	 * do it anyway even if it is not done already */
+	if (!atc_chan_is_paused(atchan)) {
+		dev_warn(chan2dev(chan),
+		"cyclic channel not paused, should be done by channel user\n");
+		atc_control(chan, DMA_PAUSE, 0);
+	}
+
+	/* now preserve additional data for cyclic operations */
+	/* next descriptor address in the cyclic list */
+	atchan->save_dscr = channel_readl(atchan, DSCR);
+
+	vdbg_dump_regs(atchan);
+}
+
 static int at_dma_suspend_noirq(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct dma_chan *chan, *_chan;
 
-	at_dma_off(platform_get_drvdata(pdev));
+	/* preserve data */
+	list_for_each_entry_safe(chan, _chan, &atdma->dma_common.channels,
+			device_node) {
+		struct at_dma_chan *atchan = to_at_dma_chan(chan);
+
+		if (atc_chan_is_cyclic(atchan))
+			atc_suspend_cyclic(atchan);
+		atchan->save_cfg = channel_readl(atchan, CFG);
+	}
+	atdma->save_imr = dma_readl(atdma, EBCIMR);
+
+	/* disable DMA controller */
+	at_dma_off(atdma);
 	clk_disable(atdma->clk);
 	return 0;
 }
 
+static void atc_resume_cyclic(struct at_dma_chan *atchan)
+{
+	struct at_dma	*atdma = to_at_dma(atchan->chan_common.device);
+
+	/* restore channel status for cyclic descriptors list:
+	 * next descriptor in the cyclic list at the time of suspend */
+	channel_writel(atchan, SADDR, 0);
+	channel_writel(atchan, DADDR, 0);
+	channel_writel(atchan, CTRLA, 0);
+	channel_writel(atchan, CTRLB, 0);
+	channel_writel(atchan, DSCR, atchan->save_dscr);
+	dma_writel(atdma, CHER, atchan->mask);
+
+	/* channel pause status should be removed by channel user
+	 * We cannot take the initiative to do it here */
+
+	vdbg_dump_regs(atchan);
+}
+
 static int at_dma_resume_noirq(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct at_dma *atdma = platform_get_drvdata(pdev);
+	struct dma_chan *chan, *_chan;
 
+	/* bring back DMA controller */
 	clk_enable(atdma->clk);
 	dma_writel(atdma, EN, AT_DMA_ENABLE);
+
+	/* clear any pending interrupt */
+	while (dma_readl(atdma, EBCISR))
+		cpu_relax();
+
+	/* restore saved data */
+	dma_writel(atdma, EBCIER, atdma->save_imr);
+	list_for_each_entry_safe(chan, _chan, &atdma->dma_common.channels,
+			device_node) {
+		struct at_dma_chan *atchan = to_at_dma_chan(chan);
+
+		channel_writel(atchan, CFG, atchan->save_cfg);
+		if (atc_chan_is_cyclic(atchan))
+			atc_resume_cyclic(atchan);
+	}
 	return 0;
 }
 
 static const struct dev_pm_ops at_dma_dev_pm_ops = {
+	.prepare = at_dma_prepare,
 	.suspend_noirq = at_dma_suspend_noirq,
 	.resume_noirq = at_dma_resume_noirq,
 };
diff --git a/drivers/dma/at_hdmac_regs.h b/drivers/dma/at_hdmac_regs.h
index 087dbf1dd39c..aa4c9aebab7c 100644
--- a/drivers/dma/at_hdmac_regs.h
+++ b/drivers/dma/at_hdmac_regs.h
@@ -204,6 +204,9 @@ enum atc_status {
  * @status: transmit status information from irq/prep* functions
  *                to tasklet (use atomic operations)
  * @tasklet: bottom half to finish transaction work
+ * @save_cfg: configuration register that is saved on suspend/resume cycle
+ * @save_dscr: for cyclic operations, preserve next descriptor address in
+ *             the cyclic list on suspend/resume cycle
  * @lock: serializes enqueue/dequeue operations to descriptors lists
  * @completed_cookie: identifier for the most recently completed operation
  * @active_list: list of descriptors dmaengine is being running on
@@ -218,6 +221,8 @@ struct at_dma_chan {
 	u8			mask;
 	unsigned long		status;
 	struct tasklet_struct	tasklet;
+	u32			save_cfg;
+	u32			save_dscr;
 
 	spinlock_t		lock;
 
@@ -248,6 +253,7 @@ static inline struct at_dma_chan *to_at_dma_chan(struct dma_chan *dchan)
  * @chan_common: common dmaengine dma_device object members
  * @ch_regs: memory mapped register base
  * @clk: dma controller clock
+ * @save_imr: interrupt mask register that is saved on suspend/resume cycle
  * @all_chan_mask: all channels availlable in a mask
  * @dma_desc_pool: base of DMA descriptor region (DMA address)
  * @chan: channels table to store at_dma_chan structures
@@ -256,6 +262,7 @@ struct at_dma {
 	struct dma_device	dma_common;
 	void __iomem		*regs;
 	struct clk		*clk;
+	u32			save_imr;
 
 	u8			all_chan_mask;
 
@@ -355,6 +362,23 @@ static inline int atc_chan_is_enabled(struct at_dma_chan *atchan)
 	return !!(dma_readl(atdma, CHSR) & atchan->mask);
 }
 
+/**
+ * atc_chan_is_paused - test channel pause/resume status
+ * @atchan: channel we want to test status
+ */
+static inline int atc_chan_is_paused(struct at_dma_chan *atchan)
+{
+	return test_bit(ATC_IS_PAUSED, &atchan->status);
+}
+
+/**
+ * atc_chan_is_cyclic - test if given channel has cyclic property set
+ * @atchan: channel we want to test status
+ */
+static inline int atc_chan_is_cyclic(struct at_dma_chan *atchan)
+{
+	return test_bit(ATC_IS_CYCLIC, &atchan->status);
+}
 
 /**
  * set_desc_eol - set end-of-link to descriptor so it will end transfer
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index 765f5ff22304..eb1d8641cf5c 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -10,6 +10,7 @@
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/dmaengine.h>
+#include <linux/freezer.h>
 #include <linux/init.h>
 #include <linux/kthread.h>
 #include <linux/module.h>
@@ -251,6 +252,7 @@ static int dmatest_func(void *data)
 	int			i;
 
 	thread_name = current->comm;
+	set_freezable_with_signal();
 
 	ret = -ENOMEM;
 
@@ -305,7 +307,8 @@ static int dmatest_func(void *data)
 		dma_addr_t dma_srcs[src_cnt];
 		dma_addr_t dma_dsts[dst_cnt];
 		struct completion cmp;
-		unsigned long tmo = msecs_to_jiffies(timeout);
+		unsigned long start, tmo, end = 0 /* compiler... */;
+		bool reload = true;
 		u8 align = 0;
 
 		total_tests++;
@@ -404,7 +407,17 @@ static int dmatest_func(void *data)
 		}
 		dma_async_issue_pending(chan);
 
-		tmo = wait_for_completion_timeout(&cmp, tmo);
+		do {
+			start = jiffies;
+			if (reload)
+				end = start + msecs_to_jiffies(timeout);
+			else if (end <= start)
+				end = start + 1;
+			tmo = wait_for_completion_interruptible_timeout(&cmp,
+								end - start);
+			reload = try_to_freeze();
+		} while (tmo == -ERESTARTSYS);
+
 		status = dma_async_is_tx_complete(chan, cookie, NULL, NULL);
 
 		if (tmo == 0) {
@@ -477,6 +490,8 @@ err_srcs:
 	pr_notice("%s: terminating after %u tests, %u failures (status %d)\n",
 			thread_name, total_tests, failed_tests, ret);
 
+	/* terminate all transfers on specified channels */
+	chan->device->device_control(chan, DMA_TERMINATE_ALL, 0);
 	if (iterations > 0)
 		while (!kthread_should_stop()) {
 			DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait_dmatest_exit);
@@ -499,6 +514,10 @@ static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
 		list_del(&thread->node);
 		kfree(thread);
 	}
+
+	/* terminate all transfers on specified channels */
+	dtc->chan->device->device_control(dtc->chan, DMA_TERMINATE_ALL, 0);
+
 	kfree(dtc);
 }
 
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
index 4d180ca9a1d8..9bfd6d360718 100644
--- a/drivers/dma/dw_dmac.c
+++ b/drivers/dma/dw_dmac.c
@@ -1407,12 +1407,11 @@ static int __init dw_probe(struct platform_device *pdev)
 	dw->all_chan_mask = (1 << pdata->nr_channels) - 1;
 
 	INIT_LIST_HEAD(&dw->dma.channels);
-	for (i = 0; i < pdata->nr_channels; i++, dw->dma.chancnt++) {
+	for (i = 0; i < pdata->nr_channels; i++) {
 		struct dw_dma_chan	*dwc = &dw->chan[i];
 
 		dwc->chan.device = &dw->dma;
 		dwc->chan.cookie = dwc->completed = 1;
-		dwc->chan.chan_id = i;
 		if (pdata->chan_allocation_order == CHAN_ALLOCATION_ASCENDING)
 			list_add_tail(&dwc->chan.device_node,
 					&dw->dma.channels);
@@ -1468,7 +1467,7 @@ static int __init dw_probe(struct platform_device *pdev)
 	dma_writel(dw, CFG, DW_CFG_DMA_EN);
 
 	printk(KERN_INFO "%s: DesignWare DMA Controller, %d channels\n",
-			dev_name(&pdev->dev), dw->dma.chancnt);
+			dev_name(&pdev->dev), pdata->nr_channels);
 
 	dma_async_device_register(&dw->dma);
 
diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c
index 5d7a49bd7c26..b47e2b803faf 100644
--- a/drivers/dma/ep93xx_dma.c
+++ b/drivers/dma/ep93xx_dma.c
@@ -22,6 +22,7 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/dmaengine.h>
+#include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index d99f71c356b5..d746899f36e1 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -14,6 +14,7 @@
  * http://www.gnu.org/copyleft/gpl.html
  */
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/types.h>
 #include <linux/mm.h>
 #include <linux/interrupt.h>
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index 7bd7e98548cd..eab1fe71259e 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/types.h>
 #include <linux/mm.h>
 #include <linux/interrupt.h>
@@ -318,6 +319,7 @@ struct sdma_engine {
 	dma_addr_t			context_phys;
 	struct dma_device		dma_device;
 	struct clk			*clk;
+	struct mutex			channel_0_lock;
 	struct sdma_script_start_addrs	*script_addrs;
 };
 
@@ -415,11 +417,15 @@ static int sdma_load_script(struct sdma_engine *sdma, void *buf, int size,
 	dma_addr_t buf_phys;
 	int ret;
 
+	mutex_lock(&sdma->channel_0_lock);
+
 	buf_virt = dma_alloc_coherent(NULL,
 			size,
 			&buf_phys, GFP_KERNEL);
-	if (!buf_virt)
-		return -ENOMEM;
+	if (!buf_virt) {
+		ret = -ENOMEM;
+		goto err_out;
+	}
 
 	bd0->mode.command = C0_SETPM;
 	bd0->mode.status = BD_DONE | BD_INTR | BD_WRAP | BD_EXTD;
@@ -433,6 +439,9 @@ static int sdma_load_script(struct sdma_engine *sdma, void *buf, int size,
 
 	dma_free_coherent(NULL, size, buf_virt, buf_phys);
 
+err_out:
+	mutex_unlock(&sdma->channel_0_lock);
+
 	return ret;
 }
 
@@ -656,6 +665,8 @@ static int sdma_load_context(struct sdma_channel *sdmac)
 	dev_dbg(sdma->dev, "event_mask0 = 0x%08x\n", sdmac->event_mask0);
 	dev_dbg(sdma->dev, "event_mask1 = 0x%08x\n", sdmac->event_mask1);
 
+	mutex_lock(&sdma->channel_0_lock);
+
 	memset(context, 0, sizeof(*context));
 	context->channel_state.pc = load_address;
 
@@ -676,6 +687,8 @@ static int sdma_load_context(struct sdma_channel *sdmac)
 
 	ret = sdma_run_channel(&sdma->channel[0]);
 
+	mutex_unlock(&sdma->channel_0_lock);
+
 	return ret;
 }
 
@@ -1131,18 +1144,17 @@ static void sdma_add_scripts(struct sdma_engine *sdma,
 			saddr_arr[i] = addr_arr[i];
 }
 
-static int __init sdma_get_firmware(struct sdma_engine *sdma,
-		const char *fw_name)
+static void sdma_load_firmware(const struct firmware *fw, void *context)
 {
-	const struct firmware *fw;
+	struct sdma_engine *sdma = context;
 	const struct sdma_firmware_header *header;
-	int ret;
 	const struct sdma_script_start_addrs *addr;
 	unsigned short *ram_code;
 
-	ret = request_firmware(&fw, fw_name, sdma->dev);
-	if (ret)
-		return ret;
+	if (!fw) {
+		dev_err(sdma->dev, "firmware not found\n");
+		return;
+	}
 
 	if (fw->size < sizeof(*header))
 		goto err_firmware;
@@ -1172,6 +1184,16 @@ static int __init sdma_get_firmware(struct sdma_engine *sdma,
 
 err_firmware:
 	release_firmware(fw);
+}
+
+static int __init sdma_get_firmware(struct sdma_engine *sdma,
+		const char *fw_name)
+{
+	int ret;
+
+	ret = request_firmware_nowait(THIS_MODULE,
+			FW_ACTION_HOTPLUG, fw_name, sdma->dev,
+			GFP_KERNEL, sdma, sdma_load_firmware);
 
 	return ret;
 }
@@ -1269,11 +1291,14 @@ static int __init sdma_probe(struct platform_device *pdev)
 	struct sdma_platform_data *pdata = pdev->dev.platform_data;
 	int i;
 	struct sdma_engine *sdma;
+	s32 *saddr_arr;
 
 	sdma = kzalloc(sizeof(*sdma), GFP_KERNEL);
 	if (!sdma)
 		return -ENOMEM;
 
+	mutex_init(&sdma->channel_0_lock);
+
 	sdma->dev = &pdev->dev;
 
 	iores = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -1310,6 +1335,11 @@ static int __init sdma_probe(struct platform_device *pdev)
 		goto err_alloc;
 	}
 
+	/* initially no scripts available */
+	saddr_arr = (s32 *)sdma->script_addrs;
+	for (i = 0; i < SDMA_SCRIPT_ADDRS_ARRAY_SIZE_V1; i++)
+		saddr_arr[i] = -EINVAL;
+
 	if (of_id)
 		pdev->id_entry = of_id->data;
 	sdma->devtype = pdev->id_entry->driver_data;
diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index 8a3fdd87db97..9e96c43a846a 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -115,16 +115,15 @@ DMAC1 interrupt Functions*/
 
 /**
  * dmac1_mask_periphral_intr -	mask the periphral interrupt
- * @midc: dma channel for which masking is required
+ * @mid: dma device for which masking is required
  *
  * Masks the DMA periphral interrupt
  * this is valid for DMAC1 family controllers only
  * This controller should have periphral mask registers already mapped
  */
-static void dmac1_mask_periphral_intr(struct intel_mid_dma_chan *midc)
+static void dmac1_mask_periphral_intr(struct middma_device *mid)
 {
 	u32 pimr;
-	struct middma_device *mid = to_middma_device(midc->chan.device);
 
 	if (mid->pimr_mask) {
 		pimr = readl(mid->mask_reg + LNW_PERIPHRAL_MASK);
@@ -184,7 +183,6 @@ static void enable_dma_interrupt(struct intel_mid_dma_chan *midc)
 static void disable_dma_interrupt(struct intel_mid_dma_chan *midc)
 {
 	/*Check LPE PISR, make sure fwd is disabled*/
-	dmac1_mask_periphral_intr(midc);
 	iowrite32(MASK_INTR_REG(midc->ch_id), midc->dma_base + MASK_BLOCK);
 	iowrite32(MASK_INTR_REG(midc->ch_id), midc->dma_base + MASK_TFR);
 	iowrite32(MASK_INTR_REG(midc->ch_id), midc->dma_base + MASK_ERR);
@@ -1114,7 +1112,6 @@ static int mid_setup_dma(struct pci_dev *pdev)
 
 		midch->chan.device = &dma->common;
 		midch->chan.cookie =  1;
-		midch->chan.chan_id = i;
 		midch->ch_id = dma->chan_base + i;
 		pr_debug("MDMA:Init CH %d, ID %d\n", i, midch->ch_id);
 
@@ -1150,7 +1147,6 @@ static int mid_setup_dma(struct pci_dev *pdev)
 	dma_cap_set(DMA_SLAVE, dma->common.cap_mask);
 	dma_cap_set(DMA_PRIVATE, dma->common.cap_mask);
 	dma->common.dev = &pdev->dev;
-	dma->common.chancnt = dma->max_chan;
 
 	dma->common.device_alloc_chan_resources =
 					intel_mid_dma_alloc_chan_resources;
@@ -1350,6 +1346,7 @@ int dma_suspend(struct pci_dev *pci, pm_message_t state)
 		if (device->ch[i].in_use)
 			return -EAGAIN;
 	}
+	dmac1_mask_periphral_intr(device);
 	device->state = SUSPENDED;
 	pci_save_state(pci);
 	pci_disable_device(pci);
diff --git a/drivers/dma/mpc512x_dma.c b/drivers/dma/mpc512x_dma.c
index b9bae94f2015..8ba4edc6185e 100644
--- a/drivers/dma/mpc512x_dma.c
+++ b/drivers/dma/mpc512x_dma.c
@@ -741,7 +741,6 @@ static int __devinit mpc_dma_probe(struct platform_device *op)
 		mchan = &mdma->channels[i];
 
 		mchan->chan.device = dma;
-		mchan->chan.chan_id = i;
 		mchan->chan.cookie = 1;
 		mchan->completed_cookie = mchan->chan.cookie;
 
diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
index be641cbd36fc..b4588bdd98bb 100644
--- a/drivers/dma/mxs-dma.c
+++ b/drivers/dma/mxs-dma.c
@@ -130,6 +130,23 @@ struct mxs_dma_engine {
 	struct mxs_dma_chan		mxs_chans[MXS_DMA_CHANNELS];
 };
 
+static inline void mxs_dma_clkgate(struct mxs_dma_chan *mxs_chan, int enable)
+{
+	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
+	int chan_id = mxs_chan->chan.chan_id;
+	int set_clr = enable ? MXS_CLR_ADDR : MXS_SET_ADDR;
+
+	/* enable apbh channel clock */
+	if (dma_is_apbh()) {
+		if (apbh_is_old())
+			writel(1 << (chan_id + BP_APBH_CTRL0_CLKGATE_CHANNEL),
+				mxs_dma->base + HW_APBHX_CTRL0 + set_clr);
+		else
+			writel(1 << chan_id,
+				mxs_dma->base + HW_APBHX_CTRL0 + set_clr);
+	}
+}
+
 static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
 {
 	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
@@ -148,38 +165,21 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
 	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
 	int chan_id = mxs_chan->chan.chan_id;
 
+	/* clkgate needs to be enabled before writing other registers */
+	mxs_dma_clkgate(mxs_chan, 1);
+
 	/* set cmd_addr up */
 	writel(mxs_chan->ccw_phys,
 		mxs_dma->base + HW_APBHX_CHn_NXTCMDAR(chan_id));
 
-	/* enable apbh channel clock */
-	if (dma_is_apbh()) {
-		if (apbh_is_old())
-			writel(1 << (chan_id + BP_APBH_CTRL0_CLKGATE_CHANNEL),
-				mxs_dma->base + HW_APBHX_CTRL0 + MXS_CLR_ADDR);
-		else
-			writel(1 << chan_id,
-				mxs_dma->base + HW_APBHX_CTRL0 + MXS_CLR_ADDR);
-	}
-
 	/* write 1 to SEMA to kick off the channel */
 	writel(1, mxs_dma->base + HW_APBHX_CHn_SEMA(chan_id));
 }
 
 static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
 {
-	struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
-	int chan_id = mxs_chan->chan.chan_id;
-
 	/* disable apbh channel clock */
-	if (dma_is_apbh()) {
-		if (apbh_is_old())
-			writel(1 << (chan_id + BP_APBH_CTRL0_CLKGATE_CHANNEL),
-				mxs_dma->base + HW_APBHX_CTRL0 + MXS_SET_ADDR);
-		else
-			writel(1 << chan_id,
-				mxs_dma->base + HW_APBHX_CTRL0 + MXS_SET_ADDR);
-	}
+	mxs_dma_clkgate(mxs_chan, 0);
 
 	mxs_chan->status = DMA_SUCCESS;
 }
@@ -338,7 +338,10 @@ static int mxs_dma_alloc_chan_resources(struct dma_chan *chan)
 	if (ret)
 		goto err_clk;
 
+	/* clkgate needs to be enabled for reset to finish */
+	mxs_dma_clkgate(mxs_chan, 1);
 	mxs_dma_reset_chan(mxs_chan);
+	mxs_dma_clkgate(mxs_chan, 0);
 
 	dma_async_tx_descriptor_init(&mxs_chan->desc, chan);
 	mxs_chan->desc.tx_submit = mxs_dma_tx_submit;
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index 1ac8d4b580b7..a6d0e3dbed07 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -60,7 +60,7 @@
 #define DMA_DESC_FOLLOW_WITHOUT_IRQ	0x2
 #define DMA_DESC_FOLLOW_WITH_IRQ	0x3
 
-#define MAX_CHAN_NR			8
+#define MAX_CHAN_NR			12
 
 #define DMA_MASK_CTL0_MODE	0x33333333
 #define DMA_MASK_CTL2_MODE	0x00003333
@@ -872,8 +872,7 @@ static int __devinit pch_dma_probe(struct pci_dev *pdev,
 	int i;
 
 	nr_channels = id->driver_data;
-	pd = kzalloc(sizeof(struct pch_dma)+
-		sizeof(struct pch_dma_chan) * nr_channels, GFP_KERNEL);
+	pd = kzalloc(sizeof(*pd), GFP_KERNEL);
 	if (!pd)
 		return -ENOMEM;
 
@@ -926,7 +925,6 @@ static int __devinit pch_dma_probe(struct pci_dev *pdev,
 	}
 
 	pd->dma.dev = &pdev->dev;
-	pd->dma.chancnt = nr_channels;
 
 	INIT_LIST_HEAD(&pd->dma.channels);
 
@@ -935,7 +933,6 @@ static int __devinit pch_dma_probe(struct pci_dev *pdev,
 
 		pd_chan->chan.device = &pd->dma;
 		pd_chan->chan.cookie = 1;
-		pd_chan->chan.chan_id = i;
 
 		pd_chan->membase = &regs->desc[i];
 
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 00eee59e8b33..571041477ab2 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -17,6 +17,8 @@
 #include <linux/interrupt.h>
 #include <linux/amba/bus.h>
 #include <linux/amba/pl330.h>
+#include <linux/pm_runtime.h>
+#include <linux/scatterlist.h>
 
 #define NR_DEFAULT_DESC	16
 
@@ -68,6 +70,14 @@ struct dma_pl330_chan {
 	 * NULL if the channel is available to be acquired.
 	 */
 	void *pl330_chid;
+
+	/* For D-to-M and M-to-D channels */
+	int burst_sz; /* the peripheral fifo width */
+	int burst_len; /* the number of burst */
+	dma_addr_t fifo_addr;
+
+	/* for cyclic capability */
+	bool cyclic;
 };
 
 struct dma_pl330_dmac {
@@ -83,6 +93,8 @@ struct dma_pl330_dmac {
 
 	/* Peripheral channels connected to this DMAC */
 	struct dma_pl330_chan *peripherals; /* keep at end */
+
+	struct clk *clk;
 };
 
 struct dma_pl330_desc {
@@ -152,6 +164,31 @@ static inline void free_desc_list(struct list_head *list)
 	spin_unlock_irqrestore(&pdmac->pool_lock, flags);
 }
 
+static inline void handle_cyclic_desc_list(struct list_head *list)
+{
+	struct dma_pl330_desc *desc;
+	struct dma_pl330_chan *pch;
+	unsigned long flags;
+
+	if (list_empty(list))
+		return;
+
+	list_for_each_entry(desc, list, node) {
+		dma_async_tx_callback callback;
+
+		/* Change status to reload it */
+		desc->status = PREP;
+		pch = desc->pchan;
+		callback = desc->txd.callback;
+		if (callback)
+			callback(desc->txd.callback_param);
+	}
+
+	spin_lock_irqsave(&pch->lock, flags);
+	list_splice_tail_init(list, &pch->work_list);
+	spin_unlock_irqrestore(&pch->lock, flags);
+}
+
 static inline void fill_queue(struct dma_pl330_chan *pch)
 {
 	struct dma_pl330_desc *desc;
@@ -205,7 +242,10 @@ static void pl330_tasklet(unsigned long data)
 
 	spin_unlock_irqrestore(&pch->lock, flags);
 
-	free_desc_list(&list);
+	if (pch->cyclic)
+		handle_cyclic_desc_list(&list);
+	else
+		free_desc_list(&list);
 }
 
 static void dma_pl330_rqcb(void *token, enum pl330_op_err err)
@@ -236,6 +276,7 @@ static int pl330_alloc_chan_resources(struct dma_chan *chan)
 	spin_lock_irqsave(&pch->lock, flags);
 
 	pch->completed = chan->cookie = 1;
+	pch->cyclic = false;
 
 	pch->pl330_chid = pl330_request_channel(&pdmac->pif);
 	if (!pch->pl330_chid) {
@@ -253,25 +294,52 @@ static int pl330_alloc_chan_resources(struct dma_chan *chan)
 static int pl330_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, unsigned long arg)
 {
 	struct dma_pl330_chan *pch = to_pchan(chan);
-	struct dma_pl330_desc *desc;
+	struct dma_pl330_desc *desc, *_dt;
 	unsigned long flags;
+	struct dma_pl330_dmac *pdmac = pch->dmac;
+	struct dma_slave_config *slave_config;
+	LIST_HEAD(list);
 
-	/* Only supports DMA_TERMINATE_ALL */
-	if (cmd != DMA_TERMINATE_ALL)
-		return -ENXIO;
-
-	spin_lock_irqsave(&pch->lock, flags);
-
-	/* FLUSH the PL330 Channel thread */
-	pl330_chan_ctrl(pch->pl330_chid, PL330_OP_FLUSH);
+	switch (cmd) {
+	case DMA_TERMINATE_ALL:
+		spin_lock_irqsave(&pch->lock, flags);
 
-	/* Mark all desc done */
-	list_for_each_entry(desc, &pch->work_list, node)
-		desc->status = DONE;
+		/* FLUSH the PL330 Channel thread */
+		pl330_chan_ctrl(pch->pl330_chid, PL330_OP_FLUSH);
 
-	spin_unlock_irqrestore(&pch->lock, flags);
+		/* Mark all desc done */
+		list_for_each_entry_safe(desc, _dt, &pch->work_list , node) {
+			desc->status = DONE;
+			pch->completed = desc->txd.cookie;
+			list_move_tail(&desc->node, &list);
+		}
 
-	pl330_tasklet((unsigned long) pch);
+		list_splice_tail_init(&list, &pdmac->desc_pool);
+		spin_unlock_irqrestore(&pch->lock, flags);
+		break;
+	case DMA_SLAVE_CONFIG:
+		slave_config = (struct dma_slave_config *)arg;
+
+		if (slave_config->direction == DMA_TO_DEVICE) {
+			if (slave_config->dst_addr)
+				pch->fifo_addr = slave_config->dst_addr;
+			if (slave_config->dst_addr_width)
+				pch->burst_sz = __ffs(slave_config->dst_addr_width);
+			if (slave_config->dst_maxburst)
+				pch->burst_len = slave_config->dst_maxburst;
+		} else if (slave_config->direction == DMA_FROM_DEVICE) {
+			if (slave_config->src_addr)
+				pch->fifo_addr = slave_config->src_addr;
+			if (slave_config->src_addr_width)
+				pch->burst_sz = __ffs(slave_config->src_addr_width);
+			if (slave_config->src_maxburst)
+				pch->burst_len = slave_config->src_maxburst;
+		}
+		break;
+	default:
+		dev_err(pch->dmac->pif.dev, "Not supported command.\n");
+		return -ENXIO;
+	}
 
 	return 0;
 }
@@ -288,6 +356,9 @@ static void pl330_free_chan_resources(struct dma_chan *chan)
 	pl330_release_channel(pch->pl330_chid);
 	pch->pl330_chid = NULL;
 
+	if (pch->cyclic)
+		list_splice_tail_init(&pch->work_list, &pch->dmac->desc_pool);
+
 	spin_unlock_irqrestore(&pch->lock, flags);
 }
 
@@ -453,7 +524,7 @@ static struct dma_pl330_desc *pl330_get_desc(struct dma_pl330_chan *pch)
 
 	if (peri) {
 		desc->req.rqtype = peri->rqtype;
-		desc->req.peri = peri->peri_id;
+		desc->req.peri = pch->chan.chan_id;
 	} else {
 		desc->req.rqtype = MEMTOMEM;
 		desc->req.peri = 0;
@@ -524,6 +595,51 @@ static inline int get_burst_len(struct dma_pl330_desc *desc, size_t len)
 	return burst_len;
 }
 
+static struct dma_async_tx_descriptor *pl330_prep_dma_cyclic(
+		struct dma_chan *chan, dma_addr_t dma_addr, size_t len,
+		size_t period_len, enum dma_data_direction direction)
+{
+	struct dma_pl330_desc *desc;
+	struct dma_pl330_chan *pch = to_pchan(chan);
+	dma_addr_t dst;
+	dma_addr_t src;
+
+	desc = pl330_get_desc(pch);
+	if (!desc) {
+		dev_err(pch->dmac->pif.dev, "%s:%d Unable to fetch desc\n",
+			__func__, __LINE__);
+		return NULL;
+	}
+
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		desc->rqcfg.src_inc = 1;
+		desc->rqcfg.dst_inc = 0;
+		src = dma_addr;
+		dst = pch->fifo_addr;
+		break;
+	case DMA_FROM_DEVICE:
+		desc->rqcfg.src_inc = 0;
+		desc->rqcfg.dst_inc = 1;
+		src = pch->fifo_addr;
+		dst = dma_addr;
+		break;
+	default:
+		dev_err(pch->dmac->pif.dev, "%s:%d Invalid dma direction\n",
+		__func__, __LINE__);
+		return NULL;
+	}
+
+	desc->rqcfg.brst_size = pch->burst_sz;
+	desc->rqcfg.brst_len = 1;
+
+	pch->cyclic = true;
+
+	fill_px(&desc->px, dst, src, period_len);
+
+	return &desc->txd;
+}
+
 static struct dma_async_tx_descriptor *
 pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst,
 		dma_addr_t src, size_t len, unsigned long flags)
@@ -579,7 +695,7 @@ pl330_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 	struct dma_pl330_peri *peri = chan->private;
 	struct scatterlist *sg;
 	unsigned long flags;
-	int i, burst_size;
+	int i;
 	dma_addr_t addr;
 
 	if (unlikely(!pch || !sgl || !sg_len || !peri))
@@ -595,8 +711,7 @@ pl330_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		return NULL;
 	}
 
-	addr = peri->fifo_addr;
-	burst_size = peri->burst_sz;
+	addr = pch->fifo_addr;
 
 	first = NULL;
 
@@ -644,7 +759,7 @@ pl330_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
 				sg_dma_address(sg), addr, sg_dma_len(sg));
 		}
 
-		desc->rqcfg.brst_size = burst_size;
+		desc->rqcfg.brst_size = pch->burst_sz;
 		desc->rqcfg.brst_len = 1;
 	}
 
@@ -696,6 +811,30 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 		goto probe_err1;
 	}
 
+	pdmac->clk = clk_get(&adev->dev, "dma");
+	if (IS_ERR(pdmac->clk)) {
+		dev_err(&adev->dev, "Cannot get operation clock.\n");
+		ret = -EINVAL;
+		goto probe_err1;
+	}
+
+	amba_set_drvdata(adev, pdmac);
+
+#ifdef CONFIG_PM_RUNTIME
+	/* to use the runtime PM helper functions */
+	pm_runtime_enable(&adev->dev);
+
+	/* enable the power domain */
+	if (pm_runtime_get_sync(&adev->dev)) {
+		dev_err(&adev->dev, "failed to get runtime pm\n");
+		ret = -ENODEV;
+		goto probe_err1;
+	}
+#else
+	/* enable dma clk */
+	clk_enable(pdmac->clk);
+#endif
+
 	irq = adev->irq[0];
 	ret = request_irq(irq, pl330_irq_handler, 0,
 			dev_name(&adev->dev), pi);
@@ -732,6 +871,7 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 			case MEMTODEV:
 			case DEVTOMEM:
 				dma_cap_set(DMA_SLAVE, pd->cap_mask);
+				dma_cap_set(DMA_CYCLIC, pd->cap_mask);
 				break;
 			default:
 				dev_err(&adev->dev, "DEVTODEV Not Supported\n");
@@ -747,11 +887,9 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 		spin_lock_init(&pch->lock);
 		pch->pl330_chid = NULL;
 		pch->chan.device = pd;
-		pch->chan.chan_id = i;
 		pch->dmac = pdmac;
 
 		/* Add the channel to the DMAC list */
-		pd->chancnt++;
 		list_add_tail(&pch->chan.device_node, &pd->channels);
 	}
 
@@ -760,6 +898,7 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 	pd->device_alloc_chan_resources = pl330_alloc_chan_resources;
 	pd->device_free_chan_resources = pl330_free_chan_resources;
 	pd->device_prep_dma_memcpy = pl330_prep_dma_memcpy;
+	pd->device_prep_dma_cyclic = pl330_prep_dma_cyclic;
 	pd->device_tx_status = pl330_tx_status;
 	pd->device_prep_slave_sg = pl330_prep_slave_sg;
 	pd->device_control = pl330_control;
@@ -771,8 +910,6 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 		goto probe_err4;
 	}
 
-	amba_set_drvdata(adev, pdmac);
-
 	dev_info(&adev->dev,
 		"Loaded driver for PL330 DMAC-%d\n", adev->periphid);
 	dev_info(&adev->dev,
@@ -833,6 +970,13 @@ static int __devexit pl330_remove(struct amba_device *adev)
 	res = &adev->res;
 	release_mem_region(res->start, resource_size(res));
 
+#ifdef CONFIG_PM_RUNTIME
+	pm_runtime_put(&adev->dev);
+	pm_runtime_disable(&adev->dev);
+#else
+	clk_disable(pdmac->clk);
+#endif
+
 	kfree(pdmac);
 
 	return 0;
@@ -846,10 +990,49 @@ static struct amba_id pl330_ids[] = {
 	{ 0, 0 },
 };
 
+#ifdef CONFIG_PM_RUNTIME
+static int pl330_runtime_suspend(struct device *dev)
+{
+	struct dma_pl330_dmac *pdmac = dev_get_drvdata(dev);
+
+	if (!pdmac) {
+		dev_err(dev, "failed to get dmac\n");
+		return -ENODEV;
+	}
+
+	clk_disable(pdmac->clk);
+
+	return 0;
+}
+
+static int pl330_runtime_resume(struct device *dev)
+{
+	struct dma_pl330_dmac *pdmac = dev_get_drvdata(dev);
+
+	if (!pdmac) {
+		dev_err(dev, "failed to get dmac\n");
+		return -ENODEV;
+	}
+
+	clk_enable(pdmac->clk);
+
+	return 0;
+}
+#else
+#define pl330_runtime_suspend	NULL
+#define pl330_runtime_resume	NULL
+#endif /* CONFIG_PM_RUNTIME */
+
+static const struct dev_pm_ops pl330_pm_ops = {
+	.runtime_suspend = pl330_runtime_suspend,
+	.runtime_resume = pl330_runtime_resume,
+};
+
 static struct amba_driver pl330_driver = {
 	.drv = {
 		.owner = THIS_MODULE,
 		.name = "dma-pl330",
+		.pm = &pl330_pm_ops,
 	},
 	.id_table = pl330_ids,
 	.probe = pl330_probe,
diff --git a/drivers/dma/shdma.c b/drivers/dma/shdma.c
index 7f49235d14b9..81809c2b46ab 100644
--- a/drivers/dma/shdma.c
+++ b/drivers/dma/shdma.c
@@ -259,14 +259,23 @@ static int dmae_set_dmars(struct sh_dmae_chan *sh_chan, u16 val)
 	return 0;
 }
 
+static void sh_chan_xfer_ld_queue(struct sh_dmae_chan *sh_chan);
+
 static dma_cookie_t sh_dmae_tx_submit(struct dma_async_tx_descriptor *tx)
 {
 	struct sh_desc *desc = tx_to_sh_desc(tx), *chunk, *last = desc, *c;
 	struct sh_dmae_chan *sh_chan = to_sh_chan(tx->chan);
+	struct sh_dmae_slave *param = tx->chan->private;
 	dma_async_tx_callback callback = tx->callback;
 	dma_cookie_t cookie;
+	bool power_up;
+
+	spin_lock_irq(&sh_chan->desc_lock);
 
-	spin_lock_bh(&sh_chan->desc_lock);
+	if (list_empty(&sh_chan->ld_queue))
+		power_up = true;
+	else
+		power_up = false;
 
 	cookie = sh_chan->common.cookie;
 	cookie++;
@@ -302,7 +311,38 @@ static dma_cookie_t sh_dmae_tx_submit(struct dma_async_tx_descriptor *tx)
 		tx->cookie, &last->async_tx, sh_chan->id,
 		desc->hw.sar, desc->hw.tcr, desc->hw.dar);
 
-	spin_unlock_bh(&sh_chan->desc_lock);
+	if (power_up) {
+		sh_chan->pm_state = DMAE_PM_BUSY;
+
+		pm_runtime_get(sh_chan->dev);
+
+		spin_unlock_irq(&sh_chan->desc_lock);
+
+		pm_runtime_barrier(sh_chan->dev);
+
+		spin_lock_irq(&sh_chan->desc_lock);
+
+		/* Have we been reset, while waiting? */
+		if (sh_chan->pm_state != DMAE_PM_ESTABLISHED) {
+			dev_dbg(sh_chan->dev, "Bring up channel %d\n",
+				sh_chan->id);
+			if (param) {
+				const struct sh_dmae_slave_config *cfg =
+					param->config;
+
+				dmae_set_dmars(sh_chan, cfg->mid_rid);
+				dmae_set_chcr(sh_chan, cfg->chcr);
+			} else {
+				dmae_init(sh_chan);
+			}
+
+			if (sh_chan->pm_state == DMAE_PM_PENDING)
+				sh_chan_xfer_ld_queue(sh_chan);
+			sh_chan->pm_state = DMAE_PM_ESTABLISHED;
+		}
+	}
+
+	spin_unlock_irq(&sh_chan->desc_lock);
 
 	return cookie;
 }
@@ -346,8 +386,6 @@ static int sh_dmae_alloc_chan_resources(struct dma_chan *chan)
 	struct sh_dmae_slave *param = chan->private;
 	int ret;
 
-	pm_runtime_get_sync(sh_chan->dev);
-
 	/*
 	 * This relies on the guarantee from dmaengine that alloc_chan_resources
 	 * never runs concurrently with itself or free_chan_resources.
@@ -367,31 +405,20 @@ static int sh_dmae_alloc_chan_resources(struct dma_chan *chan)
 		}
 
 		param->config = cfg;
-
-		dmae_set_dmars(sh_chan, cfg->mid_rid);
-		dmae_set_chcr(sh_chan, cfg->chcr);
-	} else {
-		dmae_init(sh_chan);
 	}
 
-	spin_lock_bh(&sh_chan->desc_lock);
 	while (sh_chan->descs_allocated < NR_DESCS_PER_CHANNEL) {
-		spin_unlock_bh(&sh_chan->desc_lock);
 		desc = kzalloc(sizeof(struct sh_desc), GFP_KERNEL);
-		if (!desc) {
-			spin_lock_bh(&sh_chan->desc_lock);
+		if (!desc)
 			break;
-		}
 		dma_async_tx_descriptor_init(&desc->async_tx,
 					&sh_chan->common);
 		desc->async_tx.tx_submit = sh_dmae_tx_submit;
 		desc->mark = DESC_IDLE;
 
-		spin_lock_bh(&sh_chan->desc_lock);
 		list_add(&desc->node, &sh_chan->ld_free);
 		sh_chan->descs_allocated++;
 	}
-	spin_unlock_bh(&sh_chan->desc_lock);
 
 	if (!sh_chan->descs_allocated) {
 		ret = -ENOMEM;
@@ -405,7 +432,7 @@ edescalloc:
 		clear_bit(param->slave_id, sh_dmae_slave_used);
 etestused:
 efindslave:
-	pm_runtime_put(sh_chan->dev);
+	chan->private = NULL;
 	return ret;
 }
 
@@ -417,7 +444,6 @@ static void sh_dmae_free_chan_resources(struct dma_chan *chan)
 	struct sh_dmae_chan *sh_chan = to_sh_chan(chan);
 	struct sh_desc *desc, *_desc;
 	LIST_HEAD(list);
-	int descs = sh_chan->descs_allocated;
 
 	/* Protect against ISR */
 	spin_lock_irq(&sh_chan->desc_lock);
@@ -437,15 +463,12 @@ static void sh_dmae_free_chan_resources(struct dma_chan *chan)
 		chan->private = NULL;
 	}
 
-	spin_lock_bh(&sh_chan->desc_lock);
+	spin_lock_irq(&sh_chan->desc_lock);
 
 	list_splice_init(&sh_chan->ld_free, &list);
 	sh_chan->descs_allocated = 0;
 
-	spin_unlock_bh(&sh_chan->desc_lock);
-
-	if (descs > 0)
-		pm_runtime_put(sh_chan->dev);
+	spin_unlock_irq(&sh_chan->desc_lock);
 
 	list_for_each_entry_safe(desc, _desc, &list, node)
 		kfree(desc);
@@ -534,6 +557,7 @@ static struct dma_async_tx_descriptor *sh_dmae_prep_sg(struct sh_dmae_chan *sh_c
 	struct sh_desc *first = NULL, *new = NULL /* compiler... */;
 	LIST_HEAD(tx_list);
 	int chunks = 0;
+	unsigned long irq_flags;
 	int i;
 
 	if (!sg_len)
@@ -544,7 +568,7 @@ static struct dma_async_tx_descriptor *sh_dmae_prep_sg(struct sh_dmae_chan *sh_c
 			(SH_DMA_TCR_MAX + 1);
 
 	/* Have to lock the whole loop to protect against concurrent release */
-	spin_lock_bh(&sh_chan->desc_lock);
+	spin_lock_irqsave(&sh_chan->desc_lock, irq_flags);
 
 	/*
 	 * Chaining:
@@ -590,7 +614,7 @@ static struct dma_async_tx_descriptor *sh_dmae_prep_sg(struct sh_dmae_chan *sh_c
 	/* Put them back on the free list, so, they don't get lost */
 	list_splice_tail(&tx_list, &sh_chan->ld_free);
 
-	spin_unlock_bh(&sh_chan->desc_lock);
+	spin_unlock_irqrestore(&sh_chan->desc_lock, irq_flags);
 
 	return &first->async_tx;
 
@@ -599,7 +623,7 @@ err_get_desc:
 		new->mark = DESC_IDLE;
 	list_splice(&tx_list, &sh_chan->ld_free);
 
-	spin_unlock_bh(&sh_chan->desc_lock);
+	spin_unlock_irqrestore(&sh_chan->desc_lock, irq_flags);
 
 	return NULL;
 }
@@ -661,6 +685,7 @@ static int sh_dmae_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 			   unsigned long arg)
 {
 	struct sh_dmae_chan *sh_chan = to_sh_chan(chan);
+	unsigned long flags;
 
 	/* Only supports DMA_TERMINATE_ALL */
 	if (cmd != DMA_TERMINATE_ALL)
@@ -669,7 +694,7 @@ static int sh_dmae_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 	if (!chan)
 		return -EINVAL;
 
-	spin_lock_bh(&sh_chan->desc_lock);
+	spin_lock_irqsave(&sh_chan->desc_lock, flags);
 	dmae_halt(sh_chan);
 
 	if (!list_empty(&sh_chan->ld_queue)) {
@@ -678,9 +703,8 @@ static int sh_dmae_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
 						  struct sh_desc, node);
 		desc->partial = (desc->hw.tcr - sh_dmae_readl(sh_chan, TCR)) <<
 			sh_chan->xmit_shift;
-
 	}
-	spin_unlock_bh(&sh_chan->desc_lock);
+	spin_unlock_irqrestore(&sh_chan->desc_lock, flags);
 
 	sh_dmae_chan_ld_cleanup(sh_chan, true);
 
@@ -695,8 +719,9 @@ static dma_async_tx_callback __ld_cleanup(struct sh_dmae_chan *sh_chan, bool all
 	dma_cookie_t cookie = 0;
 	dma_async_tx_callback callback = NULL;
 	void *param = NULL;
+	unsigned long flags;
 
-	spin_lock_bh(&sh_chan->desc_lock);
+	spin_lock_irqsave(&sh_chan->desc_lock, flags);
 	list_for_each_entry_safe(desc, _desc, &sh_chan->ld_queue, node) {
 		struct dma_async_tx_descriptor *tx = &desc->async_tx;
 
@@ -762,7 +787,13 @@ static dma_async_tx_callback __ld_cleanup(struct sh_dmae_chan *sh_chan, bool all
 		     async_tx_test_ack(&desc->async_tx)) || all) {
 			/* Remove from ld_queue list */
 			desc->mark = DESC_IDLE;
+
 			list_move(&desc->node, &sh_chan->ld_free);
+
+			if (list_empty(&sh_chan->ld_queue)) {
+				dev_dbg(sh_chan->dev, "Bring down channel %d\n", sh_chan->id);
+				pm_runtime_put(sh_chan->dev);
+			}
 		}
 	}
 
@@ -773,7 +804,7 @@ static dma_async_tx_callback __ld_cleanup(struct sh_dmae_chan *sh_chan, bool all
 		 */
 		sh_chan->completed_cookie = sh_chan->common.cookie;
 
-	spin_unlock_bh(&sh_chan->desc_lock);
+	spin_unlock_irqrestore(&sh_chan->desc_lock, flags);
 
 	if (callback)
 		callback(param);
@@ -792,14 +823,14 @@ static void sh_dmae_chan_ld_cleanup(struct sh_dmae_chan *sh_chan, bool all)
 		;
 }
 
+/* Called under spin_lock_irq(&sh_chan->desc_lock) */
 static void sh_chan_xfer_ld_queue(struct sh_dmae_chan *sh_chan)
 {
 	struct sh_desc *desc;
 
-	spin_lock_bh(&sh_chan->desc_lock);
 	/* DMA work check */
 	if (dmae_is_busy(sh_chan))
-		goto sh_chan_xfer_ld_queue_end;
+		return;
 
 	/* Find the first not transferred descriptor */
 	list_for_each_entry(desc, &sh_chan->ld_queue, node)
@@ -812,15 +843,18 @@ static void sh_chan_xfer_ld_queue(struct sh_dmae_chan *sh_chan)
 			dmae_start(sh_chan);
 			break;
 		}
-
-sh_chan_xfer_ld_queue_end:
-	spin_unlock_bh(&sh_chan->desc_lock);
 }
 
 static void sh_dmae_memcpy_issue_pending(struct dma_chan *chan)
 {
 	struct sh_dmae_chan *sh_chan = to_sh_chan(chan);
-	sh_chan_xfer_ld_queue(sh_chan);
+
+	spin_lock_irq(&sh_chan->desc_lock);
+	if (sh_chan->pm_state == DMAE_PM_ESTABLISHED)
+		sh_chan_xfer_ld_queue(sh_chan);
+	else
+		sh_chan->pm_state = DMAE_PM_PENDING;
+	spin_unlock_irq(&sh_chan->desc_lock);
 }
 
 static enum dma_status sh_dmae_tx_status(struct dma_chan *chan,
@@ -831,6 +865,7 @@ static enum dma_status sh_dmae_tx_status(struct dma_chan *chan,
 	dma_cookie_t last_used;
 	dma_cookie_t last_complete;
 	enum dma_status status;
+	unsigned long flags;
 
 	sh_dmae_chan_ld_cleanup(sh_chan, false);
 
@@ -841,7 +876,7 @@ static enum dma_status sh_dmae_tx_status(struct dma_chan *chan,
 	BUG_ON(last_complete < 0);
 	dma_set_tx_state(txstate, last_complete, last_used, 0);
 
-	spin_lock_bh(&sh_chan->desc_lock);
+	spin_lock_irqsave(&sh_chan->desc_lock, flags);
 
 	status = dma_async_is_complete(cookie, last_complete, last_used);
 
@@ -859,7 +894,7 @@ static enum dma_status sh_dmae_tx_status(struct dma_chan *chan,
 			}
 	}
 
-	spin_unlock_bh(&sh_chan->desc_lock);
+	spin_unlock_irqrestore(&sh_chan->desc_lock, flags);
 
 	return status;
 }
@@ -912,6 +947,12 @@ static bool sh_dmae_reset(struct sh_dmae_device *shdev)
 
 		list_splice_init(&sh_chan->ld_queue, &dl);
 
+		if (!list_empty(&dl)) {
+			dev_dbg(sh_chan->dev, "Bring down channel %d\n", sh_chan->id);
+			pm_runtime_put(sh_chan->dev);
+		}
+		sh_chan->pm_state = DMAE_PM_ESTABLISHED;
+
 		spin_unlock(&sh_chan->desc_lock);
 
 		/* Complete all  */
@@ -952,7 +993,7 @@ static void dmae_do_tasklet(unsigned long data)
 	u32 sar_buf = sh_dmae_readl(sh_chan, SAR);
 	u32 dar_buf = sh_dmae_readl(sh_chan, DAR);
 
-	spin_lock(&sh_chan->desc_lock);
+	spin_lock_irq(&sh_chan->desc_lock);
 	list_for_each_entry(desc, &sh_chan->ld_queue, node) {
 		if (desc->mark == DESC_SUBMITTED &&
 		    ((desc->direction == DMA_FROM_DEVICE &&
@@ -965,10 +1006,10 @@ static void dmae_do_tasklet(unsigned long data)
 			break;
 		}
 	}
-	spin_unlock(&sh_chan->desc_lock);
-
 	/* Next desc */
 	sh_chan_xfer_ld_queue(sh_chan);
+	spin_unlock_irq(&sh_chan->desc_lock);
+
 	sh_dmae_chan_ld_cleanup(sh_chan, false);
 }
 
@@ -1036,7 +1077,9 @@ static int __devinit sh_dmae_chan_probe(struct sh_dmae_device *shdev, int id,
 		return -ENOMEM;
 	}
 
-	/* copy struct dma_device */
+	new_sh_chan->pm_state = DMAE_PM_ESTABLISHED;
+
+	/* reference struct dma_device */
 	new_sh_chan->common.device = &shdev->common;
 
 	new_sh_chan->dev = shdev->common.dev;
diff --git a/drivers/dma/shdma.h b/drivers/dma/shdma.h
index dc56576f9fdb..2b55a276dc5b 100644
--- a/drivers/dma/shdma.h
+++ b/drivers/dma/shdma.h
@@ -23,6 +23,12 @@
 
 struct device;
 
+enum dmae_pm_state {
+	DMAE_PM_ESTABLISHED,
+	DMAE_PM_BUSY,
+	DMAE_PM_PENDING,
+};
+
 struct sh_dmae_chan {
 	dma_cookie_t completed_cookie;	/* The maximum cookie completed */
 	spinlock_t desc_lock;		/* Descriptor operation lock */
@@ -38,6 +44,7 @@ struct sh_dmae_chan {
 	u32 __iomem *base;
 	char dev_id[16];		/* unique name per DMAC of channel */
 	int pm_error;
+	enum dmae_pm_state pm_state;
 };
 
 struct sh_dmae_device {
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index f69f90a61873..a4a398f2ef61 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -753,7 +753,7 @@ static int __devinit td_probe(struct platform_device *pdev)
 
 	INIT_LIST_HEAD(&td->dma.channels);
 
-	for (i = 0; i < pdata->nr_channels; i++, td->dma.chancnt++) {
+	for (i = 0; i < pdata->nr_channels; i++) {
 		struct timb_dma_chan *td_chan = &td->channels[i];
 		struct timb_dma_platform_data_channel *pchan =
 			pdata->channels + i;
@@ -762,12 +762,11 @@ static int __devinit td_probe(struct platform_device *pdev)
 		if ((i % 2) == pchan->rx) {
 			dev_err(&pdev->dev, "Wrong channel configuration\n");
 			err = -EINVAL;
-			goto err_tasklet_kill;
+			goto err_free_irq;
 		}
 
 		td_chan->chan.device = &td->dma;
 		td_chan->chan.cookie = 1;
-		td_chan->chan.chan_id = i;
 		spin_lock_init(&td_chan->lock);
 		INIT_LIST_HEAD(&td_chan->active_list);
 		INIT_LIST_HEAD(&td_chan->queue);