Merge branch 'topic/sh' into for-linus
diff --git a/Documentation/devicetree/bindings/dma/mv-xor-v2.txt b/Documentation/devicetree/bindings/dma/mv-xor-v2.txt
new file mode 100644
index 0000000..217a90e
--- /dev/null
+++ b/Documentation/devicetree/bindings/dma/mv-xor-v2.txt
@@ -0,0 +1,24 @@
+* Marvell XOR v2 engines
+
+Required properties:
+- compatible: one of the following values:
+    "marvell,armada-7k-xor"
+    "marvell,xor-v2"
+- reg: Should contain registers location and length (two sets)
+    the first set is the DMA registers
+    the second set is the global registers
+- msi-parent: Phandle to the MSI-capable interrupt controller used for
+  interrupts.
+
+Optional properties:
+- clocks: Optional reference to the clock used by the XOR engine.
+
+Example:
+
+	xor0@400000 {
+		compatible = "marvell,xor-v2";
+		reg = <0x400000 0x1000>,
+		      <0x410000 0x1000>;
+		msi-parent = <&gic_v2m0>;
+		dma-coherent;
+	};
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 8c98779..f6c46d0 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -339,6 +339,20 @@
 	---help---
 	  Enable support for the Marvell XOR engine.
 
+config MV_XOR_V2
+	bool "Marvell XOR engine version 2 support "
+	depends on ARM64
+	select DMA_ENGINE
+	select DMA_ENGINE_RAID
+	select ASYNC_TX_ENABLE_CHANNEL_SWITCH
+	select GENERIC_MSI_IRQ_DOMAIN
+	---help---
+	  Enable support for the Marvell version 2 XOR engine.
+
+	  This engine provides acceleration for copy, XOR and RAID6
+	  operations, and is available on Marvell Armada 7K and 8K
+	  platforms.
+
 config MXS_DMA
 	bool "MXS DMA support"
 	depends on SOC_IMX23 || SOC_IMX28 || SOC_IMX6Q || SOC_IMX6UL
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index 614f28b..e4dc9ca 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -45,6 +45,7 @@
 obj-$(CONFIG_MOXART_DMA) += moxart-dma.o
 obj-$(CONFIG_MPC512X_DMA) += mpc512x_dma.o
 obj-$(CONFIG_MV_XOR) += mv_xor.o
+obj-$(CONFIG_MV_XOR_V2) += mv_xor_v2.o
 obj-$(CONFIG_MXS_DMA) += mxs-dma.o
 obj-$(CONFIG_MX3_IPU) += ipu/
 obj-$(CONFIG_NBPFAXI_DMA) += nbpfaxi.o
diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c
index 81db1c4..939a7c3 100644
--- a/drivers/dma/amba-pl08x.c
+++ b/drivers/dma/amba-pl08x.c
@@ -1443,8 +1443,6 @@
 	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);
@@ -1901,11 +1899,8 @@
 	 */
 	for (i = 0; i < channels; i++) {
 		chan = kzalloc(sizeof(*chan), GFP_KERNEL);
-		if (!chan) {
-			dev_err(&pl08x->adev->dev,
-				"%s no memory for channel\n", __func__);
+		if (!chan)
 			return -ENOMEM;
-		}
 
 		chan->host = pl08x;
 		chan->state = PL08X_CHAN_IDLE;
@@ -2360,9 +2355,6 @@
 	pl08x->phy_chans = kzalloc((vd->channels * sizeof(*pl08x->phy_chans)),
 			GFP_KERNEL);
 	if (!pl08x->phy_chans) {
-		dev_err(&adev->dev, "%s failed to allocate "
-			"physical channel holders\n",
-			__func__);
 		ret = -ENOMEM;
 		goto out_no_phychans;
 	}
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 8e304b1..0d5c7ce 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -456,7 +456,7 @@
 	return desc;
 }
 
-void at_xdmac_init_used_desc(struct at_xdmac_desc *desc)
+static void at_xdmac_init_used_desc(struct at_xdmac_desc *desc)
 {
 	memset(&desc->lld, 0, sizeof(desc->lld));
 	INIT_LIST_HEAD(&desc->descs_list);
@@ -1195,14 +1195,14 @@
 	desc->lld.mbr_cfg = chan_cc;
 
 	dev_dbg(chan2dev(chan),
-		"%s: lld: mbr_da=%pad, mbr_ds=%pad, mbr_ubc=0x%08x, mbr_cfg=0x%08x\n",
-		__func__, &desc->lld.mbr_da, &desc->lld.mbr_ds, desc->lld.mbr_ubc,
+		"%s: lld: mbr_da=%pad, mbr_ds=0x%08x, mbr_ubc=0x%08x, mbr_cfg=0x%08x\n",
+		__func__, &desc->lld.mbr_da, desc->lld.mbr_ds, desc->lld.mbr_ubc,
 		desc->lld.mbr_cfg);
 
 	return desc;
 }
 
-struct dma_async_tx_descriptor *
+static struct dma_async_tx_descriptor *
 at_xdmac_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value,
 			 size_t len, unsigned long flags)
 {
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 6149b27..e18dc59 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -393,11 +393,12 @@
 	unsigned int sg_len)
 {
 	struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
-	size_t max_len = bcm2835_dma_max_frame_length(c);
-	unsigned int i, len;
+	size_t len, max_len;
+	unsigned int i;
 	dma_addr_t addr;
 	struct scatterlist *sgent;
 
+	max_len = bcm2835_dma_max_frame_length(c);
 	for_each_sg(sgl, sgent, sg_len, i) {
 		for (addr = sg_dma_address(sgent), len = sg_dma_len(sgent);
 		     len > 0;
@@ -613,7 +614,7 @@
 	spin_unlock_irqrestore(&c->vc.lock, flags);
 }
 
-struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_memcpy(
+static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_memcpy(
 	struct dma_chan *chan, dma_addr_t dst, dma_addr_t src,
 	size_t len, unsigned long flags)
 {
diff --git a/drivers/dma/bestcomm/bestcomm.c b/drivers/dma/bestcomm/bestcomm.c
index 180fedb..7ce8437 100644
--- a/drivers/dma/bestcomm/bestcomm.c
+++ b/drivers/dma/bestcomm/bestcomm.c
@@ -397,8 +397,6 @@
 	/* Get a clean struct */
 	bcom_eng = kzalloc(sizeof(struct bcom_engine), GFP_KERNEL);
 	if (!bcom_eng) {
-		printk(KERN_ERR DRIVER_NAME ": "
-			"Can't allocate state structure\n");
 		rv = -ENOMEM;
 		goto error_sramclean;
 	}
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index c340ca9..c100616 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -2422,7 +2422,7 @@
 	enum dma_status ret;
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_COMPLETE)
+	if (ret == DMA_COMPLETE || !txstate)
 		return ret;
 
 	dma_set_residue(txstate, coh901318_get_bytes_left(chan));
diff --git a/drivers/dma/dma-axi-dmac.c b/drivers/dma/dma-axi-dmac.c
index c346809..7f0b9aa 100644
--- a/drivers/dma/dma-axi-dmac.c
+++ b/drivers/dma/dma-axi-dmac.c
@@ -270,6 +270,9 @@
 	unsigned int pending;
 
 	pending = axi_dmac_read(dmac, AXI_DMAC_REG_IRQ_PENDING);
+	if (!pending)
+		return IRQ_NONE;
+
 	axi_dmac_write(dmac, AXI_DMAC_REG_IRQ_PENDING, pending);
 
 	spin_lock(&dmac->chan.vchan.lock);
@@ -579,7 +582,9 @@
 		return -ENOMEM;
 
 	dmac->irq = platform_get_irq(pdev, 0);
-	if (dmac->irq <= 0)
+	if (dmac->irq < 0)
+		return dmac->irq;
+	if (dmac->irq == 0)
 		return -EINVAL;
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -683,6 +688,7 @@
 	{ .compatible = "adi,axi-dmac-1.00.a" },
 	{ },
 };
+MODULE_DEVICE_TABLE(of, axi_dmac_of_match_table);
 
 static struct platform_driver axi_dmac_driver = {
 	.driver = {
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 8181ed1..b95ef74 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -1069,10 +1069,8 @@
 
 	edesc = kzalloc(sizeof(*edesc) + sg_len * sizeof(edesc->pset[0]),
 			GFP_ATOMIC);
-	if (!edesc) {
-		dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
+	if (!edesc)
 		return NULL;
-	}
 
 	edesc->pset_nr = sg_len;
 	edesc->residue = 0;
@@ -1114,14 +1112,17 @@
 		edesc->absync = ret;
 		edesc->residue += sg_dma_len(sg);
 
-		/* If this is the last in a current SG set of transactions,
-		   enable interrupts so that next set is processed */
-		if (!((i+1) % MAX_NR_SG))
-			edesc->pset[i].param.opt |= TCINTEN;
-
-		/* If this is the last set, enable completion interrupt flag */
 		if (i == sg_len - 1)
+			/* Enable completion interrupt */
 			edesc->pset[i].param.opt |= TCINTEN;
+		else if (!((i+1) % MAX_NR_SG))
+			/*
+			 * Enable early completion interrupt for the
+			 * intermediateset. In this case the driver will be
+			 * notified when the paRAM set is submitted to TC. This
+			 * will allow more time to set up the next set of slots.
+			 */
+			edesc->pset[i].param.opt |= (TCINTEN | TCCMODE);
 	}
 	edesc->residue_stat = edesc->residue;
 
@@ -1173,10 +1174,8 @@
 
 	edesc = kzalloc(sizeof(*edesc) + nslots * sizeof(edesc->pset[0]),
 			GFP_ATOMIC);
-	if (!edesc) {
-		dev_dbg(dev, "Failed to allocate a descriptor\n");
+	if (!edesc)
 		return NULL;
-	}
 
 	edesc->pset_nr = nslots;
 	edesc->residue = edesc->residue_stat = len;
@@ -1298,10 +1297,8 @@
 
 	edesc = kzalloc(sizeof(*edesc) + nslots * sizeof(edesc->pset[0]),
 			GFP_ATOMIC);
-	if (!edesc) {
-		dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
+	if (!edesc)
 		return NULL;
-	}
 
 	edesc->cyclic = 1;
 	edesc->pset_nr = nslots;
@@ -2207,10 +2204,8 @@
 		return ret;
 
 	ecc = devm_kzalloc(dev, sizeof(*ecc), GFP_KERNEL);
-	if (!ecc) {
-		dev_err(dev, "Can't allocate controller\n");
+	if (!ecc)
 		return -ENOMEM;
-	}
 
 	ecc->dev = dev;
 	ecc->id = pdev->id;
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index be2e62b..cc06eea 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -852,6 +852,14 @@
 	return 0;
 }
 
+static void fsl_disable_clocks(struct fsl_edma_engine *fsl_edma)
+{
+	int i;
+
+	for (i = 0; i < DMAMUX_NR; i++)
+		clk_disable_unprepare(fsl_edma->muxclk[i]);
+}
+
 static int fsl_edma_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
@@ -897,6 +905,10 @@
 
 		ret = clk_prepare_enable(fsl_edma->muxclk[i]);
 		if (ret) {
+			/* disable only clks which were enabled on error */
+			for (; i >= 0; i--)
+				clk_disable_unprepare(fsl_edma->muxclk[i]);
+
 			dev_err(&pdev->dev, "DMAMUX clk block failed.\n");
 			return ret;
 		}
@@ -951,14 +963,18 @@
 
 	ret = dma_async_device_register(&fsl_edma->dma_dev);
 	if (ret) {
-		dev_err(&pdev->dev, "Can't register Freescale eDMA engine.\n");
+		dev_err(&pdev->dev,
+			"Can't register Freescale eDMA engine. (%d)\n", ret);
+		fsl_disable_clocks(fsl_edma);
 		return ret;
 	}
 
 	ret = of_dma_controller_register(np, fsl_edma_xlate, fsl_edma);
 	if (ret) {
-		dev_err(&pdev->dev, "Can't register Freescale eDMA of_dma.\n");
+		dev_err(&pdev->dev,
+			"Can't register Freescale eDMA of_dma. (%d)\n", ret);
 		dma_async_device_unregister(&fsl_edma->dma_dev);
+		fsl_disable_clocks(fsl_edma);
 		return ret;
 	}
 
@@ -972,13 +988,10 @@
 {
 	struct device_node *np = pdev->dev.of_node;
 	struct fsl_edma_engine *fsl_edma = platform_get_drvdata(pdev);
-	int i;
 
 	of_dma_controller_free(np);
 	dma_async_device_unregister(&fsl_edma->dma_dev);
-
-	for (i = 0; i < DMAMUX_NR; i++)
-		clk_disable_unprepare(fsl_edma->muxclk[i]);
+	fsl_disable_clocks(fsl_edma);
 
 	return 0;
 }
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index a8828ed..911b717 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -1234,7 +1234,6 @@
 	/* alloc channel */
 	chan = kzalloc(sizeof(*chan), GFP_KERNEL);
 	if (!chan) {
-		dev_err(fdev->dev, "no free memory for DMA channels!\n");
 		err = -ENOMEM;
 		goto out_return;
 	}
@@ -1340,7 +1339,6 @@
 
 	fdev = kzalloc(sizeof(*fdev), GFP_KERNEL);
 	if (!fdev) {
-		dev_err(&op->dev, "No enough memory for 'priv'\n");
 		err = -ENOMEM;
 		goto out_return;
 	}
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index 0f6fd42..ce865f6 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/init.h>
+#include <linux/iopoll.h>
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/bitops.h>
@@ -571,28 +572,20 @@
 static int sdma_run_channel0(struct sdma_engine *sdma)
 {
 	int ret;
-	unsigned long timeout = 500;
+	u32 reg;
 
 	sdma_enable_channel(sdma, 0);
 
-	while (!(ret = readl_relaxed(sdma->regs + SDMA_H_INTR) & 1)) {
-		if (timeout-- <= 0)
-			break;
-		udelay(1);
-	}
-
-	if (ret) {
-		/* Clear the interrupt status */
-		writel_relaxed(ret, sdma->regs + SDMA_H_INTR);
-	} else {
+	ret = readl_relaxed_poll_timeout_atomic(sdma->regs + SDMA_H_STATSTOP,
+						reg, !(reg & 1), 1, 500);
+	if (ret)
 		dev_err(sdma->dev, "Timeout waiting for CH0 ready\n");
-	}
 
 	/* Set bits of CONFIG register with dynamic context switching */
 	if (readl(sdma->regs + SDMA_H_CONFIG) == 0)
 		writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG);
 
-	return ret ? 0 : -ETIMEDOUT;
+	return ret;
 }
 
 static int sdma_load_script(struct sdma_engine *sdma, void *buf, int size,
@@ -727,9 +720,9 @@
 	unsigned long stat;
 
 	stat = readl_relaxed(sdma->regs + SDMA_H_INTR);
-	/* not interested in channel 0 interrupts */
-	stat &= ~1;
 	writel_relaxed(stat, sdma->regs + SDMA_H_INTR);
+	/* channel 0 is special and not handled here, see run_channel0() */
+	stat &= ~1;
 
 	while (stat) {
 		int channel = fls(stat) - 1;
diff --git a/drivers/dma/k3dma.c b/drivers/dma/k3dma.c
index 1ba2fd7..35961af 100644
--- a/drivers/dma/k3dma.c
+++ b/drivers/dma/k3dma.c
@@ -425,10 +425,9 @@
 
 	num = DIV_ROUND_UP(len, DMA_MAX_SIZE);
 	ds = kzalloc(sizeof(*ds) + num * sizeof(ds->desc_hw[0]), GFP_ATOMIC);
-	if (!ds) {
-		dev_dbg(chan->device->dev, "vchan %p: kzalloc fail\n", &c->vc);
+	if (!ds)
 		return NULL;
-	}
+
 	ds->desc_hw_lli = __virt_to_phys((unsigned long)&ds->desc_hw[0]);
 	ds->size = len;
 	ds->desc_num = num;
@@ -481,10 +480,9 @@
 	}
 
 	ds = kzalloc(sizeof(*ds) + num * sizeof(ds->desc_hw[0]), GFP_ATOMIC);
-	if (!ds) {
-		dev_dbg(chan->device->dev, "vchan %p: kzalloc fail\n", &c->vc);
+	if (!ds)
 		return NULL;
-	}
+
 	ds->desc_hw_lli = __virt_to_phys((unsigned long)&ds->desc_hw[0]);
 	ds->desc_num = num;
 	num = 0;
diff --git a/drivers/dma/mmp_tdma.c b/drivers/dma/mmp_tdma.c
index 3df0422..ba7f412 100644
--- a/drivers/dma/mmp_tdma.c
+++ b/drivers/dma/mmp_tdma.c
@@ -551,10 +551,9 @@
 
 	/* alloc channel */
 	tdmac = devm_kzalloc(tdev->dev, sizeof(*tdmac), GFP_KERNEL);
-	if (!tdmac) {
-		dev_err(tdev->dev, "no free memory for DMA channels!\n");
+	if (!tdmac)
 		return -ENOMEM;
-	}
+
 	if (irq)
 		tdmac->irq = irq;
 	tdmac->dev	   = tdev->dev;
diff --git a/drivers/dma/moxart-dma.c b/drivers/dma/moxart-dma.c
index 631c443..b3a1d9a 100644
--- a/drivers/dma/moxart-dma.c
+++ b/drivers/dma/moxart-dma.c
@@ -574,10 +574,8 @@
 	struct moxart_dmadev *mdc;
 
 	mdc = devm_kzalloc(dev, sizeof(*mdc), GFP_KERNEL);
-	if (!mdc) {
-		dev_err(dev, "can't allocate DMA container\n");
+	if (!mdc)
 		return -ENOMEM;
-	}
 
 	irq = irq_of_parse_and_map(node, 0);
 	if (irq == NO_IRQ) {
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 25d1dad..55815c1 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -1055,7 +1055,7 @@
 
 err_free_irq:
 	free_irq(mv_chan->irq, mv_chan);
- err_free_dma:
+err_free_dma:
 	dma_free_coherent(&pdev->dev, MV_XOR_POOL_SIZE,
 			  mv_chan->dma_desc_pool_virt, mv_chan->dma_desc_pool);
 	return ERR_PTR(ret);
diff --git a/drivers/dma/mv_xor_v2.c b/drivers/dma/mv_xor_v2.c
new file mode 100644
index 0000000..a28a01f
--- /dev/null
+++ b/drivers/dma/mv_xor_v2.c
@@ -0,0 +1,878 @@
+/*
+ * Copyright (C) 2015-2016 Marvell International Ltd.
+
+ * This program is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation, either version 2 of the
+ * License, or any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/msi.h>
+#include <linux/of.h>
+#include <linux/of_irq.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+
+#include "dmaengine.h"
+
+/* DMA Engine Registers */
+#define MV_XOR_V2_DMA_DESQ_BALR_OFF			0x000
+#define MV_XOR_V2_DMA_DESQ_BAHR_OFF			0x004
+#define MV_XOR_V2_DMA_DESQ_SIZE_OFF			0x008
+#define MV_XOR_V2_DMA_DESQ_DONE_OFF			0x00C
+#define   MV_XOR_V2_DMA_DESQ_DONE_PENDING_MASK		0x7FFF
+#define   MV_XOR_V2_DMA_DESQ_DONE_PENDING_SHIFT		0
+#define   MV_XOR_V2_DMA_DESQ_DONE_READ_PTR_MASK		0x1FFF
+#define   MV_XOR_V2_DMA_DESQ_DONE_READ_PTR_SHIFT	16
+#define MV_XOR_V2_DMA_DESQ_ARATTR_OFF			0x010
+#define   MV_XOR_V2_DMA_DESQ_ATTR_CACHE_MASK		0x3F3F
+#define   MV_XOR_V2_DMA_DESQ_ATTR_OUTER_SHAREABLE	0x202
+#define   MV_XOR_V2_DMA_DESQ_ATTR_CACHEABLE		0x3C3C
+#define MV_XOR_V2_DMA_IMSG_CDAT_OFF			0x014
+#define MV_XOR_V2_DMA_IMSG_THRD_OFF			0x018
+#define   MV_XOR_V2_DMA_IMSG_THRD_MASK			0x7FFF
+#define   MV_XOR_V2_DMA_IMSG_THRD_SHIFT			0x0
+#define MV_XOR_V2_DMA_DESQ_AWATTR_OFF			0x01C
+  /* Same flags as MV_XOR_V2_DMA_DESQ_ARATTR_OFF */
+#define MV_XOR_V2_DMA_DESQ_ALLOC_OFF			0x04C
+#define   MV_XOR_V2_DMA_DESQ_ALLOC_WRPTR_MASK		0xFFFF
+#define   MV_XOR_V2_DMA_DESQ_ALLOC_WRPTR_SHIFT		16
+#define MV_XOR_V2_DMA_IMSG_BALR_OFF			0x050
+#define MV_XOR_V2_DMA_IMSG_BAHR_OFF			0x054
+#define MV_XOR_V2_DMA_DESQ_CTRL_OFF			0x100
+#define	  MV_XOR_V2_DMA_DESQ_CTRL_32B			1
+#define   MV_XOR_V2_DMA_DESQ_CTRL_128B			7
+#define MV_XOR_V2_DMA_DESQ_STOP_OFF			0x800
+#define MV_XOR_V2_DMA_DESQ_DEALLOC_OFF			0x804
+#define MV_XOR_V2_DMA_DESQ_ADD_OFF			0x808
+
+/* XOR Global registers */
+#define MV_XOR_V2_GLOB_BW_CTRL				0x4
+#define   MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_RD_SHIFT	0
+#define   MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_RD_VAL	64
+#define   MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_WR_SHIFT	8
+#define   MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_WR_VAL	8
+#define   MV_XOR_V2_GLOB_BW_CTRL_RD_BURST_LEN_SHIFT	12
+#define   MV_XOR_V2_GLOB_BW_CTRL_RD_BURST_LEN_VAL	4
+#define   MV_XOR_V2_GLOB_BW_CTRL_WR_BURST_LEN_SHIFT	16
+#define	  MV_XOR_V2_GLOB_BW_CTRL_WR_BURST_LEN_VAL	4
+#define MV_XOR_V2_GLOB_PAUSE				0x014
+#define   MV_XOR_V2_GLOB_PAUSE_AXI_TIME_DIS_VAL		0x8
+#define MV_XOR_V2_GLOB_SYS_INT_CAUSE			0x200
+#define MV_XOR_V2_GLOB_SYS_INT_MASK			0x204
+#define MV_XOR_V2_GLOB_MEM_INT_CAUSE			0x220
+#define MV_XOR_V2_GLOB_MEM_INT_MASK			0x224
+
+#define MV_XOR_V2_MIN_DESC_SIZE				32
+#define MV_XOR_V2_EXT_DESC_SIZE				128
+
+#define MV_XOR_V2_DESC_RESERVED_SIZE			12
+#define MV_XOR_V2_DESC_BUFF_D_ADDR_SIZE			12
+
+#define MV_XOR_V2_CMD_LINE_NUM_MAX_D_BUF		8
+
+/*
+ * Descriptors queue size. With 32 bytes descriptors, up to 2^14
+ * descriptors are allowed, with 128 bytes descriptors, up to 2^12
+ * descriptors are allowed. This driver uses 128 bytes descriptors,
+ * but experimentation has shown that a set of 1024 descriptors is
+ * sufficient to reach a good level of performance.
+ */
+#define MV_XOR_V2_DESC_NUM				1024
+
+/**
+ * struct mv_xor_v2_descriptor - DMA HW descriptor
+ * @desc_id: used by S/W and is not affected by H/W.
+ * @flags: error and status flags
+ * @crc32_result: CRC32 calculation result
+ * @desc_ctrl: operation mode and control flags
+ * @buff_size: amount of bytes to be processed
+ * @fill_pattern_src_addr: Fill-Pattern or Source-Address and
+ * AW-Attributes
+ * @data_buff_addr: Source (and might be RAID6 destination)
+ * addresses of data buffers in RAID5 and RAID6
+ * @reserved: reserved
+ */
+struct mv_xor_v2_descriptor {
+	u16 desc_id;
+	u16 flags;
+	u32 crc32_result;
+	u32 desc_ctrl;
+
+	/* Definitions for desc_ctrl */
+#define DESC_NUM_ACTIVE_D_BUF_SHIFT	22
+#define DESC_OP_MODE_SHIFT		28
+#define DESC_OP_MODE_NOP		0	/* Idle operation */
+#define DESC_OP_MODE_MEMCPY		1	/* Pure-DMA operation */
+#define DESC_OP_MODE_MEMSET		2	/* Mem-Fill operation */
+#define DESC_OP_MODE_MEMINIT		3	/* Mem-Init operation */
+#define DESC_OP_MODE_MEM_COMPARE	4	/* Mem-Compare operation */
+#define DESC_OP_MODE_CRC32		5	/* CRC32 calculation */
+#define DESC_OP_MODE_XOR		6	/* RAID5 (XOR) operation */
+#define DESC_OP_MODE_RAID6		7	/* RAID6 P&Q-generation */
+#define DESC_OP_MODE_RAID6_REC		8	/* RAID6 Recovery */
+#define DESC_Q_BUFFER_ENABLE		BIT(16)
+#define DESC_P_BUFFER_ENABLE		BIT(17)
+#define DESC_IOD			BIT(27)
+
+	u32 buff_size;
+	u32 fill_pattern_src_addr[4];
+	u32 data_buff_addr[MV_XOR_V2_DESC_BUFF_D_ADDR_SIZE];
+	u32 reserved[MV_XOR_V2_DESC_RESERVED_SIZE];
+};
+
+/**
+ * struct mv_xor_v2_device - implements a xor device
+ * @lock: lock for the engine
+ * @dma_base: memory mapped DMA register base
+ * @glob_base: memory mapped global register base
+ * @irq_tasklet:
+ * @free_sw_desc: linked list of free SW descriptors
+ * @dmadev: dma device
+ * @dmachan: dma channel
+ * @hw_desq: HW descriptors queue
+ * @hw_desq_virt: virtual address of DESCQ
+ * @sw_desq: SW descriptors queue
+ * @desc_size: HW descriptor size
+ * @npendings: number of pending descriptors (for which tx_submit has
+ * been called, but not yet issue_pending)
+ */
+struct mv_xor_v2_device {
+	spinlock_t lock;
+	void __iomem *dma_base;
+	void __iomem *glob_base;
+	struct clk *clk;
+	struct tasklet_struct irq_tasklet;
+	struct list_head free_sw_desc;
+	struct dma_device dmadev;
+	struct dma_chan	dmachan;
+	dma_addr_t hw_desq;
+	struct mv_xor_v2_descriptor *hw_desq_virt;
+	struct mv_xor_v2_sw_desc *sw_desq;
+	int desc_size;
+	unsigned int npendings;
+};
+
+/**
+ * struct mv_xor_v2_sw_desc - implements a xor SW descriptor
+ * @idx: descriptor index
+ * @async_tx: support for the async_tx api
+ * @hw_desc: assosiated HW descriptor
+ * @free_list: node of the free SW descriprots list
+*/
+struct mv_xor_v2_sw_desc {
+	int idx;
+	struct dma_async_tx_descriptor async_tx;
+	struct mv_xor_v2_descriptor hw_desc;
+	struct list_head free_list;
+};
+
+/*
+ * Fill the data buffers to a HW descriptor
+ */
+static void mv_xor_v2_set_data_buffers(struct mv_xor_v2_device *xor_dev,
+					struct mv_xor_v2_descriptor *desc,
+					dma_addr_t src, int index)
+{
+	int arr_index = ((index >> 1) * 3);
+
+	/*
+	 * Fill the buffer's addresses to the descriptor.
+	 *
+	 * The format of the buffers address for 2 sequential buffers
+	 * X and X + 1:
+	 *
+	 *  First word:  Buffer-DX-Address-Low[31:0]
+	 *  Second word: Buffer-DX+1-Address-Low[31:0]
+	 *  Third word:  DX+1-Buffer-Address-High[47:32] [31:16]
+	 *		 DX-Buffer-Address-High[47:32] [15:0]
+	 */
+	if ((index & 0x1) == 0) {
+		desc->data_buff_addr[arr_index] = lower_32_bits(src);
+
+		desc->data_buff_addr[arr_index + 2] &= ~0xFFFF;
+		desc->data_buff_addr[arr_index + 2] |=
+			upper_32_bits(src) & 0xFFFF;
+	} else {
+		desc->data_buff_addr[arr_index + 1] =
+			lower_32_bits(src);
+
+		desc->data_buff_addr[arr_index + 2] &= ~0xFFFF0000;
+		desc->data_buff_addr[arr_index + 2] |=
+			(upper_32_bits(src) & 0xFFFF) << 16;
+	}
+}
+
+/*
+ * Return the next available index in the DESQ.
+ */
+static int mv_xor_v2_get_desq_write_ptr(struct mv_xor_v2_device *xor_dev)
+{
+	/* read the index for the next available descriptor in the DESQ */
+	u32 reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_ALLOC_OFF);
+
+	return ((reg >> MV_XOR_V2_DMA_DESQ_ALLOC_WRPTR_SHIFT)
+		& MV_XOR_V2_DMA_DESQ_ALLOC_WRPTR_MASK);
+}
+
+/*
+ * notify the engine of new descriptors, and update the available index.
+ */
+static void mv_xor_v2_add_desc_to_desq(struct mv_xor_v2_device *xor_dev,
+				       int num_of_desc)
+{
+	/* write the number of new descriptors in the DESQ. */
+	writel(num_of_desc, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_ADD_OFF);
+}
+
+/*
+ * free HW descriptors
+ */
+static void mv_xor_v2_free_desc_from_desq(struct mv_xor_v2_device *xor_dev,
+					  int num_of_desc)
+{
+	/* write the number of new descriptors in the DESQ. */
+	writel(num_of_desc, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_DEALLOC_OFF);
+}
+
+/*
+ * Set descriptor size
+ * Return the HW descriptor size in bytes
+ */
+static int mv_xor_v2_set_desc_size(struct mv_xor_v2_device *xor_dev)
+{
+	writel(MV_XOR_V2_DMA_DESQ_CTRL_128B,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_CTRL_OFF);
+
+	return MV_XOR_V2_EXT_DESC_SIZE;
+}
+
+/*
+ * Set the IMSG threshold
+ */
+static inline
+void mv_xor_v2_set_imsg_thrd(struct mv_xor_v2_device *xor_dev, int thrd_val)
+{
+	u32 reg;
+
+	reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_IMSG_THRD_OFF);
+
+	reg &= (~MV_XOR_V2_DMA_IMSG_THRD_MASK << MV_XOR_V2_DMA_IMSG_THRD_SHIFT);
+	reg |= (thrd_val << MV_XOR_V2_DMA_IMSG_THRD_SHIFT);
+
+	writel(reg, xor_dev->dma_base + MV_XOR_V2_DMA_IMSG_THRD_OFF);
+}
+
+static irqreturn_t mv_xor_v2_interrupt_handler(int irq, void *data)
+{
+	struct mv_xor_v2_device *xor_dev = data;
+	unsigned int ndescs;
+	u32 reg;
+
+	reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_DONE_OFF);
+
+	ndescs = ((reg >> MV_XOR_V2_DMA_DESQ_DONE_PENDING_SHIFT) &
+		  MV_XOR_V2_DMA_DESQ_DONE_PENDING_MASK);
+
+	/* No descriptors to process */
+	if (!ndescs)
+		return IRQ_NONE;
+
+	/*
+	 * Update IMSG threshold, to disable new IMSG interrupts until
+	 * end of the tasklet
+	 */
+	mv_xor_v2_set_imsg_thrd(xor_dev, MV_XOR_V2_DESC_NUM);
+
+	/* schedule a tasklet to handle descriptors callbacks */
+	tasklet_schedule(&xor_dev->irq_tasklet);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * submit a descriptor to the DMA engine
+ */
+static dma_cookie_t
+mv_xor_v2_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	int desq_ptr;
+	void *dest_hw_desc;
+	dma_cookie_t cookie;
+	struct mv_xor_v2_sw_desc *sw_desc =
+		container_of(tx, struct mv_xor_v2_sw_desc, async_tx);
+	struct mv_xor_v2_device *xor_dev =
+		container_of(tx->chan, struct mv_xor_v2_device, dmachan);
+
+	dev_dbg(xor_dev->dmadev.dev,
+		"%s sw_desc %p: async_tx %p\n",
+		__func__, sw_desc, &sw_desc->async_tx);
+
+	/* assign coookie */
+	spin_lock_bh(&xor_dev->lock);
+	cookie = dma_cookie_assign(tx);
+
+	/* get the next available slot in the DESQ */
+	desq_ptr = mv_xor_v2_get_desq_write_ptr(xor_dev);
+
+	/* copy the HW descriptor from the SW descriptor to the DESQ */
+	dest_hw_desc = xor_dev->hw_desq_virt + desq_ptr;
+
+	memcpy(dest_hw_desc, &sw_desc->hw_desc, xor_dev->desc_size);
+
+	xor_dev->npendings++;
+
+	spin_unlock_bh(&xor_dev->lock);
+
+	return cookie;
+}
+
+/*
+ * Prepare a SW descriptor
+ */
+static struct mv_xor_v2_sw_desc	*
+mv_xor_v2_prep_sw_desc(struct mv_xor_v2_device *xor_dev)
+{
+	struct mv_xor_v2_sw_desc *sw_desc;
+
+	/* Lock the channel */
+	spin_lock_bh(&xor_dev->lock);
+
+	if (list_empty(&xor_dev->free_sw_desc)) {
+		spin_unlock_bh(&xor_dev->lock);
+		/* schedule tasklet to free some descriptors */
+		tasklet_schedule(&xor_dev->irq_tasklet);
+		return NULL;
+	}
+
+	/* get a free SW descriptor from the SW DESQ */
+	sw_desc = list_first_entry(&xor_dev->free_sw_desc,
+				   struct mv_xor_v2_sw_desc, free_list);
+	list_del(&sw_desc->free_list);
+
+	/* Release the channel */
+	spin_unlock_bh(&xor_dev->lock);
+
+	/* set the async tx descriptor */
+	dma_async_tx_descriptor_init(&sw_desc->async_tx, &xor_dev->dmachan);
+	sw_desc->async_tx.tx_submit = mv_xor_v2_tx_submit;
+	async_tx_ack(&sw_desc->async_tx);
+
+	return sw_desc;
+}
+
+/*
+ * Prepare a HW descriptor for a memcpy operation
+ */
+static struct dma_async_tx_descriptor *
+mv_xor_v2_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest,
+			  dma_addr_t src, size_t len, unsigned long flags)
+{
+	struct mv_xor_v2_sw_desc *sw_desc;
+	struct mv_xor_v2_descriptor *hw_descriptor;
+	struct mv_xor_v2_device	*xor_dev;
+
+	xor_dev = container_of(chan, struct mv_xor_v2_device, dmachan);
+
+	dev_dbg(xor_dev->dmadev.dev,
+		"%s len: %zu src %pad dest %pad flags: %ld\n",
+		__func__, len, &src, &dest, flags);
+
+	sw_desc = mv_xor_v2_prep_sw_desc(xor_dev);
+
+	sw_desc->async_tx.flags = flags;
+
+	/* set the HW descriptor */
+	hw_descriptor = &sw_desc->hw_desc;
+
+	/* save the SW descriptor ID to restore when operation is done */
+	hw_descriptor->desc_id = sw_desc->idx;
+
+	/* Set the MEMCPY control word */
+	hw_descriptor->desc_ctrl =
+		DESC_OP_MODE_MEMCPY << DESC_OP_MODE_SHIFT;
+
+	if (flags & DMA_PREP_INTERRUPT)
+		hw_descriptor->desc_ctrl |= DESC_IOD;
+
+	/* Set source address */
+	hw_descriptor->fill_pattern_src_addr[0] = lower_32_bits(src);
+	hw_descriptor->fill_pattern_src_addr[1] =
+		upper_32_bits(src) & 0xFFFF;
+
+	/* Set Destination address */
+	hw_descriptor->fill_pattern_src_addr[2] = lower_32_bits(dest);
+	hw_descriptor->fill_pattern_src_addr[3] =
+		upper_32_bits(dest) & 0xFFFF;
+
+	/* Set buffers size */
+	hw_descriptor->buff_size = len;
+
+	/* return the async tx descriptor */
+	return &sw_desc->async_tx;
+}
+
+/*
+ * Prepare a HW descriptor for a XOR operation
+ */
+static struct dma_async_tx_descriptor *
+mv_xor_v2_prep_dma_xor(struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
+		       unsigned int src_cnt, size_t len, unsigned long flags)
+{
+	struct mv_xor_v2_sw_desc *sw_desc;
+	struct mv_xor_v2_descriptor *hw_descriptor;
+	struct mv_xor_v2_device	*xor_dev =
+		container_of(chan, struct mv_xor_v2_device, dmachan);
+	int i;
+
+	if (src_cnt > MV_XOR_V2_CMD_LINE_NUM_MAX_D_BUF || src_cnt < 1)
+		return NULL;
+
+	dev_dbg(xor_dev->dmadev.dev,
+		"%s src_cnt: %d len: %zu dest %pad flags: %ld\n",
+		__func__, src_cnt, len, &dest, flags);
+
+	sw_desc = mv_xor_v2_prep_sw_desc(xor_dev);
+
+	sw_desc->async_tx.flags = flags;
+
+	/* set the HW descriptor */
+	hw_descriptor = &sw_desc->hw_desc;
+
+	/* save the SW descriptor ID to restore when operation is done */
+	hw_descriptor->desc_id = sw_desc->idx;
+
+	/* Set the XOR control word */
+	hw_descriptor->desc_ctrl =
+		DESC_OP_MODE_XOR << DESC_OP_MODE_SHIFT;
+	hw_descriptor->desc_ctrl |= DESC_P_BUFFER_ENABLE;
+
+	if (flags & DMA_PREP_INTERRUPT)
+		hw_descriptor->desc_ctrl |= DESC_IOD;
+
+	/* Set the data buffers */
+	for (i = 0; i < src_cnt; i++)
+		mv_xor_v2_set_data_buffers(xor_dev, hw_descriptor, src[i], i);
+
+	hw_descriptor->desc_ctrl |=
+		src_cnt << DESC_NUM_ACTIVE_D_BUF_SHIFT;
+
+	/* Set Destination address */
+	hw_descriptor->fill_pattern_src_addr[2] = lower_32_bits(dest);
+	hw_descriptor->fill_pattern_src_addr[3] =
+		upper_32_bits(dest) & 0xFFFF;
+
+	/* Set buffers size */
+	hw_descriptor->buff_size = len;
+
+	/* return the async tx descriptor */
+	return &sw_desc->async_tx;
+}
+
+/*
+ * Prepare a HW descriptor for interrupt operation.
+ */
+static struct dma_async_tx_descriptor *
+mv_xor_v2_prep_dma_interrupt(struct dma_chan *chan, unsigned long flags)
+{
+	struct mv_xor_v2_sw_desc *sw_desc;
+	struct mv_xor_v2_descriptor *hw_descriptor;
+	struct mv_xor_v2_device	*xor_dev =
+		container_of(chan, struct mv_xor_v2_device, dmachan);
+
+	sw_desc = mv_xor_v2_prep_sw_desc(xor_dev);
+
+	/* set the HW descriptor */
+	hw_descriptor = &sw_desc->hw_desc;
+
+	/* save the SW descriptor ID to restore when operation is done */
+	hw_descriptor->desc_id = sw_desc->idx;
+
+	/* Set the INTERRUPT control word */
+	hw_descriptor->desc_ctrl =
+		DESC_OP_MODE_NOP << DESC_OP_MODE_SHIFT;
+	hw_descriptor->desc_ctrl |= DESC_IOD;
+
+	/* return the async tx descriptor */
+	return &sw_desc->async_tx;
+}
+
+/*
+ * push pending transactions to hardware
+ */
+static void mv_xor_v2_issue_pending(struct dma_chan *chan)
+{
+	struct mv_xor_v2_device *xor_dev =
+		container_of(chan, struct mv_xor_v2_device, dmachan);
+
+	spin_lock_bh(&xor_dev->lock);
+
+	/*
+	 * update the engine with the number of descriptors to
+	 * process
+	 */
+	mv_xor_v2_add_desc_to_desq(xor_dev, xor_dev->npendings);
+	xor_dev->npendings = 0;
+
+	/* Activate the channel */
+	writel(0, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_STOP_OFF);
+
+	spin_unlock_bh(&xor_dev->lock);
+}
+
+static inline
+int mv_xor_v2_get_pending_params(struct mv_xor_v2_device *xor_dev,
+				 int *pending_ptr)
+{
+	u32 reg;
+
+	reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_DONE_OFF);
+
+	/* get the next pending descriptor index */
+	*pending_ptr = ((reg >> MV_XOR_V2_DMA_DESQ_DONE_READ_PTR_SHIFT) &
+			MV_XOR_V2_DMA_DESQ_DONE_READ_PTR_MASK);
+
+	/* get the number of descriptors pending handle */
+	return ((reg >> MV_XOR_V2_DMA_DESQ_DONE_PENDING_SHIFT) &
+		MV_XOR_V2_DMA_DESQ_DONE_PENDING_MASK);
+}
+
+/*
+ * handle the descriptors after HW process
+ */
+static void mv_xor_v2_tasklet(unsigned long data)
+{
+	struct mv_xor_v2_device *xor_dev = (struct mv_xor_v2_device *) data;
+	int pending_ptr, num_of_pending, i;
+	struct mv_xor_v2_descriptor *next_pending_hw_desc = NULL;
+	struct mv_xor_v2_sw_desc *next_pending_sw_desc = NULL;
+
+	dev_dbg(xor_dev->dmadev.dev, "%s %d\n", __func__, __LINE__);
+
+	/* get the pending descriptors parameters */
+	num_of_pending = mv_xor_v2_get_pending_params(xor_dev, &pending_ptr);
+
+	/* next HW descriptor */
+	next_pending_hw_desc = xor_dev->hw_desq_virt + pending_ptr;
+
+	/* loop over free descriptors */
+	for (i = 0; i < num_of_pending; i++) {
+
+		if (pending_ptr > MV_XOR_V2_DESC_NUM)
+			pending_ptr = 0;
+
+		if (next_pending_sw_desc != NULL)
+			next_pending_hw_desc++;
+
+		/* get the SW descriptor related to the HW descriptor */
+		next_pending_sw_desc =
+			&xor_dev->sw_desq[next_pending_hw_desc->desc_id];
+
+		/* call the callback */
+		if (next_pending_sw_desc->async_tx.cookie > 0) {
+			/*
+			 * update the channel's completed cookie - no
+			 * lock is required the IMSG threshold provide
+			 * the locking
+			 */
+			dma_cookie_complete(&next_pending_sw_desc->async_tx);
+
+			if (next_pending_sw_desc->async_tx.callback)
+				next_pending_sw_desc->async_tx.callback(
+				next_pending_sw_desc->async_tx.callback_param);
+
+			dma_descriptor_unmap(&next_pending_sw_desc->async_tx);
+		}
+
+		dma_run_dependencies(&next_pending_sw_desc->async_tx);
+
+		/* Lock the channel */
+		spin_lock_bh(&xor_dev->lock);
+
+		/* add the SW descriptor to the free descriptors list */
+		list_add(&next_pending_sw_desc->free_list,
+			 &xor_dev->free_sw_desc);
+
+		/* Release the channel */
+		spin_unlock_bh(&xor_dev->lock);
+
+		/* increment the next descriptor */
+		pending_ptr++;
+	}
+
+	if (num_of_pending != 0) {
+		/* free the descriptores */
+		mv_xor_v2_free_desc_from_desq(xor_dev, num_of_pending);
+	}
+
+	/* Update IMSG threshold, to enable new IMSG interrupts */
+	mv_xor_v2_set_imsg_thrd(xor_dev, 0);
+}
+
+/*
+ *	Set DMA Interrupt-message (IMSG) parameters
+ */
+static void mv_xor_v2_set_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
+{
+	struct mv_xor_v2_device *xor_dev = dev_get_drvdata(desc->dev);
+
+	writel(msg->address_lo,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_IMSG_BALR_OFF);
+	writel(msg->address_hi & 0xFFFF,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_IMSG_BAHR_OFF);
+	writel(msg->data,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_IMSG_CDAT_OFF);
+}
+
+static int mv_xor_v2_descq_init(struct mv_xor_v2_device *xor_dev)
+{
+	u32 reg;
+
+	/* write the DESQ size to the DMA engine */
+	writel(MV_XOR_V2_DESC_NUM,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_SIZE_OFF);
+
+	/* write the DESQ address to the DMA enngine*/
+	writel(xor_dev->hw_desq & 0xFFFFFFFF,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_BALR_OFF);
+	writel((xor_dev->hw_desq & 0xFFFF00000000) >> 32,
+	       xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_BAHR_OFF);
+
+	/* enable the DMA engine */
+	writel(0, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_STOP_OFF);
+
+	/*
+	 * This is a temporary solution, until we activate the
+	 * SMMU. Set the attributes for reading & writing data buffers
+	 * & descriptors to:
+	 *
+	 *  - OuterShareable - Snoops will be performed on CPU caches
+	 *  - Enable cacheable - Bufferable, Modifiable, Other Allocate
+	 *    and Allocate
+	 */
+	reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_ARATTR_OFF);
+	reg &= ~MV_XOR_V2_DMA_DESQ_ATTR_CACHE_MASK;
+	reg |= MV_XOR_V2_DMA_DESQ_ATTR_OUTER_SHAREABLE |
+		MV_XOR_V2_DMA_DESQ_ATTR_CACHEABLE;
+	writel(reg, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_ARATTR_OFF);
+
+	reg = readl(xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_AWATTR_OFF);
+	reg &= ~MV_XOR_V2_DMA_DESQ_ATTR_CACHE_MASK;
+	reg |= MV_XOR_V2_DMA_DESQ_ATTR_OUTER_SHAREABLE |
+		MV_XOR_V2_DMA_DESQ_ATTR_CACHEABLE;
+	writel(reg, xor_dev->dma_base + MV_XOR_V2_DMA_DESQ_AWATTR_OFF);
+
+	/* BW CTRL - set values to optimize the XOR performance:
+	 *
+	 *  - Set WrBurstLen & RdBurstLen - the unit will issue
+	 *    maximum of 256B write/read transactions.
+	 * -  Limit the number of outstanding write & read data
+	 *    (OBB/IBB) requests to the maximal value.
+	*/
+	reg = ((MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_RD_VAL <<
+		MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_RD_SHIFT) |
+	       (MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_WR_VAL  <<
+		MV_XOR_V2_GLOB_BW_CTRL_NUM_OSTD_WR_SHIFT) |
+	       (MV_XOR_V2_GLOB_BW_CTRL_RD_BURST_LEN_VAL <<
+		MV_XOR_V2_GLOB_BW_CTRL_RD_BURST_LEN_SHIFT) |
+	       (MV_XOR_V2_GLOB_BW_CTRL_WR_BURST_LEN_VAL <<
+		MV_XOR_V2_GLOB_BW_CTRL_WR_BURST_LEN_SHIFT));
+	writel(reg, xor_dev->glob_base + MV_XOR_V2_GLOB_BW_CTRL);
+
+	/* Disable the AXI timer feature */
+	reg = readl(xor_dev->glob_base + MV_XOR_V2_GLOB_PAUSE);
+	reg |= MV_XOR_V2_GLOB_PAUSE_AXI_TIME_DIS_VAL;
+	writel(reg, xor_dev->glob_base + MV_XOR_V2_GLOB_PAUSE);
+
+	return 0;
+}
+
+static int mv_xor_v2_probe(struct platform_device *pdev)
+{
+	struct mv_xor_v2_device *xor_dev;
+	struct resource *res;
+	int i, ret = 0;
+	struct dma_device *dma_dev;
+	struct mv_xor_v2_sw_desc *sw_desc;
+	struct msi_desc *msi_desc;
+
+	BUILD_BUG_ON(sizeof(struct mv_xor_v2_descriptor) !=
+		     MV_XOR_V2_EXT_DESC_SIZE);
+
+	xor_dev = devm_kzalloc(&pdev->dev, sizeof(*xor_dev), GFP_KERNEL);
+	if (!xor_dev)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	xor_dev->dma_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(xor_dev->dma_base))
+		return PTR_ERR(xor_dev->dma_base);
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	xor_dev->glob_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(xor_dev->glob_base))
+		return PTR_ERR(xor_dev->glob_base);
+
+	platform_set_drvdata(pdev, xor_dev);
+
+	xor_dev->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(xor_dev->clk) && PTR_ERR(xor_dev->clk) == -EPROBE_DEFER)
+		return -EPROBE_DEFER;
+	if (!IS_ERR(xor_dev->clk)) {
+		ret = clk_prepare_enable(xor_dev->clk);
+		if (ret)
+			return ret;
+	}
+
+	ret = platform_msi_domain_alloc_irqs(&pdev->dev, 1,
+					     mv_xor_v2_set_msi_msg);
+	if (ret)
+		goto disable_clk;
+
+	msi_desc = first_msi_entry(&pdev->dev);
+	if (!msi_desc)
+		goto free_msi_irqs;
+
+	ret = devm_request_irq(&pdev->dev, msi_desc->irq,
+			       mv_xor_v2_interrupt_handler, 0,
+			       dev_name(&pdev->dev), xor_dev);
+	if (ret)
+		goto free_msi_irqs;
+
+	tasklet_init(&xor_dev->irq_tasklet, mv_xor_v2_tasklet,
+		     (unsigned long) xor_dev);
+
+	xor_dev->desc_size = mv_xor_v2_set_desc_size(xor_dev);
+
+	dma_cookie_init(&xor_dev->dmachan);
+
+	/*
+	 * allocate coherent memory for hardware descriptors
+	 * note: writecombine gives slightly better performance, but
+	 * requires that we explicitly flush the writes
+	 */
+	xor_dev->hw_desq_virt =
+		dma_alloc_coherent(&pdev->dev,
+				   xor_dev->desc_size * MV_XOR_V2_DESC_NUM,
+				   &xor_dev->hw_desq, GFP_KERNEL);
+	if (!xor_dev->hw_desq_virt) {
+		ret = -ENOMEM;
+		goto free_msi_irqs;
+	}
+
+	/* alloc memory for the SW descriptors */
+	xor_dev->sw_desq = devm_kzalloc(&pdev->dev, sizeof(*sw_desc) *
+					MV_XOR_V2_DESC_NUM, GFP_KERNEL);
+	if (!xor_dev->sw_desq) {
+		ret = -ENOMEM;
+		goto free_hw_desq;
+	}
+
+	spin_lock_init(&xor_dev->lock);
+
+	/* init the free SW descriptors list */
+	INIT_LIST_HEAD(&xor_dev->free_sw_desc);
+
+	/* add all SW descriptors to the free list */
+	for (i = 0; i < MV_XOR_V2_DESC_NUM; i++) {
+		xor_dev->sw_desq[i].idx = i;
+		list_add(&xor_dev->sw_desq[i].free_list,
+			 &xor_dev->free_sw_desc);
+	}
+
+	dma_dev = &xor_dev->dmadev;
+
+	/* set DMA capabilities */
+	dma_cap_zero(dma_dev->cap_mask);
+	dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
+	dma_cap_set(DMA_XOR, dma_dev->cap_mask);
+	dma_cap_set(DMA_INTERRUPT, dma_dev->cap_mask);
+
+	/* init dma link list */
+	INIT_LIST_HEAD(&dma_dev->channels);
+
+	/* set base routines */
+	dma_dev->device_tx_status = dma_cookie_status;
+	dma_dev->device_issue_pending = mv_xor_v2_issue_pending;
+	dma_dev->dev = &pdev->dev;
+
+	dma_dev->device_prep_dma_memcpy = mv_xor_v2_prep_dma_memcpy;
+	dma_dev->device_prep_dma_interrupt = mv_xor_v2_prep_dma_interrupt;
+	dma_dev->max_xor = 8;
+	dma_dev->device_prep_dma_xor = mv_xor_v2_prep_dma_xor;
+
+	xor_dev->dmachan.device = dma_dev;
+
+	list_add_tail(&xor_dev->dmachan.device_node,
+		      &dma_dev->channels);
+
+	mv_xor_v2_descq_init(xor_dev);
+
+	ret = dma_async_device_register(dma_dev);
+	if (ret)
+		goto free_hw_desq;
+
+	dev_notice(&pdev->dev, "Marvell Version 2 XOR driver\n");
+
+	return 0;
+
+free_hw_desq:
+	dma_free_coherent(&pdev->dev,
+			  xor_dev->desc_size * MV_XOR_V2_DESC_NUM,
+			  xor_dev->hw_desq_virt, xor_dev->hw_desq);
+free_msi_irqs:
+	platform_msi_domain_free_irqs(&pdev->dev);
+disable_clk:
+	if (!IS_ERR(xor_dev->clk))
+		clk_disable_unprepare(xor_dev->clk);
+	return ret;
+}
+
+static int mv_xor_v2_remove(struct platform_device *pdev)
+{
+	struct mv_xor_v2_device *xor_dev = platform_get_drvdata(pdev);
+
+	dma_async_device_unregister(&xor_dev->dmadev);
+
+	dma_free_coherent(&pdev->dev,
+			  xor_dev->desc_size * MV_XOR_V2_DESC_NUM,
+			  xor_dev->hw_desq_virt, xor_dev->hw_desq);
+
+	platform_msi_domain_free_irqs(&pdev->dev);
+
+	clk_disable_unprepare(xor_dev->clk);
+
+	return 0;
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id mv_xor_v2_dt_ids[] = {
+	{ .compatible = "marvell,xor-v2", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, mv_xor_v2_dt_ids);
+#endif
+
+static struct platform_driver mv_xor_v2_driver = {
+	.probe		= mv_xor_v2_probe,
+	.remove		= mv_xor_v2_remove,
+	.driver		= {
+		.name	= "mv_xor_v2",
+		.of_match_table = of_match_ptr(mv_xor_v2_dt_ids),
+	},
+};
+
+module_platform_driver(mv_xor_v2_driver);
+
+MODULE_DESCRIPTION("DMA engine driver for Marvell's Version 2 of XOR engine");
+MODULE_LICENSE("GPL");
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index 2b5a198..9f0e98b 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1300,10 +1300,9 @@
 
 	nbpf = devm_kzalloc(dev, sizeof(*nbpf) + num_channels *
 			    sizeof(nbpf->chan[0]), GFP_KERNEL);
-	if (!nbpf) {
-		dev_err(dev, "Memory allocation failed\n");
+	if (!nbpf)
 		return -ENOMEM;
-	}
+
 	dma_dev = &nbpf->dma_dev;
 	dma_dev->dev = dev;
 
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 372b435..c8767d3 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2828,10 +2828,8 @@
 
 	/* Allocate a new DMAC and its Channels */
 	pl330 = devm_kzalloc(&adev->dev, sizeof(*pl330), GFP_KERNEL);
-	if (!pl330) {
-		dev_err(&adev->dev, "unable to allocate mem\n");
+	if (!pl330)
 		return -ENOMEM;
-	}
 
 	pd = &pl330->ddma;
 	pd->dev = &adev->dev;
@@ -2890,7 +2888,6 @@
 	pl330->peripherals = kzalloc(num_chan * sizeof(*pch), GFP_KERNEL);
 	if (!pl330->peripherals) {
 		ret = -ENOMEM;
-		dev_err(&adev->dev, "unable to allocate pl330->peripherals\n");
 		goto probe_err2;
 	}
 
diff --git a/drivers/dma/ppc4xx/adma.c b/drivers/dma/ppc4xx/adma.c
index 9217f89..da3688b 100644
--- a/drivers/dma/ppc4xx/adma.c
+++ b/drivers/dma/ppc4xx/adma.c
@@ -4084,7 +4084,6 @@
 	/* create a device */
 	adev = kzalloc(sizeof(*adev), GFP_KERNEL);
 	if (!adev) {
-		dev_err(&ofdev->dev, "failed to allocate device\n");
 		initcode = PPC_ADMA_INIT_ALLOC;
 		ret = -ENOMEM;
 		goto err_adev_alloc;
@@ -4145,7 +4144,6 @@
 	/* create a channel */
 	chan = kzalloc(sizeof(*chan), GFP_KERNEL);
 	if (!chan) {
-		dev_err(&ofdev->dev, "can't allocate channel structure\n");
 		initcode = PPC_ADMA_INIT_CHANNEL;
 		ret = -ENOMEM;
 		goto err_chan_alloc;
diff --git a/drivers/dma/qcom/bam_dma.c b/drivers/dma/qcom/bam_dma.c
index 969b481..03c4eb3 100644
--- a/drivers/dma/qcom/bam_dma.c
+++ b/drivers/dma/qcom/bam_dma.c
@@ -48,6 +48,7 @@
 #include <linux/of_dma.h>
 #include <linux/clk.h>
 #include <linux/dmaengine.h>
+#include <linux/pm_runtime.h>
 
 #include "../dmaengine.h"
 #include "../virt-dma.h"
@@ -58,6 +59,8 @@
 	__le16 flags;
 };
 
+#define BAM_DMA_AUTOSUSPEND_DELAY 100
+
 #define DESC_FLAG_INT BIT(15)
 #define DESC_FLAG_EOT BIT(14)
 #define DESC_FLAG_EOB BIT(13)
@@ -527,12 +530,17 @@
 	struct bam_device *bdev = bchan->bdev;
 	u32 val;
 	unsigned long flags;
+	int ret;
+
+	ret = pm_runtime_get_sync(bdev->dev);
+	if (ret < 0)
+		return;
 
 	vchan_free_chan_resources(to_virt_chan(chan));
 
 	if (bchan->curr_txd) {
 		dev_err(bchan->bdev->dev, "Cannot free busy channel\n");
-		return;
+		goto err;
 	}
 
 	spin_lock_irqsave(&bchan->vc.lock, flags);
@@ -550,6 +558,10 @@
 
 	/* disable irq */
 	writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_IRQ_EN));
+
+err:
+	pm_runtime_mark_last_busy(bdev->dev);
+	pm_runtime_put_autosuspend(bdev->dev);
 }
 
 /**
@@ -696,11 +708,18 @@
 	struct bam_chan *bchan = to_bam_chan(chan);
 	struct bam_device *bdev = bchan->bdev;
 	unsigned long flag;
+	int ret;
+
+	ret = pm_runtime_get_sync(bdev->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&bchan->vc.lock, flag);
 	writel_relaxed(1, bam_addr(bdev, bchan->id, BAM_P_HALT));
 	bchan->paused = 1;
 	spin_unlock_irqrestore(&bchan->vc.lock, flag);
+	pm_runtime_mark_last_busy(bdev->dev);
+	pm_runtime_put_autosuspend(bdev->dev);
 
 	return 0;
 }
@@ -715,11 +734,18 @@
 	struct bam_chan *bchan = to_bam_chan(chan);
 	struct bam_device *bdev = bchan->bdev;
 	unsigned long flag;
+	int ret;
+
+	ret = pm_runtime_get_sync(bdev->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&bchan->vc.lock, flag);
 	writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_HALT));
 	bchan->paused = 0;
 	spin_unlock_irqrestore(&bchan->vc.lock, flag);
+	pm_runtime_mark_last_busy(bdev->dev);
+	pm_runtime_put_autosuspend(bdev->dev);
 
 	return 0;
 }
@@ -795,6 +821,7 @@
 {
 	struct bam_device *bdev = data;
 	u32 clr_mask = 0, srcs = 0;
+	int ret;
 
 	srcs |= process_channel_irqs(bdev);
 
@@ -802,6 +829,10 @@
 	if (srcs & P_IRQ)
 		tasklet_schedule(&bdev->task);
 
+	ret = pm_runtime_get_sync(bdev->dev);
+	if (ret < 0)
+		return ret;
+
 	if (srcs & BAM_IRQ) {
 		clr_mask = readl_relaxed(bam_addr(bdev, 0, BAM_IRQ_STTS));
 
@@ -814,6 +845,9 @@
 		writel_relaxed(clr_mask, bam_addr(bdev, 0, BAM_IRQ_CLR));
 	}
 
+	pm_runtime_mark_last_busy(bdev->dev);
+	pm_runtime_put_autosuspend(bdev->dev);
+
 	return IRQ_HANDLED;
 }
 
@@ -893,6 +927,7 @@
 	struct bam_desc_hw *desc;
 	struct bam_desc_hw *fifo = PTR_ALIGN(bchan->fifo_virt,
 					sizeof(struct bam_desc_hw));
+	int ret;
 
 	lockdep_assert_held(&bchan->vc.lock);
 
@@ -904,6 +939,10 @@
 	async_desc = container_of(vd, struct bam_async_desc, vd);
 	bchan->curr_txd = async_desc;
 
+	ret = pm_runtime_get_sync(bdev->dev);
+	if (ret < 0)
+		return;
+
 	/* on first use, initialize the channel hardware */
 	if (!bchan->initialized)
 		bam_chan_init_hw(bchan, async_desc->dir);
@@ -946,6 +985,9 @@
 	wmb();
 	writel_relaxed(bchan->tail * sizeof(struct bam_desc_hw),
 			bam_addr(bdev, bchan->id, BAM_P_EVNT_REG));
+
+	pm_runtime_mark_last_busy(bdev->dev);
+	pm_runtime_put_autosuspend(bdev->dev);
 }
 
 /**
@@ -970,6 +1012,7 @@
 			bam_start_dma(bchan);
 		spin_unlock_irqrestore(&bchan->vc.lock, flags);
 	}
+
 }
 
 /**
@@ -1213,6 +1256,13 @@
 	if (ret)
 		goto err_unregister_dma;
 
+	pm_runtime_irq_safe(&pdev->dev);
+	pm_runtime_set_autosuspend_delay(&pdev->dev, BAM_DMA_AUTOSUSPEND_DELAY);
+	pm_runtime_use_autosuspend(&pdev->dev);
+	pm_runtime_mark_last_busy(&pdev->dev);
+	pm_runtime_set_active(&pdev->dev);
+	pm_runtime_enable(&pdev->dev);
+
 	return 0;
 
 err_unregister_dma:
@@ -1233,6 +1283,8 @@
 	struct bam_device *bdev = platform_get_drvdata(pdev);
 	u32 i;
 
+	pm_runtime_force_suspend(&pdev->dev);
+
 	of_dma_controller_free(pdev->dev.of_node);
 	dma_async_device_unregister(&bdev->common);
 
@@ -1260,11 +1312,66 @@
 	return 0;
 }
 
+static int __maybe_unused bam_dma_runtime_suspend(struct device *dev)
+{
+	struct bam_device *bdev = dev_get_drvdata(dev);
+
+	clk_disable(bdev->bamclk);
+
+	return 0;
+}
+
+static int __maybe_unused bam_dma_runtime_resume(struct device *dev)
+{
+	struct bam_device *bdev = dev_get_drvdata(dev);
+	int ret;
+
+	ret = clk_enable(bdev->bamclk);
+	if (ret < 0) {
+		dev_err(dev, "clk_enable failed: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int __maybe_unused bam_dma_suspend(struct device *dev)
+{
+	struct bam_device *bdev = dev_get_drvdata(dev);
+
+	pm_runtime_force_suspend(dev);
+
+	clk_unprepare(bdev->bamclk);
+
+	return 0;
+}
+
+static int __maybe_unused bam_dma_resume(struct device *dev)
+{
+	struct bam_device *bdev = dev_get_drvdata(dev);
+	int ret;
+
+	ret = clk_prepare(bdev->bamclk);
+	if (ret)
+		return ret;
+
+	pm_runtime_force_resume(dev);
+
+	return 0;
+}
+
+static const struct dev_pm_ops bam_dma_pm_ops = {
+	SET_LATE_SYSTEM_SLEEP_PM_OPS(bam_dma_suspend, bam_dma_resume)
+	SET_RUNTIME_PM_OPS(bam_dma_runtime_suspend, bam_dma_runtime_resume,
+				NULL)
+};
+
 static struct platform_driver bam_dma_driver = {
 	.probe = bam_dma_probe,
 	.remove = bam_dma_remove,
 	.driver = {
 		.name = "bam-dma-engine",
+		.pm = &bam_dma_pm_ops,
 		.of_match_table = bam_of_match,
 	},
 };
diff --git a/drivers/dma/qcom/hidma_mgmt.c b/drivers/dma/qcom/hidma_mgmt.c
index c0e3653..6416ded 100644
--- a/drivers/dma/qcom/hidma_mgmt.c
+++ b/drivers/dma/qcom/hidma_mgmt.c
@@ -371,8 +371,8 @@
 		pdevinfo.size_data = 0;
 		pdevinfo.dma_mask = DMA_BIT_MASK(64);
 		new_pdev = platform_device_register_full(&pdevinfo);
-		if (!new_pdev) {
-			ret = -ENODEV;
+		if (IS_ERR(new_pdev)) {
+			ret = PTR_ERR(new_pdev);
 			goto out;
 		}
 		of_dma_configure(&new_pdev->dev, child);
diff --git a/drivers/dma/s3c24xx-dma.c b/drivers/dma/s3c24xx-dma.c
index 17ccdfd..0d2d187 100644
--- a/drivers/dma/s3c24xx-dma.c
+++ b/drivers/dma/s3c24xx-dma.c
@@ -768,16 +768,12 @@
 
 	spin_lock_irqsave(&s3cchan->vc.lock, flags);
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret == DMA_COMPLETE) {
-		spin_unlock_irqrestore(&s3cchan->vc.lock, flags);
-		return ret;
-	}
 
 	/*
 	 * There's no point calculating the residue if there's
 	 * no txstate to store the value.
 	 */
-	if (!txstate) {
+	if (ret == DMA_COMPLETE || !txstate) {
 		spin_unlock_irqrestore(&s3cchan->vc.lock, flags);
 		return ret;
 	}
@@ -1105,11 +1101,8 @@
 	 */
 	for (i = 0; i < channels; i++) {
 		chan = devm_kzalloc(dmadev->dev, sizeof(*chan), GFP_KERNEL);
-		if (!chan) {
-			dev_err(dmadev->dev,
-				"%s no memory for channel\n", __func__);
+		if (!chan)
 			return -ENOMEM;
-		}
 
 		chan->id = i;
 		chan->host = s3cdma;
diff --git a/drivers/dma/sh/shdmac.c b/drivers/dma/sh/shdmac.c
index 80d8640..c94ffab 100644
--- a/drivers/dma/sh/shdmac.c
+++ b/drivers/dma/sh/shdmac.c
@@ -532,11 +532,8 @@
 
 	sh_chan = devm_kzalloc(sdev->dma_dev.dev, sizeof(struct sh_dmae_chan),
 			       GFP_KERNEL);
-	if (!sh_chan) {
-		dev_err(sdev->dma_dev.dev,
-			"No free memory for allocating dma channels!\n");
+	if (!sh_chan)
 		return -ENOMEM;
-	}
 
 	schan = &sh_chan->shdma_chan;
 	schan->max_xfer_len = SH_DMA_TCR_MAX + 1;
@@ -732,10 +729,8 @@
 
 	shdev = devm_kzalloc(&pdev->dev, sizeof(struct sh_dmae_device),
 			     GFP_KERNEL);
-	if (!shdev) {
-		dev_err(&pdev->dev, "Not enough memory\n");
+	if (!shdev)
 		return -ENOMEM;
-	}
 
 	dma_dev = &shdev->shdma_dev.dma_dev;
 
diff --git a/drivers/dma/sh/sudmac.c b/drivers/dma/sh/sudmac.c
index 6da2eaa..69b9564 100644
--- a/drivers/dma/sh/sudmac.c
+++ b/drivers/dma/sh/sudmac.c
@@ -245,11 +245,8 @@
 	int err;
 
 	sc = devm_kzalloc(&pdev->dev, sizeof(struct sudmac_chan), GFP_KERNEL);
-	if (!sc) {
-		dev_err(sdev->dma_dev.dev,
-			"No free memory for allocating dma channels!\n");
+	if (!sc)
 		return -ENOMEM;
-	}
 
 	schan = &sc->shdma_chan;
 	schan->max_xfer_len = 64 * 1024 * 1024 - 1;
@@ -349,10 +346,8 @@
 	err = -ENOMEM;
 	su_dev = devm_kzalloc(&pdev->dev, sizeof(struct sudmac_device),
 			      GFP_KERNEL);
-	if (!su_dev) {
-		dev_err(&pdev->dev, "Not enough memory\n");
+	if (!su_dev)
 		return err;
-	}
 
 	dma_dev = &su_dev->shdma_dev.dma_dev;
 
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index e48350e..8ea51c7 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -854,10 +854,9 @@
 	int ret, i;
 
 	sdma = devm_kzalloc(dev, sizeof(*sdma), GFP_KERNEL);
-	if (!sdma) {
-		dev_err(dev, "Memory exhausted!\n");
+	if (!sdma)
 		return -ENOMEM;
-	}
+
 	data = (struct sirfsoc_dmadata *)
 		(of_match_device(op->dev.driver->of_match_table,
 				 &op->dev)->data);
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index 6fb8307..8b18e44 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -2588,7 +2588,7 @@
 	}
 
 	ret = dma_cookie_status(chan, cookie, txstate);
-	if (ret != DMA_COMPLETE)
+	if (ret != DMA_COMPLETE && txstate)
 		dma_set_residue(txstate, stedma40_residue(chan));
 
 	if (d40_is_paused(d40c))
@@ -3237,10 +3237,8 @@
 		       (num_phy_chans + num_log_chans + num_memcpy_chans) *
 		       sizeof(struct d40_chan), GFP_KERNEL);
 
-	if (base == NULL) {
-		d40_err(&pdev->dev, "Out of memory\n");
+	if (base == NULL)
 		goto failure;
-	}
 
 	base->rev = rev;
 	base->clk = clk;
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 5065ca4..3835fcd 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -865,7 +865,7 @@
 	size_t bytes = 0;
 
 	ret = dma_cookie_status(chan, cookie, state);
-	if (ret == DMA_COMPLETE)
+	if (ret == DMA_COMPLETE || !state)
 		return ret;
 
 	spin_lock_irqsave(&vchan->vc.lock, flags);
diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index 01e316f..9d70b3a 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c
@@ -300,10 +300,8 @@
 
 	/* Allocate DMA desc */
 	dma_desc = kzalloc(sizeof(*dma_desc), GFP_NOWAIT);
-	if (!dma_desc) {
-		dev_err(tdc2dev(tdc), "dma_desc alloc failed\n");
+	if (!dma_desc)
 		return NULL;
-	}
 
 	dma_async_tx_descriptor_init(&dma_desc->txd, &tdc->dma_chan);
 	dma_desc->txd.tx_submit = tegra_dma_tx_submit;
@@ -340,8 +338,7 @@
 	spin_unlock_irqrestore(&tdc->lock, flags);
 
 	sg_req = kzalloc(sizeof(struct tegra_dma_sg_req), GFP_NOWAIT);
-	if (!sg_req)
-		dev_err(tdc2dev(tdc), "sg_req alloc failed\n");
+
 	return sg_req;
 }
 
@@ -822,13 +819,8 @@
 	/* Check on wait_ack desc status */
 	list_for_each_entry(dma_desc, &tdc->free_dma_desc, node) {
 		if (dma_desc->txd.cookie == cookie) {
-			residual =  dma_desc->bytes_requested -
-					(dma_desc->bytes_transferred %
-						dma_desc->bytes_requested);
-			dma_set_residue(txstate, residual);
 			ret = dma_desc->dma_status;
-			spin_unlock_irqrestore(&tdc->lock, flags);
-			return ret;
+			goto found;
 		}
 	}
 
@@ -836,17 +828,22 @@
 	list_for_each_entry(sg_req, &tdc->pending_sg_req, node) {
 		dma_desc = sg_req->dma_desc;
 		if (dma_desc->txd.cookie == cookie) {
-			residual =  dma_desc->bytes_requested -
-					(dma_desc->bytes_transferred %
-						dma_desc->bytes_requested);
-			dma_set_residue(txstate, residual);
 			ret = dma_desc->dma_status;
-			spin_unlock_irqrestore(&tdc->lock, flags);
-			return ret;
+			goto found;
 		}
 	}
 
-	dev_dbg(tdc2dev(tdc), "cookie %d does not found\n", cookie);
+	dev_dbg(tdc2dev(tdc), "cookie %d not found\n", cookie);
+	dma_desc = NULL;
+
+found:
+	if (dma_desc && txstate) {
+		residual = dma_desc->bytes_requested -
+			   (dma_desc->bytes_transferred %
+			    dma_desc->bytes_requested);
+		dma_set_residue(txstate, residual);
+	}
+
 	spin_unlock_irqrestore(&tdc->lock, flags);
 	return ret;
 }
@@ -1319,10 +1316,8 @@
 
 	tdma = devm_kzalloc(&pdev->dev, sizeof(*tdma) + cdata->nr_channels *
 			sizeof(struct tegra_dma_channel), GFP_KERNEL);
-	if (!tdma) {
-		dev_err(&pdev->dev, "Error: memory allocation failed\n");
+	if (!tdma)
 		return -ENOMEM;
-	}
 
 	tdma->dev = &pdev->dev;
 	tdma->chip_data = cdata;
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index 559cd40..e82745a 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -337,18 +337,14 @@
 	int err;
 
 	td_desc = kzalloc(sizeof(struct timb_dma_desc), GFP_KERNEL);
-	if (!td_desc) {
-		dev_err(chan2dev(chan), "Failed to alloc descriptor\n");
+	if (!td_desc)
 		goto out;
-	}
 
 	td_desc->desc_list_len = td_chan->desc_elems * TIMB_DMA_DESC_SIZE;
 
 	td_desc->desc_list = kzalloc(td_desc->desc_list_len, GFP_KERNEL);
-	if (!td_desc->desc_list) {
-		dev_err(chan2dev(chan), "Failed to alloc descriptor\n");
+	if (!td_desc->desc_list)
 		goto err;
-	}
 
 	dma_async_tx_descriptor_init(&td_desc->txd, chan);
 	td_desc->txd.tx_submit = td_tx_submit;