diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 4839bfa74a107a1ad4cbdfb09017d86da1fd1a41..19a99743cf524670d5906400d286cc877bcf91d6 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -271,7 +271,7 @@ struct pl330_config {
 #define DMAC_MODE_NS	(1 << 0)
 	unsigned int	mode;
 	unsigned int	data_bus_width:10; /* In number of bits */
-	unsigned int	data_buf_dep:10;
+	unsigned int	data_buf_dep:11;
 	unsigned int	num_chan:4;
 	unsigned int	num_peri:6;
 	u32		peri_ns;
@@ -2336,7 +2336,7 @@ static inline int get_burst_len(struct dma_pl330_desc *desc, size_t len)
 	int burst_len;
 
 	burst_len = pl330->pcfg.data_bus_width / 8;
-	burst_len *= pl330->pcfg.data_buf_dep;
+	burst_len *= pl330->pcfg.data_buf_dep / pl330->pcfg.num_chan;
 	burst_len >>= desc->rqcfg.brst_size;
 
 	/* src/dst_burst_len can't be more than 16 */
@@ -2459,16 +2459,25 @@ pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst,
 	/* Select max possible burst size */
 	burst = pl330->pcfg.data_bus_width / 8;
 
-	while (burst > 1) {
-		if (!(len % burst))
-			break;
+	/*
+	 * Make sure we use a burst size that aligns with all the memcpy
+	 * parameters because our DMA programming algorithm doesn't cope with
+	 * transfers which straddle an entry in the DMA device's MFIFO.
+	 */
+	while ((src | dst | len) & (burst - 1))
 		burst /= 2;
-	}
 
 	desc->rqcfg.brst_size = 0;
 	while (burst != (1 << desc->rqcfg.brst_size))
 		desc->rqcfg.brst_size++;
 
+	/*
+	 * If burst size is smaller than bus width then make sure we only
+	 * transfer one at a time to avoid a burst stradling an MFIFO entry.
+	 */
+	if (desc->rqcfg.brst_size * 8 < pl330->pcfg.data_bus_width)
+		desc->rqcfg.brst_len = 1;
+
 	desc->rqcfg.brst_len = get_burst_len(desc, len);
 
 	desc->txd.flags = flags;
@@ -2732,7 +2741,7 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
 
 
 	dev_info(&adev->dev,
-		"Loaded driver for PL330 DMAC-%d\n", adev->periphid);
+		"Loaded driver for PL330 DMAC-%x\n", adev->periphid);
 	dev_info(&adev->dev,
 		"\tDBUFF-%ux%ubytes Num_Chans-%u Num_Peri-%u Num_Events-%u\n",
 		pcfg->data_buf_dep, pcfg->data_bus_width / 8, pcfg->num_chan,
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 3aa10b32825491dce9d5c0ad244ddc53f78df12e..91292f5513ff2df6d051b784fcb81d276f6e6c63 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -230,30 +230,25 @@ static inline void sun6i_dma_dump_chan_regs(struct sun6i_dma_dev *sdev,
 		readl(pchan->base + DMA_CHAN_CUR_PARA));
 }
 
-static inline int convert_burst(u32 maxburst, u8 *burst)
+static inline s8 convert_burst(u32 maxburst)
 {
 	switch (maxburst) {
 	case 1:
-		*burst = 0;
-		break;
+		return 0;
 	case 8:
-		*burst = 2;
-		break;
+		return 2;
 	default:
 		return -EINVAL;
 	}
-
-	return 0;
 }
 
-static inline int convert_buswidth(enum dma_slave_buswidth addr_width, u8 *width)
+static inline s8 convert_buswidth(enum dma_slave_buswidth addr_width)
 {
 	if ((addr_width < DMA_SLAVE_BUSWIDTH_1_BYTE) ||
 	    (addr_width > DMA_SLAVE_BUSWIDTH_4_BYTES))
 		return -EINVAL;
 
-	*width = addr_width >> 1;
-	return 0;
+	return addr_width >> 1;
 }
 
 static void *sun6i_dma_lli_add(struct sun6i_dma_lli *prev,
@@ -284,26 +279,25 @@ static inline int sun6i_dma_cfg_lli(struct sun6i_dma_lli *lli,
 				    struct dma_slave_config *config)
 {
 	u8 src_width, dst_width, src_burst, dst_burst;
-	int ret;
 
 	if (!config)
 		return -EINVAL;
 
-	ret = convert_burst(config->src_maxburst, &src_burst);
-	if (ret)
-		return ret;
+	src_burst = convert_burst(config->src_maxburst);
+	if (src_burst)
+		return src_burst;
 
-	ret = convert_burst(config->dst_maxburst, &dst_burst);
-	if (ret)
-		return ret;
+	dst_burst = convert_burst(config->dst_maxburst);
+	if (dst_burst)
+		return dst_burst;
 
-	ret = convert_buswidth(config->src_addr_width, &src_width);
-	if (ret)
-		return ret;
+	src_width = convert_buswidth(config->src_addr_width);
+	if (src_width)
+		return src_width;
 
-	ret = convert_buswidth(config->dst_addr_width, &dst_width);
-	if (ret)
-		return ret;
+	dst_width = convert_buswidth(config->dst_addr_width);
+	if (dst_width)
+		return dst_width;
 
 	lli->cfg = DMA_CHAN_CFG_SRC_BURST(src_burst) |
 		DMA_CHAN_CFG_SRC_WIDTH(src_width) |
@@ -542,11 +536,10 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 {
 	struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
 	struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
-	struct dma_slave_config *sconfig = &vchan->cfg;
 	struct sun6i_dma_lli *v_lli;
 	struct sun6i_desc *txd;
 	dma_addr_t p_lli;
-	int ret;
+	s8 burst, width;
 
 	dev_dbg(chan2dev(chan),
 		"%s; chan: %d, dest: %pad, src: %pad, len: %zu. flags: 0x%08lx\n",
@@ -565,14 +558,21 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 		goto err_txd_free;
 	}
 
-	ret = sun6i_dma_cfg_lli(v_lli, src, dest, len, sconfig);
-	if (ret)
-		goto err_dma_free;
+	v_lli->src = src;
+	v_lli->dst = dest;
+	v_lli->len = len;
+	v_lli->para = NORMAL_WAIT;
 
+	burst = convert_burst(8);
+	width = convert_buswidth(DMA_SLAVE_BUSWIDTH_4_BYTES);
 	v_lli->cfg |= DMA_CHAN_CFG_SRC_DRQ(DRQ_SDRAM) |
 		DMA_CHAN_CFG_DST_DRQ(DRQ_SDRAM) |
 		DMA_CHAN_CFG_DST_LINEAR_MODE |
-		DMA_CHAN_CFG_SRC_LINEAR_MODE;
+		DMA_CHAN_CFG_SRC_LINEAR_MODE |
+		DMA_CHAN_CFG_SRC_BURST(burst) |
+		DMA_CHAN_CFG_SRC_WIDTH(width) |
+		DMA_CHAN_CFG_DST_BURST(burst) |
+		DMA_CHAN_CFG_DST_WIDTH(width);
 
 	sun6i_dma_lli_add(NULL, v_lli, p_lli, txd);
 
@@ -580,8 +580,6 @@ static struct dma_async_tx_descriptor *sun6i_dma_prep_dma_memcpy(
 
 	return vchan_tx_prep(&vchan->vc, &txd->vd, flags);
 
-err_dma_free:
-	dma_pool_free(sdev->pool, v_lli, p_lli);
 err_txd_free:
 	kfree(txd);
 	return NULL;
@@ -915,6 +913,7 @@ static int sun6i_dma_probe(struct platform_device *pdev)
 	sdc->slave.device_prep_dma_memcpy	= sun6i_dma_prep_dma_memcpy;
 	sdc->slave.device_control		= sun6i_dma_control;
 	sdc->slave.chancnt			= NR_MAX_VCHANS;
+	sdc->slave.copy_align			= 4;
 
 	sdc->slave.dev = &pdev->dev;