Merge tag 'dmaengine-fix-4.4-rc6' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine fixes from Vinod Koul:
 "This has fixes spread thru driver, notably among them:

   - edma fixes for recent edma DT changes which went into 4.4
   - odd fixes for at_hdmac
   - minor fixes on bc dma and mic dma"

* tag 'dmaengine-fix-4.4-rc6' of git://git.infradead.org/users/vkoul/slave-dma:
  dmaengine: at_xdmac: fix at_xdmac_prep_dma_memcpy()
  dmaengine: edma: DT: Change reserved slot array from 16bit to 32bit type
  dmaengine: edma: DT: Change memcpy channel array from 16bit to 32bit type
  dmaengine: mic_x100: add missing spin_unlock
  dmaengine: bcm2835-dma: Convert to use DMA pool
  dmaengine: at_xdmac: fix bad behavior in interleaved mode
  dmaengine: at_xdmac: fix false condition for memset_sg transfers
  dmaengine: at_xdmac: fix macro typo

6 files changed, 110 insertions, 57 deletions

diff --git a/Documentation/devicetree/bindings/dma/ti-edma.txt b/Documentation/devicetree/bindings/dma/ti-edma.txt
index d3d0a4fb1c73..079b42a81d7c 100644
--- a/Documentation/devicetree/bindings/dma/ti-edma.txt
+++ b/Documentation/devicetree/bindings/dma/ti-edma.txt
@@ -22,8 +22,7 @@ Required properties:
 Optional properties:
 - ti,hwmods:	Name of the hwmods associated to the eDMA CC
 - ti,edma-memcpy-channels: List of channels allocated to be used for memcpy, iow
-		these channels will be SW triggered channels. The list must
-		contain 16 bits numbers, see example.
+		these channels will be SW triggered channels. See example.
 - ti,edma-reserved-slot-ranges: PaRAM slot ranges which should not be used by
 		the driver, they are allocated to be used by for example the
 		DSP. See example.
@@ -56,10 +55,9 @@ edma: edma@49000000 {
 	ti,tptcs = <&edma_tptc0 7>, <&edma_tptc1 7>, <&edma_tptc2 0>;
 
 	/* Channel 20 and 21 is allocated for memcpy */
-	ti,edma-memcpy-channels = /bits/ 16 <20 21>;
-	/* The following PaRAM slots are reserved: 35-45 and 100-110 */
-	ti,edma-reserved-slot-ranges = /bits/ 16 <35 10>,
-				       /bits/ 16 <100 10>;
+	ti,edma-memcpy-channels = <20 21>;
+	/* The following PaRAM slots are reserved: 35-44 and 100-109 */
+	ti,edma-reserved-slot-ranges = <35 10>, <100 10>;
 };
 
 edma_tptc0: tptc@49800000 {
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 7f039de143f0..370c661c7d7b 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -156,7 +156,7 @@
 #define		AT_XDMAC_CC_WRIP	(0x1 << 23)	/* Write in Progress (read only) */
 #define			AT_XDMAC_CC_WRIP_DONE		(0x0 << 23)
 #define			AT_XDMAC_CC_WRIP_IN_PROGRESS	(0x1 << 23)
-#define		AT_XDMAC_CC_PERID(i)	(0x7f & (h) << 24)	/* Channel Peripheral Identifier */
+#define		AT_XDMAC_CC_PERID(i)	(0x7f & (i) << 24)	/* Channel Peripheral Identifier */
 #define AT_XDMAC_CDS_MSP	0x2C	/* Channel Data Stride Memory Set Pattern */
 #define AT_XDMAC_CSUS		0x30	/* Channel Source Microblock Stride */
 #define AT_XDMAC_CDUS		0x34	/* Channel Destination Microblock Stride */
@@ -965,7 +965,9 @@ at_xdmac_prep_interleaved(struct dma_chan *chan,
 							NULL,
 							src_addr, dst_addr,
 							xt, xt->sgl);
-		for (i = 0; i < xt->numf; i++)
+
+		/* Length of the block is (BLEN+1) microblocks. */
+		for (i = 0; i < xt->numf - 1; i++)
 			at_xdmac_increment_block_count(chan, first);
 
 		dev_dbg(chan2dev(chan), "%s: add desc 0x%p to descs_list 0x%p\n",
@@ -1086,6 +1088,7 @@ at_xdmac_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
 		/* Check remaining length and change data width if needed. */
 		dwidth = at_xdmac_align_width(chan,
 					      src_addr | dst_addr | xfer_size);
+		chan_cc &= ~AT_XDMAC_CC_DWIDTH_MASK;
 		chan_cc |= AT_XDMAC_CC_DWIDTH(dwidth);
 
 		ublen = xfer_size >> dwidth;
@@ -1333,7 +1336,7 @@ at_xdmac_prep_dma_memset_sg(struct dma_chan *chan, struct scatterlist *sgl,
 		 * since we don't care about the stride anymore.
 		 */
 		if ((i == (sg_len - 1)) &&
-		    sg_dma_len(ppsg) == sg_dma_len(psg)) {
+		    sg_dma_len(psg) == sg_dma_len(sg)) {
 			dev_dbg(chan2dev(chan),
 				"%s: desc 0x%p can be merged with desc 0x%p\n",
 				__func__, desc, pdesc);
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index c92d6a70ccf3..996c4b00d323 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -31,6 +31,7 @@
  */
 #include <linux/dmaengine.h>
 #include <linux/dma-mapping.h>
+#include <linux/dmapool.h>
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
@@ -62,6 +63,11 @@ struct bcm2835_dma_cb {
 	uint32_t pad[2];
 };
 
+struct bcm2835_cb_entry {
+	struct bcm2835_dma_cb *cb;
+	dma_addr_t paddr;
+};
+
 struct bcm2835_chan {
 	struct virt_dma_chan vc;
 	struct list_head node;
@@ -72,18 +78,18 @@ struct bcm2835_chan {
 
 	int ch;
 	struct bcm2835_desc *desc;
+	struct dma_pool *cb_pool;
 
 	void __iomem *chan_base;
 	int irq_number;
 };
 
 struct bcm2835_desc {
+	struct bcm2835_chan *c;
 	struct virt_dma_desc vd;
 	enum dma_transfer_direction dir;
 
-	unsigned int control_block_size;
-	struct bcm2835_dma_cb *control_block_base;
-	dma_addr_t control_block_base_phys;
+	struct bcm2835_cb_entry *cb_list;
 
 	unsigned int frames;
 	size_t size;
@@ -143,10 +149,13 @@ static inline struct bcm2835_desc *to_bcm2835_dma_desc(
 static void bcm2835_dma_desc_free(struct virt_dma_desc *vd)
 {
 	struct bcm2835_desc *desc = container_of(vd, struct bcm2835_desc, vd);
-	dma_free_coherent(desc->vd.tx.chan->device->dev,
-			desc->control_block_size,
-			desc->control_block_base,
-			desc->control_block_base_phys);
+	int i;
+
+	for (i = 0; i < desc->frames; i++)
+		dma_pool_free(desc->c->cb_pool, desc->cb_list[i].cb,
+			      desc->cb_list[i].paddr);
+
+	kfree(desc->cb_list);
 	kfree(desc);
 }
 
@@ -199,7 +208,7 @@ static void bcm2835_dma_start_desc(struct bcm2835_chan *c)
 
 	c->desc = d = to_bcm2835_dma_desc(&vd->tx);
 
-	writel(d->control_block_base_phys, c->chan_base + BCM2835_DMA_ADDR);
+	writel(d->cb_list[0].paddr, c->chan_base + BCM2835_DMA_ADDR);
 	writel(BCM2835_DMA_ACTIVE, c->chan_base + BCM2835_DMA_CS);
 }
 
@@ -232,9 +241,16 @@ static irqreturn_t bcm2835_dma_callback(int irq, void *data)
 static int bcm2835_dma_alloc_chan_resources(struct dma_chan *chan)
 {
 	struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
+	struct device *dev = c->vc.chan.device->dev;
+
+	dev_dbg(dev, "Allocating DMA channel %d\n", c->ch);
 
-	dev_dbg(c->vc.chan.device->dev,
-			"Allocating DMA channel %d\n", c->ch);
+	c->cb_pool = dma_pool_create(dev_name(dev), dev,
+				     sizeof(struct bcm2835_dma_cb), 0, 0);
+	if (!c->cb_pool) {
+		dev_err(dev, "unable to allocate descriptor pool\n");
+		return -ENOMEM;
+	}
 
 	return request_irq(c->irq_number,
 			bcm2835_dma_callback, 0, "DMA IRQ", c);
@@ -246,6 +262,7 @@ static void bcm2835_dma_free_chan_resources(struct dma_chan *chan)
 
 	vchan_free_chan_resources(&c->vc);
 	free_irq(c->irq_number, c);
+	dma_pool_destroy(c->cb_pool);
 
 	dev_dbg(c->vc.chan.device->dev, "Freeing DMA channel %u\n", c->ch);
 }
@@ -261,8 +278,7 @@ static size_t bcm2835_dma_desc_size_pos(struct bcm2835_desc *d, dma_addr_t addr)
 	size_t size;
 
 	for (size = i = 0; i < d->frames; i++) {
-		struct bcm2835_dma_cb *control_block =
-			&d->control_block_base[i];
+		struct bcm2835_dma_cb *control_block = d->cb_list[i].cb;
 		size_t this_size = control_block->length;
 		dma_addr_t dma;
 
@@ -343,6 +359,7 @@ static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_cyclic(
 	dma_addr_t dev_addr;
 	unsigned int es, sync_type;
 	unsigned int frame;
+	int i;
 
 	/* Grab configuration */
 	if (!is_slave_direction(direction)) {
@@ -374,27 +391,31 @@ static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_cyclic(
 	if (!d)
 		return NULL;
 
+	d->c = c;
 	d->dir = direction;
 	d->frames = buf_len / period_len;
 
-	/* Allocate memory for control blocks */
-	d->control_block_size = d->frames * sizeof(struct bcm2835_dma_cb);
-	d->control_block_base = dma_zalloc_coherent(chan->device->dev,
-			d->control_block_size, &d->control_block_base_phys,
-			GFP_NOWAIT);
-
-	if (!d->control_block_base) {
+	d->cb_list = kcalloc(d->frames, sizeof(*d->cb_list), GFP_KERNEL);
+	if (!d->cb_list) {
 		kfree(d);
 		return NULL;
 	}
+	/* Allocate memory for control blocks */
+	for (i = 0; i < d->frames; i++) {
+		struct bcm2835_cb_entry *cb_entry = &d->cb_list[i];
+
+		cb_entry->cb = dma_pool_zalloc(c->cb_pool, GFP_ATOMIC,
+					       &cb_entry->paddr);
+		if (!cb_entry->cb)
+			goto error_cb;
+	}
 
 	/*
 	 * Iterate over all frames, create a control block
 	 * for each frame and link them together.
 	 */
 	for (frame = 0; frame < d->frames; frame++) {
-		struct bcm2835_dma_cb *control_block =
-			&d->control_block_base[frame];
+		struct bcm2835_dma_cb *control_block = d->cb_list[frame].cb;
 
 		/* Setup adresses */
 		if (d->dir == DMA_DEV_TO_MEM) {
@@ -428,12 +449,21 @@ static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_cyclic(
 		 * This DMA engine driver currently only supports cyclic DMA.
 		 * Therefore, wrap around at number of frames.
 		 */
-		control_block->next = d->control_block_base_phys +
-			sizeof(struct bcm2835_dma_cb)
-			* ((frame + 1) % d->frames);
+		control_block->next = d->cb_list[((frame + 1) % d->frames)].paddr;
 	}
 
 	return vchan_tx_prep(&c->vc, &d->vd, flags);
+error_cb:
+	i--;
+	for (; i >= 0; i--) {
+		struct bcm2835_cb_entry *cb_entry = &d->cb_list[i];
+
+		dma_pool_free(c->cb_pool, cb_entry->cb, cb_entry->paddr);
+	}
+
+	kfree(d->cb_list);
+	kfree(d);
+	return NULL;
 }
 
 static int bcm2835_dma_slave_config(struct dma_chan *chan,
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 0675e268d577..16fe773fb846 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -1752,16 +1752,14 @@ static enum dma_status edma_tx_status(struct dma_chan *chan,
 	return ret;
 }
 
-static bool edma_is_memcpy_channel(int ch_num, u16 *memcpy_channels)
+static bool edma_is_memcpy_channel(int ch_num, s32 *memcpy_channels)
 {
-	s16 *memcpy_ch = memcpy_channels;
-
 	if (!memcpy_channels)
 		return false;
-	while (*memcpy_ch != -1) {
-		if (*memcpy_ch == ch_num)
+	while (*memcpy_channels != -1) {
+		if (*memcpy_channels == ch_num)
 			return true;
-		memcpy_ch++;
+		memcpy_channels++;
 	}
 	return false;
 }
@@ -1775,7 +1773,7 @@ static void edma_dma_init(struct edma_cc *ecc, bool legacy_mode)
 {
 	struct dma_device *s_ddev = &ecc->dma_slave;
 	struct dma_device *m_ddev = NULL;
-	s16 *memcpy_channels = ecc->info->memcpy_channels;
+	s32 *memcpy_channels = ecc->info->memcpy_channels;
 	int i, j;
 
 	dma_cap_zero(s_ddev->cap_mask);
@@ -1996,16 +1994,16 @@ static struct edma_soc_info *edma_setup_info_from_dt(struct device *dev,
 	prop = of_find_property(dev->of_node, "ti,edma-memcpy-channels", &sz);
 	if (prop) {
 		const char pname[] = "ti,edma-memcpy-channels";
-		size_t nelm = sz / sizeof(s16);
-		s16 *memcpy_ch;
+		size_t nelm = sz / sizeof(s32);
+		s32 *memcpy_ch;
 
-		memcpy_ch = devm_kcalloc(dev, nelm + 1, sizeof(s16),
+		memcpy_ch = devm_kcalloc(dev, nelm + 1, sizeof(s32),
 					 GFP_KERNEL);
 		if (!memcpy_ch)
 			return ERR_PTR(-ENOMEM);
 
-		ret = of_property_read_u16_array(dev->of_node, pname,
-						 (u16 *)memcpy_ch, nelm);
+		ret = of_property_read_u32_array(dev->of_node, pname,
+						 (u32 *)memcpy_ch, nelm);
 		if (ret)
 			return ERR_PTR(ret);
 
@@ -2017,31 +2015,50 @@ static struct edma_soc_info *edma_setup_info_from_dt(struct device *dev,
 				&sz);
 	if (prop) {
 		const char pname[] = "ti,edma-reserved-slot-ranges";
+		u32 (*tmp)[2];
 		s16 (*rsv_slots)[2];
-		size_t nelm = sz / sizeof(*rsv_slots);
+		size_t nelm = sz / sizeof(*tmp);
 		struct edma_rsv_info *rsv_info;
+		int i;
 
 		if (!nelm)
 			return info;
 
+		tmp = kcalloc(nelm, sizeof(*tmp), GFP_KERNEL);
+		if (!tmp)
+			return ERR_PTR(-ENOMEM);
+
 		rsv_info = devm_kzalloc(dev, sizeof(*rsv_info), GFP_KERNEL);
-		if (!rsv_info)
+		if (!rsv_info) {
+			kfree(tmp);
 			return ERR_PTR(-ENOMEM);
+		}
 
 		rsv_slots = devm_kcalloc(dev, nelm + 1, sizeof(*rsv_slots),
 					 GFP_KERNEL);
-		if (!rsv_slots)
+		if (!rsv_slots) {
+			kfree(tmp);
 			return ERR_PTR(-ENOMEM);
+		}
 
-		ret = of_property_read_u16_array(dev->of_node, pname,
-						 (u16 *)rsv_slots, nelm * 2);
-		if (ret)
+		ret = of_property_read_u32_array(dev->of_node, pname,
+						 (u32 *)tmp, nelm * 2);
+		if (ret) {
+			kfree(tmp);
 			return ERR_PTR(ret);
+		}
 
+		for (i = 0; i < nelm; i++) {
+			rsv_slots[i][0] = tmp[i][0];
+			rsv_slots[i][1] = tmp[i][1];
+		}
 		rsv_slots[nelm][0] = -1;
 		rsv_slots[nelm][1] = -1;
+
 		info->rsv = rsv_info;
 		info->rsv->rsv_slots = (const s16 (*)[2])rsv_slots;
+
+		kfree(tmp);
 	}
 
 	return info;
diff --git a/drivers/dma/mic_x100_dma.c b/drivers/dma/mic_x100_dma.c
index 068e920ecb68..cddfa8dbf4bd 100644
--- a/drivers/dma/mic_x100_dma.c
+++ b/drivers/dma/mic_x100_dma.c
@@ -317,6 +317,7 @@ mic_dma_prep_memcpy_lock(struct dma_chan *ch, dma_addr_t dma_dest,
 	struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch);
 	struct device *dev = mic_dma_ch_to_device(mic_ch);
 	int result;
+	struct dma_async_tx_descriptor *tx = NULL;
 
 	if (!len && !flags)
 		return NULL;
@@ -324,10 +325,13 @@ mic_dma_prep_memcpy_lock(struct dma_chan *ch, dma_addr_t dma_dest,
 	spin_lock(&mic_ch->prep_lock);
 	result = mic_dma_do_dma(mic_ch, flags, dma_src, dma_dest, len);
 	if (result >= 0)
-		return allocate_tx(mic_ch);
-	dev_err(dev, "Error enqueueing dma, error=%d\n", result);
+		tx = allocate_tx(mic_ch);
+
+	if (!tx)
+		dev_err(dev, "Error enqueueing dma, error=%d\n", result);
+
 	spin_unlock(&mic_ch->prep_lock);
-	return NULL;
+	return tx;
 }
 
 static struct dma_async_tx_descriptor *
@@ -335,13 +339,14 @@ mic_dma_prep_interrupt_lock(struct dma_chan *ch, unsigned long flags)
 {
 	struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch);
 	int ret;
+	struct dma_async_tx_descriptor *tx = NULL;
 
 	spin_lock(&mic_ch->prep_lock);
 	ret = mic_dma_do_dma(mic_ch, flags, 0, 0, 0);
 	if (!ret)
-		return allocate_tx(mic_ch);
+		tx = allocate_tx(mic_ch);
 	spin_unlock(&mic_ch->prep_lock);
-	return NULL;
+	return tx;
 }
 
 /* Return the status of the transaction */
diff --git a/include/linux/platform_data/edma.h b/include/linux/platform_data/edma.h
index e2878baeb90e..4299f4ba03bd 100644
--- a/include/linux/platform_data/edma.h
+++ b/include/linux/platform_data/edma.h
@@ -72,7 +72,7 @@ struct edma_soc_info {
 	struct edma_rsv_info	*rsv;
 
 	/* List of channels allocated for memcpy, terminated with -1 */
-	s16			*memcpy_channels;
+	s32			*memcpy_channels;
 
 	s8	(*queue_priority_mapping)[2];
 	const s16	(*xbar_chans)[2];