Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx: (24 commits)
  I/OAT: I/OAT version 3.0 support
  I/OAT: tcp_dma_copybreak default value dependent on I/OAT version
  I/OAT: Add watchdog/reset functionality to ioatdma
  iop_adma: cleanup iop_chan_xor_slot_count
  iop_adma: document how to calculate the minimum descriptor pool size
  iop_adma: directly reclaim descriptors on allocation failure
  async_tx: make async_tx_test_ack a boolean routine
  async_tx: remove depend_tx from async_tx_sync_epilog
  async_tx: export async_tx_quiesce
  async_tx: fix handling of the "out of descriptor" condition in async_xor
  async_tx: ensure the xor destination buffer remains dma-mapped
  async_tx: list_for_each_entry_rcu() cleanup
  dmaengine: Driver for the Synopsys DesignWare DMA controller
  dmaengine: Add slave DMA interface
  dmaengine: add DMA_COMPL_SKIP_{SRC,DEST}_UNMAP flags to control dma unmap
  dmaengine: Add dma_client parameter to device_alloc_chan_resources
  dmatest: Simple DMA memcpy test client
  dmaengine: DMA engine driver for Marvell XOR engine
  iop-adma: fix platform driver hotplug/coldplug
  dmaengine: track the number of clients using a channel
  ...

Fixed up conflict in drivers/dca/dca-sysfs.c manually

33 files changed, 4590 insertions, 339 deletions

diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 021d51217184..604f44f5dd16 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -7,6 +7,7 @@
  */
 #include <linux/clk.h>
 #include <linux/delay.h>
+#include <linux/dw_dmac.h>
 #include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
@@ -594,6 +595,17 @@ static void __init genclk_init_parent(struct clk *clk)
 	clk->parent = parent;
 }
 
+static struct dw_dma_platform_data dw_dmac0_data = {
+	.nr_channels	= 3,
+};
+
+static struct resource dw_dmac0_resource[] = {
+	PBMEM(0xff200000),
+	IRQ(2),
+};
+DEFINE_DEV_DATA(dw_dmac, 0);
+DEV_CLK(hclk, dw_dmac0, hsb, 10);
+
 /* --------------------------------------------------------------------
  *  System peripherals
  * -------------------------------------------------------------------- */
@@ -708,17 +720,6 @@ static struct clk pico_clk = {
 	.users		= 1,
 };
 
-static struct resource dmaca0_resource[] = {
-	{
-		.start	= 0xff200000,
-		.end	= 0xff20ffff,
-		.flags	= IORESOURCE_MEM,
-	},
-	IRQ(2),
-};
-DEFINE_DEV(dmaca, 0);
-DEV_CLK(hclk, dmaca0, hsb, 10);
-
 /* --------------------------------------------------------------------
  * HMATRIX
  * -------------------------------------------------------------------- */
@@ -831,7 +832,7 @@ void __init at32_add_system_devices(void)
 	platform_device_register(&at32_eic0_device);
 	platform_device_register(&smc0_device);
 	platform_device_register(&pdc_device);
-	platform_device_register(&dmaca0_device);
+	platform_device_register(&dw_dmac0_device);
 
 	platform_device_register(&at32_tcb0_device);
 	platform_device_register(&at32_tcb1_device);
@@ -2032,7 +2033,7 @@ struct clk *at32_clock_list[] = {
 	&smc0_mck,
 	&pdc_hclk,
 	&pdc_pclk,
-	&dmaca0_hclk,
+	&dw_dmac0_hclk,
 	&pico_clk,
 	&pio0_mck,
 	&pio1_mck,
diff --git a/crypto/async_tx/async_memcpy.c b/crypto/async_tx/async_memcpy.c
index a5eda80e8427..ddccfb01c416 100644
--- a/crypto/async_tx/async_memcpy.c
+++ b/crypto/async_tx/async_memcpy.c
@@ -73,15 +73,7 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 		pr_debug("%s: (sync) len: %zu\n", __func__, len);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		dest_buf = kmap_atomic(dest, KM_USER0) + dest_offset;
 		src_buf = kmap_atomic(src, KM_USER1) + src_offset;
@@ -91,7 +83,7 @@ async_memcpy(struct page *dest, struct page *src, unsigned int dest_offset,
 		kunmap_atomic(dest_buf, KM_USER0);
 		kunmap_atomic(src_buf, KM_USER1);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/crypto/async_tx/async_memset.c b/crypto/async_tx/async_memset.c
index f5ff3906b035..5b5eb99bb244 100644
--- a/crypto/async_tx/async_memset.c
+++ b/crypto/async_tx/async_memset.c
@@ -72,19 +72,11 @@ async_memset(struct page *dest, int val, unsigned int offset,
 		dest_buf = (void *) (((char *) page_address(dest)) + offset);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(depend_tx->ack);
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
 		memset(dest_buf, val, len);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/crypto/async_tx/async_tx.c b/crypto/async_tx/async_tx.c
index 095c798d3170..85eaf7b1c531 100644
--- a/crypto/async_tx/async_tx.c
+++ b/crypto/async_tx/async_tx.c
@@ -295,7 +295,7 @@ dma_channel_add_remove(struct dma_client *client,
 	case DMA_RESOURCE_REMOVED:
 		found = 0;
 		spin_lock_irqsave(&async_tx_lock, flags);
-		list_for_each_entry_rcu(ref, &async_tx_master_list, node)
+		list_for_each_entry(ref, &async_tx_master_list, node)
 			if (ref->chan == chan) {
 				/* permit backing devices to go away */
 				dma_chan_put(ref->chan);
@@ -608,23 +608,34 @@ async_trigger_callback(enum async_tx_flags flags,
 		pr_debug("%s: (sync)\n", __func__);
 
 		/* wait for any prerequisite operations */
-		if (depend_tx) {
-			/* if ack is already set then we cannot be sure
-			 * we are referring to the correct operation
-			 */
-			BUG_ON(async_tx_test_ack(depend_tx));
-			if (dma_wait_for_async_tx(depend_tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for depend_tx\n",
-					__func__);
-		}
+		async_tx_quiesce(&depend_tx);
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
 }
 EXPORT_SYMBOL_GPL(async_trigger_callback);
 
+/**
+ * async_tx_quiesce - ensure tx is complete and freeable upon return
+ * @tx - transaction to quiesce
+ */
+void async_tx_quiesce(struct dma_async_tx_descriptor **tx)
+{
+	if (*tx) {
+		/* if ack is already set then we cannot be sure
+		 * we are referring to the correct operation
+		 */
+		BUG_ON(async_tx_test_ack(*tx));
+		if (dma_wait_for_async_tx(*tx) == DMA_ERROR)
+			panic("DMA_ERROR waiting for transaction\n");
+		async_tx_ack(*tx);
+		*tx = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(async_tx_quiesce);
+
 module_init(async_tx_init);
 module_exit(async_tx_exit);
 
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 3a0dddca5a10..65974c6d3d7a 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -35,74 +35,121 @@
  * 	when CONFIG_DMA_ENGINE=n
  */
 static __always_inline struct dma_async_tx_descriptor *
-do_async_xor(struct dma_device *device,
-	struct dma_chan *chan, struct page *dest, struct page **src_list,
-	unsigned int offset, unsigned int src_cnt, size_t len,
-	enum async_tx_flags flags, struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+do_async_xor(struct dma_chan *chan, struct page *dest, struct page **src_list,
+	     unsigned int offset, int src_cnt, size_t len,
+	     enum async_tx_flags flags,
+	     struct dma_async_tx_descriptor *depend_tx,
+	     dma_async_tx_callback cb_fn, void *cb_param)
 {
-	dma_addr_t dma_dest;
+	struct dma_device *dma = chan->device;
 	dma_addr_t *dma_src = (dma_addr_t *) src_list;
-	struct dma_async_tx_descriptor *tx;
+	struct dma_async_tx_descriptor *tx = NULL;
+	int src_off = 0;
 	int i;
-	unsigned long dma_prep_flags = cb_fn ? DMA_PREP_INTERRUPT : 0;
-
-	pr_debug("%s: len: %zu\n", __func__, len);
-
-	dma_dest = dma_map_page(device->dev, dest, offset, len,
-				DMA_FROM_DEVICE);
+	dma_async_tx_callback _cb_fn;
+	void *_cb_param;
+	enum async_tx_flags async_flags;
+	enum dma_ctrl_flags dma_flags;
+	int xor_src_cnt;
+	dma_addr_t dma_dest;
 
+	dma_dest = dma_map_page(dma->dev, dest, offset, len, DMA_FROM_DEVICE);
 	for (i = 0; i < src_cnt; i++)
-		dma_src[i] = dma_map_page(device->dev, src_list[i], offset,
+		dma_src[i] = dma_map_page(dma->dev, src_list[i], offset,
 					  len, DMA_TO_DEVICE);
 
-	/* Since we have clobbered the src_list we are committed
-	 * to doing this asynchronously.  Drivers force forward progress
-	 * in case they can not provide a descriptor
-	 */
-	tx = device->device_prep_dma_xor(chan, dma_dest, dma_src, src_cnt, len,
-					 dma_prep_flags);
-	if (!tx) {
-		if (depend_tx)
-			dma_wait_for_async_tx(depend_tx);
-
-		while (!tx)
-			tx = device->device_prep_dma_xor(chan, dma_dest,
-							 dma_src, src_cnt, len,
-							 dma_prep_flags);
-	}
+	while (src_cnt) {
+		async_flags = flags;
+		dma_flags = 0;
+		xor_src_cnt = min(src_cnt, dma->max_xor);
+		/* if we are submitting additional xors, leave the chain open,
+		 * clear the callback parameters, and leave the destination
+		 * buffer mapped
+		 */
+		if (src_cnt > xor_src_cnt) {
+			async_flags &= ~ASYNC_TX_ACK;
+			dma_flags = DMA_COMPL_SKIP_DEST_UNMAP;
+			_cb_fn = NULL;
+			_cb_param = NULL;
+		} else {
+			_cb_fn = cb_fn;
+			_cb_param = cb_param;
+		}
+		if (_cb_fn)
+			dma_flags |= DMA_PREP_INTERRUPT;
 
-	async_tx_submit(chan, tx, flags, depend_tx, cb_fn, cb_param);
+		/* Since we have clobbered the src_list we are committed
+		 * to doing this asynchronously.  Drivers force forward progress
+		 * in case they can not provide a descriptor
+		 */
+		tx = dma->device_prep_dma_xor(chan, dma_dest, &dma_src[src_off],
+					      xor_src_cnt, len, dma_flags);
+
+		if (unlikely(!tx))
+			async_tx_quiesce(&depend_tx);
+
+		/* spin wait for the preceeding transactions to complete */
+		while (unlikely(!tx)) {
+			dma_async_issue_pending(chan);
+			tx = dma->device_prep_dma_xor(chan, dma_dest,
+						      &dma_src[src_off],
+						      xor_src_cnt, len,
+						      dma_flags);
+		}
+
+		async_tx_submit(chan, tx, async_flags, depend_tx, _cb_fn,
+				_cb_param);
+
+		depend_tx = tx;
+		flags |= ASYNC_TX_DEP_ACK;
+
+		if (src_cnt > xor_src_cnt) {
+			/* drop completed sources */
+			src_cnt -= xor_src_cnt;
+			src_off += xor_src_cnt;
+
+			/* use the intermediate result a source */
+			dma_src[--src_off] = dma_dest;
+			src_cnt++;
+		} else
+			break;
+	}
 
 	return tx;
 }
 
 static void
 do_sync_xor(struct page *dest, struct page **src_list, unsigned int offset,
-	unsigned int src_cnt, size_t len, enum async_tx_flags flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+	    int src_cnt, size_t len, enum async_tx_flags flags,
+	    dma_async_tx_callback cb_fn, void *cb_param)
 {
-	void *_dest;
 	int i;
-
-	pr_debug("%s: len: %zu\n", __func__, len);
+	int xor_src_cnt;
+	int src_off = 0;
+	void *dest_buf;
+	void **srcs = (void **) src_list;
 
 	/* reuse the 'src_list' array to convert to buffer pointers */
 	for (i = 0; i < src_cnt; i++)
-		src_list[i] = (struct page *)
-			(page_address(src_list[i]) + offset);
+		srcs[i] = page_address(src_list[i]) + offset;
 
 	/* set destination address */
-	_dest = page_address(dest) + offset;
+	dest_buf = page_address(dest) + offset;
 
 	if (flags & ASYNC_TX_XOR_ZERO_DST)
-		memset(_dest, 0, len);
+		memset(dest_buf, 0, len);
 
-	xor_blocks(src_cnt, len, _dest,
-		(void **) src_list);
+	while (src_cnt > 0) {
+		/* process up to 'MAX_XOR_BLOCKS' sources */
+		xor_src_cnt = min(src_cnt, MAX_XOR_BLOCKS);
+		xor_blocks(xor_src_cnt, len, dest_buf, &srcs[src_off]);
 
-	async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		/* drop completed sources */
+		src_cnt -= xor_src_cnt;
+		src_off += xor_src_cnt;
+	}
+
+	async_tx_sync_epilog(cb_fn, cb_param);
 }
 
 /**
@@ -132,106 +179,34 @@ async_xor(struct page *dest, struct page **src_list, unsigned int offset,
 	struct dma_chan *chan = async_tx_find_channel(depend_tx, DMA_XOR,
 						      &dest, 1, src_list,
 						      src_cnt, len);
-	struct dma_device *device = chan ? chan->device : NULL;
-	struct dma_async_tx_descriptor *tx = NULL;
-	dma_async_tx_callback _cb_fn;
-	void *_cb_param;
-	unsigned long local_flags;
-	int xor_src_cnt;
-	int i = 0, src_off = 0;
-
 	BUG_ON(src_cnt <= 1);
 
-	while (src_cnt) {
-		local_flags = flags;
-		if (device) { /* run the xor asynchronously */
-			xor_src_cnt = min(src_cnt, device->max_xor);
-			/* if we are submitting additional xors
-			 * only set the callback on the last transaction
-			 */
-			if (src_cnt > xor_src_cnt) {
-				local_flags &= ~ASYNC_TX_ACK;
-				_cb_fn = NULL;
-				_cb_param = NULL;
-			} else {
-				_cb_fn = cb_fn;
-				_cb_param = cb_param;
-			}
-
-			tx = do_async_xor(device, chan, dest,
-					  &src_list[src_off], offset,
-					  xor_src_cnt, len, local_flags,
-					  depend_tx, _cb_fn, _cb_param);
-		} else { /* run the xor synchronously */
-			/* in the sync case the dest is an implied source
-			 * (assumes the dest is at the src_off index)
-			 */
-			if (flags & ASYNC_TX_XOR_DROP_DST) {
-				src_cnt--;
-				src_off++;
-			}
-
-			/* process up to 'MAX_XOR_BLOCKS' sources */
-			xor_src_cnt = min(src_cnt, MAX_XOR_BLOCKS);
-
-			/* if we are submitting additional xors
-			 * only set the callback on the last transaction
-			 */
-			if (src_cnt > xor_src_cnt) {
-				local_flags &= ~ASYNC_TX_ACK;
-				_cb_fn = NULL;
-				_cb_param = NULL;
-			} else {
-				_cb_fn = cb_fn;
-				_cb_param = cb_param;
-			}
-
-			/* wait for any prerequisite operations */
-			if (depend_tx) {
-				/* if ack is already set then we cannot be sure
-				 * we are referring to the correct operation
-				 */
-				BUG_ON(async_tx_test_ack(depend_tx));
-				if (dma_wait_for_async_tx(depend_tx) ==
-					DMA_ERROR)
-					panic("%s: DMA_ERROR waiting for "
-						"depend_tx\n",
-						__func__);
-			}
-
-			do_sync_xor(dest, &src_list[src_off], offset,
-				xor_src_cnt, len, local_flags, depend_tx,
-				_cb_fn, _cb_param);
-		}
+	if (chan) {
+		/* run the xor asynchronously */
+		pr_debug("%s (async): len: %zu\n", __func__, len);
 
-		/* the previous tx is hidden from the client,
-		 * so ack it
-		 */
-		if (i && depend_tx)
-			async_tx_ack(depend_tx);
+		return do_async_xor(chan, dest, src_list, offset, src_cnt, len,
+				    flags, depend_tx, cb_fn, cb_param);
+	} else {
+		/* run the xor synchronously */
+		pr_debug("%s (sync): len: %zu\n", __func__, len);
 
-		depend_tx = tx;
+		/* in the sync case the dest is an implied source
+		 * (assumes the dest is the first source)
+		 */
+		if (flags & ASYNC_TX_XOR_DROP_DST) {
+			src_cnt--;
+			src_list++;
+		}
 
-		if (src_cnt > xor_src_cnt) {
-			/* drop completed sources */
-			src_cnt -= xor_src_cnt;
-			src_off += xor_src_cnt;
+		/* wait for any prerequisite operations */
+		async_tx_quiesce(&depend_tx);
 
-			/* unconditionally preserve the destination */
-			flags &= ~ASYNC_TX_XOR_ZERO_DST;
+		do_sync_xor(dest, src_list, offset, src_cnt, len,
+			    flags, cb_fn, cb_param);
 
-			/* use the intermediate result a source, but remember
-			 * it's dropped, because it's implied, in the sync case
-			 */
-			src_list[--src_off] = dest;
-			src_cnt++;
-			flags |= ASYNC_TX_XOR_DROP_DST;
-		} else
-			src_cnt = 0;
-		i++;
+		return NULL;
 	}
-
-	return tx;
 }
 EXPORT_SYMBOL_GPL(async_xor);
 
@@ -285,11 +260,11 @@ async_xor_zero_sum(struct page *dest, struct page **src_list,
 		tx = device->device_prep_dma_zero_sum(chan, dma_src, src_cnt,
 						      len, result,
 						      dma_prep_flags);
-		if (!tx) {
-			if (depend_tx)
-				dma_wait_for_async_tx(depend_tx);
+		if (unlikely(!tx)) {
+			async_tx_quiesce(&depend_tx);
 
 			while (!tx)
+				dma_async_issue_pending(chan);
 				tx = device->device_prep_dma_zero_sum(chan,
 					dma_src, src_cnt, len, result,
 					dma_prep_flags);
@@ -307,18 +282,11 @@ async_xor_zero_sum(struct page *dest, struct page **src_list,
 		tx = async_xor(dest, src_list, offset, src_cnt, len, xor_flags,
 			depend_tx, NULL, NULL);
 
-		if (tx) {
-			if (dma_wait_for_async_tx(tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for tx\n",
-					__func__);
-			async_tx_ack(tx);
-		}
+		async_tx_quiesce(&tx);
 
 		*result = page_is_zero(dest, offset, len) ? 0 : 1;
 
-		tx = NULL;
-
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(cb_fn, cb_param);
 	}
 
 	return tx;
diff --git a/drivers/dca/dca-core.c b/drivers/dca/dca-core.c
index bf5b92f86df7..ec249d2db633 100644
--- a/drivers/dca/dca-core.c
+++ b/drivers/dca/dca-core.c
@@ -28,13 +28,29 @@
 #include <linux/device.h>
 #include <linux/dca.h>
 
-MODULE_LICENSE("GPL");
+#define DCA_VERSION "1.4"
 
-/* For now we're assuming a single, global, DCA provider for the system. */
+MODULE_VERSION(DCA_VERSION);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Intel Corporation");
 
 static DEFINE_SPINLOCK(dca_lock);
 
-static struct dca_provider *global_dca = NULL;
+static LIST_HEAD(dca_providers);
+
+static struct dca_provider *dca_find_provider_by_dev(struct device *dev)
+{
+	struct dca_provider *dca, *ret = NULL;
+
+	list_for_each_entry(dca, &dca_providers, node) {
+		if ((!dev) || (dca->ops->dev_managed(dca, dev))) {
+			ret = dca;
+			break;
+		}
+	}
+
+	return ret;
+}
 
 /**
  * dca_add_requester - add a dca client to the list
@@ -42,25 +58,39 @@ static struct dca_provider *global_dca = NULL;
  */
 int dca_add_requester(struct device *dev)
 {
-	int err, slot;
+	struct dca_provider *dca;
+	int err, slot = -ENODEV;
 
-	if (!global_dca)
-		return -ENODEV;
+	if (!dev)
+		return -EFAULT;
 
 	spin_lock(&dca_lock);
-	slot = global_dca->ops->add_requester(global_dca, dev);
-	spin_unlock(&dca_lock);
-	if (slot < 0)
+
+	/* check if the requester has not been added already */
+	dca = dca_find_provider_by_dev(dev);
+	if (dca) {
+		spin_unlock(&dca_lock);
+		return -EEXIST;
+	}
+
+	list_for_each_entry(dca, &dca_providers, node) {
+		slot = dca->ops->add_requester(dca, dev);
+		if (slot >= 0)
+			break;
+	}
+	if (slot < 0) {
+		spin_unlock(&dca_lock);
 		return slot;
+	}
 
-	err = dca_sysfs_add_req(global_dca, dev, slot);
+	err = dca_sysfs_add_req(dca, dev, slot);
 	if (err) {
-		spin_lock(&dca_lock);
-		global_dca->ops->remove_requester(global_dca, dev);
+		dca->ops->remove_requester(dca, dev);
 		spin_unlock(&dca_lock);
 		return err;
 	}
 
+	spin_unlock(&dca_lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dca_add_requester);
@@ -71,30 +101,78 @@ EXPORT_SYMBOL_GPL(dca_add_requester);
  */
 int dca_remove_requester(struct device *dev)
 {
+	struct dca_provider *dca;
 	int slot;
-	if (!global_dca)
-		return -ENODEV;
+
+	if (!dev)
+		return -EFAULT;
 
 	spin_lock(&dca_lock);
-	slot = global_dca->ops->remove_requester(global_dca, dev);
-	spin_unlock(&dca_lock);
-	if (slot < 0)
+	dca = dca_find_provider_by_dev(dev);
+	if (!dca) {
+		spin_unlock(&dca_lock);
+		return -ENODEV;
+	}
+	slot = dca->ops->remove_requester(dca, dev);
+	if (slot < 0) {
+		spin_unlock(&dca_lock);
 		return slot;
+	}
 
-	dca_sysfs_remove_req(global_dca, slot);
+	dca_sysfs_remove_req(dca, slot);
+
+	spin_unlock(&dca_lock);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(dca_remove_requester);
 
 /**
- * dca_get_tag - return the dca tag for the given cpu
+ * dca_common_get_tag - return the dca tag (serves both new and old api)
+ * @dev - the device that wants dca service
  * @cpu - the cpuid as returned by get_cpu()
  */
-u8 dca_get_tag(int cpu)
+u8 dca_common_get_tag(struct device *dev, int cpu)
 {
-	if (!global_dca)
+	struct dca_provider *dca;
+	u8 tag;
+
+	spin_lock(&dca_lock);
+
+	dca = dca_find_provider_by_dev(dev);
+	if (!dca) {
+		spin_unlock(&dca_lock);
 		return -ENODEV;
-	return global_dca->ops->get_tag(global_dca, cpu);
+	}
+	tag = dca->ops->get_tag(dca, dev, cpu);
+
+	spin_unlock(&dca_lock);
+	return tag;
+}
+
+/**
+ * dca3_get_tag - return the dca tag to the requester device
+ *                for the given cpu (new api)
+ * @dev - the device that wants dca service
+ * @cpu - the cpuid as returned by get_cpu()
+ */
+u8 dca3_get_tag(struct device *dev, int cpu)
+{
+	if (!dev)
+		return -EFAULT;
+
+	return dca_common_get_tag(dev, cpu);
+}
+EXPORT_SYMBOL_GPL(dca3_get_tag);
+
+/**
+ * dca_get_tag - return the dca tag for the given cpu (old api)
+ * @cpu - the cpuid as returned by get_cpu()
+ */
+u8 dca_get_tag(int cpu)
+{
+	struct device *dev = NULL;
+
+	return dca_common_get_tag(dev, cpu);
 }
 EXPORT_SYMBOL_GPL(dca_get_tag);
 
@@ -140,12 +218,10 @@ int register_dca_provider(struct dca_provider *dca, struct device *dev)
 {
 	int err;
 
-	if (global_dca)
-		return -EEXIST;
 	err = dca_sysfs_add_provider(dca, dev);
 	if (err)
 		return err;
-	global_dca = dca;
+	list_add(&dca->node, &dca_providers);
 	blocking_notifier_call_chain(&dca_provider_chain,
 				     DCA_PROVIDER_ADD, NULL);
 	return 0;
@@ -158,11 +234,9 @@ EXPORT_SYMBOL_GPL(register_dca_provider);
  */
 void unregister_dca_provider(struct dca_provider *dca)
 {
-	if (!global_dca)
-		return;
 	blocking_notifier_call_chain(&dca_provider_chain,
 				     DCA_PROVIDER_REMOVE, NULL);
-	global_dca = NULL;
+	list_del(&dca->node);
 	dca_sysfs_remove_provider(dca);
 }
 EXPORT_SYMBOL_GPL(unregister_dca_provider);
@@ -187,6 +261,7 @@ EXPORT_SYMBOL_GPL(dca_unregister_notify);
 
 static int __init dca_init(void)
 {
+	printk(KERN_ERR "dca service started, version %s\n", DCA_VERSION);
 	return dca_sysfs_init();
 }
 
diff --git a/drivers/dca/dca-sysfs.c b/drivers/dca/dca-sysfs.c
index 9a70377bfb34..7af4b403bd2d 100644
--- a/drivers/dca/dca-sysfs.c
+++ b/drivers/dca/dca-sysfs.c
@@ -13,10 +13,11 @@ static spinlock_t dca_idr_lock;
 int dca_sysfs_add_req(struct dca_provider *dca, struct device *dev, int slot)
 {
 	struct device *cd;
+	static int req_count;
 
 	cd = device_create_drvdata(dca_class, dca->cd,
 				   MKDEV(0, slot + 1), NULL,
-				   "requester%d", slot);
+				   "requester%d", req_count++);
 	if (IS_ERR(cd))
 		return PTR_ERR(cd);
 	return 0;
diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig
index 6239c3df30ac..cd303901eb5b 100644
--- a/drivers/dma/Kconfig
+++ b/drivers/dma/Kconfig
@@ -4,13 +4,14 @@
 
 menuconfig DMADEVICES
 	bool "DMA Engine support"
-	depends on (PCI && X86) || ARCH_IOP32X || ARCH_IOP33X || ARCH_IOP13XX || PPC
-	depends on !HIGHMEM64G
+	depends on !HIGHMEM64G && HAS_DMA
 	help
 	  DMA engines can do asynchronous data transfers without
 	  involving the host CPU.  Currently, this framework can be
 	  used to offload memory copies in the network stack and
-	  RAID operations in the MD driver.
+	  RAID operations in the MD driver.  This menu only presents
+	  DMA Device drivers supported by the configured arch, it may
+	  be empty in some cases.
 
 if DMADEVICES
 
@@ -37,6 +38,15 @@ config INTEL_IOP_ADMA
 	help
 	  Enable support for the Intel(R) IOP Series RAID engines.
 
+config DW_DMAC
+	tristate "Synopsys DesignWare AHB DMA support"
+	depends on AVR32
+	select DMA_ENGINE
+	default y if CPU_AT32AP7000
+	help
+	  Support the Synopsys DesignWare AHB DMA controller.  This
+	  can be integrated in chips such as the Atmel AT32ap7000.
+
 config FSL_DMA
 	bool "Freescale MPC85xx/MPC83xx DMA support"
 	depends on PPC
@@ -46,6 +56,14 @@ config FSL_DMA
 	  MPC8560/40, MPC8555, MPC8548 and MPC8641 processors.
 	  The MPC8349, MPC8360 is also supported.
 
+config MV_XOR
+	bool "Marvell XOR engine support"
+	depends on PLAT_ORION
+	select ASYNC_CORE
+	select DMA_ENGINE
+	---help---
+	  Enable support for the Marvell XOR engine.
+
 config DMA_ENGINE
 	bool
 
@@ -55,10 +73,19 @@ comment "DMA Clients"
 config NET_DMA
 	bool "Network: TCP receive copy offload"
 	depends on DMA_ENGINE && NET
+	default (INTEL_IOATDMA || FSL_DMA)
 	help
 	  This enables the use of DMA engines in the network stack to
 	  offload receive copy-to-user operations, freeing CPU cycles.
-	  Since this is the main user of the DMA engine, it should be enabled;
-	  say Y here.
+
+	  Say Y here if you enabled INTEL_IOATDMA or FSL_DMA, otherwise
+	  say N.
+
+config DMATEST
+	tristate "DMA Test client"
+	depends on DMA_ENGINE
+	help
+	  Simple DMA test client. Say N unless you're debugging a
+	  DMA Device driver.
 
 endif
diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
index c8036d945902..14f59527d4f6 100644
--- a/drivers/dma/Makefile
+++ b/drivers/dma/Makefile
@@ -1,6 +1,9 @@
 obj-$(CONFIG_DMA_ENGINE) += dmaengine.o
 obj-$(CONFIG_NET_DMA) += iovlock.o
+obj-$(CONFIG_DMATEST) += dmatest.o
 obj-$(CONFIG_INTEL_IOATDMA) += ioatdma.o
 ioatdma-objs := ioat.o ioat_dma.o ioat_dca.o
 obj-$(CONFIG_INTEL_IOP_ADMA) += iop-adma.o
 obj-$(CONFIG_FSL_DMA) += fsldma.o
+obj-$(CONFIG_MV_XOR) += mv_xor.o
+obj-$(CONFIG_DW_DMAC) += dw_dmac.o
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 97b329e76798..dc003a3a787d 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -169,12 +169,18 @@ static void dma_client_chan_alloc(struct dma_client *client)
 	enum dma_state_client ack;
 
 	/* Find a channel */
-	list_for_each_entry(device, &dma_device_list, global_node)
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		/* Does the client require a specific DMA controller? */
+		if (client->slave && client->slave->dma_dev
+				&& client->slave->dma_dev != device->dev)
+			continue;
+
 		list_for_each_entry(chan, &device->channels, device_node) {
 			if (!dma_chan_satisfies_mask(chan, client->cap_mask))
 				continue;
 
-			desc = chan->device->device_alloc_chan_resources(chan);
+			desc = chan->device->device_alloc_chan_resources(
+					chan, client);
 			if (desc >= 0) {
 				ack = client->event_callback(client,
 						chan,
@@ -183,12 +189,14 @@ static void dma_client_chan_alloc(struct dma_client *client)
 				/* we are done once this client rejects
 				 * an available resource
 				 */
-				if (ack == DMA_ACK)
+				if (ack == DMA_ACK) {
 					dma_chan_get(chan);
-				else if (ack == DMA_NAK)
+					chan->client_count++;
+				} else if (ack == DMA_NAK)
 					return;
 			}
 		}
+	}
 }
 
 enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie)
@@ -272,8 +280,10 @@ static void dma_clients_notify_removed(struct dma_chan *chan)
 		/* client was holding resources for this channel so
 		 * free it
 		 */
-		if (ack == DMA_ACK)
+		if (ack == DMA_ACK) {
 			dma_chan_put(chan);
+			chan->client_count--;
+		}
 	}
 
 	mutex_unlock(&dma_list_mutex);
@@ -285,6 +295,10 @@ static void dma_clients_notify_removed(struct dma_chan *chan)
  */
 void dma_async_client_register(struct dma_client *client)
 {
+	/* validate client data */
+	BUG_ON(dma_has_cap(DMA_SLAVE, client->cap_mask) &&
+		!client->slave);
+
 	mutex_lock(&dma_list_mutex);
 	list_add_tail(&client->global_node, &dma_client_list);
 	mutex_unlock(&dma_list_mutex);
@@ -313,8 +327,10 @@ void dma_async_client_unregister(struct dma_client *client)
 			ack = client->event_callback(client, chan,
 				DMA_RESOURCE_REMOVED);
 
-			if (ack == DMA_ACK)
+			if (ack == DMA_ACK) {
 				dma_chan_put(chan);
+				chan->client_count--;
+			}
 		}
 
 	list_del(&client->global_node);
@@ -359,6 +375,10 @@ int dma_async_device_register(struct dma_device *device)
 		!device->device_prep_dma_memset);
 	BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) &&
 		!device->device_prep_dma_interrupt);
+	BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
+		!device->device_prep_slave_sg);
+	BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
+		!device->device_terminate_all);
 
 	BUG_ON(!device->device_alloc_chan_resources);
 	BUG_ON(!device->device_free_chan_resources);
@@ -378,7 +398,7 @@ int dma_async_device_register(struct dma_device *device)
 
 		chan->chan_id = chancnt++;
 		chan->dev.class = &dma_devclass;
-		chan->dev.parent = NULL;
+		chan->dev.parent = device->dev;
 		snprintf(chan->dev.bus_id, BUS_ID_SIZE, "dma%dchan%d",
 		         device->dev_id, chan->chan_id);
 
@@ -394,6 +414,7 @@ int dma_async_device_register(struct dma_device *device)
 		kref_get(&device->refcount);
 		kref_get(&device->refcount);
 		kref_init(&chan->refcount);
+		chan->client_count = 0;
 		chan->slow_ref = 0;
 		INIT_RCU_HEAD(&chan->rcu);
 	}
diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
new file mode 100644
index 000000000000..a08d19704743
--- /dev/null
+++ b/drivers/dma/dmatest.c
@@ -0,0 +1,444 @@
+/*
+ * DMA Engine test module
+ *
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/init.h>
+#include <linux/kthread.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/random.h>
+#include <linux/wait.h>
+
+static unsigned int test_buf_size = 16384;
+module_param(test_buf_size, uint, S_IRUGO);
+MODULE_PARM_DESC(test_buf_size, "Size of the memcpy test buffer");
+
+static char test_channel[BUS_ID_SIZE];
+module_param_string(channel, test_channel, sizeof(test_channel), S_IRUGO);
+MODULE_PARM_DESC(channel, "Bus ID of the channel to test (default: any)");
+
+static char test_device[BUS_ID_SIZE];
+module_param_string(device, test_device, sizeof(test_device), S_IRUGO);
+MODULE_PARM_DESC(device, "Bus ID of the DMA Engine to test (default: any)");
+
+static unsigned int threads_per_chan = 1;
+module_param(threads_per_chan, uint, S_IRUGO);
+MODULE_PARM_DESC(threads_per_chan,
+		"Number of threads to start per channel (default: 1)");
+
+static unsigned int max_channels;
+module_param(max_channels, uint, S_IRUGO);
+MODULE_PARM_DESC(nr_channels,
+		"Maximum number of channels to use (default: all)");
+
+/*
+ * Initialization patterns. All bytes in the source buffer has bit 7
+ * set, all bytes in the destination buffer has bit 7 cleared.
+ *
+ * Bit 6 is set for all bytes which are to be copied by the DMA
+ * engine. Bit 5 is set for all bytes which are to be overwritten by
+ * the DMA engine.
+ *
+ * The remaining bits are the inverse of a counter which increments by
+ * one for each byte address.
+ */
+#define PATTERN_SRC		0x80
+#define PATTERN_DST		0x00
+#define PATTERN_COPY		0x40
+#define PATTERN_OVERWRITE	0x20
+#define PATTERN_COUNT_MASK	0x1f
+
+struct dmatest_thread {
+	struct list_head	node;
+	struct task_struct	*task;
+	struct dma_chan		*chan;
+	u8			*srcbuf;
+	u8			*dstbuf;
+};
+
+struct dmatest_chan {
+	struct list_head	node;
+	struct dma_chan		*chan;
+	struct list_head	threads;
+};
+
+/*
+ * These are protected by dma_list_mutex since they're only used by
+ * the DMA client event callback
+ */
+static LIST_HEAD(dmatest_channels);
+static unsigned int nr_channels;
+
+static bool dmatest_match_channel(struct dma_chan *chan)
+{
+	if (test_channel[0] == '\0')
+		return true;
+	return strcmp(chan->dev.bus_id, test_channel) == 0;
+}
+
+static bool dmatest_match_device(struct dma_device *device)
+{
+	if (test_device[0] == '\0')
+		return true;
+	return strcmp(device->dev->bus_id, test_device) == 0;
+}
+
+static unsigned long dmatest_random(void)
+{
+	unsigned long buf;
+
+	get_random_bytes(&buf, sizeof(buf));
+	return buf;
+}
+
+static void dmatest_init_srcbuf(u8 *buf, unsigned int start, unsigned int len)
+{
+	unsigned int i;
+
+	for (i = 0; i < start; i++)
+		buf[i] = PATTERN_SRC | (~i & PATTERN_COUNT_MASK);
+	for ( ; i < start + len; i++)
+		buf[i] = PATTERN_SRC | PATTERN_COPY
+			| (~i & PATTERN_COUNT_MASK);;
+	for ( ; i < test_buf_size; i++)
+		buf[i] = PATTERN_SRC | (~i & PATTERN_COUNT_MASK);
+}
+
+static void dmatest_init_dstbuf(u8 *buf, unsigned int start, unsigned int len)
+{
+	unsigned int i;
+
+	for (i = 0; i < start; i++)
+		buf[i] = PATTERN_DST | (~i & PATTERN_COUNT_MASK);
+	for ( ; i < start + len; i++)
+		buf[i] = PATTERN_DST | PATTERN_OVERWRITE
+			| (~i & PATTERN_COUNT_MASK);
+	for ( ; i < test_buf_size; i++)
+		buf[i] = PATTERN_DST | (~i & PATTERN_COUNT_MASK);
+}
+
+static void dmatest_mismatch(u8 actual, u8 pattern, unsigned int index,
+		unsigned int counter, bool is_srcbuf)
+{
+	u8		diff = actual ^ pattern;
+	u8		expected = pattern | (~counter & PATTERN_COUNT_MASK);
+	const char	*thread_name = current->comm;
+
+	if (is_srcbuf)
+		pr_warning("%s: srcbuf[0x%x] overwritten!"
+				" Expected %02x, got %02x\n",
+				thread_name, index, expected, actual);
+	else if ((pattern & PATTERN_COPY)
+			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
+		pr_warning("%s: dstbuf[0x%x] not copied!"
+				" Expected %02x, got %02x\n",
+				thread_name, index, expected, actual);
+	else if (diff & PATTERN_SRC)
+		pr_warning("%s: dstbuf[0x%x] was copied!"
+				" Expected %02x, got %02x\n",
+				thread_name, index, expected, actual);
+	else
+		pr_warning("%s: dstbuf[0x%x] mismatch!"
+				" Expected %02x, got %02x\n",
+				thread_name, index, expected, actual);
+}
+
+static unsigned int dmatest_verify(u8 *buf, unsigned int start,
+		unsigned int end, unsigned int counter, u8 pattern,
+		bool is_srcbuf)
+{
+	unsigned int i;
+	unsigned int error_count = 0;
+	u8 actual;
+
+	for (i = start; i < end; i++) {
+		actual = buf[i];
+		if (actual != (pattern | (~counter & PATTERN_COUNT_MASK))) {
+			if (error_count < 32)
+				dmatest_mismatch(actual, pattern, i, counter,
+						is_srcbuf);
+			error_count++;
+		}
+		counter++;
+	}
+
+	if (error_count > 32)
+		pr_warning("%s: %u errors suppressed\n",
+			current->comm, error_count - 32);
+
+	return error_count;
+}
+
+/*
+ * This function repeatedly tests DMA transfers of various lengths and
+ * offsets until it is told to exit by kthread_stop(). There may be
+ * multiple threads running this function in parallel for a single
+ * channel, and there may be multiple channels being tested in
+ * parallel.
+ *
+ * Before each test, the source and destination buffer is initialized
+ * with a known pattern. This pattern is different depending on
+ * whether it's in an area which is supposed to be copied or
+ * overwritten, and different in the source and destination buffers.
+ * So if the DMA engine doesn't copy exactly what we tell it to copy,
+ * we'll notice.
+ */
+static int dmatest_func(void *data)
+{
+	struct dmatest_thread	*thread = data;
+	struct dma_chan		*chan;
+	const char		*thread_name;
+	unsigned int		src_off, dst_off, len;
+	unsigned int		error_count;
+	unsigned int		failed_tests = 0;
+	unsigned int		total_tests = 0;
+	dma_cookie_t		cookie;
+	enum dma_status		status;
+	int			ret;
+
+	thread_name = current->comm;
+
+	ret = -ENOMEM;
+	thread->srcbuf = kmalloc(test_buf_size, GFP_KERNEL);
+	if (!thread->srcbuf)
+		goto err_srcbuf;
+	thread->dstbuf = kmalloc(test_buf_size, GFP_KERNEL);
+	if (!thread->dstbuf)
+		goto err_dstbuf;
+
+	smp_rmb();
+	chan = thread->chan;
+	dma_chan_get(chan);
+
+	while (!kthread_should_stop()) {
+		total_tests++;
+
+		len = dmatest_random() % test_buf_size + 1;
+		src_off = dmatest_random() % (test_buf_size - len + 1);
+		dst_off = dmatest_random() % (test_buf_size - len + 1);
+
+		dmatest_init_srcbuf(thread->srcbuf, src_off, len);
+		dmatest_init_dstbuf(thread->dstbuf, dst_off, len);
+
+		cookie = dma_async_memcpy_buf_to_buf(chan,
+				thread->dstbuf + dst_off,
+				thread->srcbuf + src_off,
+				len);
+		if (dma_submit_error(cookie)) {
+			pr_warning("%s: #%u: submit error %d with src_off=0x%x "
+					"dst_off=0x%x len=0x%x\n",
+					thread_name, total_tests - 1, cookie,
+					src_off, dst_off, len);
+			msleep(100);
+			failed_tests++;
+			continue;
+		}
+		dma_async_memcpy_issue_pending(chan);
+
+		do {
+			msleep(1);
+			status = dma_async_memcpy_complete(
+					chan, cookie, NULL, NULL);
+		} while (status == DMA_IN_PROGRESS);
+
+		if (status == DMA_ERROR) {
+			pr_warning("%s: #%u: error during copy\n",
+					thread_name, total_tests - 1);
+			failed_tests++;
+			continue;
+		}
+
+		error_count = 0;
+
+		pr_debug("%s: verifying source buffer...\n", thread_name);
+		error_count += dmatest_verify(thread->srcbuf, 0, src_off,
+				0, PATTERN_SRC, true);
+		error_count += dmatest_verify(thread->srcbuf, src_off,
+				src_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, true);
+		error_count += dmatest_verify(thread->srcbuf, src_off + len,
+				test_buf_size, src_off + len,
+				PATTERN_SRC, true);
+
+		pr_debug("%s: verifying dest buffer...\n",
+				thread->task->comm);
+		error_count += dmatest_verify(thread->dstbuf, 0, dst_off,
+				0, PATTERN_DST, false);
+		error_count += dmatest_verify(thread->dstbuf, dst_off,
+				dst_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, false);
+		error_count += dmatest_verify(thread->dstbuf, dst_off + len,
+				test_buf_size, dst_off + len,
+				PATTERN_DST, false);
+
+		if (error_count) {
+			pr_warning("%s: #%u: %u errors with "
+				"src_off=0x%x dst_off=0x%x len=0x%x\n",
+				thread_name, total_tests - 1, error_count,
+				src_off, dst_off, len);
+			failed_tests++;
+		} else {
+			pr_debug("%s: #%u: No errors with "
+				"src_off=0x%x dst_off=0x%x len=0x%x\n",
+				thread_name, total_tests - 1,
+				src_off, dst_off, len);
+		}
+	}
+
+	ret = 0;
+	dma_chan_put(chan);
+	kfree(thread->dstbuf);
+err_dstbuf:
+	kfree(thread->srcbuf);
+err_srcbuf:
+	pr_notice("%s: terminating after %u tests, %u failures (status %d)\n",
+			thread_name, total_tests, failed_tests, ret);
+	return ret;
+}
+
+static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
+{
+	struct dmatest_thread	*thread;
+	struct dmatest_thread	*_thread;
+	int			ret;
+
+	list_for_each_entry_safe(thread, _thread, &dtc->threads, node) {
+		ret = kthread_stop(thread->task);
+		pr_debug("dmatest: thread %s exited with status %d\n",
+				thread->task->comm, ret);
+		list_del(&thread->node);
+		kfree(thread);
+	}
+	kfree(dtc);
+}
+
+static enum dma_state_client dmatest_add_channel(struct dma_chan *chan)
+{
+	struct dmatest_chan	*dtc;
+	struct dmatest_thread	*thread;
+	unsigned int		i;
+
+	dtc = kmalloc(sizeof(struct dmatest_chan), GFP_ATOMIC);
+	if (!dtc) {
+		pr_warning("dmatest: No memory for %s\n", chan->dev.bus_id);
+		return DMA_NAK;
+	}
+
+	dtc->chan = chan;
+	INIT_LIST_HEAD(&dtc->threads);
+
+	for (i = 0; i < threads_per_chan; i++) {
+		thread = kzalloc(sizeof(struct dmatest_thread), GFP_KERNEL);
+		if (!thread) {
+			pr_warning("dmatest: No memory for %s-test%u\n",
+					chan->dev.bus_id, i);
+			break;
+		}
+		thread->chan = dtc->chan;
+		smp_wmb();
+		thread->task = kthread_run(dmatest_func, thread, "%s-test%u",
+				chan->dev.bus_id, i);
+		if (IS_ERR(thread->task)) {
+			pr_warning("dmatest: Failed to run thread %s-test%u\n",
+					chan->dev.bus_id, i);
+			kfree(thread);
+			break;
+		}
+
+		/* srcbuf and dstbuf are allocated by the thread itself */
+
+		list_add_tail(&thread->node, &dtc->threads);
+	}
+
+	pr_info("dmatest: Started %u threads using %s\n", i, chan->dev.bus_id);
+
+	list_add_tail(&dtc->node, &dmatest_channels);
+	nr_channels++;
+
+	return DMA_ACK;
+}
+
+static enum dma_state_client dmatest_remove_channel(struct dma_chan *chan)
+{
+	struct dmatest_chan	*dtc, *_dtc;
+
+	list_for_each_entry_safe(dtc, _dtc, &dmatest_channels, node) {
+		if (dtc->chan == chan) {
+			list_del(&dtc->node);
+			dmatest_cleanup_channel(dtc);
+			pr_debug("dmatest: lost channel %s\n",
+					chan->dev.bus_id);
+			return DMA_ACK;
+		}
+	}
+
+	return DMA_DUP;
+}
+
+/*
+ * Start testing threads as new channels are assigned to us, and kill
+ * them when the channels go away.
+ *
+ * When we unregister the client, all channels are removed so this
+ * will also take care of cleaning things up when the module is
+ * unloaded.
+ */
+static enum dma_state_client
+dmatest_event(struct dma_client *client, struct dma_chan *chan,
+		enum dma_state state)
+{
+	enum dma_state_client	ack = DMA_NAK;
+
+	switch (state) {
+	case DMA_RESOURCE_AVAILABLE:
+		if (!dmatest_match_channel(chan)
+				|| !dmatest_match_device(chan->device))
+			ack = DMA_DUP;
+		else if (max_channels && nr_channels >= max_channels)
+			ack = DMA_NAK;
+		else
+			ack = dmatest_add_channel(chan);
+		break;
+
+	case DMA_RESOURCE_REMOVED:
+		ack = dmatest_remove_channel(chan);
+		break;
+
+	default:
+		pr_info("dmatest: Unhandled event %u (%s)\n",
+				state, chan->dev.bus_id);
+		break;
+	}
+
+	return ack;
+}
+
+static struct dma_client dmatest_client = {
+	.event_callback	= dmatest_event,
+};
+
+static int __init dmatest_init(void)
+{
+	dma_cap_set(DMA_MEMCPY, dmatest_client.cap_mask);
+	dma_async_client_register(&dmatest_client);
+	dma_async_client_chan_request(&dmatest_client);
+
+	return 0;
+}
+module_init(dmatest_init);
+
+static void __exit dmatest_exit(void)
+{
+	dma_async_client_unregister(&dmatest_client);
+}
+module_exit(dmatest_exit);
+
+MODULE_AUTHOR("Haavard Skinnemoen <hskinnemoen@atmel.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/dma/dw_dmac.c b/drivers/dma/dw_dmac.c
new file mode 100644
index 000000000000..94df91771243
--- /dev/null
+++ b/drivers/dma/dw_dmac.c
@@ -0,0 +1,1122 @@
+/*
+ * Driver for the Synopsys DesignWare DMA Controller (aka DMACA on
+ * AVR32 systems.)
+ *
+ * Copyright (C) 2007-2008 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include "dw_dmac_regs.h"
+
+/*
+ * This supports the Synopsys "DesignWare AHB Central DMA Controller",
+ * (DW_ahb_dmac) which is used with various AMBA 2.0 systems (not all
+ * of which use ARM any more).  See the "Databook" from Synopsys for
+ * information beyond what licensees probably provide.
+ *
+ * The driver has currently been tested only with the Atmel AT32AP7000,
+ * which does not support descriptor writeback.
+ */
+
+/* NOTE:  DMS+SMS is system-specific. We should get this information
+ * from the platform code somehow.
+ */
+#define DWC_DEFAULT_CTLLO	(DWC_CTLL_DST_MSIZE(0)		\
+				| DWC_CTLL_SRC_MSIZE(0)		\
+				| DWC_CTLL_DMS(0)		\
+				| DWC_CTLL_SMS(1)		\
+				| DWC_CTLL_LLP_D_EN		\
+				| DWC_CTLL_LLP_S_EN)
+
+/*
+ * This is configuration-dependent and usually a funny size like 4095.
+ * Let's round it down to the nearest power of two.
+ *
+ * Note that this is a transfer count, i.e. if we transfer 32-bit
+ * words, we can do 8192 bytes per descriptor.
+ *
+ * This parameter is also system-specific.
+ */
+#define DWC_MAX_COUNT	2048U
+
+/*
+ * Number of descriptors to allocate for each channel. This should be
+ * made configurable somehow; preferably, the clients (at least the
+ * ones using slave transfers) should be able to give us a hint.
+ */
+#define NR_DESCS_PER_CHANNEL	64
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Because we're not relying on writeback from the controller (it may not
+ * even be configured into the core!) we don't need to use dma_pool.  These
+ * descriptors -- and associated data -- are cacheable.  We do need to make
+ * sure their dcache entries are written back before handing them off to
+ * the controller, though.
+ */
+
+static struct dw_desc *dwc_first_active(struct dw_dma_chan *dwc)
+{
+	return list_entry(dwc->active_list.next, struct dw_desc, desc_node);
+}
+
+static struct dw_desc *dwc_first_queued(struct dw_dma_chan *dwc)
+{
+	return list_entry(dwc->queue.next, struct dw_desc, desc_node);
+}
+
+static struct dw_desc *dwc_desc_get(struct dw_dma_chan *dwc)
+{
+	struct dw_desc *desc, *_desc;
+	struct dw_desc *ret = NULL;
+	unsigned int i = 0;
+
+	spin_lock_bh(&dwc->lock);
+	list_for_each_entry_safe(desc, _desc, &dwc->free_list, desc_node) {
+		if (async_tx_test_ack(&desc->txd)) {
+			list_del(&desc->desc_node);
+			ret = desc;
+			break;
+		}
+		dev_dbg(&dwc->chan.dev, "desc %p not ACKed\n", desc);
+		i++;
+	}
+	spin_unlock_bh(&dwc->lock);
+
+	dev_vdbg(&dwc->chan.dev, "scanned %u descriptors on freelist\n", i);
+
+	return ret;
+}
+
+static void dwc_sync_desc_for_cpu(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	struct dw_desc	*child;
+
+	list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+		dma_sync_single_for_cpu(dwc->chan.dev.parent,
+				child->txd.phys, sizeof(child->lli),
+				DMA_TO_DEVICE);
+	dma_sync_single_for_cpu(dwc->chan.dev.parent,
+			desc->txd.phys, sizeof(desc->lli),
+			DMA_TO_DEVICE);
+}
+
+/*
+ * Move a descriptor, including any children, to the free list.
+ * `desc' must not be on any lists.
+ */
+static void dwc_desc_put(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	if (desc) {
+		struct dw_desc *child;
+
+		dwc_sync_desc_for_cpu(dwc, desc);
+
+		spin_lock_bh(&dwc->lock);
+		list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+			dev_vdbg(&dwc->chan.dev,
+					"moving child desc %p to freelist\n",
+					child);
+		list_splice_init(&desc->txd.tx_list, &dwc->free_list);
+		dev_vdbg(&dwc->chan.dev, "moving desc %p to freelist\n", desc);
+		list_add(&desc->desc_node, &dwc->free_list);
+		spin_unlock_bh(&dwc->lock);
+	}
+}
+
+/* Called with dwc->lock held and bh disabled */
+static dma_cookie_t
+dwc_assign_cookie(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	dma_cookie_t cookie = dwc->chan.cookie;
+
+	if (++cookie < 0)
+		cookie = 1;
+
+	dwc->chan.cookie = cookie;
+	desc->txd.cookie = cookie;
+
+	return cookie;
+}
+
+/*----------------------------------------------------------------------*/
+
+/* Called with dwc->lock held and bh disabled */
+static void dwc_dostart(struct dw_dma_chan *dwc, struct dw_desc *first)
+{
+	struct dw_dma	*dw = to_dw_dma(dwc->chan.device);
+
+	/* ASSERT:  channel is idle */
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_err(&dwc->chan.dev,
+			"BUG: Attempted to start non-idle channel\n");
+		dev_err(&dwc->chan.dev,
+			"  SAR: 0x%x DAR: 0x%x LLP: 0x%x CTL: 0x%x:%08x\n",
+			channel_readl(dwc, SAR),
+			channel_readl(dwc, DAR),
+			channel_readl(dwc, LLP),
+			channel_readl(dwc, CTL_HI),
+			channel_readl(dwc, CTL_LO));
+
+		/* The tasklet will hopefully advance the queue... */
+		return;
+	}
+
+	channel_writel(dwc, LLP, first->txd.phys);
+	channel_writel(dwc, CTL_LO,
+			DWC_CTLL_LLP_D_EN | DWC_CTLL_LLP_S_EN);
+	channel_writel(dwc, CTL_HI, 0);
+	channel_set_bit(dw, CH_EN, dwc->mask);
+}
+
+/*----------------------------------------------------------------------*/
+
+static void
+dwc_descriptor_complete(struct dw_dma_chan *dwc, struct dw_desc *desc)
+{
+	dma_async_tx_callback		callback;
+	void				*param;
+	struct dma_async_tx_descriptor	*txd = &desc->txd;
+
+	dev_vdbg(&dwc->chan.dev, "descriptor %u complete\n", txd->cookie);
+
+	dwc->completed = txd->cookie;
+	callback = txd->callback;
+	param = txd->callback_param;
+
+	dwc_sync_desc_for_cpu(dwc, desc);
+	list_splice_init(&txd->tx_list, &dwc->free_list);
+	list_move(&desc->desc_node, &dwc->free_list);
+
+	/*
+	 * We use dma_unmap_page() regardless of how the buffers were
+	 * mapped before they were submitted...
+	 */
+	if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP))
+		dma_unmap_page(dwc->chan.dev.parent, desc->lli.dar, desc->len,
+				DMA_FROM_DEVICE);
+	if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP))
+		dma_unmap_page(dwc->chan.dev.parent, desc->lli.sar, desc->len,
+				DMA_TO_DEVICE);
+
+	/*
+	 * The API requires that no submissions are done from a
+	 * callback, so we don't need to drop the lock here
+	 */
+	if (callback)
+		callback(param);
+}
+
+static void dwc_complete_all(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	struct dw_desc *desc, *_desc;
+	LIST_HEAD(list);
+
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_err(&dwc->chan.dev,
+			"BUG: XFER bit set, but channel not idle!\n");
+
+		/* Try to continue after resetting the channel... */
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+		while (dma_readl(dw, CH_EN) & dwc->mask)
+			cpu_relax();
+	}
+
+	/*
+	 * Submit queued descriptors ASAP, i.e. before we go through
+	 * the completed ones.
+	 */
+	if (!list_empty(&dwc->queue))
+		dwc_dostart(dwc, dwc_first_queued(dwc));
+	list_splice_init(&dwc->active_list, &list);
+	list_splice_init(&dwc->queue, &dwc->active_list);
+
+	list_for_each_entry_safe(desc, _desc, &list, desc_node)
+		dwc_descriptor_complete(dwc, desc);
+}
+
+static void dwc_scan_descriptors(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	dma_addr_t llp;
+	struct dw_desc *desc, *_desc;
+	struct dw_desc *child;
+	u32 status_xfer;
+
+	/*
+	 * Clear block interrupt flag before scanning so that we don't
+	 * miss any, and read LLP before RAW_XFER to ensure it is
+	 * valid if we decide to scan the list.
+	 */
+	dma_writel(dw, CLEAR.BLOCK, dwc->mask);
+	llp = channel_readl(dwc, LLP);
+	status_xfer = dma_readl(dw, RAW.XFER);
+
+	if (status_xfer & dwc->mask) {
+		/* Everything we've submitted is done */
+		dma_writel(dw, CLEAR.XFER, dwc->mask);
+		dwc_complete_all(dw, dwc);
+		return;
+	}
+
+	dev_vdbg(&dwc->chan.dev, "scan_descriptors: llp=0x%x\n", llp);
+
+	list_for_each_entry_safe(desc, _desc, &dwc->active_list, desc_node) {
+		if (desc->lli.llp == llp)
+			/* This one is currently in progress */
+			return;
+
+		list_for_each_entry(child, &desc->txd.tx_list, desc_node)
+			if (child->lli.llp == llp)
+				/* Currently in progress */
+				return;
+
+		/*
+		 * No descriptors so far seem to be in progress, i.e.
+		 * this one must be done.
+		 */
+		dwc_descriptor_complete(dwc, desc);
+	}
+
+	dev_err(&dwc->chan.dev,
+		"BUG: All descriptors done, but channel not idle!\n");
+
+	/* Try to continue after resetting the channel... */
+	channel_clear_bit(dw, CH_EN, dwc->mask);
+	while (dma_readl(dw, CH_EN) & dwc->mask)
+		cpu_relax();
+
+	if (!list_empty(&dwc->queue)) {
+		dwc_dostart(dwc, dwc_first_queued(dwc));
+		list_splice_init(&dwc->queue, &dwc->active_list);
+	}
+}
+
+static void dwc_dump_lli(struct dw_dma_chan *dwc, struct dw_lli *lli)
+{
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"  desc: s0x%x d0x%x l0x%x c0x%x:%x\n",
+			lli->sar, lli->dar, lli->llp,
+			lli->ctlhi, lli->ctllo);
+}
+
+static void dwc_handle_error(struct dw_dma *dw, struct dw_dma_chan *dwc)
+{
+	struct dw_desc *bad_desc;
+	struct dw_desc *child;
+
+	dwc_scan_descriptors(dw, dwc);
+
+	/*
+	 * The descriptor currently at the head of the active list is
+	 * borked. Since we don't have any way to report errors, we'll
+	 * just have to scream loudly and try to carry on.
+	 */
+	bad_desc = dwc_first_active(dwc);
+	list_del_init(&bad_desc->desc_node);
+	list_splice_init(&dwc->queue, dwc->active_list.prev);
+
+	/* Clear the error flag and try to restart the controller */
+	dma_writel(dw, CLEAR.ERROR, dwc->mask);
+	if (!list_empty(&dwc->active_list))
+		dwc_dostart(dwc, dwc_first_active(dwc));
+
+	/*
+	 * KERN_CRITICAL may seem harsh, but since this only happens
+	 * when someone submits a bad physical address in a
+	 * descriptor, we should consider ourselves lucky that the
+	 * controller flagged an error instead of scribbling over
+	 * random memory locations.
+	 */
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"Bad descriptor submitted for DMA!\n");
+	dev_printk(KERN_CRIT, &dwc->chan.dev,
+			"  cookie: %d\n", bad_desc->txd.cookie);
+	dwc_dump_lli(dwc, &bad_desc->lli);
+	list_for_each_entry(child, &bad_desc->txd.tx_list, desc_node)
+		dwc_dump_lli(dwc, &child->lli);
+
+	/* Pretend the descriptor completed successfully */
+	dwc_descriptor_complete(dwc, bad_desc);
+}
+
+static void dw_dma_tasklet(unsigned long data)
+{
+	struct dw_dma *dw = (struct dw_dma *)data;
+	struct dw_dma_chan *dwc;
+	u32 status_block;
+	u32 status_xfer;
+	u32 status_err;
+	int i;
+
+	status_block = dma_readl(dw, RAW.BLOCK);
+	status_xfer = dma_readl(dw, RAW.BLOCK);
+	status_err = dma_readl(dw, RAW.ERROR);
+
+	dev_vdbg(dw->dma.dev, "tasklet: status_block=%x status_err=%x\n",
+			status_block, status_err);
+
+	for (i = 0; i < dw->dma.chancnt; i++) {
+		dwc = &dw->chan[i];
+		spin_lock(&dwc->lock);
+		if (status_err & (1 << i))
+			dwc_handle_error(dw, dwc);
+		else if ((status_block | status_xfer) & (1 << i))
+			dwc_scan_descriptors(dw, dwc);
+		spin_unlock(&dwc->lock);
+	}
+
+	/*
+	 * Re-enable interrupts. Block Complete interrupts are only
+	 * enabled if the INT_EN bit in the descriptor is set. This
+	 * will trigger a scan before the whole list is done.
+	 */
+	channel_set_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_set_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_set_bit(dw, MASK.ERROR, dw->all_chan_mask);
+}
+
+static irqreturn_t dw_dma_interrupt(int irq, void *dev_id)
+{
+	struct dw_dma *dw = dev_id;
+	u32 status;
+
+	dev_vdbg(dw->dma.dev, "interrupt: status=0x%x\n",
+			dma_readl(dw, STATUS_INT));
+
+	/*
+	 * Just disable the interrupts. We'll turn them back on in the
+	 * softirq handler.
+	 */
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	status = dma_readl(dw, STATUS_INT);
+	if (status) {
+		dev_err(dw->dma.dev,
+			"BUG: Unexpected interrupts pending: 0x%x\n",
+			status);
+
+		/* Try to recover */
+		channel_clear_bit(dw, MASK.XFER, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.BLOCK, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.SRC_TRAN, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.DST_TRAN, (1 << 8) - 1);
+		channel_clear_bit(dw, MASK.ERROR, (1 << 8) - 1);
+	}
+
+	tasklet_schedule(&dw->tasklet);
+
+	return IRQ_HANDLED;
+}
+
+/*----------------------------------------------------------------------*/
+
+static dma_cookie_t dwc_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct dw_desc		*desc = txd_to_dw_desc(tx);
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(tx->chan);
+	dma_cookie_t		cookie;
+
+	spin_lock_bh(&dwc->lock);
+	cookie = dwc_assign_cookie(dwc, desc);
+
+	/*
+	 * REVISIT: We should attempt to chain as many descriptors as
+	 * possible, perhaps even appending to those already submitted
+	 * for DMA. But this is hard to do in a race-free manner.
+	 */
+	if (list_empty(&dwc->active_list)) {
+		dev_vdbg(&tx->chan->dev, "tx_submit: started %u\n",
+				desc->txd.cookie);
+		dwc_dostart(dwc, desc);
+		list_add_tail(&desc->desc_node, &dwc->active_list);
+	} else {
+		dev_vdbg(&tx->chan->dev, "tx_submit: queued %u\n",
+				desc->txd.cookie);
+
+		list_add_tail(&desc->desc_node, &dwc->queue);
+	}
+
+	spin_unlock_bh(&dwc->lock);
+
+	return cookie;
+}
+
+static struct dma_async_tx_descriptor *
+dwc_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+		size_t len, unsigned long flags)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_desc		*desc;
+	struct dw_desc		*first;
+	struct dw_desc		*prev;
+	size_t			xfer_count;
+	size_t			offset;
+	unsigned int		src_width;
+	unsigned int		dst_width;
+	u32			ctllo;
+
+	dev_vdbg(&chan->dev, "prep_dma_memcpy d0x%x s0x%x l0x%zx f0x%lx\n",
+			dest, src, len, flags);
+
+	if (unlikely(!len)) {
+		dev_dbg(&chan->dev, "prep_dma_memcpy: length is zero!\n");
+		return NULL;
+	}
+
+	/*
+	 * We can be a lot more clever here, but this should take care
+	 * of the most common optimization.
+	 */
+	if (!((src | dest  | len) & 3))
+		src_width = dst_width = 2;
+	else if (!((src | dest | len) & 1))
+		src_width = dst_width = 1;
+	else
+		src_width = dst_width = 0;
+
+	ctllo = DWC_DEFAULT_CTLLO
+			| DWC_CTLL_DST_WIDTH(dst_width)
+			| DWC_CTLL_SRC_WIDTH(src_width)
+			| DWC_CTLL_DST_INC
+			| DWC_CTLL_SRC_INC
+			| DWC_CTLL_FC_M2M;
+	prev = first = NULL;
+
+	for (offset = 0; offset < len; offset += xfer_count << src_width) {
+		xfer_count = min_t(size_t, (len - offset) >> src_width,
+				DWC_MAX_COUNT);
+
+		desc = dwc_desc_get(dwc);
+		if (!desc)
+			goto err_desc_get;
+
+		desc->lli.sar = src + offset;
+		desc->lli.dar = dest + offset;
+		desc->lli.ctllo = ctllo;
+		desc->lli.ctlhi = xfer_count;
+
+		if (!first) {
+			first = desc;
+		} else {
+			prev->lli.llp = desc->txd.phys;
+			dma_sync_single_for_device(chan->dev.parent,
+					prev->txd.phys, sizeof(prev->lli),
+					DMA_TO_DEVICE);
+			list_add_tail(&desc->desc_node,
+					&first->txd.tx_list);
+		}
+		prev = desc;
+	}
+
+
+	if (flags & DMA_PREP_INTERRUPT)
+		/* Trigger interrupt after last block */
+		prev->lli.ctllo |= DWC_CTLL_INT_EN;
+
+	prev->lli.llp = 0;
+	dma_sync_single_for_device(chan->dev.parent,
+			prev->txd.phys, sizeof(prev->lli),
+			DMA_TO_DEVICE);
+
+	first->txd.flags = flags;
+	first->len = len;
+
+	return &first->txd;
+
+err_desc_get:
+	dwc_desc_put(dwc, first);
+	return NULL;
+}
+
+static struct dma_async_tx_descriptor *
+dwc_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
+		unsigned int sg_len, enum dma_data_direction direction,
+		unsigned long flags)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma_slave	*dws = dwc->dws;
+	struct dw_desc		*prev;
+	struct dw_desc		*first;
+	u32			ctllo;
+	dma_addr_t		reg;
+	unsigned int		reg_width;
+	unsigned int		mem_width;
+	unsigned int		i;
+	struct scatterlist	*sg;
+	size_t			total_len = 0;
+
+	dev_vdbg(&chan->dev, "prep_dma_slave\n");
+
+	if (unlikely(!dws || !sg_len))
+		return NULL;
+
+	reg_width = dws->slave.reg_width;
+	prev = first = NULL;
+
+	sg_len = dma_map_sg(chan->dev.parent, sgl, sg_len, direction);
+
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		ctllo = (DWC_DEFAULT_CTLLO
+				| DWC_CTLL_DST_WIDTH(reg_width)
+				| DWC_CTLL_DST_FIX
+				| DWC_CTLL_SRC_INC
+				| DWC_CTLL_FC_M2P);
+		reg = dws->slave.tx_reg;
+		for_each_sg(sgl, sg, sg_len, i) {
+			struct dw_desc	*desc;
+			u32		len;
+			u32		mem;
+
+			desc = dwc_desc_get(dwc);
+			if (!desc) {
+				dev_err(&chan->dev,
+					"not enough descriptors available\n");
+				goto err_desc_get;
+			}
+
+			mem = sg_phys(sg);
+			len = sg_dma_len(sg);
+			mem_width = 2;
+			if (unlikely(mem & 3 || len & 3))
+				mem_width = 0;
+
+			desc->lli.sar = mem;
+			desc->lli.dar = reg;
+			desc->lli.ctllo = ctllo | DWC_CTLL_SRC_WIDTH(mem_width);
+			desc->lli.ctlhi = len >> mem_width;
+
+			if (!first) {
+				first = desc;
+			} else {
+				prev->lli.llp = desc->txd.phys;
+				dma_sync_single_for_device(chan->dev.parent,
+						prev->txd.phys,
+						sizeof(prev->lli),
+						DMA_TO_DEVICE);
+				list_add_tail(&desc->desc_node,
+						&first->txd.tx_list);
+			}
+			prev = desc;
+			total_len += len;
+		}
+		break;
+	case DMA_FROM_DEVICE:
+		ctllo = (DWC_DEFAULT_CTLLO
+				| DWC_CTLL_SRC_WIDTH(reg_width)
+				| DWC_CTLL_DST_INC
+				| DWC_CTLL_SRC_FIX
+				| DWC_CTLL_FC_P2M);
+
+		reg = dws->slave.rx_reg;
+		for_each_sg(sgl, sg, sg_len, i) {
+			struct dw_desc	*desc;
+			u32		len;
+			u32		mem;
+
+			desc = dwc_desc_get(dwc);
+			if (!desc) {
+				dev_err(&chan->dev,
+					"not enough descriptors available\n");
+				goto err_desc_get;
+			}
+
+			mem = sg_phys(sg);
+			len = sg_dma_len(sg);
+			mem_width = 2;
+			if (unlikely(mem & 3 || len & 3))
+				mem_width = 0;
+
+			desc->lli.sar = reg;
+			desc->lli.dar = mem;
+			desc->lli.ctllo = ctllo | DWC_CTLL_DST_WIDTH(mem_width);
+			desc->lli.ctlhi = len >> reg_width;
+
+			if (!first) {
+				first = desc;
+			} else {
+				prev->lli.llp = desc->txd.phys;
+				dma_sync_single_for_device(chan->dev.parent,
+						prev->txd.phys,
+						sizeof(prev->lli),
+						DMA_TO_DEVICE);
+				list_add_tail(&desc->desc_node,
+						&first->txd.tx_list);
+			}
+			prev = desc;
+			total_len += len;
+		}
+		break;
+	default:
+		return NULL;
+	}
+
+	if (flags & DMA_PREP_INTERRUPT)
+		/* Trigger interrupt after last block */
+		prev->lli.ctllo |= DWC_CTLL_INT_EN;
+
+	prev->lli.llp = 0;
+	dma_sync_single_for_device(chan->dev.parent,
+			prev->txd.phys, sizeof(prev->lli),
+			DMA_TO_DEVICE);
+
+	first->len = total_len;
+
+	return &first->txd;
+
+err_desc_get:
+	dwc_desc_put(dwc, first);
+	return NULL;
+}
+
+static void dwc_terminate_all(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc, *_desc;
+	LIST_HEAD(list);
+
+	/*
+	 * This is only called when something went wrong elsewhere, so
+	 * we don't really care about the data. Just disable the
+	 * channel. We still have to poll the channel enable bit due
+	 * to AHB/HSB limitations.
+	 */
+	spin_lock_bh(&dwc->lock);
+
+	channel_clear_bit(dw, CH_EN, dwc->mask);
+
+	while (dma_readl(dw, CH_EN) & dwc->mask)
+		cpu_relax();
+
+	/* active_list entries will end up before queued entries */
+	list_splice_init(&dwc->queue, &list);
+	list_splice_init(&dwc->active_list, &list);
+
+	spin_unlock_bh(&dwc->lock);
+
+	/* Flush all pending and queued descriptors */
+	list_for_each_entry_safe(desc, _desc, &list, desc_node)
+		dwc_descriptor_complete(dwc, desc);
+}
+
+static enum dma_status
+dwc_is_tx_complete(struct dma_chan *chan,
+		dma_cookie_t cookie,
+		dma_cookie_t *done, dma_cookie_t *used)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	dma_cookie_t		last_used;
+	dma_cookie_t		last_complete;
+	int			ret;
+
+	last_complete = dwc->completed;
+	last_used = chan->cookie;
+
+	ret = dma_async_is_complete(cookie, last_complete, last_used);
+	if (ret != DMA_SUCCESS) {
+		dwc_scan_descriptors(to_dw_dma(chan->device), dwc);
+
+		last_complete = dwc->completed;
+		last_used = chan->cookie;
+
+		ret = dma_async_is_complete(cookie, last_complete, last_used);
+	}
+
+	if (done)
+		*done = last_complete;
+	if (used)
+		*used = last_used;
+
+	return ret;
+}
+
+static void dwc_issue_pending(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+
+	spin_lock_bh(&dwc->lock);
+	if (!list_empty(&dwc->queue))
+		dwc_scan_descriptors(to_dw_dma(chan->device), dwc);
+	spin_unlock_bh(&dwc->lock);
+}
+
+static int dwc_alloc_chan_resources(struct dma_chan *chan,
+		struct dma_client *client)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc;
+	struct dma_slave	*slave;
+	struct dw_dma_slave	*dws;
+	int			i;
+	u32			cfghi;
+	u32			cfglo;
+
+	dev_vdbg(&chan->dev, "alloc_chan_resources\n");
+
+	/* Channels doing slave DMA can only handle one client. */
+	if (dwc->dws || client->slave) {
+		if (chan->client_count)
+			return -EBUSY;
+	}
+
+	/* ASSERT:  channel is idle */
+	if (dma_readl(dw, CH_EN) & dwc->mask) {
+		dev_dbg(&chan->dev, "DMA channel not idle?\n");
+		return -EIO;
+	}
+
+	dwc->completed = chan->cookie = 1;
+
+	cfghi = DWC_CFGH_FIFO_MODE;
+	cfglo = 0;
+
+	slave = client->slave;
+	if (slave) {
+		/*
+		 * We need controller-specific data to set up slave
+		 * transfers.
+		 */
+		BUG_ON(!slave->dma_dev || slave->dma_dev != dw->dma.dev);
+
+		dws = container_of(slave, struct dw_dma_slave, slave);
+
+		dwc->dws = dws;
+		cfghi = dws->cfg_hi;
+		cfglo = dws->cfg_lo;
+	} else {
+		dwc->dws = NULL;
+	}
+
+	channel_writel(dwc, CFG_LO, cfglo);
+	channel_writel(dwc, CFG_HI, cfghi);
+
+	/*
+	 * NOTE: some controllers may have additional features that we
+	 * need to initialize here, like "scatter-gather" (which
+	 * doesn't mean what you think it means), and status writeback.
+	 */
+
+	spin_lock_bh(&dwc->lock);
+	i = dwc->descs_allocated;
+	while (dwc->descs_allocated < NR_DESCS_PER_CHANNEL) {
+		spin_unlock_bh(&dwc->lock);
+
+		desc = kzalloc(sizeof(struct dw_desc), GFP_KERNEL);
+		if (!desc) {
+			dev_info(&chan->dev,
+				"only allocated %d descriptors\n", i);
+			spin_lock_bh(&dwc->lock);
+			break;
+		}
+
+		dma_async_tx_descriptor_init(&desc->txd, chan);
+		desc->txd.tx_submit = dwc_tx_submit;
+		desc->txd.flags = DMA_CTRL_ACK;
+		INIT_LIST_HEAD(&desc->txd.tx_list);
+		desc->txd.phys = dma_map_single(chan->dev.parent, &desc->lli,
+				sizeof(desc->lli), DMA_TO_DEVICE);
+		dwc_desc_put(dwc, desc);
+
+		spin_lock_bh(&dwc->lock);
+		i = ++dwc->descs_allocated;
+	}
+
+	/* Enable interrupts */
+	channel_set_bit(dw, MASK.XFER, dwc->mask);
+	channel_set_bit(dw, MASK.BLOCK, dwc->mask);
+	channel_set_bit(dw, MASK.ERROR, dwc->mask);
+
+	spin_unlock_bh(&dwc->lock);
+
+	dev_dbg(&chan->dev,
+		"alloc_chan_resources allocated %d descriptors\n", i);
+
+	return i;
+}
+
+static void dwc_free_chan_resources(struct dma_chan *chan)
+{
+	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
+	struct dw_desc		*desc, *_desc;
+	LIST_HEAD(list);
+
+	dev_dbg(&chan->dev, "free_chan_resources (descs allocated=%u)\n",
+			dwc->descs_allocated);
+
+	/* ASSERT:  channel is idle */
+	BUG_ON(!list_empty(&dwc->active_list));
+	BUG_ON(!list_empty(&dwc->queue));
+	BUG_ON(dma_readl(to_dw_dma(chan->device), CH_EN) & dwc->mask);
+
+	spin_lock_bh(&dwc->lock);
+	list_splice_init(&dwc->free_list, &list);
+	dwc->descs_allocated = 0;
+	dwc->dws = NULL;
+
+	/* Disable interrupts */
+	channel_clear_bit(dw, MASK.XFER, dwc->mask);
+	channel_clear_bit(dw, MASK.BLOCK, dwc->mask);
+	channel_clear_bit(dw, MASK.ERROR, dwc->mask);
+
+	spin_unlock_bh(&dwc->lock);
+
+	list_for_each_entry_safe(desc, _desc, &list, desc_node) {
+		dev_vdbg(&chan->dev, "  freeing descriptor %p\n", desc);
+		dma_unmap_single(chan->dev.parent, desc->txd.phys,
+				sizeof(desc->lli), DMA_TO_DEVICE);
+		kfree(desc);
+	}
+
+	dev_vdbg(&chan->dev, "free_chan_resources done\n");
+}
+
+/*----------------------------------------------------------------------*/
+
+static void dw_dma_off(struct dw_dma *dw)
+{
+	dma_writel(dw, CFG, 0);
+
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.SRC_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.DST_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	while (dma_readl(dw, CFG) & DW_CFG_DMA_EN)
+		cpu_relax();
+}
+
+static int __init dw_probe(struct platform_device *pdev)
+{
+	struct dw_dma_platform_data *pdata;
+	struct resource		*io;
+	struct dw_dma		*dw;
+	size_t			size;
+	int			irq;
+	int			err;
+	int			i;
+
+	pdata = pdev->dev.platform_data;
+	if (!pdata || pdata->nr_channels > DW_DMA_MAX_NR_CHANNELS)
+		return -EINVAL;
+
+	io = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!io)
+		return -EINVAL;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return irq;
+
+	size = sizeof(struct dw_dma);
+	size += pdata->nr_channels * sizeof(struct dw_dma_chan);
+	dw = kzalloc(size, GFP_KERNEL);
+	if (!dw)
+		return -ENOMEM;
+
+	if (!request_mem_region(io->start, DW_REGLEN, pdev->dev.driver->name)) {
+		err = -EBUSY;
+		goto err_kfree;
+	}
+
+	memset(dw, 0, sizeof *dw);
+
+	dw->regs = ioremap(io->start, DW_REGLEN);
+	if (!dw->regs) {
+		err = -ENOMEM;
+		goto err_release_r;
+	}
+
+	dw->clk = clk_get(&pdev->dev, "hclk");
+	if (IS_ERR(dw->clk)) {
+		err = PTR_ERR(dw->clk);
+		goto err_clk;
+	}
+	clk_enable(dw->clk);
+
+	/* force dma off, just in case */
+	dw_dma_off(dw);
+
+	err = request_irq(irq, dw_dma_interrupt, 0, "dw_dmac", dw);
+	if (err)
+		goto err_irq;
+
+	platform_set_drvdata(pdev, dw);
+
+	tasklet_init(&dw->tasklet, dw_dma_tasklet, (unsigned long)dw);
+
+	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++) {
+		struct dw_dma_chan	*dwc = &dw->chan[i];
+
+		dwc->chan.device = &dw->dma;
+		dwc->chan.cookie = dwc->completed = 1;
+		dwc->chan.chan_id = i;
+		list_add_tail(&dwc->chan.device_node, &dw->dma.channels);
+
+		dwc->ch_regs = &__dw_regs(dw)->CHAN[i];
+		spin_lock_init(&dwc->lock);
+		dwc->mask = 1 << i;
+
+		INIT_LIST_HEAD(&dwc->active_list);
+		INIT_LIST_HEAD(&dwc->queue);
+		INIT_LIST_HEAD(&dwc->free_list);
+
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+	}
+
+	/* Clear/disable all interrupts on all channels. */
+	dma_writel(dw, CLEAR.XFER, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.BLOCK, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.SRC_TRAN, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.DST_TRAN, dw->all_chan_mask);
+	dma_writel(dw, CLEAR.ERROR, dw->all_chan_mask);
+
+	channel_clear_bit(dw, MASK.XFER, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.BLOCK, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.SRC_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.DST_TRAN, dw->all_chan_mask);
+	channel_clear_bit(dw, MASK.ERROR, dw->all_chan_mask);
+
+	dma_cap_set(DMA_MEMCPY, dw->dma.cap_mask);
+	dma_cap_set(DMA_SLAVE, dw->dma.cap_mask);
+	dw->dma.dev = &pdev->dev;
+	dw->dma.device_alloc_chan_resources = dwc_alloc_chan_resources;
+	dw->dma.device_free_chan_resources = dwc_free_chan_resources;
+
+	dw->dma.device_prep_dma_memcpy = dwc_prep_dma_memcpy;
+
+	dw->dma.device_prep_slave_sg = dwc_prep_slave_sg;
+	dw->dma.device_terminate_all = dwc_terminate_all;
+
+	dw->dma.device_is_tx_complete = dwc_is_tx_complete;
+	dw->dma.device_issue_pending = dwc_issue_pending;
+
+	dma_writel(dw, CFG, DW_CFG_DMA_EN);
+
+	printk(KERN_INFO "%s: DesignWare DMA Controller, %d channels\n",
+			pdev->dev.bus_id, dw->dma.chancnt);
+
+	dma_async_device_register(&dw->dma);
+
+	return 0;
+
+err_irq:
+	clk_disable(dw->clk);
+	clk_put(dw->clk);
+err_clk:
+	iounmap(dw->regs);
+	dw->regs = NULL;
+err_release_r:
+	release_resource(io);
+err_kfree:
+	kfree(dw);
+	return err;
+}
+
+static int __exit dw_remove(struct platform_device *pdev)
+{
+	struct dw_dma		*dw = platform_get_drvdata(pdev);
+	struct dw_dma_chan	*dwc, *_dwc;
+	struct resource		*io;
+
+	dw_dma_off(dw);
+	dma_async_device_unregister(&dw->dma);
+
+	free_irq(platform_get_irq(pdev, 0), dw);
+	tasklet_kill(&dw->tasklet);
+
+	list_for_each_entry_safe(dwc, _dwc, &dw->dma.channels,
+			chan.device_node) {
+		list_del(&dwc->chan.device_node);
+		channel_clear_bit(dw, CH_EN, dwc->mask);
+	}
+
+	clk_disable(dw->clk);
+	clk_put(dw->clk);
+
+	iounmap(dw->regs);
+	dw->regs = NULL;
+
+	io = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(io->start, DW_REGLEN);
+
+	kfree(dw);
+
+	return 0;
+}
+
+static void dw_shutdown(struct platform_device *pdev)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	dw_dma_off(platform_get_drvdata(pdev));
+	clk_disable(dw->clk);
+}
+
+static int dw_suspend_late(struct platform_device *pdev, pm_message_t mesg)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	dw_dma_off(platform_get_drvdata(pdev));
+	clk_disable(dw->clk);
+	return 0;
+}
+
+static int dw_resume_early(struct platform_device *pdev)
+{
+	struct dw_dma	*dw = platform_get_drvdata(pdev);
+
+	clk_enable(dw->clk);
+	dma_writel(dw, CFG, DW_CFG_DMA_EN);
+	return 0;
+
+}
+
+static struct platform_driver dw_driver = {
+	.remove		= __exit_p(dw_remove),
+	.shutdown	= dw_shutdown,
+	.suspend_late	= dw_suspend_late,
+	.resume_early	= dw_resume_early,
+	.driver = {
+		.name	= "dw_dmac",
+	},
+};
+
+static int __init dw_init(void)
+{
+	return platform_driver_probe(&dw_driver, dw_probe);
+}
+module_init(dw_init);
+
+static void __exit dw_exit(void)
+{
+	platform_driver_unregister(&dw_driver);
+}
+module_exit(dw_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("Synopsys DesignWare DMA Controller driver");
+MODULE_AUTHOR("Haavard Skinnemoen <haavard.skinnemoen@atmel.com>");
diff --git a/drivers/dma/dw_dmac_regs.h b/drivers/dma/dw_dmac_regs.h
new file mode 100644
index 000000000000..00fdd187bb0c
--- /dev/null
+++ b/drivers/dma/dw_dmac_regs.h
@@ -0,0 +1,225 @@
+/*
+ * Driver for the Synopsys DesignWare AHB DMA Controller
+ *
+ * Copyright (C) 2005-2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/dw_dmac.h>
+
+#define DW_DMA_MAX_NR_CHANNELS	8
+
+/*
+ * Redefine this macro to handle differences between 32- and 64-bit
+ * addressing, big vs. little endian, etc.
+ */
+#define DW_REG(name)		u32 name; u32 __pad_##name
+
+/* Hardware register definitions. */
+struct dw_dma_chan_regs {
+	DW_REG(SAR);		/* Source Address Register */
+	DW_REG(DAR);		/* Destination Address Register */
+	DW_REG(LLP);		/* Linked List Pointer */
+	u32	CTL_LO;		/* Control Register Low */
+	u32	CTL_HI;		/* Control Register High */
+	DW_REG(SSTAT);
+	DW_REG(DSTAT);
+	DW_REG(SSTATAR);
+	DW_REG(DSTATAR);
+	u32	CFG_LO;		/* Configuration Register Low */
+	u32	CFG_HI;		/* Configuration Register High */
+	DW_REG(SGR);
+	DW_REG(DSR);
+};
+
+struct dw_dma_irq_regs {
+	DW_REG(XFER);
+	DW_REG(BLOCK);
+	DW_REG(SRC_TRAN);
+	DW_REG(DST_TRAN);
+	DW_REG(ERROR);
+};
+
+struct dw_dma_regs {
+	/* per-channel registers */
+	struct dw_dma_chan_regs	CHAN[DW_DMA_MAX_NR_CHANNELS];
+
+	/* irq handling */
+	struct dw_dma_irq_regs	RAW;		/* r */
+	struct dw_dma_irq_regs	STATUS;		/* r (raw & mask) */
+	struct dw_dma_irq_regs	MASK;		/* rw (set = irq enabled) */
+	struct dw_dma_irq_regs	CLEAR;		/* w (ack, affects "raw") */
+
+	DW_REG(STATUS_INT);			/* r */
+
+	/* software handshaking */
+	DW_REG(REQ_SRC);
+	DW_REG(REQ_DST);
+	DW_REG(SGL_REQ_SRC);
+	DW_REG(SGL_REQ_DST);
+	DW_REG(LAST_SRC);
+	DW_REG(LAST_DST);
+
+	/* miscellaneous */
+	DW_REG(CFG);
+	DW_REG(CH_EN);
+	DW_REG(ID);
+	DW_REG(TEST);
+
+	/* optional encoded params, 0x3c8..0x3 */
+};
+
+/* Bitfields in CTL_LO */
+#define DWC_CTLL_INT_EN		(1 << 0)	/* irqs enabled? */
+#define DWC_CTLL_DST_WIDTH(n)	((n)<<1)	/* bytes per element */
+#define DWC_CTLL_SRC_WIDTH(n)	((n)<<4)
+#define DWC_CTLL_DST_INC	(0<<7)		/* DAR update/not */
+#define DWC_CTLL_DST_DEC	(1<<7)
+#define DWC_CTLL_DST_FIX	(2<<7)
+#define DWC_CTLL_SRC_INC	(0<<7)		/* SAR update/not */
+#define DWC_CTLL_SRC_DEC	(1<<9)
+#define DWC_CTLL_SRC_FIX	(2<<9)
+#define DWC_CTLL_DST_MSIZE(n)	((n)<<11)	/* burst, #elements */
+#define DWC_CTLL_SRC_MSIZE(n)	((n)<<14)
+#define DWC_CTLL_S_GATH_EN	(1 << 17)	/* src gather, !FIX */
+#define DWC_CTLL_D_SCAT_EN	(1 << 18)	/* dst scatter, !FIX */
+#define DWC_CTLL_FC_M2M		(0 << 20)	/* mem-to-mem */
+#define DWC_CTLL_FC_M2P		(1 << 20)	/* mem-to-periph */
+#define DWC_CTLL_FC_P2M		(2 << 20)	/* periph-to-mem */
+#define DWC_CTLL_FC_P2P		(3 << 20)	/* periph-to-periph */
+/* plus 4 transfer types for peripheral-as-flow-controller */
+#define DWC_CTLL_DMS(n)		((n)<<23)	/* dst master select */
+#define DWC_CTLL_SMS(n)		((n)<<25)	/* src master select */
+#define DWC_CTLL_LLP_D_EN	(1 << 27)	/* dest block chain */
+#define DWC_CTLL_LLP_S_EN	(1 << 28)	/* src block chain */
+
+/* Bitfields in CTL_HI */
+#define DWC_CTLH_DONE		0x00001000
+#define DWC_CTLH_BLOCK_TS_MASK	0x00000fff
+
+/* Bitfields in CFG_LO. Platform-configurable bits are in <linux/dw_dmac.h> */
+#define DWC_CFGL_CH_SUSP	(1 << 8)	/* pause xfer */
+#define DWC_CFGL_FIFO_EMPTY	(1 << 9)	/* pause xfer */
+#define DWC_CFGL_HS_DST		(1 << 10)	/* handshake w/dst */
+#define DWC_CFGL_HS_SRC		(1 << 11)	/* handshake w/src */
+#define DWC_CFGL_MAX_BURST(x)	((x) << 20)
+#define DWC_CFGL_RELOAD_SAR	(1 << 30)
+#define DWC_CFGL_RELOAD_DAR	(1 << 31)
+
+/* Bitfields in CFG_HI. Platform-configurable bits are in <linux/dw_dmac.h> */
+#define DWC_CFGH_DS_UPD_EN	(1 << 5)
+#define DWC_CFGH_SS_UPD_EN	(1 << 6)
+
+/* Bitfields in SGR */
+#define DWC_SGR_SGI(x)		((x) << 0)
+#define DWC_SGR_SGC(x)		((x) << 20)
+
+/* Bitfields in DSR */
+#define DWC_DSR_DSI(x)		((x) << 0)
+#define DWC_DSR_DSC(x)		((x) << 20)
+
+/* Bitfields in CFG */
+#define DW_CFG_DMA_EN		(1 << 0)
+
+#define DW_REGLEN		0x400
+
+struct dw_dma_chan {
+	struct dma_chan		chan;
+	void __iomem		*ch_regs;
+	u8			mask;
+
+	spinlock_t		lock;
+
+	/* these other elements are all protected by lock */
+	dma_cookie_t		completed;
+	struct list_head	active_list;
+	struct list_head	queue;
+	struct list_head	free_list;
+
+	struct dw_dma_slave	*dws;
+
+	unsigned int		descs_allocated;
+};
+
+static inline struct dw_dma_chan_regs __iomem *
+__dwc_regs(struct dw_dma_chan *dwc)
+{
+	return dwc->ch_regs;
+}
+
+#define channel_readl(dwc, name) \
+	__raw_readl(&(__dwc_regs(dwc)->name))
+#define channel_writel(dwc, name, val) \
+	__raw_writel((val), &(__dwc_regs(dwc)->name))
+
+static inline struct dw_dma_chan *to_dw_dma_chan(struct dma_chan *chan)
+{
+	return container_of(chan, struct dw_dma_chan, chan);
+}
+
+
+struct dw_dma {
+	struct dma_device	dma;
+	void __iomem		*regs;
+	struct tasklet_struct	tasklet;
+	struct clk		*clk;
+
+	u8			all_chan_mask;
+
+	struct dw_dma_chan	chan[0];
+};
+
+static inline struct dw_dma_regs __iomem *__dw_regs(struct dw_dma *dw)
+{
+	return dw->regs;
+}
+
+#define dma_readl(dw, name) \
+	__raw_readl(&(__dw_regs(dw)->name))
+#define dma_writel(dw, name, val) \
+	__raw_writel((val), &(__dw_regs(dw)->name))
+
+#define channel_set_bit(dw, reg, mask) \
+	dma_writel(dw, reg, ((mask) << 8) | (mask))
+#define channel_clear_bit(dw, reg, mask) \
+	dma_writel(dw, reg, ((mask) << 8) | 0)
+
+static inline struct dw_dma *to_dw_dma(struct dma_device *ddev)
+{
+	return container_of(ddev, struct dw_dma, dma);
+}
+
+/* LLI == Linked List Item; a.k.a. DMA block descriptor */
+struct dw_lli {
+	/* values that are not changed by hardware */
+	dma_addr_t	sar;
+	dma_addr_t	dar;
+	dma_addr_t	llp;		/* chain to next lli */
+	u32		ctllo;
+	/* values that may get written back: */
+	u32		ctlhi;
+	/* sstat and dstat can snapshot peripheral register state.
+	 * silicon config may discard either or both...
+	 */
+	u32		sstat;
+	u32		dstat;
+};
+
+struct dw_desc {
+	/* FIRST values the hardware uses */
+	struct dw_lli			lli;
+
+	/* THEN values for driver housekeeping */
+	struct list_head		desc_node;
+	struct dma_async_tx_descriptor	txd;
+	size_t				len;
+};
+
+static inline struct dw_desc *
+txd_to_dw_desc(struct dma_async_tx_descriptor *txd)
+{
+	return container_of(txd, struct dw_desc, txd);
+}
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 054eabffc185..c0059ca58340 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -366,7 +366,8 @@ static struct fsl_desc_sw *fsl_dma_alloc_descriptor(
  *
  * Return - The number of descriptors allocated.
  */
-static int fsl_dma_alloc_chan_resources(struct dma_chan *chan)
+static int fsl_dma_alloc_chan_resources(struct dma_chan *chan,
+					struct dma_client *client)
 {
 	struct fsl_dma_chan *fsl_chan = to_fsl_chan(chan);
 	LIST_HEAD(tmp_list);
@@ -809,8 +810,7 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 	if (!src) {
 		dev_err(fsl_chan->dev,
 				"selftest: Cannot alloc memory for test!\n");
-		err = -ENOMEM;
-		goto out;
+		return -ENOMEM;
 	}
 
 	dest = src + test_size;
@@ -820,7 +820,7 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 
 	chan = &fsl_chan->common;
 
-	if (fsl_dma_alloc_chan_resources(chan) < 1) {
+	if (fsl_dma_alloc_chan_resources(chan, NULL) < 1) {
 		dev_err(fsl_chan->dev,
 				"selftest: Cannot alloc resources for DMA\n");
 		err = -ENODEV;
@@ -842,13 +842,13 @@ static int fsl_dma_self_test(struct fsl_dma_chan *fsl_chan)
 	if (fsl_dma_is_complete(chan, cookie, NULL, NULL) != DMA_SUCCESS) {
 		dev_err(fsl_chan->dev, "selftest: Time out!\n");
 		err = -ENODEV;
-		goto out;
+		goto free_resources;
 	}
 
 	/* Test free and re-alloc channel resources */
 	fsl_dma_free_chan_resources(chan);
 
-	if (fsl_dma_alloc_chan_resources(chan) < 1) {
+	if (fsl_dma_alloc_chan_resources(chan, NULL) < 1) {
 		dev_err(fsl_chan->dev,
 				"selftest: Cannot alloc resources for DMA\n");
 		err = -ENODEV;
@@ -927,8 +927,7 @@ static int __devinit of_fsl_dma_chan_probe(struct of_device *dev,
 	if (!new_fsl_chan) {
 		dev_err(&dev->dev, "No free memory for allocating "
 				"dma channels!\n");
-		err = -ENOMEM;
-		goto err;
+		return -ENOMEM;
 	}
 
 	/* get dma channel register base */
@@ -936,7 +935,7 @@ static int __devinit of_fsl_dma_chan_probe(struct of_device *dev,
 	if (err) {
 		dev_err(&dev->dev, "Can't get %s property 'reg'\n",
 				dev->node->full_name);
-		goto err;
+		goto err_no_reg;
 	}
 
 	new_fsl_chan->feature = *(u32 *)match->data;
@@ -958,7 +957,7 @@ static int __devinit of_fsl_dma_chan_probe(struct of_device *dev,
 		dev_err(&dev->dev, "There is no %d channel!\n",
 				new_fsl_chan->id);
 		err = -EINVAL;
-		goto err;
+		goto err_no_chan;
 	}
 	fdev->chan[new_fsl_chan->id] = new_fsl_chan;
 	tasklet_init(&new_fsl_chan->tasklet, dma_do_tasklet,
@@ -997,23 +996,26 @@ static int __devinit of_fsl_dma_chan_probe(struct of_device *dev,
 		if (err) {
 			dev_err(&dev->dev, "DMA channel %s request_irq error "
 				"with return %d\n", dev->node->full_name, err);
-			goto err;
+			goto err_no_irq;
 		}
 	}
 
 	err = fsl_dma_self_test(new_fsl_chan);
 	if (err)
-		goto err;
+		goto err_self_test;
 
 	dev_info(&dev->dev, "#%d (%s), irq %d\n", new_fsl_chan->id,
 				match->compatible, new_fsl_chan->irq);
 
 	return 0;
-err:
-	dma_halt(new_fsl_chan);
-	iounmap(new_fsl_chan->reg_base);
+
+err_self_test:
 	free_irq(new_fsl_chan->irq, new_fsl_chan);
+err_no_irq:
 	list_del(&new_fsl_chan->common.device_node);
+err_no_chan:
+	iounmap(new_fsl_chan->reg_base);
+err_no_reg:
 	kfree(new_fsl_chan);
 	return err;
 }
@@ -1054,8 +1056,7 @@ static int __devinit of_fsl_dma_probe(struct of_device *dev,
 	fdev = kzalloc(sizeof(struct fsl_dma_device), GFP_KERNEL);
 	if (!fdev) {
 		dev_err(&dev->dev, "No enough memory for 'priv'\n");
-		err = -ENOMEM;
-		goto err;
+		return -ENOMEM;
 	}
 	fdev->dev = &dev->dev;
 	INIT_LIST_HEAD(&fdev->common.channels);
@@ -1065,7 +1066,7 @@ static int __devinit of_fsl_dma_probe(struct of_device *dev,
 	if (err) {
 		dev_err(&dev->dev, "Can't get %s property 'reg'\n",
 				dev->node->full_name);
-		goto err;
+		goto err_no_reg;
 	}
 
 	dev_info(&dev->dev, "Probe the Freescale DMA driver for %s "
@@ -1103,6 +1104,7 @@ static int __devinit of_fsl_dma_probe(struct of_device *dev,
 
 err:
 	iounmap(fdev->reg_base);
+err_no_reg:
 	kfree(fdev);
 	return err;
 }
diff --git a/drivers/dma/ioat.c b/drivers/dma/ioat.c
index 16e0fd8facfb..9b16a3af9a0a 100644
--- a/drivers/dma/ioat.c
+++ b/drivers/dma/ioat.c
@@ -47,6 +47,16 @@ static struct pci_device_id ioat_pci_tbl[] = {
 
 	/* I/OAT v2 platforms */
 	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB) },
+
+	/* I/OAT v3 platforms */
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG0) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG1) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG2) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG3) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG4) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG5) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG6) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_TBG7) },
 	{ 0, }
 };
 
@@ -83,6 +93,11 @@ static int ioat_setup_functionality(struct pci_dev *pdev, void __iomem *iobase)
 		if (device->dma && ioat_dca_enabled)
 			device->dca = ioat2_dca_init(pdev, iobase);
 		break;
+	case IOAT_VER_3_0:
+		device->dma = ioat_dma_probe(pdev, iobase);
+		if (device->dma && ioat_dca_enabled)
+			device->dca = ioat3_dca_init(pdev, iobase);
+		break;
 	default:
 		err = -ENODEV;
 		break;
diff --git a/drivers/dma/ioat_dca.c b/drivers/dma/ioat_dca.c
index 9e922760b7ff..6cf622da0286 100644
--- a/drivers/dma/ioat_dca.c
+++ b/drivers/dma/ioat_dca.c
@@ -37,12 +37,18 @@
 #include "ioatdma_registers.h"
 
 /*
- * Bit 16 of a tag map entry is the "valid" bit, if it is set then bits 0:15
+ * Bit 7 of a tag map entry is the "valid" bit, if it is set then bits 0:6
  * contain the bit number of the APIC ID to map into the DCA tag.  If the valid
  * bit is not set, then the value must be 0 or 1 and defines the bit in the tag.
  */
 #define DCA_TAG_MAP_VALID 0x80
 
+#define DCA3_TAG_MAP_BIT_TO_INV 0x80
+#define DCA3_TAG_MAP_BIT_TO_SEL 0x40
+#define DCA3_TAG_MAP_LITERAL_VAL 0x1
+
+#define DCA_TAG_MAP_MASK 0xDF
+
 /*
  * "Legacy" DCA systems do not implement the DCA register set in the
  * I/OAT device.  Software needs direct support for their tag mappings.
@@ -95,6 +101,7 @@ struct ioat_dca_slot {
 };
 
 #define IOAT_DCA_MAX_REQ 6
+#define IOAT3_DCA_MAX_REQ 2
 
 struct ioat_dca_priv {
 	void __iomem		*iobase;
@@ -171,7 +178,9 @@ static int ioat_dca_remove_requester(struct dca_provider *dca,
 	return -ENODEV;
 }
 
-static u8 ioat_dca_get_tag(struct dca_provider *dca, int cpu)
+static u8 ioat_dca_get_tag(struct dca_provider *dca,
+			   struct device *dev,
+			   int cpu)
 {
 	struct ioat_dca_priv *ioatdca = dca_priv(dca);
 	int i, apic_id, bit, value;
@@ -193,10 +202,26 @@ static u8 ioat_dca_get_tag(struct dca_provider *dca, int cpu)
 	return tag;
 }
 
+static int ioat_dca_dev_managed(struct dca_provider *dca,
+				struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+
+	pdev = to_pci_dev(dev);
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == pdev)
+			return 1;
+	}
+	return 0;
+}
+
 static struct dca_ops ioat_dca_ops = {
 	.add_requester		= ioat_dca_add_requester,
 	.remove_requester	= ioat_dca_remove_requester,
 	.get_tag		= ioat_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
 };
 
 
@@ -207,6 +232,8 @@ struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 	u8 *tag_map = NULL;
 	int i;
 	int err;
+	u8 version;
+	u8 max_requesters;
 
 	if (!system_has_dca_enabled(pdev))
 		return NULL;
@@ -237,15 +264,20 @@ struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 	if (tag_map == NULL)
 		return NULL;
 
+	version = readb(iobase + IOAT_VER_OFFSET);
+	if (version == IOAT_VER_3_0)
+		max_requesters = IOAT3_DCA_MAX_REQ;
+	else
+		max_requesters = IOAT_DCA_MAX_REQ;
+
 	dca = alloc_dca_provider(&ioat_dca_ops,
 			sizeof(*ioatdca) +
-			(sizeof(struct ioat_dca_slot) * IOAT_DCA_MAX_REQ));
+			(sizeof(struct ioat_dca_slot) * max_requesters));
 	if (!dca)
 		return NULL;
 
 	ioatdca = dca_priv(dca);
-	ioatdca->max_requesters = IOAT_DCA_MAX_REQ;
-
+	ioatdca->max_requesters = max_requesters;
 	ioatdca->dca_base = iobase + 0x54;
 
 	/* copy over the APIC ID to DCA tag mapping */
@@ -323,11 +355,13 @@ static int ioat2_dca_remove_requester(struct dca_provider *dca,
 	return -ENODEV;
 }
 
-static u8 ioat2_dca_get_tag(struct dca_provider *dca, int cpu)
+static u8 ioat2_dca_get_tag(struct dca_provider *dca,
+			    struct device *dev,
+			    int cpu)
 {
 	u8 tag;
 
-	tag = ioat_dca_get_tag(dca, cpu);
+	tag = ioat_dca_get_tag(dca, dev, cpu);
 	tag = (~tag) & 0x1F;
 	return tag;
 }
@@ -336,6 +370,7 @@ static struct dca_ops ioat2_dca_ops = {
 	.add_requester		= ioat2_dca_add_requester,
 	.remove_requester	= ioat2_dca_remove_requester,
 	.get_tag		= ioat2_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
 };
 
 static int ioat2_dca_count_dca_slots(void __iomem *iobase, u16 dca_offset)
@@ -425,3 +460,198 @@ struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase)
 
 	return dca;
 }
+
+static int ioat3_dca_add_requester(struct dca_provider *dca, struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+	u16 id;
+	u16 global_req_table;
+
+	/* This implementation only supports PCI-Express */
+	if (dev->bus != &pci_bus_type)
+		return -ENODEV;
+	pdev = to_pci_dev(dev);
+	id = dcaid_from_pcidev(pdev);
+
+	if (ioatdca->requester_count == ioatdca->max_requesters)
+		return -ENODEV;
+
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == NULL) {
+			/* found an empty slot */
+			ioatdca->requester_count++;
+			ioatdca->req_slots[i].pdev = pdev;
+			ioatdca->req_slots[i].rid = id;
+			global_req_table =
+			      readw(ioatdca->dca_base + IOAT3_DCA_GREQID_OFFSET);
+			writel(id | IOAT_DCA_GREQID_VALID,
+			       ioatdca->iobase + global_req_table + (i * 4));
+			return i;
+		}
+	}
+	/* Error, ioatdma->requester_count is out of whack */
+	return -EFAULT;
+}
+
+static int ioat3_dca_remove_requester(struct dca_provider *dca,
+				      struct device *dev)
+{
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	struct pci_dev *pdev;
+	int i;
+	u16 global_req_table;
+
+	/* This implementation only supports PCI-Express */
+	if (dev->bus != &pci_bus_type)
+		return -ENODEV;
+	pdev = to_pci_dev(dev);
+
+	for (i = 0; i < ioatdca->max_requesters; i++) {
+		if (ioatdca->req_slots[i].pdev == pdev) {
+			global_req_table =
+			      readw(ioatdca->dca_base + IOAT3_DCA_GREQID_OFFSET);
+			writel(0, ioatdca->iobase + global_req_table + (i * 4));
+			ioatdca->req_slots[i].pdev = NULL;
+			ioatdca->req_slots[i].rid = 0;
+			ioatdca->requester_count--;
+			return i;
+		}
+	}
+	return -ENODEV;
+}
+
+static u8 ioat3_dca_get_tag(struct dca_provider *dca,
+			    struct device *dev,
+			    int cpu)
+{
+	u8 tag;
+
+	struct ioat_dca_priv *ioatdca = dca_priv(dca);
+	int i, apic_id, bit, value;
+	u8 entry;
+
+	tag = 0;
+	apic_id = cpu_physical_id(cpu);
+
+	for (i = 0; i < IOAT_TAG_MAP_LEN; i++) {
+		entry = ioatdca->tag_map[i];
+		if (entry & DCA3_TAG_MAP_BIT_TO_SEL) {
+			bit = entry &
+				~(DCA3_TAG_MAP_BIT_TO_SEL | DCA3_TAG_MAP_BIT_TO_INV);
+			value = (apic_id & (1 << bit)) ? 1 : 0;
+		} else if (entry & DCA3_TAG_MAP_BIT_TO_INV) {
+			bit = entry & ~DCA3_TAG_MAP_BIT_TO_INV;
+			value = (apic_id & (1 << bit)) ? 0 : 1;
+		} else {
+			value = (entry & DCA3_TAG_MAP_LITERAL_VAL) ? 1 : 0;
+		}
+		tag |= (value << i);
+	}
+
+	return tag;
+}
+
+static struct dca_ops ioat3_dca_ops = {
+	.add_requester		= ioat3_dca_add_requester,
+	.remove_requester	= ioat3_dca_remove_requester,
+	.get_tag		= ioat3_dca_get_tag,
+	.dev_managed		= ioat_dca_dev_managed,
+};
+
+static int ioat3_dca_count_dca_slots(void *iobase, u16 dca_offset)
+{
+	int slots = 0;
+	u32 req;
+	u16 global_req_table;
+
+	global_req_table = readw(iobase + dca_offset + IOAT3_DCA_GREQID_OFFSET);
+	if (global_req_table == 0)
+		return 0;
+
+	do {
+		req = readl(iobase + global_req_table + (slots * sizeof(u32)));
+		slots++;
+	} while ((req & IOAT_DCA_GREQID_LASTID) == 0);
+
+	return slots;
+}
+
+struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase)
+{
+	struct dca_provider *dca;
+	struct ioat_dca_priv *ioatdca;
+	int slots;
+	int i;
+	int err;
+	u16 dca_offset;
+	u16 csi_fsb_control;
+	u16 pcie_control;
+	u8 bit;
+
+	union {
+		u64 full;
+		struct {
+			u32 low;
+			u32 high;
+		};
+	} tag_map;
+
+	if (!system_has_dca_enabled(pdev))
+		return NULL;
+
+	dca_offset = readw(iobase + IOAT_DCAOFFSET_OFFSET);
+	if (dca_offset == 0)
+		return NULL;
+
+	slots = ioat3_dca_count_dca_slots(iobase, dca_offset);
+	if (slots == 0)
+		return NULL;
+
+	dca = alloc_dca_provider(&ioat3_dca_ops,
+				 sizeof(*ioatdca)
+				      + (sizeof(struct ioat_dca_slot) * slots));
+	if (!dca)
+		return NULL;
+
+	ioatdca = dca_priv(dca);
+	ioatdca->iobase = iobase;
+	ioatdca->dca_base = iobase + dca_offset;
+	ioatdca->max_requesters = slots;
+
+	/* some bios might not know to turn these on */
+	csi_fsb_control = readw(ioatdca->dca_base + IOAT3_CSI_CONTROL_OFFSET);
+	if ((csi_fsb_control & IOAT3_CSI_CONTROL_PREFETCH) == 0) {
+		csi_fsb_control |= IOAT3_CSI_CONTROL_PREFETCH;
+		writew(csi_fsb_control,
+		       ioatdca->dca_base + IOAT3_CSI_CONTROL_OFFSET);
+	}
+	pcie_control = readw(ioatdca->dca_base + IOAT3_PCI_CONTROL_OFFSET);
+	if ((pcie_control & IOAT3_PCI_CONTROL_MEMWR) == 0) {
+		pcie_control |= IOAT3_PCI_CONTROL_MEMWR;
+		writew(pcie_control,
+		       ioatdca->dca_base + IOAT3_PCI_CONTROL_OFFSET);
+	}
+
+
+	/* TODO version, compatibility and configuration checks */
+
+	/* copy out the APIC to DCA tag map */
+	tag_map.low =
+		readl(ioatdca->dca_base + IOAT3_APICID_TAG_MAP_OFFSET_LOW);
+	tag_map.high =
+		readl(ioatdca->dca_base + IOAT3_APICID_TAG_MAP_OFFSET_HIGH);
+	for (i = 0; i < 8; i++) {
+		bit = tag_map.full >> (8 * i);
+		ioatdca->tag_map[i] = bit & DCA_TAG_MAP_MASK;
+	}
+
+	err = register_dca_provider(dca, &pdev->dev);
+	if (err) {
+		free_dca_provider(dca);
+		return NULL;
+	}
+
+	return dca;
+}
diff --git a/drivers/dma/ioat_dma.c b/drivers/dma/ioat_dma.c
index 318e8a22d814..a52156e56886 100644
--- a/drivers/dma/ioat_dma.c
+++ b/drivers/dma/ioat_dma.c
@@ -32,6 +32,7 @@
 #include <linux/dmaengine.h>
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
+#include <linux/workqueue.h>
 #include "ioatdma.h"
 #include "ioatdma_registers.h"
 #include "ioatdma_hw.h"
@@ -41,11 +42,23 @@
 #define to_ioat_desc(lh) container_of(lh, struct ioat_desc_sw, node)
 #define tx_to_ioat_desc(tx) container_of(tx, struct ioat_desc_sw, async_tx)
 
+#define chan_num(ch) ((int)((ch)->reg_base - (ch)->device->reg_base) / 0x80)
 static int ioat_pending_level = 4;
 module_param(ioat_pending_level, int, 0644);
 MODULE_PARM_DESC(ioat_pending_level,
 		 "high-water mark for pushing ioat descriptors (default: 4)");
 
+#define RESET_DELAY  msecs_to_jiffies(100)
+#define WATCHDOG_DELAY  round_jiffies(msecs_to_jiffies(2000))
+static void ioat_dma_chan_reset_part2(struct work_struct *work);
+static void ioat_dma_chan_watchdog(struct work_struct *work);
+
+/*
+ * workaround for IOAT ver.3.0 null descriptor issue
+ * (channel returns error when size is 0)
+ */
+#define NULL_DESC_BUFFER_SIZE 1
+
 /* internal functions */
 static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan);
 static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan);
@@ -122,6 +135,38 @@ static int ioat_dma_enumerate_channels(struct ioatdma_device *device)
 	int i;
 	struct ioat_dma_chan *ioat_chan;
 
+	/*
+	 * IOAT ver.3 workarounds
+	 */
+	if (device->version == IOAT_VER_3_0) {
+		u32 chan_err_mask;
+		u16 dev_id;
+		u32 dmauncerrsts;
+
+		/*
+		 * Write CHANERRMSK_INT with 3E07h to mask out the errors
+		 * that can cause stability issues for IOAT ver.3
+		 */
+		chan_err_mask = 0x3E07;
+		pci_write_config_dword(device->pdev,
+			IOAT_PCI_CHANERRMASK_INT_OFFSET,
+			chan_err_mask);
+
+		/*
+		 * Clear DMAUNCERRSTS Cfg-Reg Parity Error status bit
+		 * (workaround for spurious config parity error after restart)
+		 */
+		pci_read_config_word(device->pdev,
+			IOAT_PCI_DEVICE_ID_OFFSET,
+			&dev_id);
+		if (dev_id == PCI_DEVICE_ID_INTEL_IOAT_TBG0) {
+			dmauncerrsts = 0x10;
+			pci_write_config_dword(device->pdev,
+				IOAT_PCI_DMAUNCERRSTS_OFFSET,
+				dmauncerrsts);
+		}
+	}
+
 	device->common.chancnt = readb(device->reg_base + IOAT_CHANCNT_OFFSET);
 	xfercap_scale = readb(device->reg_base + IOAT_XFERCAP_OFFSET);
 	xfercap = (xfercap_scale == 0 ? -1 : (1UL << xfercap_scale));
@@ -137,6 +182,7 @@ static int ioat_dma_enumerate_channels(struct ioatdma_device *device)
 		ioat_chan->reg_base = device->reg_base + (0x80 * (i + 1));
 		ioat_chan->xfercap = xfercap;
 		ioat_chan->desccount = 0;
+		INIT_DELAYED_WORK(&ioat_chan->work, ioat_dma_chan_reset_part2);
 		if (ioat_chan->device->version != IOAT_VER_1_2) {
 			writel(IOAT_DCACTRL_CMPL_WRITE_ENABLE
 					| IOAT_DMA_DCA_ANY_CPU,
@@ -175,7 +221,7 @@ static void ioat1_dma_memcpy_issue_pending(struct dma_chan *chan)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
 
-	if (ioat_chan->pending != 0) {
+	if (ioat_chan->pending > 0) {
 		spin_lock_bh(&ioat_chan->desc_lock);
 		__ioat1_dma_memcpy_issue_pending(ioat_chan);
 		spin_unlock_bh(&ioat_chan->desc_lock);
@@ -194,13 +240,228 @@ static void ioat2_dma_memcpy_issue_pending(struct dma_chan *chan)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
 
-	if (ioat_chan->pending != 0) {
+	if (ioat_chan->pending > 0) {
 		spin_lock_bh(&ioat_chan->desc_lock);
 		__ioat2_dma_memcpy_issue_pending(ioat_chan);
 		spin_unlock_bh(&ioat_chan->desc_lock);
 	}
 }
 
+
+/**
+ * ioat_dma_chan_reset_part2 - reinit the channel after a reset
+ */
+static void ioat_dma_chan_reset_part2(struct work_struct *work)
+{
+	struct ioat_dma_chan *ioat_chan =
+		container_of(work, struct ioat_dma_chan, work.work);
+	struct ioat_desc_sw *desc;
+
+	spin_lock_bh(&ioat_chan->cleanup_lock);
+	spin_lock_bh(&ioat_chan->desc_lock);
+
+	ioat_chan->completion_virt->low = 0;
+	ioat_chan->completion_virt->high = 0;
+	ioat_chan->pending = 0;
+
+	/*
+	 * count the descriptors waiting, and be sure to do it
+	 * right for both the CB1 line and the CB2 ring
+	 */
+	ioat_chan->dmacount = 0;
+	if (ioat_chan->used_desc.prev) {
+		desc = to_ioat_desc(ioat_chan->used_desc.prev);
+		do {
+			ioat_chan->dmacount++;
+			desc = to_ioat_desc(desc->node.next);
+		} while (&desc->node != ioat_chan->used_desc.next);
+	}
+
+	/*
+	 * write the new starting descriptor address
+	 * this puts channel engine into ARMED state
+	 */
+	desc = to_ioat_desc(ioat_chan->used_desc.prev);
+	switch (ioat_chan->device->version) {
+	case IOAT_VER_1_2:
+		writel(((u64) desc->async_tx.phys) & 0x00000000FFFFFFFF,
+		       ioat_chan->reg_base + IOAT1_CHAINADDR_OFFSET_LOW);
+		writel(((u64) desc->async_tx.phys) >> 32,
+		       ioat_chan->reg_base + IOAT1_CHAINADDR_OFFSET_HIGH);
+
+		writeb(IOAT_CHANCMD_START, ioat_chan->reg_base
+			+ IOAT_CHANCMD_OFFSET(ioat_chan->device->version));
+		break;
+	case IOAT_VER_2_0:
+		writel(((u64) desc->async_tx.phys) & 0x00000000FFFFFFFF,
+		       ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_LOW);
+		writel(((u64) desc->async_tx.phys) >> 32,
+		       ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_HIGH);
+
+		/* tell the engine to go with what's left to be done */
+		writew(ioat_chan->dmacount,
+		       ioat_chan->reg_base + IOAT_CHAN_DMACOUNT_OFFSET);
+
+		break;
+	}
+	dev_err(&ioat_chan->device->pdev->dev,
+		"chan%d reset - %d descs waiting, %d total desc\n",
+		chan_num(ioat_chan), ioat_chan->dmacount, ioat_chan->desccount);
+
+	spin_unlock_bh(&ioat_chan->desc_lock);
+	spin_unlock_bh(&ioat_chan->cleanup_lock);
+}
+
+/**
+ * ioat_dma_reset_channel - restart a channel
+ * @ioat_chan: IOAT DMA channel handle
+ */
+static void ioat_dma_reset_channel(struct ioat_dma_chan *ioat_chan)
+{
+	u32 chansts, chanerr;
+
+	if (!ioat_chan->used_desc.prev)
+		return;
+
+	chanerr = readl(ioat_chan->reg_base + IOAT_CHANERR_OFFSET);
+	chansts = (ioat_chan->completion_virt->low
+					& IOAT_CHANSTS_DMA_TRANSFER_STATUS);
+	if (chanerr) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"chan%d, CHANSTS = 0x%08x CHANERR = 0x%04x, clearing\n",
+			chan_num(ioat_chan), chansts, chanerr);
+		writel(chanerr, ioat_chan->reg_base + IOAT_CHANERR_OFFSET);
+	}
+
+	/*
+	 * whack it upside the head with a reset
+	 * and wait for things to settle out.
+	 * force the pending count to a really big negative
+	 * to make sure no one forces an issue_pending
+	 * while we're waiting.
+	 */
+
+	spin_lock_bh(&ioat_chan->desc_lock);
+	ioat_chan->pending = INT_MIN;
+	writeb(IOAT_CHANCMD_RESET,
+	       ioat_chan->reg_base
+	       + IOAT_CHANCMD_OFFSET(ioat_chan->device->version));
+	spin_unlock_bh(&ioat_chan->desc_lock);
+
+	/* schedule the 2nd half instead of sleeping a long time */
+	schedule_delayed_work(&ioat_chan->work, RESET_DELAY);
+}
+
+/**
+ * ioat_dma_chan_watchdog - watch for stuck channels
+ */
+static void ioat_dma_chan_watchdog(struct work_struct *work)
+{
+	struct ioatdma_device *device =
+		container_of(work, struct ioatdma_device, work.work);
+	struct ioat_dma_chan *ioat_chan;
+	int i;
+
+	union {
+		u64 full;
+		struct {
+			u32 low;
+			u32 high;
+		};
+	} completion_hw;
+	unsigned long compl_desc_addr_hw;
+
+	for (i = 0; i < device->common.chancnt; i++) {
+		ioat_chan = ioat_lookup_chan_by_index(device, i);
+
+		if (ioat_chan->device->version == IOAT_VER_1_2
+			/* have we started processing anything yet */
+		    && ioat_chan->last_completion
+			/* have we completed any since last watchdog cycle? */
+		    && (ioat_chan->last_completion ==
+				ioat_chan->watchdog_completion)
+			/* has TCP stuck on one cookie since last watchdog? */
+		    && (ioat_chan->watchdog_tcp_cookie ==
+				ioat_chan->watchdog_last_tcp_cookie)
+		    && (ioat_chan->watchdog_tcp_cookie !=
+				ioat_chan->completed_cookie)
+			/* is there something in the chain to be processed? */
+			/* CB1 chain always has at least the last one processed */
+		    && (ioat_chan->used_desc.prev != ioat_chan->used_desc.next)
+		    && ioat_chan->pending == 0) {
+
+			/*
+			 * check CHANSTS register for completed
+			 * descriptor address.
+			 * if it is different than completion writeback,
+			 * it is not zero
+			 * and it has changed since the last watchdog
+			 *     we can assume that channel
+			 *     is still working correctly
+			 *     and the problem is in completion writeback.
+			 *     update completion writeback
+			 *     with actual CHANSTS value
+			 * else
+			 *     try resetting the channel
+			 */
+
+			completion_hw.low = readl(ioat_chan->reg_base +
+				IOAT_CHANSTS_OFFSET_LOW(ioat_chan->device->version));
+			completion_hw.high = readl(ioat_chan->reg_base +
+				IOAT_CHANSTS_OFFSET_HIGH(ioat_chan->device->version));
+#if (BITS_PER_LONG == 64)
+			compl_desc_addr_hw =
+				completion_hw.full
+				& IOAT_CHANSTS_COMPLETED_DESCRIPTOR_ADDR;
+#else
+			compl_desc_addr_hw =
+				completion_hw.low & IOAT_LOW_COMPLETION_MASK;
+#endif
+
+			if ((compl_desc_addr_hw != 0)
+			   && (compl_desc_addr_hw != ioat_chan->watchdog_completion)
+			   && (compl_desc_addr_hw != ioat_chan->last_compl_desc_addr_hw)) {
+				ioat_chan->last_compl_desc_addr_hw = compl_desc_addr_hw;
+				ioat_chan->completion_virt->low = completion_hw.low;
+				ioat_chan->completion_virt->high = completion_hw.high;
+			} else {
+				ioat_dma_reset_channel(ioat_chan);
+				ioat_chan->watchdog_completion = 0;
+				ioat_chan->last_compl_desc_addr_hw = 0;
+			}
+
+		/*
+		 * for version 2.0 if there are descriptors yet to be processed
+		 * and the last completed hasn't changed since the last watchdog
+		 *      if they haven't hit the pending level
+		 *          issue the pending to push them through
+		 *      else
+		 *          try resetting the channel
+		 */
+		} else if (ioat_chan->device->version == IOAT_VER_2_0
+		    && ioat_chan->used_desc.prev
+		    && ioat_chan->last_completion
+		    && ioat_chan->last_completion == ioat_chan->watchdog_completion) {
+
+			if (ioat_chan->pending < ioat_pending_level)
+				ioat2_dma_memcpy_issue_pending(&ioat_chan->common);
+			else {
+				ioat_dma_reset_channel(ioat_chan);
+				ioat_chan->watchdog_completion = 0;
+			}
+		} else {
+			ioat_chan->last_compl_desc_addr_hw = 0;
+			ioat_chan->watchdog_completion
+					= ioat_chan->last_completion;
+		}
+
+		ioat_chan->watchdog_last_tcp_cookie =
+			ioat_chan->watchdog_tcp_cookie;
+	}
+
+	schedule_delayed_work(&device->work, WATCHDOG_DELAY);
+}
+
 static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(tx->chan);
@@ -250,6 +511,13 @@ static dma_cookie_t ioat1_tx_submit(struct dma_async_tx_descriptor *tx)
 		prev = new;
 	} while (len && (new = ioat1_dma_get_next_descriptor(ioat_chan)));
 
+	if (!new) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"tx submit failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return -ENOMEM;
+	}
+
 	hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
 	if (new->async_tx.callback) {
 		hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_INT_GN;
@@ -335,7 +603,14 @@ static dma_cookie_t ioat2_tx_submit(struct dma_async_tx_descriptor *tx)
 		desc_count++;
 	} while (len && (new = ioat2_dma_get_next_descriptor(ioat_chan)));
 
-	hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
+	if (!new) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"tx submit failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return -ENOMEM;
+	}
+
+	hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
 	if (new->async_tx.callback) {
 		hw->ctl |= IOAT_DMA_DESCRIPTOR_CTL_INT_GN;
 		if (first != new) {
@@ -406,6 +681,7 @@ static struct ioat_desc_sw *ioat_dma_alloc_descriptor(
 		desc_sw->async_tx.tx_submit = ioat1_tx_submit;
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		desc_sw->async_tx.tx_submit = ioat2_tx_submit;
 		break;
 	}
@@ -452,7 +728,8 @@ static void ioat2_dma_massage_chan_desc(struct ioat_dma_chan *ioat_chan)
  * ioat_dma_alloc_chan_resources - returns the number of allocated descriptors
  * @chan: the channel to be filled out
  */
-static int ioat_dma_alloc_chan_resources(struct dma_chan *chan)
+static int ioat_dma_alloc_chan_resources(struct dma_chan *chan,
+					 struct dma_client *client)
 {
 	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
 	struct ioat_desc_sw *desc;
@@ -555,6 +832,7 @@ static void ioat_dma_free_chan_resources(struct dma_chan *chan)
 		}
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		list_for_each_entry_safe(desc, _desc,
 					 ioat_chan->free_desc.next, node) {
 			list_del(&desc->node);
@@ -585,6 +863,10 @@ static void ioat_dma_free_chan_resources(struct dma_chan *chan)
 	ioat_chan->last_completion = ioat_chan->completion_addr = 0;
 	ioat_chan->pending = 0;
 	ioat_chan->dmacount = 0;
+	ioat_chan->watchdog_completion = 0;
+	ioat_chan->last_compl_desc_addr_hw = 0;
+	ioat_chan->watchdog_tcp_cookie =
+		ioat_chan->watchdog_last_tcp_cookie = 0;
 }
 
 /**
@@ -640,7 +922,8 @@ ioat2_dma_get_next_descriptor(struct ioat_dma_chan *ioat_chan)
 
 		/* set up the noop descriptor */
 		noop_desc = to_ioat_desc(ioat_chan->used_desc.next);
-		noop_desc->hw->size = 0;
+		/* set size to non-zero value (channel returns error when size is 0) */
+		noop_desc->hw->size = NULL_DESC_BUFFER_SIZE;
 		noop_desc->hw->ctl = IOAT_DMA_DESCRIPTOR_NUL;
 		noop_desc->hw->src_addr = 0;
 		noop_desc->hw->dst_addr = 0;
@@ -690,6 +973,7 @@ static struct ioat_desc_sw *ioat_dma_get_next_descriptor(
 		return ioat1_dma_get_next_descriptor(ioat_chan);
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		return ioat2_dma_get_next_descriptor(ioat_chan);
 		break;
 	}
@@ -716,8 +1000,12 @@ static struct dma_async_tx_descriptor *ioat1_dma_prep_memcpy(
 		new->src = dma_src;
 		new->async_tx.flags = flags;
 		return &new->async_tx;
-	} else
+	} else {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"chan%d - get_next_desc failed: %d descs waiting, %d total desc\n",
+			chan_num(ioat_chan), ioat_chan->dmacount, ioat_chan->desccount);
 		return NULL;
+	}
 }
 
 static struct dma_async_tx_descriptor *ioat2_dma_prep_memcpy(
@@ -744,8 +1032,13 @@ static struct dma_async_tx_descriptor *ioat2_dma_prep_memcpy(
 		new->src = dma_src;
 		new->async_tx.flags = flags;
 		return &new->async_tx;
-	} else
+	} else {
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		dev_err(&ioat_chan->device->pdev->dev,
+			"chan%d - get_next_desc failed: %d descs waiting, %d total desc\n",
+			chan_num(ioat_chan), ioat_chan->dmacount, ioat_chan->desccount);
 		return NULL;
+	}
 }
 
 static void ioat_dma_cleanup_tasklet(unsigned long data)
@@ -756,6 +1049,27 @@ static void ioat_dma_cleanup_tasklet(unsigned long data)
 	       chan->reg_base + IOAT_CHANCTRL_OFFSET);
 }
 
+static void
+ioat_dma_unmap(struct ioat_dma_chan *ioat_chan, struct ioat_desc_sw *desc)
+{
+	/*
+	 * yes we are unmapping both _page and _single
+	 * alloc'd regions with unmap_page. Is this
+	 * *really* that bad?
+	 */
+	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_DEST_UNMAP))
+		pci_unmap_page(ioat_chan->device->pdev,
+				pci_unmap_addr(desc, dst),
+				pci_unmap_len(desc, len),
+				PCI_DMA_FROMDEVICE);
+
+	if (!(desc->async_tx.flags & DMA_COMPL_SKIP_SRC_UNMAP))
+		pci_unmap_page(ioat_chan->device->pdev,
+				pci_unmap_addr(desc, src),
+				pci_unmap_len(desc, len),
+				PCI_DMA_TODEVICE);
+}
+
 /**
  * ioat_dma_memcpy_cleanup - cleanup up finished descriptors
  * @chan: ioat channel to be cleaned up
@@ -799,11 +1113,27 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 
 	if (phys_complete == ioat_chan->last_completion) {
 		spin_unlock_bh(&ioat_chan->cleanup_lock);
+		/*
+		 * perhaps we're stuck so hard that the watchdog can't go off?
+		 * try to catch it after 2 seconds
+		 */
+		if (ioat_chan->device->version != IOAT_VER_3_0) {
+			if (time_after(jiffies,
+				       ioat_chan->last_completion_time + HZ*WATCHDOG_DELAY)) {
+				ioat_dma_chan_watchdog(&(ioat_chan->device->work.work));
+				ioat_chan->last_completion_time = jiffies;
+			}
+		}
 		return;
 	}
+	ioat_chan->last_completion_time = jiffies;
 
 	cookie = 0;
-	spin_lock_bh(&ioat_chan->desc_lock);
+	if (!spin_trylock_bh(&ioat_chan->desc_lock)) {
+		spin_unlock_bh(&ioat_chan->cleanup_lock);
+		return;
+	}
+
 	switch (ioat_chan->device->version) {
 	case IOAT_VER_1_2:
 		list_for_each_entry_safe(desc, _desc,
@@ -816,21 +1146,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 			 */
 			if (desc->async_tx.cookie) {
 				cookie = desc->async_tx.cookie;
-
-				/*
-				 * yes we are unmapping both _page and _single
-				 * alloc'd regions with unmap_page. Is this
-				 * *really* that bad?
-				 */
-				pci_unmap_page(ioat_chan->device->pdev,
-						pci_unmap_addr(desc, dst),
-						pci_unmap_len(desc, len),
-						PCI_DMA_FROMDEVICE);
-				pci_unmap_page(ioat_chan->device->pdev,
-						pci_unmap_addr(desc, src),
-						pci_unmap_len(desc, len),
-						PCI_DMA_TODEVICE);
-
+				ioat_dma_unmap(ioat_chan, desc);
 				if (desc->async_tx.callback) {
 					desc->async_tx.callback(desc->async_tx.callback_param);
 					desc->async_tx.callback = NULL;
@@ -862,6 +1178,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 		}
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		/* has some other thread has already cleaned up? */
 		if (ioat_chan->used_desc.prev == NULL)
 			break;
@@ -889,16 +1206,7 @@ static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan)
 				if (desc->async_tx.cookie) {
 					cookie = desc->async_tx.cookie;
 					desc->async_tx.cookie = 0;
-
-					pci_unmap_page(ioat_chan->device->pdev,
-						      pci_unmap_addr(desc, dst),
-						      pci_unmap_len(desc, len),
-						      PCI_DMA_FROMDEVICE);
-					pci_unmap_page(ioat_chan->device->pdev,
-						      pci_unmap_addr(desc, src),
-						      pci_unmap_len(desc, len),
-						      PCI_DMA_TODEVICE);
-
+					ioat_dma_unmap(ioat_chan, desc);
 					if (desc->async_tx.callback) {
 						desc->async_tx.callback(desc->async_tx.callback_param);
 						desc->async_tx.callback = NULL;
@@ -943,6 +1251,7 @@ static enum dma_status ioat_dma_is_complete(struct dma_chan *chan,
 
 	last_used = chan->cookie;
 	last_complete = ioat_chan->completed_cookie;
+	ioat_chan->watchdog_tcp_cookie = cookie;
 
 	if (done)
 		*done = last_complete;
@@ -973,10 +1282,19 @@ static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan)
 	spin_lock_bh(&ioat_chan->desc_lock);
 
 	desc = ioat_dma_get_next_descriptor(ioat_chan);
+
+	if (!desc) {
+		dev_err(&ioat_chan->device->pdev->dev,
+			"Unable to start null desc - get next desc failed\n");
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		return;
+	}
+
 	desc->hw->ctl = IOAT_DMA_DESCRIPTOR_NUL
 				| IOAT_DMA_DESCRIPTOR_CTL_INT_GN
 				| IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
-	desc->hw->size = 0;
+	/* set size to non-zero value (channel returns error when size is 0) */
+	desc->hw->size = NULL_DESC_BUFFER_SIZE;
 	desc->hw->src_addr = 0;
 	desc->hw->dst_addr = 0;
 	async_tx_ack(&desc->async_tx);
@@ -994,6 +1312,7 @@ static void ioat_dma_start_null_desc(struct ioat_dma_chan *ioat_chan)
 			+ IOAT_CHANCMD_OFFSET(ioat_chan->device->version));
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		writel(((u64) desc->async_tx.phys) & 0x00000000FFFFFFFF,
 		       ioat_chan->reg_base + IOAT2_CHAINADDR_OFFSET_LOW);
 		writel(((u64) desc->async_tx.phys) >> 32,
@@ -1049,7 +1368,7 @@ static int ioat_dma_self_test(struct ioatdma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (device->common.device_alloc_chan_resources(dma_chan) < 1) {
+	if (device->common.device_alloc_chan_resources(dma_chan, NULL) < 1) {
 		dev_err(&device->pdev->dev,
 			"selftest cannot allocate chan resource\n");
 		err = -ENODEV;
@@ -1312,6 +1631,7 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 						ioat1_dma_memcpy_issue_pending;
 		break;
 	case IOAT_VER_2_0:
+	case IOAT_VER_3_0:
 		device->common.device_prep_dma_memcpy = ioat2_dma_prep_memcpy;
 		device->common.device_issue_pending =
 						ioat2_dma_memcpy_issue_pending;
@@ -1331,8 +1651,16 @@ struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 	if (err)
 		goto err_self_test;
 
+	ioat_set_tcp_copy_break(device);
+
 	dma_async_device_register(&device->common);
 
+	if (device->version != IOAT_VER_3_0) {
+		INIT_DELAYED_WORK(&device->work, ioat_dma_chan_watchdog);
+		schedule_delayed_work(&device->work,
+				      WATCHDOG_DELAY);
+	}
+
 	return device;
 
 err_self_test:
@@ -1365,6 +1693,10 @@ void ioat_dma_remove(struct ioatdma_device *device)
 	pci_release_regions(device->pdev);
 	pci_disable_device(device->pdev);
 
+	if (device->version != IOAT_VER_3_0) {
+		cancel_delayed_work(&device->work);
+	}
+
 	list_for_each_entry_safe(chan, _chan,
 				 &device->common.channels, device_node) {
 		ioat_chan = to_ioat_chan(chan);
diff --git a/drivers/dma/ioatdma.h b/drivers/dma/ioatdma.h
index f2c7fedbf009..a3306d0e1372 100644
--- a/drivers/dma/ioatdma.h
+++ b/drivers/dma/ioatdma.h
@@ -27,8 +27,9 @@
 #include <linux/dmapool.h>
 #include <linux/cache.h>
 #include <linux/pci_ids.h>
+#include <net/tcp.h>
 
-#define IOAT_DMA_VERSION  "2.04"
+#define IOAT_DMA_VERSION  "3.30"
 
 enum ioat_interrupt {
 	none = 0,
@@ -40,6 +41,7 @@ enum ioat_interrupt {
 
 #define IOAT_LOW_COMPLETION_MASK	0xffffffc0
 #define IOAT_DMA_DCA_ANY_CPU		~0
+#define IOAT_WATCHDOG_PERIOD		(2 * HZ)
 
 
 /**
@@ -62,6 +64,7 @@ struct ioatdma_device {
 	struct dma_device common;
 	u8 version;
 	enum ioat_interrupt irq_mode;
+	struct delayed_work work;
 	struct msix_entry msix_entries[4];
 	struct ioat_dma_chan *idx[4];
 };
@@ -75,6 +78,7 @@ struct ioat_dma_chan {
 
 	dma_cookie_t completed_cookie;
 	unsigned long last_completion;
+	unsigned long last_completion_time;
 
 	size_t xfercap;	/* XFERCAP register value expanded out */
 
@@ -82,6 +86,10 @@ struct ioat_dma_chan {
 	spinlock_t desc_lock;
 	struct list_head free_desc;
 	struct list_head used_desc;
+	unsigned long watchdog_completion;
+	int watchdog_tcp_cookie;
+	u32 watchdog_last_tcp_cookie;
+	struct delayed_work work;
 
 	int pending;
 	int dmacount;
@@ -98,6 +106,7 @@ struct ioat_dma_chan {
 			u32 high;
 		};
 	} *completion_virt;
+	unsigned long last_compl_desc_addr_hw;
 	struct tasklet_struct cleanup_task;
 };
 
@@ -121,17 +130,34 @@ struct ioat_desc_sw {
 	struct dma_async_tx_descriptor async_tx;
 };
 
+static inline void ioat_set_tcp_copy_break(struct ioatdma_device *dev)
+{
+	#ifdef CONFIG_NET_DMA
+	switch (dev->version) {
+	case IOAT_VER_1_2:
+	case IOAT_VER_3_0:
+		sysctl_tcp_dma_copybreak = 4096;
+		break;
+	case IOAT_VER_2_0:
+		sysctl_tcp_dma_copybreak = 2048;
+		break;
+	}
+	#endif
+}
+
 #if defined(CONFIG_INTEL_IOATDMA) || defined(CONFIG_INTEL_IOATDMA_MODULE)
 struct ioatdma_device *ioat_dma_probe(struct pci_dev *pdev,
 				      void __iomem *iobase);
 void ioat_dma_remove(struct ioatdma_device *device);
 struct dca_provider *ioat_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 struct dca_provider *ioat2_dca_init(struct pci_dev *pdev, void __iomem *iobase);
+struct dca_provider *ioat3_dca_init(struct pci_dev *pdev, void __iomem *iobase);
 #else
 #define ioat_dma_probe(pdev, iobase)    NULL
 #define ioat_dma_remove(device)         do { } while (0)
 #define ioat_dca_init(pdev, iobase)	NULL
 #define ioat2_dca_init(pdev, iobase)	NULL
+#define ioat3_dca_init(pdev, iobase)	NULL
 #endif
 
 #endif /* IOATDMA_H */
diff --git a/drivers/dma/ioatdma_hw.h b/drivers/dma/ioatdma_hw.h
index dd470fa91d86..f1ae2c776f74 100644
--- a/drivers/dma/ioatdma_hw.h
+++ b/drivers/dma/ioatdma_hw.h
@@ -35,6 +35,7 @@
 #define IOAT_PCI_SID            0x8086
 #define IOAT_VER_1_2            0x12    /* Version 1.2 */
 #define IOAT_VER_2_0            0x20    /* Version 2.0 */
+#define IOAT_VER_3_0            0x30    /* Version 3.0 */
 
 struct ioat_dma_descriptor {
 	uint32_t	size;
diff --git a/drivers/dma/ioatdma_registers.h b/drivers/dma/ioatdma_registers.h
index 9832d7ebd931..827cb503cac6 100644
--- a/drivers/dma/ioatdma_registers.h
+++ b/drivers/dma/ioatdma_registers.h
@@ -25,6 +25,10 @@
 #define IOAT_PCI_DMACTRL_DMA_EN			0x00000001
 #define IOAT_PCI_DMACTRL_MSI_EN			0x00000002
 
+#define IOAT_PCI_DEVICE_ID_OFFSET		0x02
+#define IOAT_PCI_DMAUNCERRSTS_OFFSET		0x148
+#define IOAT_PCI_CHANERRMASK_INT_OFFSET		0x184
+
 /* MMIO Device Registers */
 #define IOAT_CHANCNT_OFFSET			0x00	/*  8-bit */
 
@@ -149,7 +153,23 @@
 #define IOAT_DCA_GREQID_VALID       0x20000000
 #define IOAT_DCA_GREQID_LASTID      0x80000000
 
+#define IOAT3_CSI_CAPABILITY_OFFSET 0x08
+#define IOAT3_CSI_CAPABILITY_PREFETCH    0x1
+
+#define IOAT3_PCI_CAPABILITY_OFFSET 0x0A
+#define IOAT3_PCI_CAPABILITY_MEMWR  0x1
+
+#define IOAT3_CSI_CONTROL_OFFSET    0x0C
+#define IOAT3_CSI_CONTROL_PREFETCH  0x1
+
+#define IOAT3_PCI_CONTROL_OFFSET    0x0E
+#define IOAT3_PCI_CONTROL_MEMWR     0x1
+
+#define IOAT3_APICID_TAG_MAP_OFFSET 0x10
+#define IOAT3_APICID_TAG_MAP_OFFSET_LOW  0x10
+#define IOAT3_APICID_TAG_MAP_OFFSET_HIGH 0x14
 
+#define IOAT3_DCA_GREQID_OFFSET     0x02
 
 #define IOAT1_CHAINADDR_OFFSET		0x0C	/* 64-bit Descriptor Chain Address Register */
 #define IOAT2_CHAINADDR_OFFSET		0x10	/* 64-bit Descriptor Chain Address Register */
diff --git a/drivers/dma/iop-adma.c b/drivers/dma/iop-adma.c
index 0ec0f431e6a1..85bfeba4d85e 100644
--- a/drivers/dma/iop-adma.c
+++ b/drivers/dma/iop-adma.c
@@ -82,17 +82,24 @@ iop_adma_run_tx_complete_actions(struct iop_adma_desc_slot *desc,
 			struct device *dev =
 				&iop_chan->device->pdev->dev;
 			u32 len = unmap->unmap_len;
-			u32 src_cnt = unmap->unmap_src_cnt;
-			dma_addr_t addr = iop_desc_get_dest_addr(unmap,
-				iop_chan);
-
-			dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
-			while (src_cnt--) {
-				addr = iop_desc_get_src_addr(unmap,
-							iop_chan,
-							src_cnt);
-				dma_unmap_page(dev, addr, len,
-					DMA_TO_DEVICE);
+			enum dma_ctrl_flags flags = desc->async_tx.flags;
+			u32 src_cnt;
+			dma_addr_t addr;
+
+			if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
+				addr = iop_desc_get_dest_addr(unmap, iop_chan);
+				dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
+			}
+
+			if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
+				src_cnt = unmap->unmap_src_cnt;
+				while (src_cnt--) {
+					addr = iop_desc_get_src_addr(unmap,
+								     iop_chan,
+								     src_cnt);
+					dma_unmap_page(dev, addr, len,
+						       DMA_TO_DEVICE);
+				}
 			}
 			desc->group_head = NULL;
 		}
@@ -366,8 +373,8 @@ retry:
 	if (!retry++)
 		goto retry;
 
-	/* try to free some slots if the allocation fails */
-	tasklet_schedule(&iop_chan->irq_tasklet);
+	/* perform direct reclaim if the allocation fails */
+	__iop_adma_slot_cleanup(iop_chan);
 
 	return NULL;
 }
@@ -443,8 +450,18 @@ iop_adma_tx_submit(struct dma_async_tx_descriptor *tx)
 static void iop_chan_start_null_memcpy(struct iop_adma_chan *iop_chan);
 static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan);
 
-/* returns the number of allocated descriptors */
-static int iop_adma_alloc_chan_resources(struct dma_chan *chan)
+/**
+ * iop_adma_alloc_chan_resources -  returns the number of allocated descriptors
+ * @chan - allocate descriptor resources for this channel
+ * @client - current client requesting the channel be ready for requests
+ *
+ * Note: We keep the slots for 1 operation on iop_chan->chain at all times.  To
+ * avoid deadlock, via async_xor, num_descs_in_pool must at a minimum be
+ * greater than 2x the number slots needed to satisfy a device->max_xor
+ * request.
+ * */
+static int iop_adma_alloc_chan_resources(struct dma_chan *chan,
+					 struct dma_client *client)
 {
 	char *hw_desc;
 	int idx;
@@ -838,7 +855,7 @@ static int __devinit iop_adma_memcpy_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -936,7 +953,7 @@ iop_adma_xor_zero_sum_self_test(struct iop_adma_device *device)
 	dma_chan = container_of(device->common.channels.next,
 				struct dma_chan,
 				device_node);
-	if (iop_adma_alloc_chan_resources(dma_chan) < 1) {
+	if (iop_adma_alloc_chan_resources(dma_chan, NULL) < 1) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -1387,6 +1404,8 @@ static void iop_chan_start_null_xor(struct iop_adma_chan *iop_chan)
 	spin_unlock_bh(&iop_chan->lock);
 }
 
+MODULE_ALIAS("platform:iop-adma");
+
 static struct platform_driver iop_adma_driver = {
 	.probe		= iop_adma_probe,
 	.remove		= iop_adma_remove,
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
new file mode 100644
index 000000000000..a4e4494663bf
--- /dev/null
+++ b/drivers/dma/mv_xor.c
@@ -0,0 +1,1375 @@
+/*
+ * offload engine driver for the Marvell XOR engine
+ * Copyright (C) 2007, 2008, Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/async_tx.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/memory.h>
+#include <asm/plat-orion/mv_xor.h>
+#include "mv_xor.h"
+
+static void mv_xor_issue_pending(struct dma_chan *chan);
+
+#define to_mv_xor_chan(chan)		\
+	container_of(chan, struct mv_xor_chan, common)
+
+#define to_mv_xor_device(dev)		\
+	container_of(dev, struct mv_xor_device, common)
+
+#define to_mv_xor_slot(tx)		\
+	container_of(tx, struct mv_xor_desc_slot, async_tx)
+
+static void mv_desc_init(struct mv_xor_desc_slot *desc, unsigned long flags)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+
+	hw_desc->status = (1 << 31);
+	hw_desc->phy_next_desc = 0;
+	hw_desc->desc_command = (1 << 31);
+}
+
+static u32 mv_desc_get_dest_addr(struct mv_xor_desc_slot *desc)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	return hw_desc->phy_dest_addr;
+}
+
+static u32 mv_desc_get_src_addr(struct mv_xor_desc_slot *desc,
+				int src_idx)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	return hw_desc->phy_src_addr[src_idx];
+}
+
+
+static void mv_desc_set_byte_count(struct mv_xor_desc_slot *desc,
+				   u32 byte_count)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	hw_desc->byte_count = byte_count;
+}
+
+static void mv_desc_set_next_desc(struct mv_xor_desc_slot *desc,
+				  u32 next_desc_addr)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	BUG_ON(hw_desc->phy_next_desc);
+	hw_desc->phy_next_desc = next_desc_addr;
+}
+
+static void mv_desc_clear_next_desc(struct mv_xor_desc_slot *desc)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	hw_desc->phy_next_desc = 0;
+}
+
+static void mv_desc_set_block_fill_val(struct mv_xor_desc_slot *desc, u32 val)
+{
+	desc->value = val;
+}
+
+static void mv_desc_set_dest_addr(struct mv_xor_desc_slot *desc,
+				  dma_addr_t addr)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	hw_desc->phy_dest_addr = addr;
+}
+
+static int mv_chan_memset_slot_count(size_t len)
+{
+	return 1;
+}
+
+#define mv_chan_memcpy_slot_count(c) mv_chan_memset_slot_count(c)
+
+static void mv_desc_set_src_addr(struct mv_xor_desc_slot *desc,
+				 int index, dma_addr_t addr)
+{
+	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	hw_desc->phy_src_addr[index] = addr;
+	if (desc->type == DMA_XOR)
+		hw_desc->desc_command |= (1 << index);
+}
+
+static u32 mv_chan_get_current_desc(struct mv_xor_chan *chan)
+{
+	return __raw_readl(XOR_CURR_DESC(chan));
+}
+
+static void mv_chan_set_next_descriptor(struct mv_xor_chan *chan,
+					u32 next_desc_addr)
+{
+	__raw_writel(next_desc_addr, XOR_NEXT_DESC(chan));
+}
+
+static void mv_chan_set_dest_pointer(struct mv_xor_chan *chan, u32 desc_addr)
+{
+	__raw_writel(desc_addr, XOR_DEST_POINTER(chan));
+}
+
+static void mv_chan_set_block_size(struct mv_xor_chan *chan, u32 block_size)
+{
+	__raw_writel(block_size, XOR_BLOCK_SIZE(chan));
+}
+
+static void mv_chan_set_value(struct mv_xor_chan *chan, u32 value)
+{
+	__raw_writel(value, XOR_INIT_VALUE_LOW(chan));
+	__raw_writel(value, XOR_INIT_VALUE_HIGH(chan));
+}
+
+static void mv_chan_unmask_interrupts(struct mv_xor_chan *chan)
+{
+	u32 val = __raw_readl(XOR_INTR_MASK(chan));
+	val |= XOR_INTR_MASK_VALUE << (chan->idx * 16);
+	__raw_writel(val, XOR_INTR_MASK(chan));
+}
+
+static u32 mv_chan_get_intr_cause(struct mv_xor_chan *chan)
+{
+	u32 intr_cause = __raw_readl(XOR_INTR_CAUSE(chan));
+	intr_cause = (intr_cause >> (chan->idx * 16)) & 0xFFFF;
+	return intr_cause;
+}
+
+static int mv_is_err_intr(u32 intr_cause)
+{
+	if (intr_cause & ((1<<4)|(1<<5)|(1<<6)|(1<<7)|(1<<8)|(1<<9)))
+		return 1;
+
+	return 0;
+}
+
+static void mv_xor_device_clear_eoc_cause(struct mv_xor_chan *chan)
+{
+	u32 val = (1 << (1 + (chan->idx * 16)));
+	dev_dbg(chan->device->common.dev, "%s, val 0x%08x\n", __func__, val);
+	__raw_writel(val, XOR_INTR_CAUSE(chan));
+}
+
+static void mv_xor_device_clear_err_status(struct mv_xor_chan *chan)
+{
+	u32 val = 0xFFFF0000 >> (chan->idx * 16);
+	__raw_writel(val, XOR_INTR_CAUSE(chan));
+}
+
+static int mv_can_chain(struct mv_xor_desc_slot *desc)
+{
+	struct mv_xor_desc_slot *chain_old_tail = list_entry(
+		desc->chain_node.prev, struct mv_xor_desc_slot, chain_node);
+
+	if (chain_old_tail->type != desc->type)
+		return 0;
+	if (desc->type == DMA_MEMSET)
+		return 0;
+
+	return 1;
+}
+
+static void mv_set_mode(struct mv_xor_chan *chan,
+			       enum dma_transaction_type type)
+{
+	u32 op_mode;
+	u32 config = __raw_readl(XOR_CONFIG(chan));
+
+	switch (type) {
+	case DMA_XOR:
+		op_mode = XOR_OPERATION_MODE_XOR;
+		break;
+	case DMA_MEMCPY:
+		op_mode = XOR_OPERATION_MODE_MEMCPY;
+		break;
+	case DMA_MEMSET:
+		op_mode = XOR_OPERATION_MODE_MEMSET;
+		break;
+	default:
+		dev_printk(KERN_ERR, chan->device->common.dev,
+			   "error: unsupported operation %d.\n",
+			   type);
+		BUG();
+		return;
+	}
+
+	config &= ~0x7;
+	config |= op_mode;
+	__raw_writel(config, XOR_CONFIG(chan));
+	chan->current_type = type;
+}
+
+static void mv_chan_activate(struct mv_xor_chan *chan)
+{
+	u32 activation;
+
+	dev_dbg(chan->device->common.dev, " activate chan.\n");
+	activation = __raw_readl(XOR_ACTIVATION(chan));
+	activation |= 0x1;
+	__raw_writel(activation, XOR_ACTIVATION(chan));
+}
+
+static char mv_chan_is_busy(struct mv_xor_chan *chan)
+{
+	u32 state = __raw_readl(XOR_ACTIVATION(chan));
+
+	state = (state >> 4) & 0x3;
+
+	return (state == 1) ? 1 : 0;
+}
+
+static int mv_chan_xor_slot_count(size_t len, int src_cnt)
+{
+	return 1;
+}
+
+/**
+ * mv_xor_free_slots - flags descriptor slots for reuse
+ * @slot: Slot to free
+ * Caller must hold &mv_chan->lock while calling this function
+ */
+static void mv_xor_free_slots(struct mv_xor_chan *mv_chan,
+			      struct mv_xor_desc_slot *slot)
+{
+	dev_dbg(mv_chan->device->common.dev, "%s %d slot %p\n",
+		__func__, __LINE__, slot);
+
+	slot->slots_per_op = 0;
+
+}
+
+/*
+ * mv_xor_start_new_chain - program the engine to operate on new chain headed by
+ * sw_desc
+ * Caller must hold &mv_chan->lock while calling this function
+ */
+static void mv_xor_start_new_chain(struct mv_xor_chan *mv_chan,
+				   struct mv_xor_desc_slot *sw_desc)
+{
+	dev_dbg(mv_chan->device->common.dev, "%s %d: sw_desc %p\n",
+		__func__, __LINE__, sw_desc);
+	if (sw_desc->type != mv_chan->current_type)
+		mv_set_mode(mv_chan, sw_desc->type);
+
+	if (sw_desc->type == DMA_MEMSET) {
+		/* for memset requests we need to program the engine, no
+		 * descriptors used.
+		 */
+		struct mv_xor_desc *hw_desc = sw_desc->hw_desc;
+		mv_chan_set_dest_pointer(mv_chan, hw_desc->phy_dest_addr);
+		mv_chan_set_block_size(mv_chan, sw_desc->unmap_len);
+		mv_chan_set_value(mv_chan, sw_desc->value);
+	} else {
+		/* set the hardware chain */
+		mv_chan_set_next_descriptor(mv_chan, sw_desc->async_tx.phys);
+	}
+	mv_chan->pending += sw_desc->slot_cnt;
+	mv_xor_issue_pending(&mv_chan->common);
+}
+
+static dma_cookie_t
+mv_xor_run_tx_complete_actions(struct mv_xor_desc_slot *desc,
+	struct mv_xor_chan *mv_chan, dma_cookie_t cookie)
+{
+	BUG_ON(desc->async_tx.cookie < 0);
+
+	if (desc->async_tx.cookie > 0) {
+		cookie = desc->async_tx.cookie;
+
+		/* call the callback (must not sleep or submit new
+		 * operations to this channel)
+		 */
+		if (desc->async_tx.callback)
+			desc->async_tx.callback(
+				desc->async_tx.callback_param);
+
+		/* unmap dma addresses
+		 * (unmap_single vs unmap_page?)
+		 */
+		if (desc->group_head && desc->unmap_len) {
+			struct mv_xor_desc_slot *unmap = desc->group_head;
+			struct device *dev =
+				&mv_chan->device->pdev->dev;
+			u32 len = unmap->unmap_len;
+			enum dma_ctrl_flags flags = desc->async_tx.flags;
+			u32 src_cnt;
+			dma_addr_t addr;
+
+			if (!(flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
+				addr = mv_desc_get_dest_addr(unmap);
+				dma_unmap_page(dev, addr, len, DMA_FROM_DEVICE);
+			}
+
+			if (!(flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
+				src_cnt = unmap->unmap_src_cnt;
+				while (src_cnt--) {
+					addr = mv_desc_get_src_addr(unmap,
+								    src_cnt);
+					dma_unmap_page(dev, addr, len,
+						       DMA_TO_DEVICE);
+				}
+			}
+			desc->group_head = NULL;
+		}
+	}
+
+	/* run dependent operations */
+	async_tx_run_dependencies(&desc->async_tx);
+
+	return cookie;
+}
+
+static int
+mv_xor_clean_completed_slots(struct mv_xor_chan *mv_chan)
+{
+	struct mv_xor_desc_slot *iter, *_iter;
+
+	dev_dbg(mv_chan->device->common.dev, "%s %d\n", __func__, __LINE__);
+	list_for_each_entry_safe(iter, _iter, &mv_chan->completed_slots,
+				 completed_node) {
+
+		if (async_tx_test_ack(&iter->async_tx)) {
+			list_del(&iter->completed_node);
+			mv_xor_free_slots(mv_chan, iter);
+		}
+	}
+	return 0;
+}
+
+static int
+mv_xor_clean_slot(struct mv_xor_desc_slot *desc,
+	struct mv_xor_chan *mv_chan)
+{
+	dev_dbg(mv_chan->device->common.dev, "%s %d: desc %p flags %d\n",
+		__func__, __LINE__, desc, desc->async_tx.flags);
+	list_del(&desc->chain_node);
+	/* the client is allowed to attach dependent operations
+	 * until 'ack' is set
+	 */
+	if (!async_tx_test_ack(&desc->async_tx)) {
+		/* move this slot to the completed_slots */
+		list_add_tail(&desc->completed_node, &mv_chan->completed_slots);
+		return 0;
+	}
+
+	mv_xor_free_slots(mv_chan, desc);
+	return 0;
+}
+
+static void __mv_xor_slot_cleanup(struct mv_xor_chan *mv_chan)
+{
+	struct mv_xor_desc_slot *iter, *_iter;
+	dma_cookie_t cookie = 0;
+	int busy = mv_chan_is_busy(mv_chan);
+	u32 current_desc = mv_chan_get_current_desc(mv_chan);
+	int seen_current = 0;
+
+	dev_dbg(mv_chan->device->common.dev, "%s %d\n", __func__, __LINE__);
+	dev_dbg(mv_chan->device->common.dev, "current_desc %x\n", current_desc);
+	mv_xor_clean_completed_slots(mv_chan);
+
+	/* free completed slots from the chain starting with
+	 * the oldest descriptor
+	 */
+
+	list_for_each_entry_safe(iter, _iter, &mv_chan->chain,
+					chain_node) {
+		prefetch(_iter);
+		prefetch(&_iter->async_tx);
+
+		/* do not advance past the current descriptor loaded into the
+		 * hardware channel, subsequent descriptors are either in
+		 * process or have not been submitted
+		 */
+		if (seen_current)
+			break;
+
+		/* stop the search if we reach the current descriptor and the
+		 * channel is busy
+		 */
+		if (iter->async_tx.phys == current_desc) {
+			seen_current = 1;
+			if (busy)
+				break;
+		}
+
+		cookie = mv_xor_run_tx_complete_actions(iter, mv_chan, cookie);
+
+		if (mv_xor_clean_slot(iter, mv_chan))
+			break;
+	}
+
+	if ((busy == 0) && !list_empty(&mv_chan->chain)) {
+		struct mv_xor_desc_slot *chain_head;
+		chain_head = list_entry(mv_chan->chain.next,
+					struct mv_xor_desc_slot,
+					chain_node);
+
+		mv_xor_start_new_chain(mv_chan, chain_head);
+	}
+
+	if (cookie > 0)
+		mv_chan->completed_cookie = cookie;
+}
+
+static void
+mv_xor_slot_cleanup(struct mv_xor_chan *mv_chan)
+{
+	spin_lock_bh(&mv_chan->lock);
+	__mv_xor_slot_cleanup(mv_chan);
+	spin_unlock_bh(&mv_chan->lock);
+}
+
+static void mv_xor_tasklet(unsigned long data)
+{
+	struct mv_xor_chan *chan = (struct mv_xor_chan *) data;
+	__mv_xor_slot_cleanup(chan);
+}
+
+static struct mv_xor_desc_slot *
+mv_xor_alloc_slots(struct mv_xor_chan *mv_chan, int num_slots,
+		    int slots_per_op)
+{
+	struct mv_xor_desc_slot *iter, *_iter, *alloc_start = NULL;
+	LIST_HEAD(chain);
+	int slots_found, retry = 0;
+
+	/* start search from the last allocated descrtiptor
+	 * if a contiguous allocation can not be found start searching
+	 * from the beginning of the list
+	 */
+retry:
+	slots_found = 0;
+	if (retry == 0)
+		iter = mv_chan->last_used;
+	else
+		iter = list_entry(&mv_chan->all_slots,
+			struct mv_xor_desc_slot,
+			slot_node);
+
+	list_for_each_entry_safe_continue(
+		iter, _iter, &mv_chan->all_slots, slot_node) {
+		prefetch(_iter);
+		prefetch(&_iter->async_tx);
+		if (iter->slots_per_op) {
+			/* give up after finding the first busy slot
+			 * on the second pass through the list
+			 */
+			if (retry)
+				break;
+
+			slots_found = 0;
+			continue;
+		}
+
+		/* start the allocation if the slot is correctly aligned */
+		if (!slots_found++)
+			alloc_start = iter;
+
+		if (slots_found == num_slots) {
+			struct mv_xor_desc_slot *alloc_tail = NULL;
+			struct mv_xor_desc_slot *last_used = NULL;
+			iter = alloc_start;
+			while (num_slots) {
+				int i;
+
+				/* pre-ack all but the last descriptor */
+				async_tx_ack(&iter->async_tx);
+
+				list_add_tail(&iter->chain_node, &chain);
+				alloc_tail = iter;
+				iter->async_tx.cookie = 0;
+				iter->slot_cnt = num_slots;
+				iter->xor_check_result = NULL;
+				for (i = 0; i < slots_per_op; i++) {
+					iter->slots_per_op = slots_per_op - i;
+					last_used = iter;
+					iter = list_entry(iter->slot_node.next,
+						struct mv_xor_desc_slot,
+						slot_node);
+				}
+				num_slots -= slots_per_op;
+			}
+			alloc_tail->group_head = alloc_start;
+			alloc_tail->async_tx.cookie = -EBUSY;
+			list_splice(&chain, &alloc_tail->async_tx.tx_list);
+			mv_chan->last_used = last_used;
+			mv_desc_clear_next_desc(alloc_start);
+			mv_desc_clear_next_desc(alloc_tail);
+			return alloc_tail;
+		}
+	}
+	if (!retry++)
+		goto retry;
+
+	/* try to free some slots if the allocation fails */
+	tasklet_schedule(&mv_chan->irq_tasklet);
+
+	return NULL;
+}
+
+static dma_cookie_t
+mv_desc_assign_cookie(struct mv_xor_chan *mv_chan,
+		      struct mv_xor_desc_slot *desc)
+{
+	dma_cookie_t cookie = mv_chan->common.cookie;
+
+	if (++cookie < 0)
+		cookie = 1;
+	mv_chan->common.cookie = desc->async_tx.cookie = cookie;
+	return cookie;
+}
+
+/************************ DMA engine API functions ****************************/
+static dma_cookie_t
+mv_xor_tx_submit(struct dma_async_tx_descriptor *tx)
+{
+	struct mv_xor_desc_slot *sw_desc = to_mv_xor_slot(tx);
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(tx->chan);
+	struct mv_xor_desc_slot *grp_start, *old_chain_tail;
+	dma_cookie_t cookie;
+	int new_hw_chain = 1;
+
+	dev_dbg(mv_chan->device->common.dev,
+		"%s sw_desc %p: async_tx %p\n",
+		__func__, sw_desc, &sw_desc->async_tx);
+
+	grp_start = sw_desc->group_head;
+
+	spin_lock_bh(&mv_chan->lock);
+	cookie = mv_desc_assign_cookie(mv_chan, sw_desc);
+
+	if (list_empty(&mv_chan->chain))
+		list_splice_init(&sw_desc->async_tx.tx_list, &mv_chan->chain);
+	else {
+		new_hw_chain = 0;
+
+		old_chain_tail = list_entry(mv_chan->chain.prev,
+					    struct mv_xor_desc_slot,
+					    chain_node);
+		list_splice_init(&grp_start->async_tx.tx_list,
+				 &old_chain_tail->chain_node);
+
+		if (!mv_can_chain(grp_start))
+			goto submit_done;
+
+		dev_dbg(mv_chan->device->common.dev, "Append to last desc %x\n",
+			old_chain_tail->async_tx.phys);
+
+		/* fix up the hardware chain */
+		mv_desc_set_next_desc(old_chain_tail, grp_start->async_tx.phys);
+
+		/* if the channel is not busy */
+		if (!mv_chan_is_busy(mv_chan)) {
+			u32 current_desc = mv_chan_get_current_desc(mv_chan);
+			/*
+			 * and the curren desc is the end of the chain before
+			 * the append, then we need to start the channel
+			 */
+			if (current_desc == old_chain_tail->async_tx.phys)
+				new_hw_chain = 1;
+		}
+	}
+
+	if (new_hw_chain)
+		mv_xor_start_new_chain(mv_chan, grp_start);
+
+submit_done:
+	spin_unlock_bh(&mv_chan->lock);
+
+	return cookie;
+}
+
+/* returns the number of allocated descriptors */
+static int mv_xor_alloc_chan_resources(struct dma_chan *chan,
+				       struct dma_client *client)
+{
+	char *hw_desc;
+	int idx;
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+	struct mv_xor_desc_slot *slot = NULL;
+	struct mv_xor_platform_data *plat_data =
+		mv_chan->device->pdev->dev.platform_data;
+	int num_descs_in_pool = plat_data->pool_size/MV_XOR_SLOT_SIZE;
+
+	/* Allocate descriptor slots */
+	idx = mv_chan->slots_allocated;
+	while (idx < num_descs_in_pool) {
+		slot = kzalloc(sizeof(*slot), GFP_KERNEL);
+		if (!slot) {
+			printk(KERN_INFO "MV XOR Channel only initialized"
+				" %d descriptor slots", idx);
+			break;
+		}
+		hw_desc = (char *) mv_chan->device->dma_desc_pool_virt;
+		slot->hw_desc = (void *) &hw_desc[idx * MV_XOR_SLOT_SIZE];
+
+		dma_async_tx_descriptor_init(&slot->async_tx, chan);
+		slot->async_tx.tx_submit = mv_xor_tx_submit;
+		INIT_LIST_HEAD(&slot->chain_node);
+		INIT_LIST_HEAD(&slot->slot_node);
+		INIT_LIST_HEAD(&slot->async_tx.tx_list);
+		hw_desc = (char *) mv_chan->device->dma_desc_pool;
+		slot->async_tx.phys =
+			(dma_addr_t) &hw_desc[idx * MV_XOR_SLOT_SIZE];
+		slot->idx = idx++;
+
+		spin_lock_bh(&mv_chan->lock);
+		mv_chan->slots_allocated = idx;
+		list_add_tail(&slot->slot_node, &mv_chan->all_slots);
+		spin_unlock_bh(&mv_chan->lock);
+	}
+
+	if (mv_chan->slots_allocated && !mv_chan->last_used)
+		mv_chan->last_used = list_entry(mv_chan->all_slots.next,
+					struct mv_xor_desc_slot,
+					slot_node);
+
+	dev_dbg(mv_chan->device->common.dev,
+		"allocated %d descriptor slots last_used: %p\n",
+		mv_chan->slots_allocated, mv_chan->last_used);
+
+	return mv_chan->slots_allocated ? : -ENOMEM;
+}
+
+static struct dma_async_tx_descriptor *
+mv_xor_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+		size_t len, unsigned long flags)
+{
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+	struct mv_xor_desc_slot *sw_desc, *grp_start;
+	int slot_cnt;
+
+	dev_dbg(mv_chan->device->common.dev,
+		"%s dest: %x src %x len: %u flags: %ld\n",
+		__func__, dest, src, len, flags);
+	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
+		return NULL;
+
+	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+
+	spin_lock_bh(&mv_chan->lock);
+	slot_cnt = mv_chan_memcpy_slot_count(len);
+	sw_desc = mv_xor_alloc_slots(mv_chan, slot_cnt, 1);
+	if (sw_desc) {
+		sw_desc->type = DMA_MEMCPY;
+		sw_desc->async_tx.flags = flags;
+		grp_start = sw_desc->group_head;
+		mv_desc_init(grp_start, flags);
+		mv_desc_set_byte_count(grp_start, len);
+		mv_desc_set_dest_addr(sw_desc->group_head, dest);
+		mv_desc_set_src_addr(grp_start, 0, src);
+		sw_desc->unmap_src_cnt = 1;
+		sw_desc->unmap_len = len;
+	}
+	spin_unlock_bh(&mv_chan->lock);
+
+	dev_dbg(mv_chan->device->common.dev,
+		"%s sw_desc %p async_tx %p\n",
+		__func__, sw_desc, sw_desc ? &sw_desc->async_tx : 0);
+
+	return sw_desc ? &sw_desc->async_tx : NULL;
+}
+
+static struct dma_async_tx_descriptor *
+mv_xor_prep_dma_memset(struct dma_chan *chan, dma_addr_t dest, int value,
+		       size_t len, unsigned long flags)
+{
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+	struct mv_xor_desc_slot *sw_desc, *grp_start;
+	int slot_cnt;
+
+	dev_dbg(mv_chan->device->common.dev,
+		"%s dest: %x len: %u flags: %ld\n",
+		__func__, dest, len, flags);
+	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
+		return NULL;
+
+	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+
+	spin_lock_bh(&mv_chan->lock);
+	slot_cnt = mv_chan_memset_slot_count(len);
+	sw_desc = mv_xor_alloc_slots(mv_chan, slot_cnt, 1);
+	if (sw_desc) {
+		sw_desc->type = DMA_MEMSET;
+		sw_desc->async_tx.flags = flags;
+		grp_start = sw_desc->group_head;
+		mv_desc_init(grp_start, flags);
+		mv_desc_set_byte_count(grp_start, len);
+		mv_desc_set_dest_addr(sw_desc->group_head, dest);
+		mv_desc_set_block_fill_val(grp_start, value);
+		sw_desc->unmap_src_cnt = 1;
+		sw_desc->unmap_len = len;
+	}
+	spin_unlock_bh(&mv_chan->lock);
+	dev_dbg(mv_chan->device->common.dev,
+		"%s sw_desc %p async_tx %p \n",
+		__func__, sw_desc, &sw_desc->async_tx);
+	return sw_desc ? &sw_desc->async_tx : NULL;
+}
+
+static struct dma_async_tx_descriptor *
+mv_xor_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_chan *mv_chan = to_mv_xor_chan(chan);
+	struct mv_xor_desc_slot *sw_desc, *grp_start;
+	int slot_cnt;
+
+	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
+		return NULL;
+
+	BUG_ON(unlikely(len > MV_XOR_MAX_BYTE_COUNT));
+
+	dev_dbg(mv_chan->device->common.dev,
+		"%s src_cnt: %d len: dest %x %u flags: %ld\n",
+		__func__, src_cnt, len, dest, flags);
+
+	spin_lock_bh(&mv_chan->lock);
+	slot_cnt = mv_chan_xor_slot_count(len, src_cnt);
+	sw_desc = mv_xor_alloc_slots(mv_chan, slot_cnt, 1);
+	if (sw_desc) {
+		sw_desc->type = DMA_XOR;
+		sw_desc->async_tx.flags = flags;
+		grp_start = sw_desc->group_head;
+		mv_desc_init(grp_start, flags);
+		/* the byte count field is the same as in memcpy desc*/
+		mv_desc_set_byte_count(grp_start, len);
+		mv_desc_set_dest_addr(sw_desc->group_head, dest);
+		sw_desc->unmap_src_cnt = src_cnt;
+		sw_desc->unmap_len = len;
+		while (src_cnt--)
+			mv_desc_set_src_addr(grp_start, src_cnt, src[src_cnt]);
+	}
+	spin_unlock_bh(&mv_chan->lock);
+	dev_dbg(mv_chan->device->common.dev,
+		"%s sw_desc %p async_tx %p \n",
+		__func__, sw_desc, &sw_desc->async_tx);
+	return sw_desc ? &sw_desc->async_tx : NULL;
+}
+
+static void mv_xor_free_chan_resources(struct dma_chan *chan)
+{
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+	struct mv_xor_desc_slot *iter, *_iter;
+	int in_use_descs = 0;
+
+	mv_xor_slot_cleanup(mv_chan);
+
+	spin_lock_bh(&mv_chan->lock);
+	list_for_each_entry_safe(iter, _iter, &mv_chan->chain,
+					chain_node) {
+		in_use_descs++;
+		list_del(&iter->chain_node);
+	}
+	list_for_each_entry_safe(iter, _iter, &mv_chan->completed_slots,
+				 completed_node) {
+		in_use_descs++;
+		list_del(&iter->completed_node);
+	}
+	list_for_each_entry_safe_reverse(
+		iter, _iter, &mv_chan->all_slots, slot_node) {
+		list_del(&iter->slot_node);
+		kfree(iter);
+		mv_chan->slots_allocated--;
+	}
+	mv_chan->last_used = NULL;
+
+	dev_dbg(mv_chan->device->common.dev, "%s slots_allocated %d\n",
+		__func__, mv_chan->slots_allocated);
+	spin_unlock_bh(&mv_chan->lock);
+
+	if (in_use_descs)
+		dev_err(mv_chan->device->common.dev,
+			"freeing %d in use descriptors!\n", in_use_descs);
+}
+
+/**
+ * mv_xor_is_complete - poll the status of an XOR transaction
+ * @chan: XOR channel handle
+ * @cookie: XOR transaction identifier
+ */
+static enum dma_status mv_xor_is_complete(struct dma_chan *chan,
+					  dma_cookie_t cookie,
+					  dma_cookie_t *done,
+					  dma_cookie_t *used)
+{
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+	dma_cookie_t last_used;
+	dma_cookie_t last_complete;
+	enum dma_status ret;
+
+	last_used = chan->cookie;
+	last_complete = mv_chan->completed_cookie;
+	mv_chan->is_complete_cookie = cookie;
+	if (done)
+		*done = last_complete;
+	if (used)
+		*used = last_used;
+
+	ret = dma_async_is_complete(cookie, last_complete, last_used);
+	if (ret == DMA_SUCCESS) {
+		mv_xor_clean_completed_slots(mv_chan);
+		return ret;
+	}
+	mv_xor_slot_cleanup(mv_chan);
+
+	last_used = chan->cookie;
+	last_complete = mv_chan->completed_cookie;
+
+	if (done)
+		*done = last_complete;
+	if (used)
+		*used = last_used;
+
+	return dma_async_is_complete(cookie, last_complete, last_used);
+}
+
+static void mv_dump_xor_regs(struct mv_xor_chan *chan)
+{
+	u32 val;
+
+	val = __raw_readl(XOR_CONFIG(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "config       0x%08x.\n", val);
+
+	val = __raw_readl(XOR_ACTIVATION(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "activation   0x%08x.\n", val);
+
+	val = __raw_readl(XOR_INTR_CAUSE(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "intr cause   0x%08x.\n", val);
+
+	val = __raw_readl(XOR_INTR_MASK(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "intr mask    0x%08x.\n", val);
+
+	val = __raw_readl(XOR_ERROR_CAUSE(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "error cause  0x%08x.\n", val);
+
+	val = __raw_readl(XOR_ERROR_ADDR(chan));
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "error addr   0x%08x.\n", val);
+}
+
+static void mv_xor_err_interrupt_handler(struct mv_xor_chan *chan,
+					 u32 intr_cause)
+{
+	if (intr_cause & (1 << 4)) {
+	     dev_dbg(chan->device->common.dev,
+		     "ignore this error\n");
+	     return;
+	}
+
+	dev_printk(KERN_ERR, chan->device->common.dev,
+		   "error on chan %d. intr cause 0x%08x.\n",
+		   chan->idx, intr_cause);
+
+	mv_dump_xor_regs(chan);
+	BUG();
+}
+
+static irqreturn_t mv_xor_interrupt_handler(int irq, void *data)
+{
+	struct mv_xor_chan *chan = data;
+	u32 intr_cause = mv_chan_get_intr_cause(chan);
+
+	dev_dbg(chan->device->common.dev, "intr cause %x\n", intr_cause);
+
+	if (mv_is_err_intr(intr_cause))
+		mv_xor_err_interrupt_handler(chan, intr_cause);
+
+	tasklet_schedule(&chan->irq_tasklet);
+
+	mv_xor_device_clear_eoc_cause(chan);
+
+	return IRQ_HANDLED;
+}
+
+static void mv_xor_issue_pending(struct dma_chan *chan)
+{
+	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
+
+	if (mv_chan->pending >= MV_XOR_THRESHOLD) {
+		mv_chan->pending = 0;
+		mv_chan_activate(mv_chan);
+	}
+}
+
+/*
+ * Perform a transaction to verify the HW works.
+ */
+#define MV_XOR_TEST_SIZE 2000
+
+static int __devinit mv_xor_memcpy_self_test(struct mv_xor_device *device)
+{
+	int i;
+	void *src, *dest;
+	dma_addr_t src_dma, dest_dma;
+	struct dma_chan *dma_chan;
+	dma_cookie_t cookie;
+	struct dma_async_tx_descriptor *tx;
+	int err = 0;
+	struct mv_xor_chan *mv_chan;
+
+	src = kmalloc(sizeof(u8) * MV_XOR_TEST_SIZE, GFP_KERNEL);
+	if (!src)
+		return -ENOMEM;
+
+	dest = kzalloc(sizeof(u8) * MV_XOR_TEST_SIZE, GFP_KERNEL);
+	if (!dest) {
+		kfree(src);
+		return -ENOMEM;
+	}
+
+	/* Fill in src buffer */
+	for (i = 0; i < MV_XOR_TEST_SIZE; i++)
+		((u8 *) src)[i] = (u8)i;
+
+	/* Start copy, using first DMA channel */
+	dma_chan = container_of(device->common.channels.next,
+				struct dma_chan,
+				device_node);
+	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
+		err = -ENODEV;
+		goto out;
+	}
+
+	dest_dma = dma_map_single(dma_chan->device->dev, dest,
+				  MV_XOR_TEST_SIZE, DMA_FROM_DEVICE);
+
+	src_dma = dma_map_single(dma_chan->device->dev, src,
+				 MV_XOR_TEST_SIZE, DMA_TO_DEVICE);
+
+	tx = mv_xor_prep_dma_memcpy(dma_chan, dest_dma, src_dma,
+				    MV_XOR_TEST_SIZE, 0);
+	cookie = mv_xor_tx_submit(tx);
+	mv_xor_issue_pending(dma_chan);
+	async_tx_ack(tx);
+	msleep(1);
+
+	if (mv_xor_is_complete(dma_chan, cookie, NULL, NULL) !=
+	    DMA_SUCCESS) {
+		dev_printk(KERN_ERR, dma_chan->device->dev,
+			   "Self-test copy timed out, disabling\n");
+		err = -ENODEV;
+		goto free_resources;
+	}
+
+	mv_chan = to_mv_xor_chan(dma_chan);
+	dma_sync_single_for_cpu(&mv_chan->device->pdev->dev, dest_dma,
+				MV_XOR_TEST_SIZE, DMA_FROM_DEVICE);
+	if (memcmp(src, dest, MV_XOR_TEST_SIZE)) {
+		dev_printk(KERN_ERR, dma_chan->device->dev,
+			   "Self-test copy failed compare, disabling\n");
+		err = -ENODEV;
+		goto free_resources;
+	}
+
+free_resources:
+	mv_xor_free_chan_resources(dma_chan);
+out:
+	kfree(src);
+	kfree(dest);
+	return err;
+}
+
+#define MV_XOR_NUM_SRC_TEST 4 /* must be <= 15 */
+static int __devinit
+mv_xor_xor_self_test(struct mv_xor_device *device)
+{
+	int i, src_idx;
+	struct page *dest;
+	struct page *xor_srcs[MV_XOR_NUM_SRC_TEST];
+	dma_addr_t dma_srcs[MV_XOR_NUM_SRC_TEST];
+	dma_addr_t dest_dma;
+	struct dma_async_tx_descriptor *tx;
+	struct dma_chan *dma_chan;
+	dma_cookie_t cookie;
+	u8 cmp_byte = 0;
+	u32 cmp_word;
+	int err = 0;
+	struct mv_xor_chan *mv_chan;
+
+	for (src_idx = 0; src_idx < MV_XOR_NUM_SRC_TEST; src_idx++) {
+		xor_srcs[src_idx] = alloc_page(GFP_KERNEL);
+		if (!xor_srcs[src_idx])
+			while (src_idx--) {
+				__free_page(xor_srcs[src_idx]);
+				return -ENOMEM;
+			}
+	}
+
+	dest = alloc_page(GFP_KERNEL);
+	if (!dest)
+		while (src_idx--) {
+			__free_page(xor_srcs[src_idx]);
+			return -ENOMEM;
+		}
+
+	/* Fill in src buffers */
+	for (src_idx = 0; src_idx < MV_XOR_NUM_SRC_TEST; src_idx++) {
+		u8 *ptr = page_address(xor_srcs[src_idx]);
+		for (i = 0; i < PAGE_SIZE; i++)
+			ptr[i] = (1 << src_idx);
+	}
+
+	for (src_idx = 0; src_idx < MV_XOR_NUM_SRC_TEST; src_idx++)
+		cmp_byte ^= (u8) (1 << src_idx);
+
+	cmp_word = (cmp_byte << 24) | (cmp_byte << 16) |
+		(cmp_byte << 8) | cmp_byte;
+
+	memset(page_address(dest), 0, PAGE_SIZE);
+
+	dma_chan = container_of(device->common.channels.next,
+				struct dma_chan,
+				device_node);
+	if (mv_xor_alloc_chan_resources(dma_chan, NULL) < 1) {
+		err = -ENODEV;
+		goto out;
+	}
+
+	/* test xor */
+	dest_dma = dma_map_page(dma_chan->device->dev, dest, 0, PAGE_SIZE,
+				DMA_FROM_DEVICE);
+
+	for (i = 0; i < MV_XOR_NUM_SRC_TEST; i++)
+		dma_srcs[i] = dma_map_page(dma_chan->device->dev, xor_srcs[i],
+					   0, PAGE_SIZE, DMA_TO_DEVICE);
+
+	tx = mv_xor_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
+				 MV_XOR_NUM_SRC_TEST, PAGE_SIZE, 0);
+
+	cookie = mv_xor_tx_submit(tx);
+	mv_xor_issue_pending(dma_chan);
+	async_tx_ack(tx);
+	msleep(8);
+
+	if (mv_xor_is_complete(dma_chan, cookie, NULL, NULL) !=
+	    DMA_SUCCESS) {
+		dev_printk(KERN_ERR, dma_chan->device->dev,
+			   "Self-test xor timed out, disabling\n");
+		err = -ENODEV;
+		goto free_resources;
+	}
+
+	mv_chan = to_mv_xor_chan(dma_chan);
+	dma_sync_single_for_cpu(&mv_chan->device->pdev->dev, dest_dma,
+				PAGE_SIZE, DMA_FROM_DEVICE);
+	for (i = 0; i < (PAGE_SIZE / sizeof(u32)); i++) {
+		u32 *ptr = page_address(dest);
+		if (ptr[i] != cmp_word) {
+			dev_printk(KERN_ERR, dma_chan->device->dev,
+				   "Self-test xor failed compare, disabling."
+				   " index %d, data %x, expected %x\n", i,
+				   ptr[i], cmp_word);
+			err = -ENODEV;
+			goto free_resources;
+		}
+	}
+
+free_resources:
+	mv_xor_free_chan_resources(dma_chan);
+out:
+	src_idx = MV_XOR_NUM_SRC_TEST;
+	while (src_idx--)
+		__free_page(xor_srcs[src_idx]);
+	__free_page(dest);
+	return err;
+}
+
+static int __devexit mv_xor_remove(struct platform_device *dev)
+{
+	struct mv_xor_device *device = platform_get_drvdata(dev);
+	struct dma_chan *chan, *_chan;
+	struct mv_xor_chan *mv_chan;
+	struct mv_xor_platform_data *plat_data = dev->dev.platform_data;
+
+	dma_async_device_unregister(&device->common);
+
+	dma_free_coherent(&dev->dev, plat_data->pool_size,
+			device->dma_desc_pool_virt, device->dma_desc_pool);
+
+	list_for_each_entry_safe(chan, _chan, &device->common.channels,
+				device_node) {
+		mv_chan = to_mv_xor_chan(chan);
+		list_del(&chan->device_node);
+	}
+
+	return 0;
+}
+
+static int __devinit mv_xor_probe(struct platform_device *pdev)
+{
+	int ret = 0;
+	int irq;
+	struct mv_xor_device *adev;
+	struct mv_xor_chan *mv_chan;
+	struct dma_device *dma_dev;
+	struct mv_xor_platform_data *plat_data = pdev->dev.platform_data;
+
+
+	adev = devm_kzalloc(&pdev->dev, sizeof(*adev), GFP_KERNEL);
+	if (!adev)
+		return -ENOMEM;
+
+	dma_dev = &adev->common;
+
+	/* allocate coherent memory for hardware descriptors
+	 * note: writecombine gives slightly better performance, but
+	 * requires that we explicitly flush the writes
+	 */
+	adev->dma_desc_pool_virt = dma_alloc_writecombine(&pdev->dev,
+							  plat_data->pool_size,
+							  &adev->dma_desc_pool,
+							  GFP_KERNEL);
+	if (!adev->dma_desc_pool_virt)
+		return -ENOMEM;
+
+	adev->id = plat_data->hw_id;
+
+	/* discover transaction capabilites from the platform data */
+	dma_dev->cap_mask = plat_data->cap_mask;
+	adev->pdev = pdev;
+	platform_set_drvdata(pdev, adev);
+
+	adev->shared = platform_get_drvdata(plat_data->shared);
+
+	INIT_LIST_HEAD(&dma_dev->channels);
+
+	/* set base routines */
+	dma_dev->device_alloc_chan_resources = mv_xor_alloc_chan_resources;
+	dma_dev->device_free_chan_resources = mv_xor_free_chan_resources;
+	dma_dev->device_is_tx_complete = mv_xor_is_complete;
+	dma_dev->device_issue_pending = mv_xor_issue_pending;
+	dma_dev->dev = &pdev->dev;
+
+	/* set prep routines based on capability */
+	if (dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask))
+		dma_dev->device_prep_dma_memcpy = mv_xor_prep_dma_memcpy;
+	if (dma_has_cap(DMA_MEMSET, dma_dev->cap_mask))
+		dma_dev->device_prep_dma_memset = mv_xor_prep_dma_memset;
+	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask)) {
+		dma_dev->max_xor = 8;                  ;
+		dma_dev->device_prep_dma_xor = mv_xor_prep_dma_xor;
+	}
+
+	mv_chan = devm_kzalloc(&pdev->dev, sizeof(*mv_chan), GFP_KERNEL);
+	if (!mv_chan) {
+		ret = -ENOMEM;
+		goto err_free_dma;
+	}
+	mv_chan->device = adev;
+	mv_chan->idx = plat_data->hw_id;
+	mv_chan->mmr_base = adev->shared->xor_base;
+
+	if (!mv_chan->mmr_base) {
+		ret = -ENOMEM;
+		goto err_free_dma;
+	}
+	tasklet_init(&mv_chan->irq_tasklet, mv_xor_tasklet, (unsigned long)
+		     mv_chan);
+
+	/* clear errors before enabling interrupts */
+	mv_xor_device_clear_err_status(mv_chan);
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		ret = irq;
+		goto err_free_dma;
+	}
+	ret = devm_request_irq(&pdev->dev, irq,
+			       mv_xor_interrupt_handler,
+			       0, dev_name(&pdev->dev), mv_chan);
+	if (ret)
+		goto err_free_dma;
+
+	mv_chan_unmask_interrupts(mv_chan);
+
+	mv_set_mode(mv_chan, DMA_MEMCPY);
+
+	spin_lock_init(&mv_chan->lock);
+	INIT_LIST_HEAD(&mv_chan->chain);
+	INIT_LIST_HEAD(&mv_chan->completed_slots);
+	INIT_LIST_HEAD(&mv_chan->all_slots);
+	INIT_RCU_HEAD(&mv_chan->common.rcu);
+	mv_chan->common.device = dma_dev;
+
+	list_add_tail(&mv_chan->common.device_node, &dma_dev->channels);
+
+	if (dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask)) {
+		ret = mv_xor_memcpy_self_test(adev);
+		dev_dbg(&pdev->dev, "memcpy self test returned %d\n", ret);
+		if (ret)
+			goto err_free_dma;
+	}
+
+	if (dma_has_cap(DMA_XOR, dma_dev->cap_mask)) {
+		ret = mv_xor_xor_self_test(adev);
+		dev_dbg(&pdev->dev, "xor self test returned %d\n", ret);
+		if (ret)
+			goto err_free_dma;
+	}
+
+	dev_printk(KERN_INFO, &pdev->dev, "Marvell XOR: "
+	  "( %s%s%s%s)\n",
+	  dma_has_cap(DMA_XOR, dma_dev->cap_mask) ? "xor " : "",
+	  dma_has_cap(DMA_MEMSET, dma_dev->cap_mask)  ? "fill " : "",
+	  dma_has_cap(DMA_MEMCPY, dma_dev->cap_mask) ? "cpy " : "",
+	  dma_has_cap(DMA_INTERRUPT, dma_dev->cap_mask) ? "intr " : "");
+
+	dma_async_device_register(dma_dev);
+	goto out;
+
+ err_free_dma:
+	dma_free_coherent(&adev->pdev->dev, plat_data->pool_size,
+			adev->dma_desc_pool_virt, adev->dma_desc_pool);
+ out:
+	return ret;
+}
+
+static void
+mv_xor_conf_mbus_windows(struct mv_xor_shared_private *msp,
+			 struct mbus_dram_target_info *dram)
+{
+	void __iomem *base = msp->xor_base;
+	u32 win_enable = 0;
+	int i;
+
+	for (i = 0; i < 8; i++) {
+		writel(0, base + WINDOW_BASE(i));
+		writel(0, base + WINDOW_SIZE(i));
+		if (i < 4)
+			writel(0, base + WINDOW_REMAP_HIGH(i));
+	}
+
+	for (i = 0; i < dram->num_cs; i++) {
+		struct mbus_dram_window *cs = dram->cs + i;
+
+		writel((cs->base & 0xffff0000) |
+		       (cs->mbus_attr << 8) |
+		       dram->mbus_dram_target_id, base + WINDOW_BASE(i));
+		writel((cs->size - 1) & 0xffff0000, base + WINDOW_SIZE(i));
+
+		win_enable |= (1 << i);
+		win_enable |= 3 << (16 + (2 * i));
+	}
+
+	writel(win_enable, base + WINDOW_BAR_ENABLE(0));
+	writel(win_enable, base + WINDOW_BAR_ENABLE(1));
+}
+
+static struct platform_driver mv_xor_driver = {
+	.probe		= mv_xor_probe,
+	.remove		= mv_xor_remove,
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= MV_XOR_NAME,
+	},
+};
+
+static int mv_xor_shared_probe(struct platform_device *pdev)
+{
+	struct mv_xor_platform_shared_data *msd = pdev->dev.platform_data;
+	struct mv_xor_shared_private *msp;
+	struct resource *res;
+
+	dev_printk(KERN_NOTICE, &pdev->dev, "Marvell shared XOR driver\n");
+
+	msp = devm_kzalloc(&pdev->dev, sizeof(*msp), GFP_KERNEL);
+	if (!msp)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENODEV;
+
+	msp->xor_base = devm_ioremap(&pdev->dev, res->start,
+				     res->end - res->start + 1);
+	if (!msp->xor_base)
+		return -EBUSY;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (!res)
+		return -ENODEV;
+
+	msp->xor_high_base = devm_ioremap(&pdev->dev, res->start,
+					  res->end - res->start + 1);
+	if (!msp->xor_high_base)
+		return -EBUSY;
+
+	platform_set_drvdata(pdev, msp);
+
+	/*
+	 * (Re-)program MBUS remapping windows if we are asked to.
+	 */
+	if (msd != NULL && msd->dram != NULL)
+		mv_xor_conf_mbus_windows(msp, msd->dram);
+
+	return 0;
+}
+
+static int mv_xor_shared_remove(struct platform_device *pdev)
+{
+	return 0;
+}
+
+static struct platform_driver mv_xor_shared_driver = {
+	.probe		= mv_xor_shared_probe,
+	.remove		= mv_xor_shared_remove,
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= MV_XOR_SHARED_NAME,
+	},
+};
+
+
+static int __init mv_xor_init(void)
+{
+	int rc;
+
+	rc = platform_driver_register(&mv_xor_shared_driver);
+	if (!rc) {
+		rc = platform_driver_register(&mv_xor_driver);
+		if (rc)
+			platform_driver_unregister(&mv_xor_shared_driver);
+	}
+	return rc;
+}
+module_init(mv_xor_init);
+
+/* it's currently unsafe to unload this module */
+#if 0
+static void __exit mv_xor_exit(void)
+{
+	platform_driver_unregister(&mv_xor_driver);
+	platform_driver_unregister(&mv_xor_shared_driver);
+	return;
+}
+
+module_exit(mv_xor_exit);
+#endif
+
+MODULE_AUTHOR("Saeed Bishara <saeed@marvell.com>");
+MODULE_DESCRIPTION("DMA engine driver for Marvell's XOR engine");
+MODULE_LICENSE("GPL");
diff --git a/drivers/dma/mv_xor.h b/drivers/dma/mv_xor.h
new file mode 100644
index 000000000000..06cafe1ef521
--- /dev/null
+++ b/drivers/dma/mv_xor.h
@@ -0,0 +1,183 @@
+/*
+ * Copyright (C) 2007, 2008, Marvell International Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#ifndef MV_XOR_H
+#define MV_XOR_H
+
+#include <linux/types.h>
+#include <linux/io.h>
+#include <linux/dmaengine.h>
+#include <linux/interrupt.h>
+
+#define USE_TIMER
+#define MV_XOR_SLOT_SIZE		64
+#define MV_XOR_THRESHOLD		1
+
+#define XOR_OPERATION_MODE_XOR		0
+#define XOR_OPERATION_MODE_MEMCPY	2
+#define XOR_OPERATION_MODE_MEMSET	4
+
+#define XOR_CURR_DESC(chan)	(chan->mmr_base + 0x210 + (chan->idx * 4))
+#define XOR_NEXT_DESC(chan)	(chan->mmr_base + 0x200 + (chan->idx * 4))
+#define XOR_BYTE_COUNT(chan)	(chan->mmr_base + 0x220 + (chan->idx * 4))
+#define XOR_DEST_POINTER(chan)	(chan->mmr_base + 0x2B0 + (chan->idx * 4))
+#define XOR_BLOCK_SIZE(chan)	(chan->mmr_base + 0x2C0 + (chan->idx * 4))
+#define XOR_INIT_VALUE_LOW(chan)	(chan->mmr_base + 0x2E0)
+#define XOR_INIT_VALUE_HIGH(chan)	(chan->mmr_base + 0x2E4)
+
+#define XOR_CONFIG(chan)	(chan->mmr_base + 0x10 + (chan->idx * 4))
+#define XOR_ACTIVATION(chan)	(chan->mmr_base + 0x20 + (chan->idx * 4))
+#define XOR_INTR_CAUSE(chan)	(chan->mmr_base + 0x30)
+#define XOR_INTR_MASK(chan)	(chan->mmr_base + 0x40)
+#define XOR_ERROR_CAUSE(chan)	(chan->mmr_base + 0x50)
+#define XOR_ERROR_ADDR(chan)	(chan->mmr_base + 0x60)
+#define XOR_INTR_MASK_VALUE	0x3F5
+
+#define WINDOW_BASE(w)		(0x250 + ((w) << 2))
+#define WINDOW_SIZE(w)		(0x270 + ((w) << 2))
+#define WINDOW_REMAP_HIGH(w)	(0x290 + ((w) << 2))
+#define WINDOW_BAR_ENABLE(chan)	(0x240 + ((chan) << 2))
+
+struct mv_xor_shared_private {
+	void __iomem	*xor_base;
+	void __iomem	*xor_high_base;
+};
+
+
+/**
+ * struct mv_xor_device - internal representation of a XOR device
+ * @pdev: Platform device
+ * @id: HW XOR Device selector
+ * @dma_desc_pool: base of DMA descriptor region (DMA address)
+ * @dma_desc_pool_virt: base of DMA descriptor region (CPU address)
+ * @common: embedded struct dma_device
+ */
+struct mv_xor_device {
+	struct platform_device		*pdev;
+	int				id;
+	dma_addr_t			dma_desc_pool;
+	void				*dma_desc_pool_virt;
+	struct dma_device		common;
+	struct mv_xor_shared_private	*shared;
+};
+
+/**
+ * struct mv_xor_chan - internal representation of a XOR channel
+ * @pending: allows batching of hardware operations
+ * @completed_cookie: identifier for the most recently completed operation
+ * @lock: serializes enqueue/dequeue operations to the descriptors pool
+ * @mmr_base: memory mapped register base
+ * @idx: the index of the xor channel
+ * @chain: device chain view of the descriptors
+ * @completed_slots: slots completed by HW but still need to be acked
+ * @device: parent device
+ * @common: common dmaengine channel object members
+ * @last_used: place holder for allocation to continue from where it left off
+ * @all_slots: complete domain of slots usable by the channel
+ * @slots_allocated: records the actual size of the descriptor slot pool
+ * @irq_tasklet: bottom half where mv_xor_slot_cleanup runs
+ */
+struct mv_xor_chan {
+	int			pending;
+	dma_cookie_t		completed_cookie;
+	spinlock_t		lock; /* protects the descriptor slot pool */
+	void __iomem		*mmr_base;
+	unsigned int		idx;
+	enum dma_transaction_type	current_type;
+	struct list_head	chain;
+	struct list_head	completed_slots;
+	struct mv_xor_device	*device;
+	struct dma_chan		common;
+	struct mv_xor_desc_slot	*last_used;
+	struct list_head	all_slots;
+	int			slots_allocated;
+	struct tasklet_struct	irq_tasklet;
+#ifdef USE_TIMER
+	unsigned long		cleanup_time;
+	u32			current_on_last_cleanup;
+	dma_cookie_t		is_complete_cookie;
+#endif
+};
+
+/**
+ * struct mv_xor_desc_slot - software descriptor
+ * @slot_node: node on the mv_xor_chan.all_slots list
+ * @chain_node: node on the mv_xor_chan.chain list
+ * @completed_node: node on the mv_xor_chan.completed_slots list
+ * @hw_desc: virtual address of the hardware descriptor chain
+ * @phys: hardware address of the hardware descriptor chain
+ * @group_head: first operation in a transaction
+ * @slot_cnt: total slots used in an transaction (group of operations)
+ * @slots_per_op: number of slots per operation
+ * @idx: pool index
+ * @unmap_src_cnt: number of xor sources
+ * @unmap_len: transaction bytecount
+ * @async_tx: support for the async_tx api
+ * @group_list: list of slots that make up a multi-descriptor transaction
+ *	for example transfer lengths larger than the supported hw max
+ * @xor_check_result: result of zero sum
+ * @crc32_result: result crc calculation
+ */
+struct mv_xor_desc_slot {
+	struct list_head	slot_node;
+	struct list_head	chain_node;
+	struct list_head	completed_node;
+	enum dma_transaction_type	type;
+	void			*hw_desc;
+	struct mv_xor_desc_slot	*group_head;
+	u16			slot_cnt;
+	u16			slots_per_op;
+	u16			idx;
+	u16			unmap_src_cnt;
+	u32			value;
+	size_t			unmap_len;
+	struct dma_async_tx_descriptor	async_tx;
+	union {
+		u32		*xor_check_result;
+		u32		*crc32_result;
+	};
+#ifdef USE_TIMER
+	unsigned long		arrival_time;
+	struct timer_list	timeout;
+#endif
+};
+
+/* This structure describes XOR descriptor size 64bytes	*/
+struct mv_xor_desc {
+	u32 status;		/* descriptor execution status */
+	u32 crc32_result;	/* result of CRC-32 calculation */
+	u32 desc_command;	/* type of operation to be carried out */
+	u32 phy_next_desc;	/* next descriptor address pointer */
+	u32 byte_count;		/* size of src/dst blocks in bytes */
+	u32 phy_dest_addr;	/* destination block address */
+	u32 phy_src_addr[8];	/* source block addresses */
+	u32 reserved0;
+	u32 reserved1;
+};
+
+#define to_mv_sw_desc(addr_hw_desc)		\
+	container_of(addr_hw_desc, struct mv_xor_desc_slot, hw_desc)
+
+#define mv_hw_desc_slot_idx(hw_desc, idx)	\
+	((void *)(((unsigned long)hw_desc) + ((idx) << 5)))
+
+#define MV_XOR_MIN_BYTE_COUNT	(128)
+#define XOR_MAX_BYTE_COUNT	((16 * 1024 * 1024) - 1)
+#define MV_XOR_MAX_BYTE_COUNT	XOR_MAX_BYTE_COUNT
+
+
+#endif
diff --git a/include/asm-arm/arch-iop13xx/adma.h b/include/asm-arm/arch-iop13xx/adma.h
index 90d14ee564f5..ef4f5da2029f 100644
--- a/include/asm-arm/arch-iop13xx/adma.h
+++ b/include/asm-arm/arch-iop13xx/adma.h
@@ -198,17 +198,13 @@ iop_chan_memset_slot_count(size_t len, int *slots_per_op)
 static inline int
 iop_chan_xor_slot_count(size_t len, int src_cnt, int *slots_per_op)
 {
-	int num_slots;
-	/* slots_to_find = 1 for basic descriptor + 1 per 4 sources above 1
-	 * (1 source => 8 bytes) (1 slot => 32 bytes)
-	 */
-	num_slots = 1 + (((src_cnt - 1) << 3) >> 5);
-	if (((src_cnt - 1) << 3) & 0x1f)
-		num_slots++;
-
-	*slots_per_op = num_slots;
-
-	return num_slots;
+	static const char slot_count_table[] = { 1, 2, 2, 2,
+						 2, 3, 3, 3,
+						 3, 4, 4, 4,
+						 4, 5, 5, 5,
+						};
+	*slots_per_op = slot_count_table[src_cnt - 1];
+	return *slots_per_op;
 }
 
 #define ADMA_MAX_BYTE_COUNT	(16 * 1024 * 1024)
diff --git a/include/asm-arm/hardware/iop3xx-adma.h b/include/asm-arm/hardware/iop3xx-adma.h
index a32b86ac62aa..af64676650a2 100644
--- a/include/asm-arm/hardware/iop3xx-adma.h
+++ b/include/asm-arm/hardware/iop3xx-adma.h
@@ -260,7 +260,7 @@ static inline int iop_chan_memset_slot_count(size_t len, int *slots_per_op)
 static inline int iop3xx_aau_xor_slot_count(size_t len, int src_cnt,
 					int *slots_per_op)
 {
-	static const int slot_count_table[] = { 0,
+	static const char slot_count_table[] = {
 						1, 1, 1, 1, /* 01 - 04 */
 						2, 2, 2, 2, /* 05 - 08 */
 						4, 4, 4, 4, /* 09 - 12 */
@@ -270,7 +270,7 @@ static inline int iop3xx_aau_xor_slot_count(size_t len, int src_cnt,
 						8, 8, 8, 8, /* 25 - 28 */
 						8, 8, 8, 8, /* 29 - 32 */
 					      };
-	*slots_per_op = slot_count_table[src_cnt];
+	*slots_per_op = slot_count_table[src_cnt - 1];
 	return *slots_per_op;
 }
 
diff --git a/include/asm-arm/plat-orion/mv_xor.h b/include/asm-arm/plat-orion/mv_xor.h
new file mode 100644
index 000000000000..c349e8ff5cc0
--- /dev/null
+++ b/include/asm-arm/plat-orion/mv_xor.h
@@ -0,0 +1,28 @@
+/*
+ * Marvell XOR platform device data definition file.
+ */
+
+#ifndef __ASM_PLAT_ORION_MV_XOR_H
+#define __ASM_PLAT_ORION_MV_XOR_H
+
+#include <linux/dmaengine.h>
+#include <linux/mbus.h>
+
+#define MV_XOR_SHARED_NAME	"mv_xor_shared"
+#define MV_XOR_NAME		"mv_xor"
+
+struct mbus_dram_target_info;
+
+struct mv_xor_platform_shared_data {
+	struct mbus_dram_target_info	*dram;
+};
+
+struct mv_xor_platform_data {
+	struct platform_device		*shared;
+	int				hw_id;
+	dma_cap_mask_t			cap_mask;
+	size_t				pool_size;
+};
+
+
+#endif
diff --git a/include/asm-avr32/arch-at32ap/at32ap700x.h b/include/asm-avr32/arch-at32ap/at32ap700x.h
index 31e48b0e7324..d18a3053be0d 100644
--- a/include/asm-avr32/arch-at32ap/at32ap700x.h
+++ b/include/asm-avr32/arch-at32ap/at32ap700x.h
@@ -30,4 +30,20 @@
 #define GPIO_PIN_PD(N)	(GPIO_PIOD_BASE + (N))
 #define GPIO_PIN_PE(N)	(GPIO_PIOE_BASE + (N))
 
+
+/*
+ * DMAC peripheral hardware handshaking interfaces, used with dw_dmac
+ */
+#define DMAC_MCI_RX		0
+#define DMAC_MCI_TX		1
+#define DMAC_DAC_TX		2
+#define DMAC_AC97_A_RX		3
+#define DMAC_AC97_A_TX		4
+#define DMAC_AC97_B_RX		5
+#define DMAC_AC97_B_TX		6
+#define DMAC_DMAREQ_0		7
+#define DMAC_DMAREQ_1		8
+#define DMAC_DMAREQ_2		9
+#define DMAC_DMAREQ_3		10
+
 #endif /* __ASM_ARCH_AT32AP700X_H__ */
diff --git a/include/linux/async_tx.h b/include/linux/async_tx.h
index eb640f0acfac..0f50d4cc4360 100644
--- a/include/linux/async_tx.h
+++ b/include/linux/async_tx.h
@@ -101,21 +101,14 @@ async_tx_find_channel(struct dma_async_tx_descriptor *depend_tx,
 
 /**
  * async_tx_sync_epilog - actions to take if an operation is run synchronously
- * @flags: async_tx flags
- * @depend_tx: transaction depends on depend_tx
  * @cb_fn: function to call when the transaction completes
  * @cb_fn_param: parameter to pass to the callback routine
  */
 static inline void
-async_tx_sync_epilog(unsigned long flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_fn_param)
+async_tx_sync_epilog(dma_async_tx_callback cb_fn, void *cb_fn_param)
 {
 	if (cb_fn)
 		cb_fn(cb_fn_param);
-
-	if (depend_tx && (flags & ASYNC_TX_DEP_ACK))
-		async_tx_ack(depend_tx);
 }
 
 void
@@ -152,4 +145,6 @@ struct dma_async_tx_descriptor *
 async_trigger_callback(enum async_tx_flags flags,
 	struct dma_async_tx_descriptor *depend_tx,
 	dma_async_tx_callback cb_fn, void *cb_fn_param);
+
+void async_tx_quiesce(struct dma_async_tx_descriptor **tx);
 #endif /* _ASYNC_TX_H_ */
diff --git a/include/linux/dca.h b/include/linux/dca.h
index af61cd1f37e9..b00a753eda53 100644
--- a/include/linux/dca.h
+++ b/include/linux/dca.h
@@ -10,6 +10,7 @@ void dca_unregister_notify(struct notifier_block *nb);
 #define DCA_PROVIDER_REMOVE  0x0002
 
 struct dca_provider {
+	struct list_head	node;
 	struct dca_ops		*ops;
 	struct device 		*cd;
 	int			 id;
@@ -18,7 +19,9 @@ struct dca_provider {
 struct dca_ops {
 	int	(*add_requester)    (struct dca_provider *, struct device *);
 	int	(*remove_requester) (struct dca_provider *, struct device *);
-	u8	(*get_tag)	    (struct dca_provider *, int cpu);
+	u8	(*get_tag)	    (struct dca_provider *, struct device *,
+				     int cpu);
+	int	(*dev_managed)      (struct dca_provider *, struct device *);
 };
 
 struct dca_provider *alloc_dca_provider(struct dca_ops *ops, int priv_size);
@@ -32,9 +35,11 @@ static inline void *dca_priv(struct dca_provider *dca)
 }
 
 /* Requester API */
+#define DCA_GET_TAG_TWO_ARGS
 int dca_add_requester(struct device *dev);
 int dca_remove_requester(struct device *dev);
 u8 dca_get_tag(int cpu);
+u8 dca3_get_tag(struct device *dev, int cpu);
 
 /* internal stuff */
 int __init dca_sysfs_init(void);
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index d08a5c5eb928..adb0b084eb5a 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -89,10 +89,23 @@ enum dma_transaction_type {
 	DMA_MEMSET,
 	DMA_MEMCPY_CRC32C,
 	DMA_INTERRUPT,
+	DMA_SLAVE,
 };
 
 /* last transaction type for creation of the capabilities mask */
-#define DMA_TX_TYPE_END (DMA_INTERRUPT + 1)
+#define DMA_TX_TYPE_END (DMA_SLAVE + 1)
+
+/**
+ * enum dma_slave_width - DMA slave register access width.
+ * @DMA_SLAVE_WIDTH_8BIT: Do 8-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_16BIT: Do 16-bit slave register accesses
+ * @DMA_SLAVE_WIDTH_32BIT: Do 32-bit slave register accesses
+ */
+enum dma_slave_width {
+	DMA_SLAVE_WIDTH_8BIT,
+	DMA_SLAVE_WIDTH_16BIT,
+	DMA_SLAVE_WIDTH_32BIT,
+};
 
 /**
  * enum dma_ctrl_flags - DMA flags to augment operation preparation,
@@ -102,10 +115,14 @@ enum dma_transaction_type {
  * @DMA_CTRL_ACK - the descriptor cannot be reused until the client
  * 	acknowledges receipt, i.e. has has a chance to establish any
  * 	dependency chains
+ * @DMA_COMPL_SKIP_SRC_UNMAP - set to disable dma-unmapping the source buffer(s)
+ * @DMA_COMPL_SKIP_DEST_UNMAP - set to disable dma-unmapping the destination(s)
  */
 enum dma_ctrl_flags {
 	DMA_PREP_INTERRUPT = (1 << 0),
 	DMA_CTRL_ACK = (1 << 1),
+	DMA_COMPL_SKIP_SRC_UNMAP = (1 << 2),
+	DMA_COMPL_SKIP_DEST_UNMAP = (1 << 3),
 };
 
 /**
@@ -115,6 +132,32 @@ enum dma_ctrl_flags {
 typedef struct { DECLARE_BITMAP(bits, DMA_TX_TYPE_END); } dma_cap_mask_t;
 
 /**
+ * struct dma_slave - Information about a DMA slave
+ * @dev: device acting as DMA slave
+ * @dma_dev: required DMA master device. If non-NULL, the client can not be
+ *	bound to other masters than this.
+ * @tx_reg: physical address of data register used for
+ *	memory-to-peripheral transfers
+ * @rx_reg: physical address of data register used for
+ *	peripheral-to-memory transfers
+ * @reg_width: peripheral register width
+ *
+ * If dma_dev is non-NULL, the client can not be bound to other DMA
+ * masters than the one corresponding to this device. The DMA master
+ * driver may use this to determine if there is controller-specific
+ * data wrapped around this struct. Drivers of platform code that sets
+ * the dma_dev field must therefore make sure to use an appropriate
+ * controller-specific dma slave structure wrapping this struct.
+ */
+struct dma_slave {
+	struct device		*dev;
+	struct device		*dma_dev;
+	dma_addr_t		tx_reg;
+	dma_addr_t		rx_reg;
+	enum dma_slave_width	reg_width;
+};
+
+/**
  * struct dma_chan_percpu - the per-CPU part of struct dma_chan
  * @refcount: local_t used for open-coded "bigref" counting
  * @memcpy_count: transaction counter
@@ -139,6 +182,7 @@ struct dma_chan_percpu {
  * @rcu: the DMA channel's RCU head
  * @device_node: used to add this to the device chan list
  * @local: per-cpu pointer to a struct dma_chan_percpu
+ * @client-count: how many clients are using this channel
  */
 struct dma_chan {
 	struct dma_device *device;
@@ -154,6 +198,7 @@ struct dma_chan {
 
 	struct list_head device_node;
 	struct dma_chan_percpu *local;
+	int client_count;
 };
 
 #define to_dma_chan(p) container_of(p, struct dma_chan, dev)
@@ -202,11 +247,14 @@ typedef enum dma_state_client (*dma_event_callback) (struct dma_client *client,
  * @event_callback: func ptr to call when something happens
  * @cap_mask: only return channels that satisfy the requested capabilities
  *  a value of zero corresponds to any capability
+ * @slave: data for preparing slave transfer. Must be non-NULL iff the
+ *  DMA_SLAVE capability is requested.
  * @global_node: list_head for global dma_client_list
  */
 struct dma_client {
 	dma_event_callback	event_callback;
 	dma_cap_mask_t		cap_mask;
+	struct dma_slave	*slave;
 	struct list_head	global_node;
 };
 
@@ -263,6 +311,8 @@ struct dma_async_tx_descriptor {
  * @device_prep_dma_zero_sum: prepares a zero_sum operation
  * @device_prep_dma_memset: prepares a memset operation
  * @device_prep_dma_interrupt: prepares an end of chain interrupt operation
+ * @device_prep_slave_sg: prepares a slave dma operation
+ * @device_terminate_all: terminate all pending operations
  * @device_issue_pending: push pending transactions to hardware
  */
 struct dma_device {
@@ -279,7 +329,8 @@ struct dma_device {
 	int dev_id;
 	struct device *dev;
 
-	int (*device_alloc_chan_resources)(struct dma_chan *chan);
+	int (*device_alloc_chan_resources)(struct dma_chan *chan,
+			struct dma_client *client);
 	void (*device_free_chan_resources)(struct dma_chan *chan);
 
 	struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
@@ -297,6 +348,12 @@ struct dma_device {
 	struct dma_async_tx_descriptor *(*device_prep_dma_interrupt)(
 		struct dma_chan *chan, unsigned long flags);
 
+	struct dma_async_tx_descriptor *(*device_prep_slave_sg)(
+		struct dma_chan *chan, struct scatterlist *sgl,
+		unsigned int sg_len, enum dma_data_direction direction,
+		unsigned long flags);
+	void (*device_terminate_all)(struct dma_chan *chan);
+
 	enum dma_status (*device_is_tx_complete)(struct dma_chan *chan,
 			dma_cookie_t cookie, dma_cookie_t *last,
 			dma_cookie_t *used);
@@ -318,16 +375,14 @@ dma_cookie_t dma_async_memcpy_pg_to_pg(struct dma_chan *chan,
 void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx,
 	struct dma_chan *chan);
 
-static inline void
-async_tx_ack(struct dma_async_tx_descriptor *tx)
+static inline void async_tx_ack(struct dma_async_tx_descriptor *tx)
 {
 	tx->flags |= DMA_CTRL_ACK;
 }
 
-static inline int
-async_tx_test_ack(struct dma_async_tx_descriptor *tx)
+static inline bool async_tx_test_ack(struct dma_async_tx_descriptor *tx)
 {
-	return tx->flags & DMA_CTRL_ACK;
+	return (tx->flags & DMA_CTRL_ACK) == DMA_CTRL_ACK;
 }
 
 #define first_dma_cap(mask) __first_dma_cap(&(mask))
diff --git a/include/linux/dw_dmac.h b/include/linux/dw_dmac.h
new file mode 100644
index 000000000000..04d217b442bf
--- /dev/null
+++ b/include/linux/dw_dmac.h
@@ -0,0 +1,62 @@
+/*
+ * Driver for the Synopsys DesignWare DMA Controller (aka DMACA on
+ * AVR32 systems.)
+ *
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef DW_DMAC_H
+#define DW_DMAC_H
+
+#include <linux/dmaengine.h>
+
+/**
+ * struct dw_dma_platform_data - Controller configuration parameters
+ * @nr_channels: Number of channels supported by hardware (max 8)
+ */
+struct dw_dma_platform_data {
+	unsigned int	nr_channels;
+};
+
+/**
+ * struct dw_dma_slave - Controller-specific information about a slave
+ * @slave: Generic information about the slave
+ * @ctl_lo: Platform-specific initializer for the CTL_LO register
+ * @cfg_hi: Platform-specific initializer for the CFG_HI register
+ * @cfg_lo: Platform-specific initializer for the CFG_LO register
+ */
+struct dw_dma_slave {
+	struct dma_slave	slave;
+	u32			cfg_hi;
+	u32			cfg_lo;
+};
+
+/* Platform-configurable bits in CFG_HI */
+#define DWC_CFGH_FCMODE		(1 << 0)
+#define DWC_CFGH_FIFO_MODE	(1 << 1)
+#define DWC_CFGH_PROTCTL(x)	((x) << 2)
+#define DWC_CFGH_SRC_PER(x)	((x) << 7)
+#define DWC_CFGH_DST_PER(x)	((x) << 11)
+
+/* Platform-configurable bits in CFG_LO */
+#define DWC_CFGL_PRIO(x)	((x) << 5)	/* priority */
+#define DWC_CFGL_LOCK_CH_XFER	(0 << 12)	/* scope of LOCK_CH */
+#define DWC_CFGL_LOCK_CH_BLOCK	(1 << 12)
+#define DWC_CFGL_LOCK_CH_XACT	(2 << 12)
+#define DWC_CFGL_LOCK_BUS_XFER	(0 << 14)	/* scope of LOCK_BUS */
+#define DWC_CFGL_LOCK_BUS_BLOCK	(1 << 14)
+#define DWC_CFGL_LOCK_BUS_XACT	(2 << 14)
+#define DWC_CFGL_LOCK_CH	(1 << 15)	/* channel lockout */
+#define DWC_CFGL_LOCK_BUS	(1 << 16)	/* busmaster lockout */
+#define DWC_CFGL_HS_DST_POL	(1 << 18)	/* dst handshake active low */
+#define DWC_CFGL_HS_SRC_POL	(1 << 19)	/* src handshake active low */
+
+static inline struct dw_dma_slave *to_dw_dma_slave(struct dma_slave *slave)
+{
+	return container_of(slave, struct dw_dma_slave, slave);
+}
+
+#endif /* DW_DMAC_H */
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index d8507eb394cf..119ae7b8f028 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2371,6 +2371,14 @@
 #define PCI_DEVICE_ID_INTEL_ICH9_7	0x2916
 #define PCI_DEVICE_ID_INTEL_ICH9_8	0x2918
 #define PCI_DEVICE_ID_INTEL_82855PM_HB	0x3340
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG4	0x3429
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG5	0x342a
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG6	0x342b
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG7	0x342c
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG0	0x3430
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG1	0x3431
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG2	0x3432
+#define PCI_DEVICE_ID_INTEL_IOAT_TBG3	0x3433
 #define PCI_DEVICE_ID_INTEL_82830_HB	0x3575
 #define PCI_DEVICE_ID_INTEL_82830_CGC	0x3577
 #define PCI_DEVICE_ID_INTEL_82855GM_HB	0x3580
diff --git a/net/core/user_dma.c b/net/core/user_dma.c
index c77aff9c6eb3..8c6b706963ff 100644
--- a/net/core/user_dma.c
+++ b/net/core/user_dma.c
@@ -34,6 +34,7 @@
 #define NET_DMA_DEFAULT_COPYBREAK 4096
 
 int sysctl_tcp_dma_copybreak = NET_DMA_DEFAULT_COPYBREAK;
+EXPORT_SYMBOL(sysctl_tcp_dma_copybreak);
 
 /**
  *	dma_skb_copy_datagram_iovec - Copy a datagram to an iovec.