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

Pull slave-dmaengine updates from Vinod Koul:
 "This is fairly big pull by my standards as I had missed last merge
  window.  So we have the support for device tree for slave-dmaengine,
  large updates to dw_dmac driver from Andy for reusing on different
  architectures.  Along with this we have fixes on bunch of the drivers"

Fix up trivial conflicts, usually due to #include line movement next to
each other.

* 'next' of git://git.infradead.org/users/vkoul/slave-dma: (111 commits)
  Revert "ARM: SPEAr13xx: Pass DW DMAC platform data from DT"
  ARM: dts: pl330: Add #dma-cells for generic dma binding support
  DMA: PL330: Register the DMA controller with the generic DMA helpers
  DMA: PL330: Add xlate function
  DMA: PL330: Add new pl330 filter for DT case.
  dma: tegra20-apb-dma: remove unnecessary assignment
  edma: do not waste memory for dma_mask
  dma: coh901318: set residue only if dma is in progress
  dma: coh901318: avoid unbalanced locking
  dmaengine.h: remove redundant else keyword
  dma: of-dma: protect list write operation by spin_lock
  dmaengine: ste_dma40: do not remove descriptors for cyclic transfers
  dma: of-dma.c: fix memory leakage
  dw_dmac: apply default dma_mask if needed
  dmaengine: ioat - fix spare sparse complain
  dmaengine: move drivers/of/dma.c -> drivers/dma/of-dma.c
  ioatdma: fix race between updating ioat->head and IOAT_COMPLETION_PENDING
  dw_dmac: add support for Lynxpoint DMA controllers
  dw_dmac: return proper residue value
  dw_dmac: fill individual length of descriptor
  ...

6 files changed, 227 insertions, 136 deletions

diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 1a68a8ba87e6..1879a5942bfc 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -833,14 +833,14 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 
 	dma_src = dma_map_single(dev, src, IOAT_TEST_SIZE, DMA_TO_DEVICE);
 	dma_dest = dma_map_single(dev, dest, IOAT_TEST_SIZE, DMA_FROM_DEVICE);
-	flags = DMA_COMPL_SRC_UNMAP_SINGLE | DMA_COMPL_DEST_UNMAP_SINGLE |
+	flags = DMA_COMPL_SKIP_SRC_UNMAP | DMA_COMPL_SKIP_DEST_UNMAP |
 		DMA_PREP_INTERRUPT;
 	tx = device->common.device_prep_dma_memcpy(dma_chan, dma_dest, dma_src,
 						   IOAT_TEST_SIZE, flags);
 	if (!tx) {
 		dev_err(dev, "Self-test prep failed, disabling\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto unmap_dma;
 	}
 
 	async_tx_ack(tx);
@@ -851,7 +851,7 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 	if (cookie < 0) {
 		dev_err(dev, "Self-test setup failed, disabling\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto unmap_dma;
 	}
 	dma->device_issue_pending(dma_chan);
 
@@ -862,7 +862,7 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 					!= DMA_SUCCESS) {
 		dev_err(dev, "Self-test copy timed out, disabling\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto unmap_dma;
 	}
 	if (memcmp(src, dest, IOAT_TEST_SIZE)) {
 		dev_err(dev, "Self-test copy failed compare, disabling\n");
@@ -870,6 +870,9 @@ int ioat_dma_self_test(struct ioatdma_device *device)
 		goto free_resources;
 	}
 
+unmap_dma:
+	dma_unmap_single(dev, dma_src, IOAT_TEST_SIZE, DMA_TO_DEVICE);
+	dma_unmap_single(dev, dma_dest, IOAT_TEST_SIZE, DMA_FROM_DEVICE);
 free_resources:
 	dma->device_free_chan_resources(dma_chan);
 out:
diff --git a/drivers/dma/ioat/dma.h b/drivers/dma/ioat/dma.h
index 087935f1565f..53a4cbb78f47 100644
--- a/drivers/dma/ioat/dma.h
+++ b/drivers/dma/ioat/dma.h
@@ -97,6 +97,7 @@ struct ioat_chan_common {
 	#define IOAT_KOBJ_INIT_FAIL 3
 	#define IOAT_RESHAPE_PENDING 4
 	#define IOAT_RUN 5
+	#define IOAT_CHAN_ACTIVE 6
 	struct timer_list timer;
 	#define COMPLETION_TIMEOUT msecs_to_jiffies(100)
 	#define IDLE_TIMEOUT msecs_to_jiffies(2000)
diff --git a/drivers/dma/ioat/dma_v2.c b/drivers/dma/ioat/dma_v2.c
index 82d4e306c32e..b925e1b1d139 100644
--- a/drivers/dma/ioat/dma_v2.c
+++ b/drivers/dma/ioat/dma_v2.c
@@ -269,61 +269,22 @@ static void ioat2_restart_channel(struct ioat2_dma_chan *ioat)
 	__ioat2_restart_chan(ioat);
 }
 
-void ioat2_timer_event(unsigned long data)
+static void check_active(struct ioat2_dma_chan *ioat)
 {
-	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
 	struct ioat_chan_common *chan = &ioat->base;
 
-	if (test_bit(IOAT_COMPLETION_PENDING, &chan->state)) {
-		dma_addr_t phys_complete;
-		u64 status;
-
-		status = ioat_chansts(chan);
-
-		/* when halted due to errors check for channel
-		 * programming errors before advancing the completion state
-		 */
-		if (is_ioat_halted(status)) {
-			u32 chanerr;
-
-			chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
-			dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
-				__func__, chanerr);
-			if (test_bit(IOAT_RUN, &chan->state))
-				BUG_ON(is_ioat_bug(chanerr));
-			else /* we never got off the ground */
-				return;
-		}
-
-		/* if we haven't made progress and we have already
-		 * acknowledged a pending completion once, then be more
-		 * forceful with a restart
-		 */
-		spin_lock_bh(&chan->cleanup_lock);
-		if (ioat_cleanup_preamble(chan, &phys_complete)) {
-			__cleanup(ioat, phys_complete);
-		} else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
-			spin_lock_bh(&ioat->prep_lock);
-			ioat2_restart_channel(ioat);
-			spin_unlock_bh(&ioat->prep_lock);
-		} else {
-			set_bit(IOAT_COMPLETION_ACK, &chan->state);
-			mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
-		}
-		spin_unlock_bh(&chan->cleanup_lock);
-	} else {
-		u16 active;
+	if (ioat2_ring_active(ioat)) {
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+		return;
+	}
 
+	if (test_and_clear_bit(IOAT_CHAN_ACTIVE, &chan->state))
+		mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+	else if (ioat->alloc_order > ioat_get_alloc_order()) {
 		/* if the ring is idle, empty, and oversized try to step
 		 * down the size
 		 */
-		spin_lock_bh(&chan->cleanup_lock);
-		spin_lock_bh(&ioat->prep_lock);
-		active = ioat2_ring_active(ioat);
-		if (active == 0 && ioat->alloc_order > ioat_get_alloc_order())
-			reshape_ring(ioat, ioat->alloc_order-1);
-		spin_unlock_bh(&ioat->prep_lock);
-		spin_unlock_bh(&chan->cleanup_lock);
+		reshape_ring(ioat, ioat->alloc_order - 1);
 
 		/* keep shrinking until we get back to our minimum
 		 * default size
@@ -331,6 +292,60 @@ void ioat2_timer_event(unsigned long data)
 		if (ioat->alloc_order > ioat_get_alloc_order())
 			mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
 	}
+
+}
+
+void ioat2_timer_event(unsigned long data)
+{
+	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+	struct ioat_chan_common *chan = &ioat->base;
+	dma_addr_t phys_complete;
+	u64 status;
+
+	status = ioat_chansts(chan);
+
+	/* when halted due to errors check for channel
+	 * programming errors before advancing the completion state
+	 */
+	if (is_ioat_halted(status)) {
+		u32 chanerr;
+
+		chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+		dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
+			__func__, chanerr);
+		if (test_bit(IOAT_RUN, &chan->state))
+			BUG_ON(is_ioat_bug(chanerr));
+		else /* we never got off the ground */
+			return;
+	}
+
+	/* if we haven't made progress and we have already
+	 * acknowledged a pending completion once, then be more
+	 * forceful with a restart
+	 */
+	spin_lock_bh(&chan->cleanup_lock);
+	if (ioat_cleanup_preamble(chan, &phys_complete))
+		__cleanup(ioat, phys_complete);
+	else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
+		spin_lock_bh(&ioat->prep_lock);
+		ioat2_restart_channel(ioat);
+		spin_unlock_bh(&ioat->prep_lock);
+		spin_unlock_bh(&chan->cleanup_lock);
+		return;
+	} else {
+		set_bit(IOAT_COMPLETION_ACK, &chan->state);
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+	}
+
+
+	if (ioat2_ring_active(ioat))
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+	else {
+		spin_lock_bh(&ioat->prep_lock);
+		check_active(ioat);
+		spin_unlock_bh(&ioat->prep_lock);
+	}
+	spin_unlock_bh(&chan->cleanup_lock);
 }
 
 static int ioat2_reset_hw(struct ioat_chan_common *chan)
@@ -404,7 +419,7 @@ static dma_cookie_t ioat2_tx_submit_unlock(struct dma_async_tx_descriptor *tx)
 	cookie = dma_cookie_assign(tx);
 	dev_dbg(to_dev(&ioat->base), "%s: cookie: %d\n", __func__, cookie);
 
-	if (!test_and_set_bit(IOAT_COMPLETION_PENDING, &chan->state))
+	if (!test_and_set_bit(IOAT_CHAN_ACTIVE, &chan->state))
 		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
 
 	/* make descriptor updates visible before advancing ioat->head,
diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c
index 3e9d66920eb3..e8336cce360b 100644
--- a/drivers/dma/ioat/dma_v3.c
+++ b/drivers/dma/ioat/dma_v3.c
@@ -342,61 +342,22 @@ static void ioat3_restart_channel(struct ioat2_dma_chan *ioat)
 	__ioat2_restart_chan(ioat);
 }
 
-static void ioat3_timer_event(unsigned long data)
+static void check_active(struct ioat2_dma_chan *ioat)
 {
-	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
 	struct ioat_chan_common *chan = &ioat->base;
 
-	if (test_bit(IOAT_COMPLETION_PENDING, &chan->state)) {
-		dma_addr_t phys_complete;
-		u64 status;
-
-		status = ioat_chansts(chan);
-
-		/* when halted due to errors check for channel
-		 * programming errors before advancing the completion state
-		 */
-		if (is_ioat_halted(status)) {
-			u32 chanerr;
-
-			chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
-			dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
-				__func__, chanerr);
-			if (test_bit(IOAT_RUN, &chan->state))
-				BUG_ON(is_ioat_bug(chanerr));
-			else /* we never got off the ground */
-				return;
-		}
-
-		/* if we haven't made progress and we have already
-		 * acknowledged a pending completion once, then be more
-		 * forceful with a restart
-		 */
-		spin_lock_bh(&chan->cleanup_lock);
-		if (ioat_cleanup_preamble(chan, &phys_complete))
-			__cleanup(ioat, phys_complete);
-		else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
-			spin_lock_bh(&ioat->prep_lock);
-			ioat3_restart_channel(ioat);
-			spin_unlock_bh(&ioat->prep_lock);
-		} else {
-			set_bit(IOAT_COMPLETION_ACK, &chan->state);
-			mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
-		}
-		spin_unlock_bh(&chan->cleanup_lock);
-	} else {
-		u16 active;
+	if (ioat2_ring_active(ioat)) {
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+		return;
+	}
 
+	if (test_and_clear_bit(IOAT_CHAN_ACTIVE, &chan->state))
+		mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
+	else if (ioat->alloc_order > ioat_get_alloc_order()) {
 		/* if the ring is idle, empty, and oversized try to step
 		 * down the size
 		 */
-		spin_lock_bh(&chan->cleanup_lock);
-		spin_lock_bh(&ioat->prep_lock);
-		active = ioat2_ring_active(ioat);
-		if (active == 0 && ioat->alloc_order > ioat_get_alloc_order())
-			reshape_ring(ioat, ioat->alloc_order-1);
-		spin_unlock_bh(&ioat->prep_lock);
-		spin_unlock_bh(&chan->cleanup_lock);
+		reshape_ring(ioat, ioat->alloc_order - 1);
 
 		/* keep shrinking until we get back to our minimum
 		 * default size
@@ -404,6 +365,60 @@ static void ioat3_timer_event(unsigned long data)
 		if (ioat->alloc_order > ioat_get_alloc_order())
 			mod_timer(&chan->timer, jiffies + IDLE_TIMEOUT);
 	}
+
+}
+
+static void ioat3_timer_event(unsigned long data)
+{
+	struct ioat2_dma_chan *ioat = to_ioat2_chan((void *) data);
+	struct ioat_chan_common *chan = &ioat->base;
+	dma_addr_t phys_complete;
+	u64 status;
+
+	status = ioat_chansts(chan);
+
+	/* when halted due to errors check for channel
+	 * programming errors before advancing the completion state
+	 */
+	if (is_ioat_halted(status)) {
+		u32 chanerr;
+
+		chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
+		dev_err(to_dev(chan), "%s: Channel halted (%x)\n",
+			__func__, chanerr);
+		if (test_bit(IOAT_RUN, &chan->state))
+			BUG_ON(is_ioat_bug(chanerr));
+		else /* we never got off the ground */
+			return;
+	}
+
+	/* if we haven't made progress and we have already
+	 * acknowledged a pending completion once, then be more
+	 * forceful with a restart
+	 */
+	spin_lock_bh(&chan->cleanup_lock);
+	if (ioat_cleanup_preamble(chan, &phys_complete))
+		__cleanup(ioat, phys_complete);
+	else if (test_bit(IOAT_COMPLETION_ACK, &chan->state)) {
+		spin_lock_bh(&ioat->prep_lock);
+		ioat3_restart_channel(ioat);
+		spin_unlock_bh(&ioat->prep_lock);
+		spin_unlock_bh(&chan->cleanup_lock);
+		return;
+	} else {
+		set_bit(IOAT_COMPLETION_ACK, &chan->state);
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+	}
+
+
+	if (ioat2_ring_active(ioat))
+		mod_timer(&chan->timer, jiffies + COMPLETION_TIMEOUT);
+	else {
+		spin_lock_bh(&ioat->prep_lock);
+		check_active(ioat);
+		spin_unlock_bh(&ioat->prep_lock);
+	}
+	spin_unlock_bh(&chan->cleanup_lock);
 }
 
 static enum dma_status
@@ -863,6 +878,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	unsigned long tmo;
 	struct device *dev = &device->pdev->dev;
 	struct dma_device *dma = &device->common;
+	u8 op = 0;
 
 	dev_dbg(dev, "%s\n", __func__);
 
@@ -908,18 +924,22 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	}
 
 	/* test xor */
+	op = IOAT_OP_XOR;
+
 	dest_dma = dma_map_page(dev, dest, 0, PAGE_SIZE, DMA_FROM_DEVICE);
 	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
 		dma_srcs[i] = dma_map_page(dev, xor_srcs[i], 0, PAGE_SIZE,
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor(dma_chan, dest_dma, dma_srcs,
 				      IOAT_NUM_SRC_TEST, PAGE_SIZE,
-				      DMA_PREP_INTERRUPT);
+				      DMA_PREP_INTERRUPT |
+				      DMA_COMPL_SKIP_SRC_UNMAP |
+				      DMA_COMPL_SKIP_DEST_UNMAP);
 
 	if (!tx) {
 		dev_err(dev, "Self-test xor prep failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
 	async_tx_ack(tx);
@@ -930,7 +950,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (cookie < 0) {
 		dev_err(dev, "Self-test xor setup failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 	dma->device_issue_pending(dma_chan);
 
@@ -939,9 +959,13 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
 		dev_err(dev, "Self-test xor timed out\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
+	dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
+	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
+		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
 	dma_sync_single_for_cpu(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
 	for (i = 0; i < (PAGE_SIZE / sizeof(u32)); i++) {
 		u32 *ptr = page_address(dest);
@@ -957,6 +981,8 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (!dma_has_cap(DMA_XOR_VAL, dma_chan->device->cap_mask))
 		goto free_resources;
 
+	op = IOAT_OP_XOR_VAL;
+
 	/* validate the sources with the destintation page */
 	for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
 		xor_val_srcs[i] = xor_srcs[i];
@@ -969,11 +995,13 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
 					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
-					  &xor_val_result, DMA_PREP_INTERRUPT);
+					  &xor_val_result, DMA_PREP_INTERRUPT |
+					  DMA_COMPL_SKIP_SRC_UNMAP |
+					  DMA_COMPL_SKIP_DEST_UNMAP);
 	if (!tx) {
 		dev_err(dev, "Self-test zero prep failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
 	async_tx_ack(tx);
@@ -984,7 +1012,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (cookie < 0) {
 		dev_err(dev, "Self-test zero setup failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 	dma->device_issue_pending(dma_chan);
 
@@ -993,9 +1021,12 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
 		dev_err(dev, "Self-test validate timed out\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
+	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
 	if (xor_val_result != 0) {
 		dev_err(dev, "Self-test validate failed compare\n");
 		err = -ENODEV;
@@ -1007,14 +1038,18 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 		goto free_resources;
 
 	/* test memset */
+	op = IOAT_OP_FILL;
+
 	dma_addr = dma_map_page(dev, dest, 0,
 			PAGE_SIZE, DMA_FROM_DEVICE);
 	tx = dma->device_prep_dma_memset(dma_chan, dma_addr, 0, PAGE_SIZE,
-					 DMA_PREP_INTERRUPT);
+					 DMA_PREP_INTERRUPT |
+					 DMA_COMPL_SKIP_SRC_UNMAP |
+					 DMA_COMPL_SKIP_DEST_UNMAP);
 	if (!tx) {
 		dev_err(dev, "Self-test memset prep failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
 	async_tx_ack(tx);
@@ -1025,7 +1060,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (cookie < 0) {
 		dev_err(dev, "Self-test memset setup failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 	dma->device_issue_pending(dma_chan);
 
@@ -1034,9 +1069,11 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
 		dev_err(dev, "Self-test memset timed out\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
+	dma_unmap_page(dev, dma_addr, PAGE_SIZE, DMA_FROM_DEVICE);
+
 	for (i = 0; i < PAGE_SIZE/sizeof(u32); i++) {
 		u32 *ptr = page_address(dest);
 		if (ptr[i]) {
@@ -1047,17 +1084,21 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	}
 
 	/* test for non-zero parity sum */
+	op = IOAT_OP_XOR_VAL;
+
 	xor_val_result = 0;
 	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
 		dma_srcs[i] = dma_map_page(dev, xor_val_srcs[i], 0, PAGE_SIZE,
 					   DMA_TO_DEVICE);
 	tx = dma->device_prep_dma_xor_val(dma_chan, dma_srcs,
 					  IOAT_NUM_SRC_TEST + 1, PAGE_SIZE,
-					  &xor_val_result, DMA_PREP_INTERRUPT);
+					  &xor_val_result, DMA_PREP_INTERRUPT |
+					  DMA_COMPL_SKIP_SRC_UNMAP |
+					  DMA_COMPL_SKIP_DEST_UNMAP);
 	if (!tx) {
 		dev_err(dev, "Self-test 2nd zero prep failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
 	async_tx_ack(tx);
@@ -1068,7 +1109,7 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (cookie < 0) {
 		dev_err(dev, "Self-test  2nd zero setup failed\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 	dma->device_issue_pending(dma_chan);
 
@@ -1077,15 +1118,31 @@ static int ioat_xor_val_self_test(struct ioatdma_device *device)
 	if (dma->device_tx_status(dma_chan, cookie, NULL) != DMA_SUCCESS) {
 		dev_err(dev, "Self-test 2nd validate timed out\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
 	if (xor_val_result != SUM_CHECK_P_RESULT) {
 		dev_err(dev, "Self-test validate failed compare\n");
 		err = -ENODEV;
-		goto free_resources;
+		goto dma_unmap;
 	}
 
+	for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+		dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE, DMA_TO_DEVICE);
+
+	goto free_resources;
+dma_unmap:
+	if (op == IOAT_OP_XOR) {
+		dma_unmap_page(dev, dest_dma, PAGE_SIZE, DMA_FROM_DEVICE);
+		for (i = 0; i < IOAT_NUM_SRC_TEST; i++)
+			dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
+				       DMA_TO_DEVICE);
+	} else if (op == IOAT_OP_XOR_VAL) {
+		for (i = 0; i < IOAT_NUM_SRC_TEST + 1; i++)
+			dma_unmap_page(dev, dma_srcs[i], PAGE_SIZE,
+				       DMA_TO_DEVICE);
+	} else if (op == IOAT_OP_FILL)
+		dma_unmap_page(dev, dma_addr, PAGE_SIZE, DMA_FROM_DEVICE);
 free_resources:
 	dma->device_free_chan_resources(dma_chan);
 out:
@@ -1126,12 +1183,7 @@ static int ioat3_reset_hw(struct ioat_chan_common *chan)
 	chanerr = readl(chan->reg_base + IOAT_CHANERR_OFFSET);
 	writel(chanerr, chan->reg_base + IOAT_CHANERR_OFFSET);
 
-	/* -= IOAT ver.3 workarounds =- */
-	/* Write CHANERRMSK_INT with 3E07h to mask out the errors
-	 * that can cause stability issues for IOAT ver.3, and clear any
-	 * pending errors
-	 */
-	pci_write_config_dword(pdev, IOAT_PCI_CHANERRMASK_INT_OFFSET, 0x3e07);
+	/* clear any pending errors */
 	err = pci_read_config_dword(pdev, IOAT_PCI_CHANERR_INT_OFFSET, &chanerr);
 	if (err) {
 		dev_err(&pdev->dev, "channel error register unreachable\n");
@@ -1187,6 +1239,26 @@ static bool is_snb_ioat(struct pci_dev *pdev)
 	}
 }
 
+static bool is_ivb_ioat(struct pci_dev *pdev)
+{
+	switch (pdev->device) {
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB0:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB1:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB2:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB3:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB4:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB5:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB6:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB7:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB8:
+	case PCI_DEVICE_ID_INTEL_IOAT_IVB9:
+		return true;
+	default:
+		return false;
+	}
+
+}
+
 int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 {
 	struct pci_dev *pdev = device->pdev;
@@ -1207,7 +1279,7 @@ int ioat3_dma_probe(struct ioatdma_device *device, int dca)
 	dma->device_alloc_chan_resources = ioat2_alloc_chan_resources;
 	dma->device_free_chan_resources = ioat2_free_chan_resources;
 
-	if (is_jf_ioat(pdev) || is_snb_ioat(pdev))
+	if (is_jf_ioat(pdev) || is_snb_ioat(pdev) || is_ivb_ioat(pdev))
 		dma->copy_align = 6;
 
 	dma_cap_set(DMA_INTERRUPT, dma->cap_mask);
diff --git a/drivers/dma/ioat/hw.h b/drivers/dma/ioat/hw.h
index d2ff3fda0b18..7cb74c62c719 100644
--- a/drivers/dma/ioat/hw.h
+++ b/drivers/dma/ioat/hw.h
@@ -35,6 +35,17 @@
 #define IOAT_VER_3_0            0x30    /* Version 3.0 */
 #define IOAT_VER_3_2            0x32    /* Version 3.2 */
 
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB0	0x0e20
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB1	0x0e21
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB2	0x0e22
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB3	0x0e23
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB4	0x0e24
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB5	0x0e25
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB6	0x0e26
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB7	0x0e27
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB8	0x0e2e
+#define PCI_DEVICE_ID_INTEL_IOAT_IVB9	0x0e2f
+
 int system_has_dca_enabled(struct pci_dev *pdev);
 
 struct ioat_dma_descriptor {
diff --git a/drivers/dma/ioat/pci.c b/drivers/dma/ioat/pci.c
index 4f686c527ab6..71c7ecd80fac 100644
--- a/drivers/dma/ioat/pci.c
+++ b/drivers/dma/ioat/pci.c
@@ -40,17 +40,6 @@ MODULE_VERSION(IOAT_DMA_VERSION);
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_AUTHOR("Intel Corporation");
 
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB0	0x0e20
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB1	0x0e21
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB2	0x0e22
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB3	0x0e23
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB4	0x0e24
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB5	0x0e25
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB6	0x0e26
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB7	0x0e27
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB8	0x0e2e
-#define PCI_DEVICE_ID_INTEL_IOAT_IVB9	0x0e2f
-
 static struct pci_device_id ioat_pci_tbl[] = {
 	/* I/OAT v1 platforms */
 	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT) },