Merge tag 'dma-mapping-4.14' of git://git.infradead.org/users/hch/dma-mapping

Pull dma-mapping updates from Christoph Hellwig:

 - removal of the old dma_alloc_noncoherent interface

 - remove unused flags to dma_declare_coherent_memory

 - restrict OF DMA configuration to specific physical busses

 - use the iommu mailing list for dma-mapping questions and patches

* tag 'dma-mapping-4.14' of git://git.infradead.org/users/hch/dma-mapping:
  dma-coherent: fix dma_declare_coherent_memory() logic error
  ARM: imx: mx31moboard: Remove unused 'dma' variable
  dma-coherent: remove an unused variable
  MAINTAINERS: use the iommu list for the dma-mapping subsystem
  dma-coherent: remove the DMA_MEMORY_MAP and DMA_MEMORY_IO flags
  dma-coherent: remove the DMA_MEMORY_INCLUDES_CHILDREN flag
  of: restrict DMA configuration
  dma-mapping: remove dma_alloc_noncoherent and dma_free_noncoherent
  i825xx: switch to switch to dma_alloc_attrs
  au1000_eth: switch to dma_alloc_attrs
  sgiseeq: switch to dma_alloc_attrs
  dma-mapping: reduce dma_mapping_error inline bloat

22 files changed, 156 insertions, 210 deletions

diff --git a/Documentation/DMA-API.txt b/Documentation/DMA-API.txt
index 45b29326d719..ac66ae2509a9 100644
--- a/Documentation/DMA-API.txt
+++ b/Documentation/DMA-API.txt
@@ -515,14 +515,15 @@ API at all.
 ::
 
 	void *
-	dma_alloc_noncoherent(struct device *dev, size_t size,
-			      dma_addr_t *dma_handle, gfp_t flag)
+	dma_alloc_attrs(struct device *dev, size_t size, dma_addr_t *dma_handle,
+			gfp_t flag, unsigned long attrs)
 
-Identical to dma_alloc_coherent() except that the platform will
-choose to return either consistent or non-consistent memory as it sees
-fit.  By using this API, you are guaranteeing to the platform that you
-have all the correct and necessary sync points for this memory in the
-driver should it choose to return non-consistent memory.
+Identical to dma_alloc_coherent() except that when the
+DMA_ATTR_NON_CONSISTENT flags is passed in the attrs argument, the
+platform will choose to return either consistent or non-consistent memory
+as it sees fit.  By using this API, you are guaranteeing to the platform
+that you have all the correct and necessary sync points for this memory
+in the driver should it choose to return non-consistent memory.
 
 Note: where the platform can return consistent memory, it will
 guarantee that the sync points become nops.
@@ -535,12 +536,13 @@ that simply cannot make consistent memory.
 ::
 
 	void
-	dma_free_noncoherent(struct device *dev, size_t size, void *cpu_addr,
-			     dma_addr_t dma_handle)
+	dma_free_attrs(struct device *dev, size_t size, void *cpu_addr,
+		       dma_addr_t dma_handle, unsigned long attrs)
 
-Free memory allocated by the nonconsistent API.  All parameters must
-be identical to those passed in (and returned by
-dma_alloc_noncoherent()).
+Free memory allocated by the dma_alloc_attrs().  All parameters common
+parameters must identical to those otherwise passed to dma_fre_coherent,
+and the attrs argument must be identical to the attrs passed to
+dma_alloc_attrs().
 
 ::
 
@@ -564,8 +566,8 @@ memory or doing partial flushes.
 	dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 		       enum dma_data_direction direction)
 
-Do a partial sync of memory that was allocated by
-dma_alloc_noncoherent(), starting at virtual address vaddr and
+Do a partial sync of memory that was allocated by dma_alloc_attrs() with
+the DMA_ATTR_NON_CONSISTENT flag starting at virtual address vaddr and
 continuing on for size.  Again, you *must* observe the cache line
 boundaries when doing this.
 
@@ -590,34 +592,11 @@ size is the size of the area (must be multiples of PAGE_SIZE).
 
 flags can be ORed together and are:
 
-- DMA_MEMORY_MAP - request that the memory returned from
-  dma_alloc_coherent() be directly writable.
-
-- DMA_MEMORY_IO - request that the memory returned from
-  dma_alloc_coherent() be addressable using read()/write()/memcpy_toio() etc.
-
-One or both of these flags must be present.
-
-- DMA_MEMORY_INCLUDES_CHILDREN - make the declared memory be allocated by
-  dma_alloc_coherent of any child devices of this one (for memory residing
-  on a bridge).
-
 - DMA_MEMORY_EXCLUSIVE - only allocate memory from the declared regions.
   Do not allow dma_alloc_coherent() to fall back to system memory when
   it's out of memory in the declared region.
 
-The return value will be either DMA_MEMORY_MAP or DMA_MEMORY_IO and
-must correspond to a passed in flag (i.e. no returning DMA_MEMORY_IO
-if only DMA_MEMORY_MAP were passed in) for success or zero for
-failure.
-
-Note, for DMA_MEMORY_IO returns, all subsequent memory returned by
-dma_alloc_coherent() may no longer be accessed directly, but instead
-must be accessed using the correct bus functions.  If your driver
-isn't prepared to handle this contingency, it should not specify
-DMA_MEMORY_IO in the input flags.
-
-As a simplification for the platforms, only **one** such region of
+As a simplification for the platforms, only *one* such region of
 memory may be declared per device.
 
 For reasons of efficiency, most platforms choose to track the declared
diff --git a/MAINTAINERS b/MAINTAINERS
index e57a4eaec077..f46a3225e398 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4219,7 +4219,7 @@ DMA MAPPING HELPERS
 M:	Christoph Hellwig <hch@lst.de>
 M:	Marek Szyprowski <m.szyprowski@samsung.com>
 R:	Robin Murphy <robin.murphy@arm.com>
-L:	linux-kernel@vger.kernel.org
+L:	iommu@lists.linux-foundation.org
 T:	git git://git.infradead.org/users/hch/dma-mapping.git
 W:	http://git.infradead.org/users/hch/dma-mapping.git
 S:	Supported
diff --git a/arch/arm/mach-imx/mach-imx27_visstrim_m10.c b/arch/arm/mach-imx/mach-imx27_visstrim_m10.c
index dd75a4756761..5169dfba9718 100644
--- a/arch/arm/mach-imx/mach-imx27_visstrim_m10.c
+++ b/arch/arm/mach-imx/mach-imx27_visstrim_m10.c
@@ -245,7 +245,6 @@ static phys_addr_t mx2_camera_base __initdata;
 static void __init visstrim_analog_camera_init(void)
 {
 	struct platform_device *pdev;
-	int dma;
 
 	gpio_set_value(TVP5150_PWDN, 1);
 	ndelay(1);
@@ -258,12 +257,9 @@ static void __init visstrim_analog_camera_init(void)
 	if (IS_ERR(pdev))
 		return;
 
-	dma = dma_declare_coherent_memory(&pdev->dev,
-				mx2_camera_base, mx2_camera_base,
-				MX2_CAMERA_BUF_SIZE,
-				DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE);
-	if (!(dma & DMA_MEMORY_MAP))
-		return;
+	dma_declare_coherent_memory(&pdev->dev, mx2_camera_base,
+				    mx2_camera_base, MX2_CAMERA_BUF_SIZE,
+				    DMA_MEMORY_EXCLUSIVE);
 }
 
 static void __init visstrim_reserve(void)
@@ -444,16 +440,13 @@ static const struct imx_ssi_platform_data visstrim_m10_ssi_pdata __initconst = {
 static void __init visstrim_coda_init(void)
 {
 	struct platform_device *pdev;
-	int dma;
 
 	pdev = imx27_add_coda();
-	dma = dma_declare_coherent_memory(&pdev->dev,
-					  mx2_camera_base + MX2_CAMERA_BUF_SIZE,
-					  mx2_camera_base + MX2_CAMERA_BUF_SIZE,
-					  MX2_CAMERA_BUF_SIZE,
-					  DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE);
-	if (!(dma & DMA_MEMORY_MAP))
-		return;
+	dma_declare_coherent_memory(&pdev->dev,
+				    mx2_camera_base + MX2_CAMERA_BUF_SIZE,
+				    mx2_camera_base + MX2_CAMERA_BUF_SIZE,
+				    MX2_CAMERA_BUF_SIZE,
+				    DMA_MEMORY_EXCLUSIVE);
 }
 
 /* DMA deinterlace */
@@ -466,24 +459,21 @@ static void __init visstrim_deinterlace_init(void)
 {
 	int ret = -ENOMEM;
 	struct platform_device *pdev = &visstrim_deinterlace;
-	int dma;
 
 	ret = platform_device_register(pdev);
 
-	dma = dma_declare_coherent_memory(&pdev->dev,
-					  mx2_camera_base + 2 * MX2_CAMERA_BUF_SIZE,
-					  mx2_camera_base + 2 * MX2_CAMERA_BUF_SIZE,
-					  MX2_CAMERA_BUF_SIZE,
-					  DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE);
-	if (!(dma & DMA_MEMORY_MAP))
-		return;
+	dma_declare_coherent_memory(&pdev->dev,
+				    mx2_camera_base + 2 * MX2_CAMERA_BUF_SIZE,
+				    mx2_camera_base + 2 * MX2_CAMERA_BUF_SIZE,
+				    MX2_CAMERA_BUF_SIZE,
+				    DMA_MEMORY_EXCLUSIVE);
 }
 
 /* Emma-PrP for format conversion */
 static void __init visstrim_emmaprp_init(void)
 {
 	struct platform_device *pdev;
-	int dma;
+	int ret;
 
 	pdev = imx27_add_mx2_emmaprp();
 	if (IS_ERR(pdev))
@@ -493,11 +483,11 @@ static void __init visstrim_emmaprp_init(void)
 	 * Use the same memory area as the analog camera since both
 	 * devices are, by nature, exclusive.
 	 */
-	dma = dma_declare_coherent_memory(&pdev->dev,
+	ret = dma_declare_coherent_memory(&pdev->dev,
 				mx2_camera_base, mx2_camera_base,
 				MX2_CAMERA_BUF_SIZE,
-				DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE);
-	if (!(dma & DMA_MEMORY_MAP))
+				DMA_MEMORY_EXCLUSIVE);
+	if (ret)
 		pr_err("Failed to declare memory for emmaprp\n");
 }
 
diff --git a/arch/arm/mach-imx/mach-mx31moboard.c b/arch/arm/mach-imx/mach-mx31moboard.c
index bde9a9af6714..7716f83aecdd 100644
--- a/arch/arm/mach-imx/mach-mx31moboard.c
+++ b/arch/arm/mach-imx/mach-mx31moboard.c
@@ -475,7 +475,7 @@ static phys_addr_t mx3_camera_base __initdata;
 
 static int __init mx31moboard_init_cam(void)
 {
-	int dma, ret = -ENOMEM;
+	int ret;
 	struct platform_device *pdev;
 
 	imx31_add_ipu_core();
@@ -484,11 +484,11 @@ static int __init mx31moboard_init_cam(void)
 	if (IS_ERR(pdev))
 		return PTR_ERR(pdev);
 
-	dma = dma_declare_coherent_memory(&pdev->dev,
-					mx3_camera_base, mx3_camera_base,
-					MX3_CAMERA_BUF_SIZE,
-					DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE);
-	if (!(dma & DMA_MEMORY_MAP))
+	ret = dma_declare_coherent_memory(&pdev->dev,
+					  mx3_camera_base, mx3_camera_base,
+					  MX3_CAMERA_BUF_SIZE,
+					  DMA_MEMORY_EXCLUSIVE);
+	if (ret)
 		goto err;
 
 	ret = platform_device_add(pdev);
diff --git a/arch/metag/include/asm/dma-mapping.h b/arch/metag/include/asm/dma-mapping.h
index fad3dc3cb210..ea573be2b6d0 100644
--- a/arch/metag/include/asm/dma-mapping.h
+++ b/arch/metag/include/asm/dma-mapping.h
@@ -9,7 +9,7 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 }
 
 /*
- * dma_alloc_noncoherent() returns non-cacheable memory, so there's no need to
+ * dma_alloc_attrs() always returns non-cacheable memory, so there's no need to
  * do any flushing here.
  */
 static inline void
diff --git a/arch/nios2/include/asm/dma-mapping.h b/arch/nios2/include/asm/dma-mapping.h
index 7b3c6f280293..f8dc62222741 100644
--- a/arch/nios2/include/asm/dma-mapping.h
+++ b/arch/nios2/include/asm/dma-mapping.h
@@ -18,7 +18,7 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 }
 
 /*
- * dma_alloc_noncoherent() returns non-cacheable memory, so there's no need to
+ * dma_alloc_attrs() always returns non-cacheable memory, so there's no need to
  * do any flushing here.
  */
 static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
diff --git a/arch/sh/drivers/pci/fixups-dreamcast.c b/arch/sh/drivers/pci/fixups-dreamcast.c
index 9d597f7ab8dd..48aaefd8f5d6 100644
--- a/arch/sh/drivers/pci/fixups-dreamcast.c
+++ b/arch/sh/drivers/pci/fixups-dreamcast.c
@@ -63,11 +63,10 @@ static void gapspci_fixup_resources(struct pci_dev *dev)
 		res.end = GAPSPCI_DMA_BASE + GAPSPCI_DMA_SIZE - 1;
 		res.flags = IORESOURCE_MEM;
 		pcibios_resource_to_bus(dev->bus, &region, &res);
-		BUG_ON(!dma_declare_coherent_memory(&dev->dev,
+		BUG_ON(dma_declare_coherent_memory(&dev->dev,
 						res.start,
 						region.start,
 						resource_size(&res),
-						DMA_MEMORY_MAP |
 						DMA_MEMORY_EXCLUSIVE));
 		break;
 	default:
diff --git a/arch/tile/include/asm/dma-mapping.h b/arch/tile/include/asm/dma-mapping.h
index bbc71a29b2c6..7061dc8af43a 100644
--- a/arch/tile/include/asm/dma-mapping.h
+++ b/arch/tile/include/asm/dma-mapping.h
@@ -68,8 +68,8 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 int dma_set_mask(struct device *dev, u64 mask);
 
 /*
- * dma_alloc_noncoherent() is #defined to return coherent memory,
- * so there's no need to do any flushing here.
+ * dma_alloc_attrs() always returns non-cacheable memory, so there's no need to
+ * do any flushing here.
  */
 static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 				  enum dma_data_direction direction)
diff --git a/drivers/base/dma-coherent.c b/drivers/base/dma-coherent.c
index 1c152aed6b82..a39b2166b145 100644
--- a/drivers/base/dma-coherent.c
+++ b/drivers/base/dma-coherent.c
@@ -37,7 +37,7 @@ static inline dma_addr_t dma_get_device_base(struct device *dev,
 		return mem->device_base;
 }
 
-static bool dma_init_coherent_memory(
+static int dma_init_coherent_memory(
 	phys_addr_t phys_addr, dma_addr_t device_addr, size_t size, int flags,
 	struct dma_coherent_mem **mem)
 {
@@ -45,25 +45,28 @@ static bool dma_init_coherent_memory(
 	void __iomem *mem_base = NULL;
 	int pages = size >> PAGE_SHIFT;
 	int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
+	int ret;
 
-	if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
-		goto out;
-	if (!size)
+	if (!size) {
+		ret = -EINVAL;
 		goto out;
+	}
 
-	if (flags & DMA_MEMORY_MAP)
-		mem_base = memremap(phys_addr, size, MEMREMAP_WC);
-	else
-		mem_base = ioremap(phys_addr, size);
-	if (!mem_base)
+	mem_base = memremap(phys_addr, size, MEMREMAP_WC);
+	if (!mem_base) {
+		ret = -EINVAL;
 		goto out;
-
+	}
 	dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
-	if (!dma_mem)
+	if (!dma_mem) {
+		ret = -ENOMEM;
 		goto out;
+	}
 	dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
-	if (!dma_mem->bitmap)
+	if (!dma_mem->bitmap) {
+		ret = -ENOMEM;
 		goto out;
+	}
 
 	dma_mem->virt_base = mem_base;
 	dma_mem->device_base = device_addr;
@@ -73,17 +76,13 @@ static bool dma_init_coherent_memory(
 	spin_lock_init(&dma_mem->spinlock);
 
 	*mem = dma_mem;
-	return true;
+	return 0;
 
 out:
 	kfree(dma_mem);
-	if (mem_base) {
-		if (flags & DMA_MEMORY_MAP)
-			memunmap(mem_base);
-		else
-			iounmap(mem_base);
-	}
-	return false;
+	if (mem_base)
+		memunmap(mem_base);
+	return ret;
 }
 
 static void dma_release_coherent_memory(struct dma_coherent_mem *mem)
@@ -91,10 +90,7 @@ static void dma_release_coherent_memory(struct dma_coherent_mem *mem)
 	if (!mem)
 		return;
 
-	if (mem->flags & DMA_MEMORY_MAP)
-		memunmap(mem->virt_base);
-	else
-		iounmap(mem->virt_base);
+	memunmap(mem->virt_base);
 	kfree(mem->bitmap);
 	kfree(mem);
 }
@@ -109,8 +105,6 @@ static int dma_assign_coherent_memory(struct device *dev,
 		return -EBUSY;
 
 	dev->dma_mem = mem;
-	/* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
-
 	return 0;
 }
 
@@ -118,16 +112,16 @@ int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
 				dma_addr_t device_addr, size_t size, int flags)
 {
 	struct dma_coherent_mem *mem;
+	int ret;
 
-	if (!dma_init_coherent_memory(phys_addr, device_addr, size, flags,
-				      &mem))
-		return 0;
-
-	if (dma_assign_coherent_memory(dev, mem) == 0)
-		return flags & DMA_MEMORY_MAP ? DMA_MEMORY_MAP : DMA_MEMORY_IO;
+	ret = dma_init_coherent_memory(phys_addr, device_addr, size, flags, &mem);
+	if (ret)
+		return ret;
 
-	dma_release_coherent_memory(mem);
-	return 0;
+	ret = dma_assign_coherent_memory(dev, mem);
+	if (ret)
+		dma_release_coherent_memory(mem);
+	return ret;
 }
 EXPORT_SYMBOL(dma_declare_coherent_memory);
 
@@ -171,7 +165,6 @@ static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
 	int order = get_order(size);
 	unsigned long flags;
 	int pageno;
-	int dma_memory_map;
 	void *ret;
 
 	spin_lock_irqsave(&mem->spinlock, flags);
@@ -188,15 +181,9 @@ static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
 	 */
 	*dma_handle = mem->device_base + (pageno << PAGE_SHIFT);
 	ret = mem->virt_base + (pageno << PAGE_SHIFT);
-	dma_memory_map = (mem->flags & DMA_MEMORY_MAP);
 	spin_unlock_irqrestore(&mem->spinlock, flags);
-	if (dma_memory_map)
-		memset(ret, 0, size);
-	else
-		memset_io(ret, 0, size);
-
+	memset(ret, 0, size);
 	return ret;
-
 err:
 	spin_unlock_irqrestore(&mem->spinlock, flags);
 	return NULL;
@@ -359,14 +346,18 @@ static struct reserved_mem *dma_reserved_default_memory __initdata;
 static int rmem_dma_device_init(struct reserved_mem *rmem, struct device *dev)
 {
 	struct dma_coherent_mem *mem = rmem->priv;
+	int ret;
 
-	if (!mem &&
-	    !dma_init_coherent_memory(rmem->base, rmem->base, rmem->size,
-				      DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE,
-				      &mem)) {
+	if (!mem)
+		return -ENODEV;
+
+	ret = dma_init_coherent_memory(rmem->base, rmem->base, rmem->size,
+				       DMA_MEMORY_EXCLUSIVE, &mem);
+
+	if (ret) {
 		pr_err("Reserved memory: failed to init DMA memory pool at %pa, size %ld MiB\n",
 			&rmem->base, (unsigned long)rmem->size / SZ_1M);
-		return -ENODEV;
+		return ret;
 	}
 	mem->use_dev_dma_pfn_offset = true;
 	rmem->priv = mem;
diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c
index b555ff9dd8fc..e584eddef0a7 100644
--- a/drivers/base/dma-mapping.c
+++ b/drivers/base/dma-mapping.c
@@ -176,13 +176,10 @@ int dmam_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
 
 	rc = dma_declare_coherent_memory(dev, phys_addr, device_addr, size,
 					 flags);
-	if (rc) {
+	if (!rc)
 		devres_add(dev, res);
-		rc = 0;
-	} else {
+	else
 		devres_free(res);
-		rc = -ENOMEM;
-	}
 
 	return rc;
 }
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index 23f33f95d4a6..d1aed2513bd9 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -451,9 +451,6 @@ static struct port_buffer *alloc_buf(struct virtqueue *vq, size_t buf_size,
 		 * device is created by remoteproc, the DMA memory is
 		 * associated with the grandparent device:
 		 * vdev => rproc => platform-dev.
-		 * The code here would have been less quirky if
-		 * DMA_MEMORY_INCLUDES_CHILDREN had been supported
-		 * in dma-coherent.c
 		 */
 		if (!vq->vdev->dev.parent || !vq->vdev->dev.parent->parent)
 			goto free_buf;
diff --git a/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c b/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
index 96dc01750bc0..36762ec954e7 100644
--- a/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
+++ b/drivers/media/platform/soc_camera/sh_mobile_ceu_camera.c
@@ -1708,11 +1708,10 @@ static int sh_mobile_ceu_probe(struct platform_device *pdev)
 		err = dma_declare_coherent_memory(&pdev->dev, res->start,
 						  res->start,
 						  resource_size(res),
-						  DMA_MEMORY_MAP |
 						  DMA_MEMORY_EXCLUSIVE);
-		if (!err) {
+		if (err) {
 			dev_err(&pdev->dev, "Unable to declare CEU memory.\n");
-			return -ENXIO;
+			return err;
 		}
 
 		pcdev->video_limit = resource_size(res);
diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c
index a3c90fe5de00..73ca8879ada7 100644
--- a/drivers/net/ethernet/amd/au1000_eth.c
+++ b/drivers/net/ethernet/amd/au1000_eth.c
@@ -1180,9 +1180,10 @@ static int au1000_probe(struct platform_device *pdev)
 	/* Allocate the data buffers
 	 * Snooping works fine with eth on all au1xxx
 	 */
-	aup->vaddr = (u32)dma_alloc_noncoherent(NULL, MAX_BUF_SIZE *
-						(NUM_TX_BUFFS + NUM_RX_BUFFS),
-						&aup->dma_addr,	0);
+	aup->vaddr = (u32)dma_alloc_attrs(NULL, MAX_BUF_SIZE *
+					  (NUM_TX_BUFFS + NUM_RX_BUFFS),
+					  &aup->dma_addr, 0,
+					  DMA_ATTR_NON_CONSISTENT);
 	if (!aup->vaddr) {
 		dev_err(&pdev->dev, "failed to allocate data buffers\n");
 		err = -ENOMEM;
@@ -1361,8 +1362,9 @@ err_remap3:
 err_remap2:
 	iounmap(aup->mac);
 err_remap1:
-	dma_free_noncoherent(NULL, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
-			     (void *)aup->vaddr, aup->dma_addr);
+	dma_free_attrs(NULL, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
+			(void *)aup->vaddr, aup->dma_addr,
+			DMA_ATTR_NON_CONSISTENT);
 err_vaddr:
 	free_netdev(dev);
 err_alloc:
@@ -1394,9 +1396,9 @@ static int au1000_remove(struct platform_device *pdev)
 		if (aup->tx_db_inuse[i])
 			au1000_ReleaseDB(aup, aup->tx_db_inuse[i]);
 
-	dma_free_noncoherent(NULL, MAX_BUF_SIZE *
-			(NUM_TX_BUFFS + NUM_RX_BUFFS),
-			(void *)aup->vaddr, aup->dma_addr);
+	dma_free_attrs(NULL, MAX_BUF_SIZE * (NUM_TX_BUFFS + NUM_RX_BUFFS),
+			(void *)aup->vaddr, aup->dma_addr,
+			DMA_ATTR_NON_CONSISTENT);
 
 	iounmap(aup->macdma);
 	iounmap(aup->mac);
diff --git a/drivers/net/ethernet/i825xx/lasi_82596.c b/drivers/net/ethernet/i825xx/lasi_82596.c
index aa22e108f09b..b69c622ba8b2 100644
--- a/drivers/net/ethernet/i825xx/lasi_82596.c
+++ b/drivers/net/ethernet/i825xx/lasi_82596.c
@@ -96,8 +96,6 @@
 
 #define OPT_SWAP_PORT	0x0001	/* Need to wordswp on the MPU port */
 
-#define DMA_ALLOC                        dma_alloc_noncoherent
-#define DMA_FREE                         dma_free_noncoherent
 #define DMA_WBACK(ndev, addr, len) \
 	do { dma_cache_sync((ndev)->dev.parent, (void *)addr, len, DMA_TO_DEVICE); } while (0)
 
@@ -200,8 +198,8 @@ static int __exit lan_remove_chip(struct parisc_device *pdev)
 	struct i596_private *lp = netdev_priv(dev);
 
 	unregister_netdev (dev);
-	DMA_FREE(&pdev->dev, sizeof(struct i596_private),
-		 (void *)lp->dma, lp->dma_addr);
+	dma_free_attrs(&pdev->dev, sizeof(struct i596_private), lp->dma,
+		       lp->dma_addr, DMA_ATTR_NON_CONSISTENT);
 	free_netdev (dev);
 	return 0;
 }
diff --git a/drivers/net/ethernet/i825xx/lib82596.c b/drivers/net/ethernet/i825xx/lib82596.c
index 8449c58f01fd..f00a1dc2128c 100644
--- a/drivers/net/ethernet/i825xx/lib82596.c
+++ b/drivers/net/ethernet/i825xx/lib82596.c
@@ -1063,8 +1063,9 @@ static int i82596_probe(struct net_device *dev)
 	if (!dev->base_addr || !dev->irq)
 		return -ENODEV;
 
-	dma = (struct i596_dma *) DMA_ALLOC(dev->dev.parent,
-		sizeof(struct i596_dma), &lp->dma_addr, GFP_KERNEL);
+	dma = dma_alloc_attrs(dev->dev.parent, sizeof(struct i596_dma),
+			      &lp->dma_addr, GFP_KERNEL,
+			      DMA_ATTR_NON_CONSISTENT);
 	if (!dma) {
 		printk(KERN_ERR "%s: Couldn't get shared memory\n", __FILE__);
 		return -ENOMEM;
@@ -1085,8 +1086,8 @@ static int i82596_probe(struct net_device *dev)
 
 	i = register_netdev(dev);
 	if (i) {
-		DMA_FREE(dev->dev.parent, sizeof(struct i596_dma),
-				    (void *)dma, lp->dma_addr);
+		dma_free_attrs(dev->dev.parent, sizeof(struct i596_dma),
+			       dma, lp->dma_addr, DMA_ATTR_NON_CONSISTENT);
 		return i;
 	}
 
diff --git a/drivers/net/ethernet/i825xx/sni_82596.c b/drivers/net/ethernet/i825xx/sni_82596.c
index 2af7f77345fb..b2c04a789744 100644
--- a/drivers/net/ethernet/i825xx/sni_82596.c
+++ b/drivers/net/ethernet/i825xx/sni_82596.c
@@ -23,8 +23,6 @@
 
 static const char sni_82596_string[] = "snirm_82596";
 
-#define DMA_ALLOC                      dma_alloc_coherent
-#define DMA_FREE                       dma_free_coherent
 #define DMA_WBACK(priv, addr, len)     do { } while (0)
 #define DMA_INV(priv, addr, len)       do { } while (0)
 #define DMA_WBACK_INV(priv, addr, len) do { } while (0)
@@ -152,8 +150,8 @@ static int sni_82596_driver_remove(struct platform_device *pdev)
 	struct i596_private *lp = netdev_priv(dev);
 
 	unregister_netdev(dev);
-	DMA_FREE(dev->dev.parent, sizeof(struct i596_private),
-		 lp->dma, lp->dma_addr);
+	dma_free_attrs(dev->dev.parent, sizeof(struct i596_private), lp->dma,
+		       lp->dma_addr, DMA_ATTR_NON_CONSISTENT);
 	iounmap(lp->ca);
 	iounmap(lp->mpu_port);
 	free_netdev (dev);
diff --git a/drivers/net/ethernet/seeq/sgiseeq.c b/drivers/net/ethernet/seeq/sgiseeq.c
index 70347720fdf9..573691bc3b71 100644
--- a/drivers/net/ethernet/seeq/sgiseeq.c
+++ b/drivers/net/ethernet/seeq/sgiseeq.c
@@ -737,8 +737,8 @@ static int sgiseeq_probe(struct platform_device *pdev)
 	sp = netdev_priv(dev);
 
 	/* Make private data page aligned */
-	sr = dma_alloc_noncoherent(&pdev->dev, sizeof(*sp->srings),
-				&sp->srings_dma, GFP_KERNEL);
+	sr = dma_alloc_attrs(&pdev->dev, sizeof(*sp->srings), &sp->srings_dma,
+			     GFP_KERNEL, DMA_ATTR_NON_CONSISTENT);
 	if (!sr) {
 		printk(KERN_ERR "Sgiseeq: Page alloc failed, aborting.\n");
 		err = -ENOMEM;
@@ -813,8 +813,8 @@ static int sgiseeq_remove(struct platform_device *pdev)
 	struct sgiseeq_private *sp = netdev_priv(dev);
 
 	unregister_netdev(dev);
-	dma_free_noncoherent(&pdev->dev, sizeof(*sp->srings), sp->srings,
-			     sp->srings_dma);
+	dma_free_attrs(&pdev->dev, sizeof(*sp->srings), sp->srings,
+		       sp->srings_dma, DMA_ATTR_NON_CONSISTENT);
 	free_netdev(dev);
 
 	return 0;
diff --git a/drivers/of/device.c b/drivers/of/device.c
index 17b66e9715d2..64b710265d39 100644
--- a/drivers/of/device.c
+++ b/drivers/of/device.c
@@ -9,6 +9,9 @@
 #include <linux/module.h>
 #include <linux/mod_devicetable.h>
 #include <linux/slab.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/amba/bus.h>
 
 #include <asm/errno.h>
 #include "of_private.h"
@@ -84,31 +87,28 @@ int of_device_add(struct platform_device *ofdev)
  */
 int of_dma_configure(struct device *dev, struct device_node *np)
 {
-	u64 dma_addr, paddr, size;
+	u64 dma_addr, paddr, size = 0;
 	int ret;
 	bool coherent;
 	unsigned long offset;
 	const struct iommu_ops *iommu;
 	u64 mask;
 
-	/*
-	 * Set default coherent_dma_mask to 32 bit.  Drivers are expected to
-	 * setup the correct supported mask.
-	 */
-	if (!dev->coherent_dma_mask)
-		dev->coherent_dma_mask = DMA_BIT_MASK(32);
-
-	/*
-	 * Set it to coherent_dma_mask by default if the architecture
-	 * code has not set it.
-	 */
-	if (!dev->dma_mask)
-		dev->dma_mask = &dev->coherent_dma_mask;
-
 	ret = of_dma_get_range(np, &dma_addr, &paddr, &size);
 	if (ret < 0) {
+		/*
+		 * For legacy reasons, we have to assume some devices need
+		 * DMA configuration regardless of whether "dma-ranges" is
+		 * correctly specified or not.
+		 */
+		if (!dev_is_pci(dev) &&
+#ifdef CONFIG_ARM_AMBA
+		    dev->bus != &amba_bustype &&
+#endif
+		    dev->bus != &platform_bus_type)
+			return ret == -ENODEV ? 0 : ret;
+
 		dma_addr = offset = 0;
-		size = max(dev->coherent_dma_mask, dev->coherent_dma_mask + 1);
 	} else {
 		offset = PFN_DOWN(paddr - dma_addr);
 
@@ -129,6 +129,22 @@ int of_dma_configure(struct device *dev, struct device_node *np)
 		dev_dbg(dev, "dma_pfn_offset(%#08lx)\n", offset);
 	}
 
+	/*
+	 * Set default coherent_dma_mask to 32 bit.  Drivers are expected to
+	 * setup the correct supported mask.
+	 */
+	if (!dev->coherent_dma_mask)
+		dev->coherent_dma_mask = DMA_BIT_MASK(32);
+	/*
+	 * Set it to coherent_dma_mask by default if the architecture
+	 * code has not set it.
+	 */
+	if (!dev->dma_mask)
+		dev->dma_mask = &dev->coherent_dma_mask;
+
+	if (!size)
+		size = max(dev->coherent_dma_mask, dev->coherent_dma_mask + 1);
+
 	dev->dma_pfn_offset = offset;
 
 	/*
diff --git a/drivers/scsi/NCR_Q720.c b/drivers/scsi/NCR_Q720.c
index 05835bf1bf9c..54e7d26908ee 100644
--- a/drivers/scsi/NCR_Q720.c
+++ b/drivers/scsi/NCR_Q720.c
@@ -217,8 +217,7 @@ NCR_Q720_probe(struct device *dev)
 	}
 	
 	if (dma_declare_coherent_memory(dev, base_addr, base_addr,
-					mem_size, DMA_MEMORY_MAP)
-	    != DMA_MEMORY_MAP) {
+					mem_size, 0)) {
 		printk(KERN_ERR "NCR_Q720: DMA declare memory failed\n");
 		goto out_release_region;
 	}
diff --git a/drivers/usb/host/ohci-sm501.c b/drivers/usb/host/ohci-sm501.c
index a8b8d8b8d9f3..d4e0f7cd96fa 100644
--- a/drivers/usb/host/ohci-sm501.c
+++ b/drivers/usb/host/ohci-sm501.c
@@ -123,13 +123,12 @@ static int ohci_hcd_sm501_drv_probe(struct platform_device *pdev)
 	 * regular memory. The HCD_LOCAL_MEM flag does just that.
 	 */
 
-	if (!dma_declare_coherent_memory(dev, mem->start,
+	retval = dma_declare_coherent_memory(dev, mem->start,
 					 mem->start - mem->parent->start,
 					 resource_size(mem),
-					 DMA_MEMORY_MAP |
-					 DMA_MEMORY_EXCLUSIVE)) {
+					 DMA_MEMORY_EXCLUSIVE);
+	if (retval) {
 		dev_err(dev, "cannot declare coherent memory\n");
-		retval = -ENXIO;
 		goto err1;
 	}
 
diff --git a/drivers/usb/host/ohci-tmio.c b/drivers/usb/host/ohci-tmio.c
index cfcfadfc94fc..16d081a093bb 100644
--- a/drivers/usb/host/ohci-tmio.c
+++ b/drivers/usb/host/ohci-tmio.c
@@ -227,13 +227,10 @@ static int ohci_hcd_tmio_drv_probe(struct platform_device *dev)
 		goto err_ioremap_regs;
 	}
 
-	if (!dma_declare_coherent_memory(&dev->dev, sram->start,
-				sram->start,
-				resource_size(sram),
-				DMA_MEMORY_MAP | DMA_MEMORY_EXCLUSIVE)) {
-		ret = -EBUSY;
+	ret = dma_declare_coherent_memory(&dev->dev, sram->start, sram->start,
+				resource_size(sram), DMA_MEMORY_EXCLUSIVE);
+	if (ret)
 		goto err_dma_declare;
-	}
 
 	if (cell->enable) {
 		ret = cell->enable(dev);
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index 2189c79cde5d..29ce9815da87 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -550,26 +550,13 @@ static inline void dma_free_coherent(struct device *dev, size_t size,
 	return dma_free_attrs(dev, size, cpu_addr, dma_handle, 0);
 }
 
-static inline void *dma_alloc_noncoherent(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp)
-{
-	return dma_alloc_attrs(dev, size, dma_handle, gfp,
-			       DMA_ATTR_NON_CONSISTENT);
-}
-
-static inline void dma_free_noncoherent(struct device *dev, size_t size,
-		void *cpu_addr, dma_addr_t dma_handle)
-{
-	dma_free_attrs(dev, size, cpu_addr, dma_handle,
-		       DMA_ATTR_NON_CONSISTENT);
-}
-
 static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
-	debug_dma_mapping_error(dev, dma_addr);
+	const struct dma_map_ops *ops = get_dma_ops(dev);
 
-	if (get_dma_ops(dev)->mapping_error)
-		return get_dma_ops(dev)->mapping_error(dev, dma_addr);
+	debug_dma_mapping_error(dev, dma_addr);
+	if (ops->mapping_error)
+		return ops->mapping_error(dev, dma_addr);
 	return 0;
 }
 
@@ -720,10 +707,7 @@ static inline int dma_get_cache_alignment(void)
 #endif
 
 /* flags for the coherent memory api */
-#define	DMA_MEMORY_MAP			0x01
-#define DMA_MEMORY_IO			0x02
-#define DMA_MEMORY_INCLUDES_CHILDREN	0x04
-#define DMA_MEMORY_EXCLUSIVE		0x08
+#define DMA_MEMORY_EXCLUSIVE		0x01
 
 #ifdef CONFIG_HAVE_GENERIC_DMA_COHERENT
 int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
@@ -736,7 +720,7 @@ static inline int
 dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
 			    dma_addr_t device_addr, size_t size, int flags)
 {
-	return 0;
+	return -ENOSYS;
 }
 
 static inline void