Merge tag 'v5.2-rc6' into generic-dma-ops

Linux 5.2-rc6
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 5992eb9..9c588a1 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -1,24 +1,13 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
- * SWIOTLB-based DMA API implementation
- *
  * Copyright (C) 2012 ARM Ltd.
  * Author: Catalin Marinas <catalin.marinas@arm.com>
  */
 
 #include <linux/gfp.h>
-#include <linux/acpi.h>
-#include <linux/memblock.h>
 #include <linux/cache.h>
-#include <linux/export.h>
-#include <linux/slab.h>
-#include <linux/genalloc.h>
-#include <linux/dma-direct.h>
 #include <linux/dma-noncoherent.h>
-#include <linux/dma-contiguous.h>
-#include <linux/vmalloc.h>
-#include <linux/swiotlb.h>
-#include <linux/pci.h>
+#include <linux/dma-iommu.h>
 
 #include <asm/cacheflush.h>
 
@@ -47,37 +36,6 @@ void arch_dma_prep_coherent(struct page *page, size_t size)
 	__dma_flush_area(page_address(page), size);
 }
 
-#ifdef CONFIG_IOMMU_DMA
-static int __swiotlb_get_sgtable_page(struct sg_table *sgt,
-				      struct page *page, size_t size)
-{
-	int ret = sg_alloc_table(sgt, 1, GFP_KERNEL);
-
-	if (!ret)
-		sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0);
-
-	return ret;
-}
-
-static int __swiotlb_mmap_pfn(struct vm_area_struct *vma,
-			      unsigned long pfn, size_t size)
-{
-	int ret = -ENXIO;
-	unsigned long nr_vma_pages = vma_pages(vma);
-	unsigned long nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	unsigned long off = vma->vm_pgoff;
-
-	if (off < nr_pages && nr_vma_pages <= (nr_pages - off)) {
-		ret = remap_pfn_range(vma, vma->vm_start,
-				      pfn + off,
-				      vma->vm_end - vma->vm_start,
-				      vma->vm_page_prot);
-	}
-
-	return ret;
-}
-#endif /* CONFIG_IOMMU_DMA */
-
 static int __init arm64_dma_init(void)
 {
 	WARN_TAINT(ARCH_DMA_MINALIGN < cache_line_size(),
@@ -89,380 +47,18 @@ static int __init arm64_dma_init(void)
 arch_initcall(arm64_dma_init);
 
 #ifdef CONFIG_IOMMU_DMA
-#include <linux/dma-iommu.h>
-#include <linux/platform_device.h>
-#include <linux/amba/bus.h>
-
-/* Thankfully, all cache ops are by VA so we can ignore phys here */
-static void flush_page(struct device *dev, const void *virt, phys_addr_t phys)
-{
-	__dma_flush_area(virt, PAGE_SIZE);
-}
-
-static void *__iommu_alloc_attrs(struct device *dev, size_t size,
-				 dma_addr_t *handle, gfp_t gfp,
-				 unsigned long attrs)
-{
-	bool coherent = dev_is_dma_coherent(dev);
-	int ioprot = dma_info_to_prot(DMA_BIDIRECTIONAL, coherent, attrs);
-	size_t iosize = size;
-	void *addr;
-
-	if (WARN(!dev, "cannot create IOMMU mapping for unknown device\n"))
-		return NULL;
-
-	size = PAGE_ALIGN(size);
-
-	/*
-	 * Some drivers rely on this, and we probably don't want the
-	 * possibility of stale kernel data being read by devices anyway.
-	 */
-	gfp |= __GFP_ZERO;
-
-	if (!gfpflags_allow_blocking(gfp)) {
-		struct page *page;
-		/*
-		 * In atomic context we can't remap anything, so we'll only
-		 * get the virtually contiguous buffer we need by way of a
-		 * physically contiguous allocation.
-		 */
-		if (coherent) {
-			page = alloc_pages(gfp, get_order(size));
-			addr = page ? page_address(page) : NULL;
-		} else {
-			addr = dma_alloc_from_pool(size, &page, gfp);
-		}
-		if (!addr)
-			return NULL;
-
-		*handle = iommu_dma_map_page(dev, page, 0, iosize, ioprot);
-		if (*handle == DMA_MAPPING_ERROR) {
-			if (coherent)
-				__free_pages(page, get_order(size));
-			else
-				dma_free_from_pool(addr, size);
-			addr = NULL;
-		}
-	} else if (attrs & DMA_ATTR_FORCE_CONTIGUOUS) {
-		pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);
-		struct page *page;
-
-		page = dma_alloc_from_contiguous(dev, size >> PAGE_SHIFT,
-					get_order(size), gfp & __GFP_NOWARN);
-		if (!page)
-			return NULL;
-
-		*handle = iommu_dma_map_page(dev, page, 0, iosize, ioprot);
-		if (*handle == DMA_MAPPING_ERROR) {
-			dma_release_from_contiguous(dev, page,
-						    size >> PAGE_SHIFT);
-			return NULL;
-		}
-		addr = dma_common_contiguous_remap(page, size, VM_USERMAP,
-						   prot,
-						   __builtin_return_address(0));
-		if (addr) {
-			if (!coherent)
-				__dma_flush_area(page_to_virt(page), iosize);
-			memset(addr, 0, size);
-		} else {
-			iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs);
-			dma_release_from_contiguous(dev, page,
-						    size >> PAGE_SHIFT);
-		}
-	} else {
-		pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);
-		struct page **pages;
-
-		pages = iommu_dma_alloc(dev, iosize, gfp, attrs, ioprot,
-					handle, flush_page);
-		if (!pages)
-			return NULL;
-
-		addr = dma_common_pages_remap(pages, size, VM_USERMAP, prot,
-					      __builtin_return_address(0));
-		if (!addr)
-			iommu_dma_free(dev, pages, iosize, handle);
-	}
-	return addr;
-}
-
-static void __iommu_free_attrs(struct device *dev, size_t size, void *cpu_addr,
-			       dma_addr_t handle, unsigned long attrs)
-{
-	size_t iosize = size;
-
-	size = PAGE_ALIGN(size);
-	/*
-	 * @cpu_addr will be one of 4 things depending on how it was allocated:
-	 * - A remapped array of pages for contiguous allocations.
-	 * - A remapped array of pages from iommu_dma_alloc(), for all
-	 *   non-atomic allocations.
-	 * - A non-cacheable alias from the atomic pool, for atomic
-	 *   allocations by non-coherent devices.
-	 * - A normal lowmem address, for atomic allocations by
-	 *   coherent devices.
-	 * Hence how dodgy the below logic looks...
-	 */
-	if (dma_in_atomic_pool(cpu_addr, size)) {
-		iommu_dma_unmap_page(dev, handle, iosize, 0, 0);
-		dma_free_from_pool(cpu_addr, size);
-	} else if (attrs & DMA_ATTR_FORCE_CONTIGUOUS) {
-		struct page *page = vmalloc_to_page(cpu_addr);
-
-		iommu_dma_unmap_page(dev, handle, iosize, 0, attrs);
-		dma_release_from_contiguous(dev, page, size >> PAGE_SHIFT);
-		dma_common_free_remap(cpu_addr, size, VM_USERMAP);
-	} else if (is_vmalloc_addr(cpu_addr)){
-		struct vm_struct *area = find_vm_area(cpu_addr);
-
-		if (WARN_ON(!area || !area->pages))
-			return;
-		iommu_dma_free(dev, area->pages, iosize, &handle);
-		dma_common_free_remap(cpu_addr, size, VM_USERMAP);
-	} else {
-		iommu_dma_unmap_page(dev, handle, iosize, 0, 0);
-		__free_pages(virt_to_page(cpu_addr), get_order(size));
-	}
-}
-
-static int __iommu_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
-			      void *cpu_addr, dma_addr_t dma_addr, size_t size,
-			      unsigned long attrs)
-{
-	struct vm_struct *area;
-	int ret;
-
-	vma->vm_page_prot = arch_dma_mmap_pgprot(dev, vma->vm_page_prot, attrs);
-
-	if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret))
-		return ret;
-
-	if (!is_vmalloc_addr(cpu_addr)) {
-		unsigned long pfn = page_to_pfn(virt_to_page(cpu_addr));
-		return __swiotlb_mmap_pfn(vma, pfn, size);
-	}
-
-	if (attrs & DMA_ATTR_FORCE_CONTIGUOUS) {
-		/*
-		 * DMA_ATTR_FORCE_CONTIGUOUS allocations are always remapped,
-		 * hence in the vmalloc space.
-		 */
-		unsigned long pfn = vmalloc_to_pfn(cpu_addr);
-		return __swiotlb_mmap_pfn(vma, pfn, size);
-	}
-
-	area = find_vm_area(cpu_addr);
-	if (WARN_ON(!area || !area->pages))
-		return -ENXIO;
-
-	return iommu_dma_mmap(area->pages, size, vma);
-}
-
-static int __iommu_get_sgtable(struct device *dev, struct sg_table *sgt,
-			       void *cpu_addr, dma_addr_t dma_addr,
-			       size_t size, unsigned long attrs)
-{
-	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	struct vm_struct *area = find_vm_area(cpu_addr);
-
-	if (!is_vmalloc_addr(cpu_addr)) {
-		struct page *page = virt_to_page(cpu_addr);
-		return __swiotlb_get_sgtable_page(sgt, page, size);
-	}
-
-	if (attrs & DMA_ATTR_FORCE_CONTIGUOUS) {
-		/*
-		 * DMA_ATTR_FORCE_CONTIGUOUS allocations are always remapped,
-		 * hence in the vmalloc space.
-		 */
-		struct page *page = vmalloc_to_page(cpu_addr);
-		return __swiotlb_get_sgtable_page(sgt, page, size);
-	}
-
-	if (WARN_ON(!area || !area->pages))
-		return -ENXIO;
-
-	return sg_alloc_table_from_pages(sgt, area->pages, count, 0, size,
-					 GFP_KERNEL);
-}
-
-static void __iommu_sync_single_for_cpu(struct device *dev,
-					dma_addr_t dev_addr, size_t size,
-					enum dma_data_direction dir)
-{
-	phys_addr_t phys;
-
-	if (dev_is_dma_coherent(dev))
-		return;
-
-	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dev_addr);
-	arch_sync_dma_for_cpu(dev, phys, size, dir);
-}
-
-static void __iommu_sync_single_for_device(struct device *dev,
-					   dma_addr_t dev_addr, size_t size,
-					   enum dma_data_direction dir)
-{
-	phys_addr_t phys;
-
-	if (dev_is_dma_coherent(dev))
-		return;
-
-	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dev_addr);
-	arch_sync_dma_for_device(dev, phys, size, dir);
-}
-
-static dma_addr_t __iommu_map_page(struct device *dev, struct page *page,
-				   unsigned long offset, size_t size,
-				   enum dma_data_direction dir,
-				   unsigned long attrs)
-{
-	bool coherent = dev_is_dma_coherent(dev);
-	int prot = dma_info_to_prot(dir, coherent, attrs);
-	dma_addr_t dev_addr = iommu_dma_map_page(dev, page, offset, size, prot);
-
-	if (!coherent && !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
-	    dev_addr != DMA_MAPPING_ERROR)
-		__dma_map_area(page_address(page) + offset, size, dir);
-
-	return dev_addr;
-}
-
-static void __iommu_unmap_page(struct device *dev, dma_addr_t dev_addr,
-			       size_t size, enum dma_data_direction dir,
-			       unsigned long attrs)
-{
-	if ((attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0)
-		__iommu_sync_single_for_cpu(dev, dev_addr, size, dir);
-
-	iommu_dma_unmap_page(dev, dev_addr, size, dir, attrs);
-}
-
-static void __iommu_sync_sg_for_cpu(struct device *dev,
-				    struct scatterlist *sgl, int nelems,
-				    enum dma_data_direction dir)
-{
-	struct scatterlist *sg;
-	int i;
-
-	if (dev_is_dma_coherent(dev))
-		return;
-
-	for_each_sg(sgl, sg, nelems, i)
-		arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length, dir);
-}
-
-static void __iommu_sync_sg_for_device(struct device *dev,
-				       struct scatterlist *sgl, int nelems,
-				       enum dma_data_direction dir)
-{
-	struct scatterlist *sg;
-	int i;
-
-	if (dev_is_dma_coherent(dev))
-		return;
-
-	for_each_sg(sgl, sg, nelems, i)
-		arch_sync_dma_for_device(dev, sg_phys(sg), sg->length, dir);
-}
-
-static int __iommu_map_sg_attrs(struct device *dev, struct scatterlist *sgl,
-				int nelems, enum dma_data_direction dir,
-				unsigned long attrs)
-{
-	bool coherent = dev_is_dma_coherent(dev);
-
-	if ((attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0)
-		__iommu_sync_sg_for_device(dev, sgl, nelems, dir);
-
-	return iommu_dma_map_sg(dev, sgl, nelems,
-				dma_info_to_prot(dir, coherent, attrs));
-}
-
-static void __iommu_unmap_sg_attrs(struct device *dev,
-				   struct scatterlist *sgl, int nelems,
-				   enum dma_data_direction dir,
-				   unsigned long attrs)
-{
-	if ((attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0)
-		__iommu_sync_sg_for_cpu(dev, sgl, nelems, dir);
-
-	iommu_dma_unmap_sg(dev, sgl, nelems, dir, attrs);
-}
-
-static const struct dma_map_ops iommu_dma_ops = {
-	.alloc = __iommu_alloc_attrs,
-	.free = __iommu_free_attrs,
-	.mmap = __iommu_mmap_attrs,
-	.get_sgtable = __iommu_get_sgtable,
-	.map_page = __iommu_map_page,
-	.unmap_page = __iommu_unmap_page,
-	.map_sg = __iommu_map_sg_attrs,
-	.unmap_sg = __iommu_unmap_sg_attrs,
-	.sync_single_for_cpu = __iommu_sync_single_for_cpu,
-	.sync_single_for_device = __iommu_sync_single_for_device,
-	.sync_sg_for_cpu = __iommu_sync_sg_for_cpu,
-	.sync_sg_for_device = __iommu_sync_sg_for_device,
-	.map_resource = iommu_dma_map_resource,
-	.unmap_resource = iommu_dma_unmap_resource,
-};
-
-static int __init __iommu_dma_init(void)
-{
-	return iommu_dma_init();
-}
-arch_initcall(__iommu_dma_init);
-
-static void __iommu_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				  const struct iommu_ops *ops)
-{
-	struct iommu_domain *domain;
-
-	if (!ops)
-		return;
-
-	/*
-	 * The IOMMU core code allocates the default DMA domain, which the
-	 * underlying IOMMU driver needs to support via the dma-iommu layer.
-	 */
-	domain = iommu_get_domain_for_dev(dev);
-
-	if (!domain)
-		goto out_err;
-
-	if (domain->type == IOMMU_DOMAIN_DMA) {
-		if (iommu_dma_init_domain(domain, dma_base, size, dev))
-			goto out_err;
-
-		dev->dma_ops = &iommu_dma_ops;
-	}
-
-	return;
-
-out_err:
-	 pr_warn("Failed to set up IOMMU for device %s; retaining platform DMA ops\n",
-		 dev_name(dev));
-}
-
 void arch_teardown_dma_ops(struct device *dev)
 {
 	dev->dma_ops = NULL;
 }
-
-#else
-
-static void __iommu_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				  const struct iommu_ops *iommu)
-{ }
-
-#endif  /* CONFIG_IOMMU_DMA */
+#endif
 
 void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 			const struct iommu_ops *iommu, bool coherent)
 {
 	dev->dma_coherent = coherent;
-	__iommu_setup_dma_ops(dev, dma_base, size, iommu);
+	if (iommu)
+		iommu_setup_dma_ops(dev, dma_base, size);
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
index 3793182..0ba108e 100644
--- a/drivers/iommu/dma-iommu.c
+++ b/drivers/iommu/dma-iommu.c
@@ -10,7 +10,9 @@
 
 #include <linux/acpi_iort.h>
 #include <linux/device.h>
+#include <linux/dma-contiguous.h>
 #include <linux/dma-iommu.h>
+#include <linux/dma-noncoherent.h>
 #include <linux/gfp.h>
 #include <linux/huge_mm.h>
 #include <linux/iommu.h>
@@ -67,11 +69,6 @@ static struct iommu_dma_cookie *cookie_alloc(enum iommu_dma_cookie_type type)
 	return cookie;
 }
 
-int iommu_dma_init(void)
-{
-	return iova_cache_get();
-}
-
 /**
  * iommu_get_dma_cookie - Acquire DMA-API resources for a domain
  * @domain: IOMMU domain to prepare for DMA-API usage
@@ -302,7 +299,7 @@ static void iommu_dma_flush_iotlb_all(struct iova_domain *iovad)
  * to ensure it is an invalid IOVA. It is safe to reinitialise a domain, but
  * any change which could make prior IOVAs invalid will fail.
  */
-int iommu_dma_init_domain(struct iommu_domain *domain, dma_addr_t base,
+static int iommu_dma_init_domain(struct iommu_domain *domain, dma_addr_t base,
 		u64 size, struct device *dev)
 {
 	struct iommu_dma_cookie *cookie = domain->iova_cookie;
@@ -353,7 +350,6 @@ int iommu_dma_init_domain(struct iommu_domain *domain, dma_addr_t base,
 
 	return iova_reserve_iommu_regions(dev, domain);
 }
-EXPORT_SYMBOL(iommu_dma_init_domain);
 
 /**
  * dma_info_to_prot - Translate DMA API directions and attributes to IOMMU API
@@ -364,7 +360,7 @@ EXPORT_SYMBOL(iommu_dma_init_domain);
  *
  * Return: corresponding IOMMU API page protection flags
  */
-int dma_info_to_prot(enum dma_data_direction dir, bool coherent,
+static int dma_info_to_prot(enum dma_data_direction dir, bool coherent,
 		     unsigned long attrs)
 {
 	int prot = coherent ? IOMMU_CACHE : 0;
@@ -441,9 +437,10 @@ static void iommu_dma_free_iova(struct iommu_dma_cookie *cookie,
 				size >> iova_shift(iovad));
 }
 
-static void __iommu_dma_unmap(struct iommu_domain *domain, dma_addr_t dma_addr,
+static void __iommu_dma_unmap(struct device *dev, dma_addr_t dma_addr,
 		size_t size)
 {
+	struct iommu_domain *domain = iommu_get_dma_domain(dev);
 	struct iommu_dma_cookie *cookie = domain->iova_cookie;
 	struct iova_domain *iovad = &cookie->iovad;
 	size_t iova_off = iova_offset(iovad, dma_addr);
@@ -457,6 +454,30 @@ static void __iommu_dma_unmap(struct iommu_domain *domain, dma_addr_t dma_addr,
 	iommu_dma_free_iova(cookie, dma_addr, size);
 }
 
+static dma_addr_t __iommu_dma_map(struct device *dev, phys_addr_t phys,
+		size_t size, int prot)
+{
+	struct iommu_domain *domain = iommu_get_dma_domain(dev);
+	struct iommu_dma_cookie *cookie = domain->iova_cookie;
+	size_t iova_off = 0;
+	dma_addr_t iova;
+
+	if (cookie->type == IOMMU_DMA_IOVA_COOKIE) {
+		iova_off = iova_offset(&cookie->iovad, phys);
+		size = iova_align(&cookie->iovad, size + iova_off);
+	}
+
+	iova = iommu_dma_alloc_iova(domain, size, dma_get_mask(dev), dev);
+	if (!iova)
+		return DMA_MAPPING_ERROR;
+
+	if (iommu_map(domain, iova, phys - iova_off, size, prot)) {
+		iommu_dma_free_iova(cookie, iova, size);
+		return DMA_MAPPING_ERROR;
+	}
+	return iova + iova_off;
+}
+
 static void __iommu_dma_free_pages(struct page **pages, int count)
 {
 	while (count--)
@@ -522,55 +543,45 @@ static struct page **__iommu_dma_alloc_pages(struct device *dev,
 	return pages;
 }
 
-/**
- * iommu_dma_free - Free a buffer allocated by iommu_dma_alloc()
- * @dev: Device which owns this buffer
- * @pages: Array of buffer pages as returned by iommu_dma_alloc()
- * @size: Size of buffer in bytes
- * @handle: DMA address of buffer
- *
- * Frees both the pages associated with the buffer, and the array
- * describing them
- */
-void iommu_dma_free(struct device *dev, struct page **pages, size_t size,
-		dma_addr_t *handle)
+static struct page **__iommu_dma_get_pages(void *cpu_addr)
 {
-	__iommu_dma_unmap(iommu_get_dma_domain(dev), *handle, size);
-	__iommu_dma_free_pages(pages, PAGE_ALIGN(size) >> PAGE_SHIFT);
-	*handle = DMA_MAPPING_ERROR;
+	struct vm_struct *area = find_vm_area(cpu_addr);
+
+	if (!area || !area->pages)
+		return NULL;
+	return area->pages;
 }
 
 /**
- * iommu_dma_alloc - Allocate and map a buffer contiguous in IOVA space
+ * iommu_dma_alloc_remap - Allocate and map a buffer contiguous in IOVA space
  * @dev: Device to allocate memory for. Must be a real device
  *	 attached to an iommu_dma_domain
  * @size: Size of buffer in bytes
+ * @dma_handle: Out argument for allocated DMA handle
  * @gfp: Allocation flags
  * @attrs: DMA attributes for this allocation
- * @prot: IOMMU mapping flags
- * @handle: Out argument for allocated DMA handle
- * @flush_page: Arch callback which must ensure PAGE_SIZE bytes from the
- *		given VA/PA are visible to the given non-coherent device.
  *
  * If @size is less than PAGE_SIZE, then a full CPU page will be allocated,
  * but an IOMMU which supports smaller pages might not map the whole thing.
  *
- * Return: Array of struct page pointers describing the buffer,
- *	   or NULL on failure.
+ * Return: Mapped virtual address, or NULL on failure.
  */
-struct page **iommu_dma_alloc(struct device *dev, size_t size, gfp_t gfp,
-		unsigned long attrs, int prot, dma_addr_t *handle,
-		void (*flush_page)(struct device *, const void *, phys_addr_t))
+static void *iommu_dma_alloc_remap(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
 	struct iommu_domain *domain = iommu_get_dma_domain(dev);
 	struct iommu_dma_cookie *cookie = domain->iova_cookie;
 	struct iova_domain *iovad = &cookie->iovad;
+	bool coherent = dev_is_dma_coherent(dev);
+	int ioprot = dma_info_to_prot(DMA_BIDIRECTIONAL, coherent, attrs);
+	pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);
+	unsigned int count, min_size, alloc_sizes = domain->pgsize_bitmap;
 	struct page **pages;
 	struct sg_table sgt;
 	dma_addr_t iova;
-	unsigned int count, min_size, alloc_sizes = domain->pgsize_bitmap;
+	void *vaddr;
 
-	*handle = DMA_MAPPING_ERROR;
+	*dma_handle = DMA_MAPPING_ERROR;
 
 	min_size = alloc_sizes & -alloc_sizes;
 	if (min_size < PAGE_SIZE) {
@@ -596,26 +607,29 @@ struct page **iommu_dma_alloc(struct device *dev, size_t size, gfp_t gfp,
 	if (sg_alloc_table_from_pages(&sgt, pages, count, 0, size, GFP_KERNEL))
 		goto out_free_iova;
 
-	if (!(prot & IOMMU_CACHE)) {
-		struct sg_mapping_iter miter;
-		/*
-		 * The CPU-centric flushing implied by SG_MITER_TO_SG isn't
-		 * sufficient here, so skip it by using the "wrong" direction.
-		 */
-		sg_miter_start(&miter, sgt.sgl, sgt.orig_nents, SG_MITER_FROM_SG);
-		while (sg_miter_next(&miter))
-			flush_page(dev, miter.addr, page_to_phys(miter.page));
-		sg_miter_stop(&miter);
+	if (!(ioprot & IOMMU_CACHE)) {
+		struct scatterlist *sg;
+		int i;
+
+		for_each_sg(sgt.sgl, sg, sgt.orig_nents, i)
+			arch_dma_prep_coherent(sg_page(sg), sg->length);
 	}
 
-	if (iommu_map_sg(domain, iova, sgt.sgl, sgt.orig_nents, prot)
+	if (iommu_map_sg(domain, iova, sgt.sgl, sgt.orig_nents, ioprot)
 			< size)
 		goto out_free_sg;
 
-	*handle = iova;
-	sg_free_table(&sgt);
-	return pages;
+	vaddr = dma_common_pages_remap(pages, size, VM_USERMAP, prot,
+			__builtin_return_address(0));
+	if (!vaddr)
+		goto out_unmap;
 
+	*dma_handle = iova;
+	sg_free_table(&sgt);
+	return vaddr;
+
+out_unmap:
+	__iommu_dma_unmap(dev, iova, size);
 out_free_sg:
 	sg_free_table(&sgt);
 out_free_iova:
@@ -626,54 +640,94 @@ struct page **iommu_dma_alloc(struct device *dev, size_t size, gfp_t gfp,
 }
 
 /**
- * iommu_dma_mmap - Map a buffer into provided user VMA
- * @pages: Array representing buffer from iommu_dma_alloc()
+ * __iommu_dma_mmap - Map a buffer into provided user VMA
+ * @pages: Array representing buffer from __iommu_dma_alloc()
  * @size: Size of buffer in bytes
  * @vma: VMA describing requested userspace mapping
  *
  * Maps the pages of the buffer in @pages into @vma. The caller is responsible
  * for verifying the correct size and protection of @vma beforehand.
  */
-
-int iommu_dma_mmap(struct page **pages, size_t size, struct vm_area_struct *vma)
+static int __iommu_dma_mmap(struct page **pages, size_t size,
+		struct vm_area_struct *vma)
 {
 	return vm_map_pages(vma, pages, PAGE_ALIGN(size) >> PAGE_SHIFT);
 }
 
-static dma_addr_t __iommu_dma_map(struct device *dev, phys_addr_t phys,
-		size_t size, int prot, struct iommu_domain *domain)
+static void iommu_dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
 {
-	struct iommu_dma_cookie *cookie = domain->iova_cookie;
-	size_t iova_off = 0;
-	dma_addr_t iova;
+	phys_addr_t phys;
 
-	if (cookie->type == IOMMU_DMA_IOVA_COOKIE) {
-		iova_off = iova_offset(&cookie->iovad, phys);
-		size = iova_align(&cookie->iovad, size + iova_off);
-	}
+	if (dev_is_dma_coherent(dev))
+		return;
 
-	iova = iommu_dma_alloc_iova(domain, size, dma_get_mask(dev), dev);
-	if (!iova)
-		return DMA_MAPPING_ERROR;
-
-	if (iommu_map(domain, iova, phys - iova_off, size, prot)) {
-		iommu_dma_free_iova(cookie, iova, size);
-		return DMA_MAPPING_ERROR;
-	}
-	return iova + iova_off;
+	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
+	arch_sync_dma_for_cpu(dev, phys, size, dir);
 }
 
-dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
-		unsigned long offset, size_t size, int prot)
+static void iommu_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
 {
-	return __iommu_dma_map(dev, page_to_phys(page) + offset, size, prot,
-			iommu_get_dma_domain(dev));
+	phys_addr_t phys;
+
+	if (dev_is_dma_coherent(dev))
+		return;
+
+	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
+	arch_sync_dma_for_device(dev, phys, size, dir);
 }
 
-void iommu_dma_unmap_page(struct device *dev, dma_addr_t handle, size_t size,
-		enum dma_data_direction dir, unsigned long attrs)
+static void iommu_dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sgl, int nelems,
+		enum dma_data_direction dir)
 {
-	__iommu_dma_unmap(iommu_get_dma_domain(dev), handle, size);
+	struct scatterlist *sg;
+	int i;
+
+	if (dev_is_dma_coherent(dev))
+		return;
+
+	for_each_sg(sgl, sg, nelems, i)
+		arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length, dir);
+}
+
+static void iommu_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sgl, int nelems,
+		enum dma_data_direction dir)
+{
+	struct scatterlist *sg;
+	int i;
+
+	if (dev_is_dma_coherent(dev))
+		return;
+
+	for_each_sg(sgl, sg, nelems, i)
+		arch_sync_dma_for_device(dev, sg_phys(sg), sg->length, dir);
+}
+
+static dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		unsigned long attrs)
+{
+	phys_addr_t phys = page_to_phys(page) + offset;
+	bool coherent = dev_is_dma_coherent(dev);
+	int prot = dma_info_to_prot(dir, coherent, attrs);
+	dma_addr_t dma_handle;
+
+	dma_handle =__iommu_dma_map(dev, phys, size, prot);
+	if (!coherent && !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
+	    dma_handle != DMA_MAPPING_ERROR)
+		arch_sync_dma_for_device(dev, phys, size, dir);
+	return dma_handle;
+}
+
+static void iommu_dma_unmap_page(struct device *dev, dma_addr_t dma_handle,
+		size_t size, enum dma_data_direction dir, unsigned long attrs)
+{
+	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		iommu_dma_sync_single_for_cpu(dev, dma_handle, size, dir);
+	__iommu_dma_unmap(dev, dma_handle, size);
 }
 
 /*
@@ -758,18 +812,22 @@ static void __invalidate_sg(struct scatterlist *sg, int nents)
  * impedance-matching, to be able to hand off a suitably-aligned list,
  * but still preserve the original offsets and sizes for the caller.
  */
-int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
-		int nents, int prot)
+static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
+		int nents, enum dma_data_direction dir, unsigned long attrs)
 {
 	struct iommu_domain *domain = iommu_get_dma_domain(dev);
 	struct iommu_dma_cookie *cookie = domain->iova_cookie;
 	struct iova_domain *iovad = &cookie->iovad;
 	struct scatterlist *s, *prev = NULL;
+	int prot = dma_info_to_prot(dir, dev_is_dma_coherent(dev), attrs);
 	dma_addr_t iova;
 	size_t iova_len = 0;
 	unsigned long mask = dma_get_seg_boundary(dev);
 	int i;
 
+	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		iommu_dma_sync_sg_for_device(dev, sg, nents, dir);
+
 	/*
 	 * Work out how much IOVA space we need, and align the segments to
 	 * IOVA granules for the IOMMU driver to handle. With some clever
@@ -829,12 +887,16 @@ int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
 	return 0;
 }
 
-void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nents,
-		enum dma_data_direction dir, unsigned long attrs)
+static void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+		int nents, enum dma_data_direction dir, unsigned long attrs)
 {
 	dma_addr_t start, end;
 	struct scatterlist *tmp;
 	int i;
+
+	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
+		iommu_dma_sync_sg_for_cpu(dev, sg, nents, dir);
+
 	/*
 	 * The scatterlist segments are mapped into a single
 	 * contiguous IOVA allocation, so this is incredibly easy.
@@ -846,21 +908,231 @@ void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nents,
 		sg = tmp;
 	}
 	end = sg_dma_address(sg) + sg_dma_len(sg);
-	__iommu_dma_unmap(iommu_get_dma_domain(dev), start, end - start);
+	__iommu_dma_unmap(dev, start, end - start);
 }
 
-dma_addr_t iommu_dma_map_resource(struct device *dev, phys_addr_t phys,
+static dma_addr_t iommu_dma_map_resource(struct device *dev, phys_addr_t phys,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	return __iommu_dma_map(dev, phys, size,
-			dma_info_to_prot(dir, false, attrs) | IOMMU_MMIO,
-			iommu_get_dma_domain(dev));
+			dma_info_to_prot(dir, false, attrs) | IOMMU_MMIO);
 }
 
-void iommu_dma_unmap_resource(struct device *dev, dma_addr_t handle,
+static void iommu_dma_unmap_resource(struct device *dev, dma_addr_t handle,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
-	__iommu_dma_unmap(iommu_get_dma_domain(dev), handle, size);
+	__iommu_dma_unmap(dev, handle, size);
+}
+
+static void __iommu_dma_free(struct device *dev, size_t size, void *cpu_addr)
+{
+	size_t alloc_size = PAGE_ALIGN(size);
+	int count = alloc_size >> PAGE_SHIFT;
+	struct page *page = NULL, **pages = NULL;
+
+	/* Non-coherent atomic allocation? Easy */
+	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
+	    dma_free_from_pool(cpu_addr, alloc_size))
+		return;
+
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
+		/*
+		 * If it the address is remapped, then it's either non-coherent
+		 * or highmem CMA, or an iommu_dma_alloc_remap() construction.
+		 */
+		pages = __iommu_dma_get_pages(cpu_addr);
+		if (!pages)
+			page = vmalloc_to_page(cpu_addr);
+		dma_common_free_remap(cpu_addr, alloc_size, VM_USERMAP);
+	} else {
+		/* Lowmem means a coherent atomic or CMA allocation */
+		page = virt_to_page(cpu_addr);
+	}
+
+	if (pages)
+		__iommu_dma_free_pages(pages, count);
+	if (page && !dma_release_from_contiguous(dev, page, count))
+		__free_pages(page, get_order(alloc_size));
+}
+
+static void iommu_dma_free(struct device *dev, size_t size, void *cpu_addr,
+		dma_addr_t handle, unsigned long attrs)
+{
+	__iommu_dma_unmap(dev, handle, size);
+	__iommu_dma_free(dev, size, cpu_addr);
+}
+
+static void *iommu_dma_alloc_pages(struct device *dev, size_t size,
+		struct page **pagep, gfp_t gfp, unsigned long attrs)
+{
+	bool coherent = dev_is_dma_coherent(dev);
+	size_t alloc_size = PAGE_ALIGN(size);
+	struct page *page = NULL;
+	void *cpu_addr;
+
+	if (gfpflags_allow_blocking(gfp))
+		page = dma_alloc_from_contiguous(dev, alloc_size >> PAGE_SHIFT,
+						 get_order(alloc_size),
+						 gfp & __GFP_NOWARN);
+	if (!page)
+		page = alloc_pages(gfp, get_order(alloc_size));
+	if (!page)
+		return NULL;
+
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && (!coherent || PageHighMem(page))) {
+		pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);
+
+		cpu_addr = dma_common_contiguous_remap(page, alloc_size,
+				VM_USERMAP, prot, __builtin_return_address(0));
+		if (!cpu_addr)
+			goto out_free_pages;
+
+		if (!coherent)
+			arch_dma_prep_coherent(page, size);
+	} else {
+		cpu_addr = page_address(page);
+	}
+
+	*pagep = page;
+	memset(cpu_addr, 0, alloc_size);
+	return cpu_addr;
+out_free_pages:
+	if (!dma_release_from_contiguous(dev, page, alloc_size >> PAGE_SHIFT))
+		__free_pages(page, get_order(alloc_size));
+	return NULL;
+}
+
+static void *iommu_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *handle, gfp_t gfp, unsigned long attrs)
+{
+	bool coherent = dev_is_dma_coherent(dev);
+	int ioprot = dma_info_to_prot(DMA_BIDIRECTIONAL, coherent, attrs);
+	struct page *page = NULL;
+	void *cpu_addr;
+
+	gfp |= __GFP_ZERO;
+
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && gfpflags_allow_blocking(gfp) &&
+	    !(attrs & DMA_ATTR_FORCE_CONTIGUOUS))
+		return iommu_dma_alloc_remap(dev, size, handle, gfp, attrs);
+
+	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
+	    !gfpflags_allow_blocking(gfp) && !coherent)
+		cpu_addr = dma_alloc_from_pool(PAGE_ALIGN(size), &page, gfp);
+	else
+		cpu_addr = iommu_dma_alloc_pages(dev, size, &page, gfp, attrs);
+	if (!cpu_addr)
+		return NULL;
+
+	*handle = __iommu_dma_map(dev, page_to_phys(page), size, ioprot);
+	if (*handle == DMA_MAPPING_ERROR) {
+		__iommu_dma_free(dev, size, cpu_addr);
+		return NULL;
+	}
+
+	return cpu_addr;
+}
+
+static int iommu_dma_mmap(struct device *dev, struct vm_area_struct *vma,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs)
+{
+	unsigned long nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned long pfn, off = vma->vm_pgoff;
+	int ret;
+
+	vma->vm_page_prot = arch_dma_mmap_pgprot(dev, vma->vm_page_prot, attrs);
+
+	if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret))
+		return ret;
+
+	if (off >= nr_pages || vma_pages(vma) > nr_pages - off)
+		return -ENXIO;
+
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
+		struct page **pages = __iommu_dma_get_pages(cpu_addr);
+
+		if (pages)
+			return __iommu_dma_mmap(pages, size, vma);
+		pfn = vmalloc_to_pfn(cpu_addr);
+	} else {
+		pfn = page_to_pfn(virt_to_page(cpu_addr));
+	}
+
+	return remap_pfn_range(vma, vma->vm_start, pfn + off,
+			       vma->vm_end - vma->vm_start,
+			       vma->vm_page_prot);
+}
+
+static int iommu_dma_get_sgtable(struct device *dev, struct sg_table *sgt,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs)
+{
+	struct page *page;
+	int ret;
+
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
+		struct page **pages = __iommu_dma_get_pages(cpu_addr);
+
+		if (pages) {
+			return sg_alloc_table_from_pages(sgt, pages,
+					PAGE_ALIGN(size) >> PAGE_SHIFT,
+					0, size, GFP_KERNEL);
+		}
+
+		page = vmalloc_to_page(cpu_addr);
+	} else {
+		page = virt_to_page(cpu_addr);
+	}
+
+	ret = sg_alloc_table(sgt, 1, GFP_KERNEL);
+	if (!ret)
+		sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0);
+	return ret;
+}
+
+static const struct dma_map_ops iommu_dma_ops = {
+	.alloc			= iommu_dma_alloc,
+	.free			= iommu_dma_free,
+	.mmap			= iommu_dma_mmap,
+	.get_sgtable		= iommu_dma_get_sgtable,
+	.map_page		= iommu_dma_map_page,
+	.unmap_page		= iommu_dma_unmap_page,
+	.map_sg			= iommu_dma_map_sg,
+	.unmap_sg		= iommu_dma_unmap_sg,
+	.sync_single_for_cpu	= iommu_dma_sync_single_for_cpu,
+	.sync_single_for_device	= iommu_dma_sync_single_for_device,
+	.sync_sg_for_cpu	= iommu_dma_sync_sg_for_cpu,
+	.sync_sg_for_device	= iommu_dma_sync_sg_for_device,
+	.map_resource		= iommu_dma_map_resource,
+	.unmap_resource		= iommu_dma_unmap_resource,
+};
+
+/*
+ * The IOMMU core code allocates the default DMA domain, which the underlying
+ * IOMMU driver needs to support via the dma-iommu layer.
+ */
+void iommu_setup_dma_ops(struct device *dev, u64 dma_base, u64 size)
+{
+	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);
+
+	if (!domain)
+		goto out_err;
+
+	/*
+	 * The IOMMU core code allocates the default DMA domain, which the
+	 * underlying IOMMU driver needs to support via the dma-iommu layer.
+	 */
+	if (domain->type == IOMMU_DOMAIN_DMA) {
+		if (iommu_dma_init_domain(domain, dma_base, size, dev))
+			goto out_err;
+		dev->dma_ops = &iommu_dma_ops;
+	}
+
+	return;
+out_err:
+	 pr_warn("Failed to set up IOMMU for device %s; retaining platform DMA ops\n",
+		 dev_name(dev));
 }
 
 static struct iommu_dma_msi_page *iommu_dma_get_msi_page(struct device *dev,
@@ -881,7 +1153,7 @@ static struct iommu_dma_msi_page *iommu_dma_get_msi_page(struct device *dev,
 	if (!msi_page)
 		return NULL;
 
-	iova = __iommu_dma_map(dev, msi_addr, size, prot, domain);
+	iova = __iommu_dma_map(dev, msi_addr, size, prot);
 	if (iova == DMA_MAPPING_ERROR)
 		goto out_free_page;
 
@@ -943,3 +1215,9 @@ void iommu_dma_compose_msi_msg(struct msi_desc *desc,
 	msg->address_lo &= cookie_msi_granule(domain->iova_cookie) - 1;
 	msg->address_lo += lower_32_bits(msi_page->iova);
 }
+
+static int iommu_dma_init(void)
+{
+	return iova_cache_get();
+}
+arch_initcall(iommu_dma_init);
diff --git a/include/linux/dma-iommu.h b/include/linux/dma-iommu.h
index 37258c8..2112f21 100644
--- a/include/linux/dma-iommu.h
+++ b/include/linux/dma-iommu.h
@@ -5,59 +5,21 @@
 #ifndef __DMA_IOMMU_H
 #define __DMA_IOMMU_H
 
-#ifdef __KERNEL__
+#include <linux/errno.h>
 #include <linux/types.h>
-#include <asm/errno.h>
 
 #ifdef CONFIG_IOMMU_DMA
 #include <linux/dma-mapping.h>
 #include <linux/iommu.h>
 #include <linux/msi.h>
 
-int iommu_dma_init(void);
-
 /* Domain management interface for IOMMU drivers */
 int iommu_get_dma_cookie(struct iommu_domain *domain);
 int iommu_get_msi_cookie(struct iommu_domain *domain, dma_addr_t base);
 void iommu_put_dma_cookie(struct iommu_domain *domain);
 
 /* Setup call for arch DMA mapping code */
-int iommu_dma_init_domain(struct iommu_domain *domain, dma_addr_t base,
-		u64 size, struct device *dev);
-
-/* General helpers for DMA-API <-> IOMMU-API interaction */
-int dma_info_to_prot(enum dma_data_direction dir, bool coherent,
-		     unsigned long attrs);
-
-/*
- * These implement the bulk of the relevant DMA mapping callbacks, but require
- * the arch code to take care of attributes and cache maintenance
- */
-struct page **iommu_dma_alloc(struct device *dev, size_t size, gfp_t gfp,
-		unsigned long attrs, int prot, dma_addr_t *handle,
-		void (*flush_page)(struct device *, const void *, phys_addr_t));
-void iommu_dma_free(struct device *dev, struct page **pages, size_t size,
-		dma_addr_t *handle);
-
-int iommu_dma_mmap(struct page **pages, size_t size, struct vm_area_struct *vma);
-
-dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
-		unsigned long offset, size_t size, int prot);
-int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
-		int nents, int prot);
-
-/*
- * Arch code with no special attribute handling may use these
- * directly as DMA mapping callbacks for simplicity
- */
-void iommu_dma_unmap_page(struct device *dev, dma_addr_t handle, size_t size,
-		enum dma_data_direction dir, unsigned long attrs);
-void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nents,
-		enum dma_data_direction dir, unsigned long attrs);
-dma_addr_t iommu_dma_map_resource(struct device *dev, phys_addr_t phys,
-		size_t size, enum dma_data_direction dir, unsigned long attrs);
-void iommu_dma_unmap_resource(struct device *dev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir, unsigned long attrs);
+void iommu_setup_dma_ops(struct device *dev, u64 dma_base, u64 size);
 
 /* The DMA API isn't _quite_ the whole story, though... */
 /*
@@ -75,16 +37,16 @@ void iommu_dma_compose_msi_msg(struct msi_desc *desc,
 
 void iommu_dma_get_resv_regions(struct device *dev, struct list_head *list);
 
-#else
+#else /* CONFIG_IOMMU_DMA */
 
 struct iommu_domain;
 struct msi_desc;
 struct msi_msg;
 struct device;
 
-static inline int iommu_dma_init(void)
+static inline void iommu_setup_dma_ops(struct device *dev, u64 dma_base,
+		u64 size)
 {
-	return 0;
 }
 
 static inline int iommu_get_dma_cookie(struct iommu_domain *domain)
@@ -117,5 +79,4 @@ static inline void iommu_dma_get_resv_regions(struct device *dev, struct list_he
 }
 
 #endif	/* CONFIG_IOMMU_DMA */
-#endif	/* __KERNEL__ */
 #endif	/* __DMA_IOMMU_H */