Merge tag 'dma-mapping-4.16' of git://git.infradead.org/users/hch/dma-mapping

Pull dma mapping updates from Christoph Hellwig:
 "Except for a runtime warning fix from Christian this is all about
  consolidation of the generic no-IOMMU code, a well as the glue code
  for swiotlb.

  All the code is based on the x86 implementation with hooks to allow
  all architectures that aren't cache coherent to use it.

  The x86 conversion itself has been deferred because the x86
  maintainers were a little busy in the last months"

* tag 'dma-mapping-4.16' of git://git.infradead.org/users/hch/dma-mapping: (57 commits)
  MAINTAINERS: add the iommu list for swiotlb and xen-swiotlb
  arm64: use swiotlb_alloc and swiotlb_free
  arm64: replace ZONE_DMA with ZONE_DMA32
  mips: use swiotlb_{alloc,free}
  mips/netlogic: remove swiotlb support
  tile: use generic swiotlb_ops
  tile: replace ZONE_DMA with ZONE_DMA32
  unicore32: use generic swiotlb_ops
  ia64: remove an ifdef around the content of pci-dma.c
  ia64: clean up swiotlb support
  ia64: use generic swiotlb_ops
  ia64: replace ZONE_DMA with ZONE_DMA32
  swiotlb: remove various exports
  swiotlb: refactor coherent buffer allocation
  swiotlb: refactor coherent buffer freeing
  swiotlb: wire up ->dma_supported in swiotlb_dma_ops
  swiotlb: add common swiotlb_map_ops
  swiotlb: rename swiotlb_free to swiotlb_exit
  x86: rename swiotlb_dma_ops
  powerpc: rename swiotlb_dma_ops
  ...

119 files changed, 728 insertions, 1266 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index 98ee6febcf45..7b5ef9a58c38 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4343,10 +4343,12 @@ T:	git git://git.infradead.org/users/hch/dma-mapping.git
 W:	http://git.infradead.org/users/hch/dma-mapping.git
 S:	Supported
 F:	lib/dma-debug.c
-F:	lib/dma-noop.c
+F:	lib/dma-direct.c
 F:	lib/dma-virt.c
 F:	drivers/base/dma-mapping.c
 F:	drivers/base/dma-coherent.c
+F:	include/asm-generic/dma-mapping.h
+F:	include/linux/dma-direct.h
 F:	include/linux/dma-mapping.h
 
 DME1737 HARDWARE MONITOR DRIVER
@@ -13071,7 +13073,7 @@ F:	arch/x86/boot/video*
 
 SWIOTLB SUBSYSTEM
 M:	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
-L:	linux-kernel@vger.kernel.org
+L:	iommu@lists.linux-foundation.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb.git
 S:	Supported
 F:	lib/swiotlb.c
@@ -15026,6 +15028,7 @@ F:	include/xen/interface/io/vscsiif.h
 XEN SWIOTLB SUBSYSTEM
 M:	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
 L:	xen-devel@lists.xenproject.org (moderated for non-subscribers)
+L:	iommu@lists.linux-foundation.org
 S:	Supported
 F:	arch/x86/xen/*swiotlb*
 F:	drivers/xen/*swiotlb*
diff --git a/arch/Kconfig b/arch/Kconfig
index a26d6f8ab967..1b560e7f06dc 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -938,6 +938,10 @@ config STRICT_MODULE_RWX
 	  and non-text memory will be made non-executable. This provides
 	  protection against certain security exploits (e.g. writing to text)
 
+# select if the architecture provides an asm/dma-direct.h header
+config ARCH_HAS_PHYS_TO_DMA
+	bool
+
 config ARCH_HAS_REFCOUNT
 	bool
 	help
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index b31b974a03cb..e96adcbcab41 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -209,6 +209,7 @@ config ALPHA_EIGER
 
 config ALPHA_JENSEN
 	bool "Jensen"
+	depends on BROKEN
 	help
 	  DEC PC 150 AXP (aka Jensen): This is a very old Digital system - one
 	  of the first-generation Alpha systems. A number of these systems
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 9d5fd00d9e91..f3a80cf164cc 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -463,9 +463,6 @@ config ARCH_PHYS_ADDR_T_64BIT
 config ARCH_DMA_ADDR_T_64BIT
 	bool
 
-config ARC_PLAT_NEEDS_PHYS_TO_DMA
-	bool
-
 config ARC_KVADDR_SIZE
 	int "Kernel Virtual Address Space size (MB)"
 	range 0 512
diff --git a/arch/arc/include/asm/dma-mapping.h b/arch/arc/include/asm/dma-mapping.h
index 94285031c4fb..7a16824bfe98 100644
--- a/arch/arc/include/asm/dma-mapping.h
+++ b/arch/arc/include/asm/dma-mapping.h
@@ -11,13 +11,6 @@
 #ifndef ASM_ARC_DMA_MAPPING_H
 #define ASM_ARC_DMA_MAPPING_H
 
-#ifndef CONFIG_ARC_PLAT_NEEDS_PHYS_TO_DMA
-#define plat_dma_to_phys(dev, dma_handle) ((phys_addr_t)(dma_handle))
-#define plat_phys_to_dma(dev, paddr) ((dma_addr_t)(paddr))
-#else
-#include <plat/dma.h>
-#endif
-
 extern const struct dma_map_ops arc_dma_ops;
 
 static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
diff --git a/arch/arc/mm/dma.c b/arch/arc/mm/dma.c
index e9d93604ad0f..1dcc404b5aec 100644
--- a/arch/arc/mm/dma.c
+++ b/arch/arc/mm/dma.c
@@ -60,7 +60,7 @@ static void *arc_dma_alloc(struct device *dev, size_t size,
 	/* This is linear addr (0x8000_0000 based) */
 	paddr = page_to_phys(page);
 
-	*dma_handle = plat_phys_to_dma(dev, paddr);
+	*dma_handle = paddr;
 
 	/* This is kernel Virtual address (0x7000_0000 based) */
 	if (need_kvaddr) {
@@ -92,7 +92,7 @@ static void *arc_dma_alloc(struct device *dev, size_t size,
 static void arc_dma_free(struct device *dev, size_t size, void *vaddr,
 		dma_addr_t dma_handle, unsigned long attrs)
 {
-	phys_addr_t paddr = plat_dma_to_phys(dev, dma_handle);
+	phys_addr_t paddr = dma_handle;
 	struct page *page = virt_to_page(paddr);
 	int is_non_coh = 1;
 
@@ -111,7 +111,7 @@ static int arc_dma_mmap(struct device *dev, struct vm_area_struct *vma,
 {
 	unsigned long user_count = vma_pages(vma);
 	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	unsigned long pfn = __phys_to_pfn(plat_dma_to_phys(dev, dma_addr));
+	unsigned long pfn = __phys_to_pfn(dma_addr);
 	unsigned long off = vma->vm_pgoff;
 	int ret = -ENXIO;
 
@@ -175,7 +175,7 @@ static dma_addr_t arc_dma_map_page(struct device *dev, struct page *page,
 	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
 		_dma_cache_sync(paddr, size, dir);
 
-	return plat_phys_to_dma(dev, paddr);
+	return paddr;
 }
 
 /*
@@ -190,7 +190,7 @@ static void arc_dma_unmap_page(struct device *dev, dma_addr_t handle,
 			       size_t size, enum dma_data_direction dir,
 			       unsigned long attrs)
 {
-	phys_addr_t paddr = plat_dma_to_phys(dev, handle);
+	phys_addr_t paddr = handle;
 
 	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
 		_dma_cache_sync(paddr, size, dir);
@@ -224,13 +224,13 @@ static void arc_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
 static void arc_dma_sync_single_for_cpu(struct device *dev,
 		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
 {
-	_dma_cache_sync(plat_dma_to_phys(dev, dma_handle), size, DMA_FROM_DEVICE);
+	_dma_cache_sync(dma_handle, size, DMA_FROM_DEVICE);
 }
 
 static void arc_dma_sync_single_for_device(struct device *dev,
 		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
 {
-	_dma_cache_sync(plat_dma_to_phys(dev, dma_handle), size, DMA_TO_DEVICE);
+	_dma_cache_sync(dma_handle, size, DMA_TO_DEVICE);
 }
 
 static void arc_dma_sync_sg_for_cpu(struct device *dev,
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 51c8df561077..430a0aa710d6 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -8,6 +8,7 @@ config ARM
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_SET_MEMORY
+	select ARCH_HAS_PHYS_TO_DMA
 	select ARCH_HAS_STRICT_KERNEL_RWX if MMU && !XIP_KERNEL
 	select ARCH_HAS_STRICT_MODULE_RWX if MMU
 	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
@@ -24,7 +25,7 @@ config ARM
 	select CLONE_BACKWARDS
 	select CPU_PM if (SUSPEND || CPU_IDLE)
 	select DCACHE_WORD_ACCESS if HAVE_EFFICIENT_UNALIGNED_ACCESS
-	select DMA_NOOP_OPS if !MMU
+	select DMA_DIRECT_OPS if !MMU
 	select EDAC_SUPPORT
 	select EDAC_ATOMIC_SCRUB
 	select GENERIC_ALLOCATOR
diff --git a/arch/arm/include/asm/dma-direct.h b/arch/arm/include/asm/dma-direct.h
new file mode 100644
index 000000000000..5b0a8a421894
--- /dev/null
+++ b/arch/arm/include/asm/dma-direct.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef ASM_ARM_DMA_DIRECT_H
+#define ASM_ARM_DMA_DIRECT_H 1
+
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	unsigned int offset = paddr & ~PAGE_MASK;
+	return pfn_to_dma(dev, __phys_to_pfn(paddr)) + offset;
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
+{
+	unsigned int offset = dev_addr & ~PAGE_MASK;
+	return __pfn_to_phys(dma_to_pfn(dev, dev_addr)) + offset;
+}
+
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	u64 limit, mask;
+
+	if (!dev->dma_mask)
+		return 0;
+
+	mask = *dev->dma_mask;
+
+	limit = (mask + 1) & ~mask;
+	if (limit && size > limit)
+		return 0;
+
+	if ((addr | (addr + size - 1)) & ~mask)
+		return 0;
+
+	return 1;
+}
+
+#endif /* ASM_ARM_DMA_DIRECT_H */
diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index daf837423a76..8436f6ade57d 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h
@@ -18,7 +18,7 @@ extern const struct dma_map_ops arm_coherent_dma_ops;
 
 static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 {
-	return IS_ENABLED(CONFIG_MMU) ? &arm_dma_ops : &dma_noop_ops;
+	return IS_ENABLED(CONFIG_MMU) ? &arm_dma_ops : &dma_direct_ops;
 }
 
 #ifdef __arch_page_to_dma
@@ -109,39 +109,6 @@ static inline bool is_device_dma_coherent(struct device *dev)
 	return dev->archdata.dma_coherent;
 }
 
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	unsigned int offset = paddr & ~PAGE_MASK;
-	return pfn_to_dma(dev, __phys_to_pfn(paddr)) + offset;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
-{
-	unsigned int offset = dev_addr & ~PAGE_MASK;
-	return __pfn_to_phys(dma_to_pfn(dev, dev_addr)) + offset;
-}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	u64 limit, mask;
-
-	if (!dev->dma_mask)
-		return 0;
-
-	mask = *dev->dma_mask;
-
-	limit = (mask + 1) & ~mask;
-	if (limit && size > limit)
-		return 0;
-
-	if ((addr | (addr + size - 1)) & ~mask)
-		return 0;
-
-	return 1;
-}
-
-static inline void dma_mark_clean(void *addr, size_t size) { }
-
 /**
  * arm_dma_alloc - allocate consistent memory for DMA
  * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
diff --git a/arch/arm/mm/dma-mapping-nommu.c b/arch/arm/mm/dma-mapping-nommu.c
index 6db5fc26d154..619f24a42d09 100644
--- a/arch/arm/mm/dma-mapping-nommu.c
+++ b/arch/arm/mm/dma-mapping-nommu.c
@@ -11,7 +11,7 @@
 
 #include <linux/export.h>
 #include <linux/mm.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/scatterlist.h>
 
 #include <asm/cachetype.h>
@@ -22,7 +22,7 @@
 #include "dma.h"
 
 /*
- *  dma_noop_ops is used if
+ *  dma_direct_ops is used if
  *   - MMU/MPU is off
  *   - cpu is v7m w/o cache support
  *   - device is coherent
@@ -39,7 +39,6 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
 				 unsigned long attrs)
 
 {
-	const struct dma_map_ops *ops = &dma_noop_ops;
 	void *ret;
 
 	/*
@@ -48,7 +47,7 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
 	 */
 
 	if (attrs & DMA_ATTR_NON_CONSISTENT)
-		return ops->alloc(dev, size, dma_handle, gfp, attrs);
+		return dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
 
 	ret = dma_alloc_from_global_coherent(size, dma_handle);
 
@@ -70,10 +69,8 @@ static void arm_nommu_dma_free(struct device *dev, size_t size,
 			       void *cpu_addr, dma_addr_t dma_addr,
 			       unsigned long attrs)
 {
-	const struct dma_map_ops *ops = &dma_noop_ops;
-
 	if (attrs & DMA_ATTR_NON_CONSISTENT) {
-		ops->free(dev, size, cpu_addr, dma_addr, attrs);
+		dma_direct_free(dev, size, cpu_addr, dma_addr, attrs);
 	} else {
 		int ret = dma_release_from_global_coherent(get_order(size),
 							   cpu_addr);
@@ -213,7 +210,7 @@ EXPORT_SYMBOL(arm_nommu_dma_ops);
 
 static const struct dma_map_ops *arm_nommu_get_dma_map_ops(bool coherent)
 {
-	return coherent ? &dma_noop_ops : &arm_nommu_dma_ops;
+	return coherent ? &dma_direct_ops : &arm_nommu_dma_ops;
 }
 
 void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index b488076d63c2..b2b95f79c746 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -59,6 +59,7 @@ config ARM64
 	select COMMON_CLK
 	select CPU_PM if (SUSPEND || CPU_IDLE)
 	select DCACHE_WORD_ACCESS
+	select DMA_DIRECT_OPS
 	select EDAC_SUPPORT
 	select FRAME_POINTER
 	select GENERIC_ALLOCATOR
@@ -227,7 +228,7 @@ config GENERIC_CSUM
 config GENERIC_CALIBRATE_DELAY
 	def_bool y
 
-config ZONE_DMA
+config ZONE_DMA32
 	def_bool y
 
 config HAVE_GENERIC_GUP
diff --git a/arch/arm64/include/asm/dma-mapping.h b/arch/arm64/include/asm/dma-mapping.h
index 0df756b24863..b7847eb8a7bb 100644
--- a/arch/arm64/include/asm/dma-mapping.h
+++ b/arch/arm64/include/asm/dma-mapping.h
@@ -50,40 +50,5 @@ static inline bool is_device_dma_coherent(struct device *dev)
 	return dev->archdata.dma_coherent;
 }
 
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	dma_addr_t dev_addr = (dma_addr_t)paddr;
-
-	return dev_addr - ((dma_addr_t)dev->dma_pfn_offset << PAGE_SHIFT);
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
-{
-	phys_addr_t paddr = (phys_addr_t)dev_addr;
-
-	return paddr + ((phys_addr_t)dev->dma_pfn_offset << PAGE_SHIFT);
-}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return false;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-static inline void dma_mark_clean(void *addr, size_t size)
-{
-}
-
-/* Override for dma_max_pfn() */
-static inline unsigned long dma_max_pfn(struct device *dev)
-{
-	dma_addr_t dma_max = (dma_addr_t)*dev->dma_mask;
-
-	return (ulong)dma_to_phys(dev, dma_max) >> PAGE_SHIFT;
-}
-#define dma_max_pfn(dev) dma_max_pfn(dev)
-
 #endif	/* __KERNEL__ */
 #endif	/* __ASM_DMA_MAPPING_H */
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index b45c5bcaeccb..a96ec0181818 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -24,7 +24,7 @@
 #include <linux/export.h>
 #include <linux/slab.h>
 #include <linux/genalloc.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
@@ -91,46 +91,6 @@ static int __free_from_pool(void *start, size_t size)
 	return 1;
 }
 
-static void *__dma_alloc_coherent(struct device *dev, size_t size,
-				  dma_addr_t *dma_handle, gfp_t flags,
-				  unsigned long attrs)
-{
-	if (IS_ENABLED(CONFIG_ZONE_DMA) &&
-	    dev->coherent_dma_mask <= DMA_BIT_MASK(32))
-		flags |= GFP_DMA;
-	if (dev_get_cma_area(dev) && gfpflags_allow_blocking(flags)) {
-		struct page *page;
-		void *addr;
-
-		page = dma_alloc_from_contiguous(dev, size >> PAGE_SHIFT,
-						 get_order(size), flags);
-		if (!page)
-			return NULL;
-
-		*dma_handle = phys_to_dma(dev, page_to_phys(page));
-		addr = page_address(page);
-		memset(addr, 0, size);
-		return addr;
-	} else {
-		return swiotlb_alloc_coherent(dev, size, dma_handle, flags);
-	}
-}
-
-static void __dma_free_coherent(struct device *dev, size_t size,
-				void *vaddr, dma_addr_t dma_handle,
-				unsigned long attrs)
-{
-	bool freed;
-	phys_addr_t paddr = dma_to_phys(dev, dma_handle);
-
-
-	freed = dma_release_from_contiguous(dev,
-					phys_to_page(paddr),
-					size >> PAGE_SHIFT);
-	if (!freed)
-		swiotlb_free_coherent(dev, size, vaddr, dma_handle);
-}
-
 static void *__dma_alloc(struct device *dev, size_t size,
 			 dma_addr_t *dma_handle, gfp_t flags,
 			 unsigned long attrs)
@@ -152,7 +112,7 @@ static void *__dma_alloc(struct device *dev, size_t size,
 		return addr;
 	}
 
-	ptr = __dma_alloc_coherent(dev, size, dma_handle, flags, attrs);
+	ptr = swiotlb_alloc(dev, size, dma_handle, flags, attrs);
 	if (!ptr)
 		goto no_mem;
 
@@ -173,7 +133,7 @@ static void *__dma_alloc(struct device *dev, size_t size,
 	return coherent_ptr;
 
 no_map:
-	__dma_free_coherent(dev, size, ptr, *dma_handle, attrs);
+	swiotlb_free(dev, size, ptr, *dma_handle, attrs);
 no_mem:
 	return NULL;
 }
@@ -191,7 +151,7 @@ static void __dma_free(struct device *dev, size_t size,
 			return;
 		vunmap(vaddr);
 	}
-	__dma_free_coherent(dev, size, swiotlb_addr, dma_handle, attrs);
+	swiotlb_free(dev, size, swiotlb_addr, dma_handle, attrs);
 }
 
 static dma_addr_t __swiotlb_map_page(struct device *dev, struct page *page,
@@ -368,7 +328,7 @@ static int __swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t addr)
 	return 0;
 }
 
-static const struct dma_map_ops swiotlb_dma_ops = {
+static const struct dma_map_ops arm64_swiotlb_dma_ops = {
 	.alloc = __dma_alloc,
 	.free = __dma_free,
 	.mmap = __swiotlb_mmap,
@@ -397,7 +357,7 @@ static int __init atomic_pool_init(void)
 		page = dma_alloc_from_contiguous(NULL, nr_pages,
 						 pool_size_order, GFP_KERNEL);
 	else
-		page = alloc_pages(GFP_DMA, pool_size_order);
+		page = alloc_pages(GFP_DMA32, pool_size_order);
 
 	if (page) {
 		int ret;
@@ -923,7 +883,7 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 			const struct iommu_ops *iommu, bool coherent)
 {
 	if (!dev->dma_ops)
-		dev->dma_ops = &swiotlb_dma_ops;
+		dev->dma_ops = &arm64_swiotlb_dma_ops;
 
 	dev->archdata.dma_coherent = coherent;
 	__iommu_setup_dma_ops(dev, dma_base, size, iommu);
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index c903f7ccbdd2..9f3c47acf8ff 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -217,7 +217,7 @@ static void __init reserve_elfcorehdr(void)
 }
 #endif /* CONFIG_CRASH_DUMP */
 /*
- * Return the maximum physical address for ZONE_DMA (DMA_BIT_MASK(32)). It
+ * Return the maximum physical address for ZONE_DMA32 (DMA_BIT_MASK(32)). It
  * currently assumes that for memory starting above 4G, 32-bit devices will
  * use a DMA offset.
  */
@@ -233,8 +233,8 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 {
 	unsigned long max_zone_pfns[MAX_NR_ZONES]  = {0};
 
-	if (IS_ENABLED(CONFIG_ZONE_DMA))
-		max_zone_pfns[ZONE_DMA] = PFN_DOWN(max_zone_dma_phys());
+	if (IS_ENABLED(CONFIG_ZONE_DMA32))
+		max_zone_pfns[ZONE_DMA32] = PFN_DOWN(max_zone_dma_phys());
 	max_zone_pfns[ZONE_NORMAL] = max;
 
 	free_area_init_nodes(max_zone_pfns);
@@ -251,9 +251,9 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 	memset(zone_size, 0, sizeof(zone_size));
 
 	/* 4GB maximum for 32-bit only capable devices */
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
 	max_dma = PFN_DOWN(arm64_dma_phys_limit);
-	zone_size[ZONE_DMA] = max_dma - min;
+	zone_size[ZONE_DMA32] = max_dma - min;
 #endif
 	zone_size[ZONE_NORMAL] = max - max_dma;
 
@@ -266,10 +266,10 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 		if (start >= max)
 			continue;
 
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
 		if (start < max_dma) {
 			unsigned long dma_end = min(end, max_dma);
-			zhole_size[ZONE_DMA] -= dma_end - start;
+			zhole_size[ZONE_DMA32] -= dma_end - start;
 		}
 #endif
 		if (end > max_dma) {
@@ -470,7 +470,7 @@ void __init arm64_memblock_init(void)
 	early_init_fdt_scan_reserved_mem();
 
 	/* 4GB maximum for 32-bit only capable devices */
-	if (IS_ENABLED(CONFIG_ZONE_DMA))
+	if (IS_ENABLED(CONFIG_ZONE_DMA32))
 		arm64_dma_phys_limit = max_zone_dma_phys();
 	else
 		arm64_dma_phys_limit = PHYS_MASK + 1;
diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig
index 54d3f426763b..cd5a0865c97f 100644
--- a/arch/cris/Kconfig
+++ b/arch/cris/Kconfig
@@ -33,6 +33,9 @@ config GENERIC_CALIBRATE_DELAY
 config NO_IOPORT_MAP
 	def_bool y if !PCI
 
+config NO_DMA
+	def_bool y if !PCI
+
 config FORCE_MAX_ZONEORDER
 	int
 	default 6
@@ -72,6 +75,7 @@ config CRIS
 	select GENERIC_SCHED_CLOCK if ETRAX_ARCH_V32
 	select HAVE_DEBUG_BUGVERBOSE if ETRAX_ARCH_V32
 	select HAVE_NMI
+	select DMA_DIRECT_OPS if PCI
 
 config HZ
 	int
diff --git a/arch/cris/arch-v32/drivers/pci/Makefile b/arch/cris/arch-v32/drivers/pci/Makefile
index bff7482f2444..93c8be6170b1 100644
--- a/arch/cris/arch-v32/drivers/pci/Makefile
+++ b/arch/cris/arch-v32/drivers/pci/Makefile
@@ -2,4 +2,4 @@
 # Makefile for Etrax cardbus driver
 #
 
-obj-$(CONFIG_ETRAX_CARDBUS)        += bios.o dma.o
+obj-$(CONFIG_ETRAX_CARDBUS)        += bios.o
diff --git a/arch/cris/arch-v32/drivers/pci/dma.c b/arch/cris/arch-v32/drivers/pci/dma.c
deleted file mode 100644
index dbbd3816cc0b..000000000000
--- a/arch/cris/arch-v32/drivers/pci/dma.c
+++ /dev/null
@@ -1,80 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Dynamic DMA mapping support.
- *
- * On cris there is no hardware dynamic DMA address translation,
- * so consistent alloc/free are merely page allocation/freeing.
- * The rest of the dynamic DMA mapping interface is implemented
- * in asm/pci.h.
- *
- * Borrowed from i386.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <linux/gfp.h>
-#include <asm/io.h>
-
-static void *v32_dma_alloc(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
-{
-	void *ret;
-
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
-	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
-		gfp |= GFP_DMA;
-
-	ret = (void *)__get_free_pages(gfp,  get_order(size));
-
-	if (ret != NULL) {
-		memset(ret, 0, size);
-		*dma_handle = virt_to_phys(ret);
-	}
-	return ret;
-}
-
-static void v32_dma_free(struct device *dev, size_t size, void *vaddr,
-		dma_addr_t dma_handle, unsigned long attrs)
-{
-	free_pages((unsigned long)vaddr, get_order(size));
-}
-
-static inline dma_addr_t v32_dma_map_page(struct device *dev,
-		struct page *page, unsigned long offset, size_t size,
-		enum dma_data_direction direction, unsigned long attrs)
-{
-	return page_to_phys(page) + offset;
-}
-
-static inline int v32_dma_map_sg(struct device *dev, struct scatterlist *sg,
-		int nents, enum dma_data_direction direction,
-		unsigned long attrs)
-{
-	printk("Map sg\n");
-	return nents;
-}
-
-static inline int v32_dma_supported(struct device *dev, u64 mask)
-{
-        /*
-         * we fall back to GFP_DMA when the mask isn't all 1s,
-         * so we can't guarantee allocations that must be
-         * within a tighter range than GFP_DMA..
-         */
-        if (mask < 0x00ffffff)
-                return 0;
-	return 1;
-}
-
-const struct dma_map_ops v32_dma_ops = {
-	.alloc			= v32_dma_alloc,
-	.free			= v32_dma_free,
-	.map_page		= v32_dma_map_page,
-	.map_sg                 = v32_dma_map_sg,
-	.dma_supported		= v32_dma_supported,
-};
-EXPORT_SYMBOL(v32_dma_ops);
diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild
index 460349cb147f..8cf45ac30c1b 100644
--- a/arch/cris/include/asm/Kbuild
+++ b/arch/cris/include/asm/Kbuild
@@ -5,6 +5,7 @@ generic-y += cmpxchg.h
 generic-y += current.h
 generic-y += device.h
 generic-y += div64.h
+generic-y += dma-mapping.h
 generic-y += emergency-restart.h
 generic-y += exec.h
 generic-y += extable.h
diff --git a/arch/cris/include/asm/dma-mapping.h b/arch/cris/include/asm/dma-mapping.h
deleted file mode 100644
index 1553bdb30a0c..000000000000
--- a/arch/cris/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_CRIS_DMA_MAPPING_H
-#define _ASM_CRIS_DMA_MAPPING_H
-
-#ifdef CONFIG_PCI
-extern const struct dma_map_ops v32_dma_ops;
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return &v32_dma_ops;
-}
-#else
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	BUG();
-	return NULL;
-}
-#endif
-
-#endif
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index f8d3fde08190..091d6d04b5e5 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -23,6 +23,7 @@ config H8300
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_HASH
 	select CPU_NO_EFFICIENT_FFS
+	select DMA_DIRECT_OPS
 
 config CPU_BIG_ENDIAN
 	def_bool y
diff --git a/arch/h8300/include/asm/Kbuild b/arch/h8300/include/asm/Kbuild
index bc077491d299..642752c94306 100644
--- a/arch/h8300/include/asm/Kbuild
+++ b/arch/h8300/include/asm/Kbuild
@@ -9,6 +9,7 @@ generic-y += delay.h
 generic-y += device.h
 generic-y += div64.h
 generic-y += dma.h
+generic-y += dma-mapping.h
 generic-y += emergency-restart.h
 generic-y += exec.h
 generic-y += extable.h
diff --git a/arch/h8300/include/asm/dma-mapping.h b/arch/h8300/include/asm/dma-mapping.h
deleted file mode 100644
index 21bb1fc3a6f1..000000000000
--- a/arch/h8300/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _H8300_DMA_MAPPING_H
-#define _H8300_DMA_MAPPING_H
-
-extern const struct dma_map_ops h8300_dma_map_ops;
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return &h8300_dma_map_ops;
-}
-
-#endif
diff --git a/arch/h8300/kernel/Makefile b/arch/h8300/kernel/Makefile
index b62e830525c6..307aa51576dd 100644
--- a/arch/h8300/kernel/Makefile
+++ b/arch/h8300/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := vmlinux.lds
 
 obj-y := process.o traps.o ptrace.o \
 	 signal.o setup.o syscalls.o \
-	 irq.o entry.o dma.o
+	 irq.o entry.o
 
 obj-$(CONFIG_ROMKERNEL) += head_rom.o
 obj-$(CONFIG_RAMKERNEL) += head_ram.o
diff --git a/arch/h8300/kernel/dma.c b/arch/h8300/kernel/dma.c
deleted file mode 100644
index 225dd0a188dc..000000000000
--- a/arch/h8300/kernel/dma.c
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file COPYING in the main directory of this archive
- * for more details.
- */
-
-#include <linux/dma-mapping.h>
-#include <linux/kernel.h>
-#include <linux/scatterlist.h>
-#include <linux/module.h>
-#include <asm/pgalloc.h>
-
-static void *dma_alloc(struct device *dev, size_t size,
-		       dma_addr_t *dma_handle, gfp_t gfp,
-		       unsigned long attrs)
-{
-	void *ret;
-
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
-	if (dev == NULL || (*dev->dma_mask < 0xffffffff))
-		gfp |= GFP_DMA;
-	ret = (void *)__get_free_pages(gfp, get_order(size));
-
-	if (ret != NULL) {
-		memset(ret, 0, size);
-		*dma_handle = virt_to_phys(ret);
-	}
-	return ret;
-}
-
-static void dma_free(struct device *dev, size_t size,
-		     void *vaddr, dma_addr_t dma_handle,
-		     unsigned long attrs)
-
-{
-	free_pages((unsigned long)vaddr, get_order(size));
-}
-
-static dma_addr_t map_page(struct device *dev, struct page *page,
-				  unsigned long offset, size_t size,
-				  enum dma_data_direction direction,
-				  unsigned long attrs)
-{
-	return page_to_phys(page) + offset;
-}
-
-static int map_sg(struct device *dev, struct scatterlist *sgl,
-		  int nents, enum dma_data_direction direction,
-		  unsigned long attrs)
-{
-	struct scatterlist *sg;
-	int i;
-
-	for_each_sg(sgl, sg, nents, i) {
-		sg->dma_address = sg_phys(sg);
-	}
-
-	return nents;
-}
-
-const struct dma_map_ops h8300_dma_map_ops = {
-	.alloc = dma_alloc,
-	.free = dma_free,
-	.map_page = map_page,
-	.map_sg = map_sg,
-};
-EXPORT_SYMBOL(h8300_dma_map_ops);
diff --git a/arch/hexagon/include/asm/dma-mapping.h b/arch/hexagon/include/asm/dma-mapping.h
index 5208de242e79..263f6acbfb0f 100644
--- a/arch/hexagon/include/asm/dma-mapping.h
+++ b/arch/hexagon/include/asm/dma-mapping.h
@@ -37,11 +37,4 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return dma_ops;
 }
 
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return 0;
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
 #endif
diff --git a/arch/hexagon/include/asm/io.h b/arch/hexagon/include/asm/io.h
index 66f5e9a61efc..9e8621d94ee9 100644
--- a/arch/hexagon/include/asm/io.h
+++ b/arch/hexagon/include/asm/io.h
@@ -330,8 +330,6 @@ static inline void outsl(unsigned long port, const void *buffer, int count)
 	}
 }
 
-#define flush_write_buffers() do { } while (0)
-
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/arch/hexagon/kernel/dma.c b/arch/hexagon/kernel/dma.c
index 546792d176a4..ad8347c29dcf 100644
--- a/arch/hexagon/kernel/dma.c
+++ b/arch/hexagon/kernel/dma.c
@@ -19,6 +19,7 @@
  */
 
 #include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/bootmem.h>
 #include <linux/genalloc.h>
 #include <asm/dma-mapping.h>
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 315c51f58811..bbe12a038d21 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -33,6 +33,7 @@ config IA64
 	select HAVE_MEMBLOCK
 	select HAVE_MEMBLOCK_NODE_MAP
 	select HAVE_VIRT_CPU_ACCOUNTING
+	select ARCH_HAS_DMA_MARK_CLEAN
 	select ARCH_HAS_SG_CHAIN
 	select VIRT_TO_BUS
 	select ARCH_DISCARD_MEMBLOCK
@@ -65,7 +66,7 @@ config 64BIT
 	select ATA_NONSTANDARD if ATA
 	default y
 
-config ZONE_DMA
+config ZONE_DMA32
 	def_bool y
 	depends on !IA64_SGI_SN2
 
@@ -145,6 +146,7 @@ config IA64_GENERIC
 	bool "generic"
 	select NUMA
 	select ACPI_NUMA
+	select DMA_DIRECT_OPS
 	select SWIOTLB
 	select PCI_MSI
 	help
@@ -165,6 +167,7 @@ config IA64_GENERIC
 
 config IA64_DIG
 	bool "DIG-compliant"
+	select DMA_DIRECT_OPS
 	select SWIOTLB
 
 config IA64_DIG_VTD
@@ -180,6 +183,7 @@ config IA64_HP_ZX1
 
 config IA64_HP_ZX1_SWIOTLB
 	bool "HP-zx1/sx1000 with software I/O TLB"
+	select DMA_DIRECT_OPS
 	select SWIOTLB
 	help
 	  Build a kernel that runs on HP zx1 and sx1000 systems even when they
@@ -203,6 +207,7 @@ config IA64_SGI_UV
 	bool "SGI-UV"
 	select NUMA
 	select ACPI_NUMA
+	select DMA_DIRECT_OPS
 	select SWIOTLB
 	help
 	  Selecting this option will optimize the kernel for use on UV based
@@ -213,6 +218,7 @@ config IA64_SGI_UV
 
 config IA64_HP_SIM
 	bool "Ski-simulator"
+	select DMA_DIRECT_OPS
 	select SWIOTLB
 	depends on !PM
 
diff --git a/arch/ia64/hp/common/hwsw_iommu.c b/arch/ia64/hp/common/hwsw_iommu.c
index 63d8e1d2477f..58969039bed2 100644
--- a/arch/ia64/hp/common/hwsw_iommu.c
+++ b/arch/ia64/hp/common/hwsw_iommu.c
@@ -19,7 +19,7 @@
 #include <linux/export.h>
 #include <asm/machvec.h>
 
-extern const struct dma_map_ops sba_dma_ops, swiotlb_dma_ops;
+extern const struct dma_map_ops sba_dma_ops;
 
 /* swiotlb declarations & definitions: */
 extern int swiotlb_late_init_with_default_size (size_t size);
diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h
index c1bab526a046..76e4d6632d68 100644
--- a/arch/ia64/include/asm/dma-mapping.h
+++ b/arch/ia64/include/asm/dma-mapping.h
@@ -8,7 +8,6 @@
  */
 #include <asm/machvec.h>
 #include <linux/scatterlist.h>
-#include <asm/swiotlb.h>
 #include <linux/dma-debug.h>
 
 #define ARCH_HAS_DMA_GET_REQUIRED_MASK
@@ -27,22 +26,4 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return platform_dma_get_ops(NULL);
 }
 
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return 0;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return paddr;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return daddr;
-}
-
 #endif /* _ASM_IA64_DMA_MAPPING_H */
diff --git a/arch/ia64/include/asm/dma.h b/arch/ia64/include/asm/dma.h
index 186850eec934..23604d6a2cb2 100644
--- a/arch/ia64/include/asm/dma.h
+++ b/arch/ia64/include/asm/dma.h
@@ -20,6 +20,4 @@ extern unsigned long MAX_DMA_ADDRESS;
 
 #define free_dma(x)
 
-void dma_mark_clean(void *addr, size_t size);
-
 #endif /* _ASM_IA64_DMA_H */
diff --git a/arch/ia64/include/asm/swiotlb.h b/arch/ia64/include/asm/swiotlb.h
deleted file mode 100644
index 841e2c7d0b21..000000000000
--- a/arch/ia64/include/asm/swiotlb.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef ASM_IA64__SWIOTLB_H
-#define ASM_IA64__SWIOTLB_H
-
-#include <linux/dma-mapping.h>
-#include <linux/swiotlb.h>
-
-#ifdef CONFIG_SWIOTLB
-extern int swiotlb;
-extern void pci_swiotlb_init(void);
-#else
-#define swiotlb 0
-static inline void pci_swiotlb_init(void)
-{
-}
-#endif
-
-#endif /* ASM_IA64__SWIOTLB_H */
diff --git a/arch/ia64/kernel/dma-mapping.c b/arch/ia64/kernel/dma-mapping.c
index 7a82c9259609..f2d57e66fd86 100644
--- a/arch/ia64/kernel/dma-mapping.c
+++ b/arch/ia64/kernel/dma-mapping.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/dma-mapping.h>
+#include <linux/swiotlb.h>
 #include <linux/export.h>
 
 /* Set this to 1 if there is a HW IOMMU in the system */
@@ -23,3 +24,11 @@ const struct dma_map_ops *dma_get_ops(struct device *dev)
 	return dma_ops;
 }
 EXPORT_SYMBOL(dma_get_ops);
+
+#ifdef CONFIG_SWIOTLB
+void __init swiotlb_dma_init(void)
+{
+	dma_ops = &swiotlb_dma_ops;
+	swiotlb_init(1);
+}
+#endif
diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index 3ba87c22dfbc..b5df084c0af4 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -12,12 +12,7 @@
 #include <asm/iommu.h>
 #include <asm/machvec.h>
 #include <linux/dma-mapping.h>
-
-
-#ifdef CONFIG_INTEL_IOMMU
-
 #include <linux/kernel.h>
-
 #include <asm/page.h>
 
 dma_addr_t bad_dma_address __read_mostly;
@@ -104,8 +99,14 @@ void __init pci_iommu_alloc(void)
 	detect_intel_iommu();
 
 #ifdef CONFIG_SWIOTLB
-	pci_swiotlb_init();
-#endif
+	if (!iommu_detected) {
+#ifdef CONFIG_IA64_GENERIC
+		printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
+		machvec_init("dig");
+		swiotlb_dma_init();
+#else
+		panic("Unable to find Intel IOMMU");
+#endif /* CONFIG_IA64_GENERIC */
+	}
+#endif /* CONFIG_SWIOTLB */
 }
-
-#endif
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
deleted file mode 100644
index 5e50939aa03e..000000000000
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ /dev/null
@@ -1,68 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/* Glue code to lib/swiotlb.c */
-
-#include <linux/pci.h>
-#include <linux/gfp.h>
-#include <linux/cache.h>
-#include <linux/module.h>
-#include <linux/dma-mapping.h>
-
-#include <asm/swiotlb.h>
-#include <asm/dma.h>
-#include <asm/iommu.h>
-#include <asm/machvec.h>
-
-int swiotlb __read_mostly;
-EXPORT_SYMBOL(swiotlb);
-
-static void *ia64_swiotlb_alloc_coherent(struct device *dev, size_t size,
-					 dma_addr_t *dma_handle, gfp_t gfp,
-					 unsigned long attrs)
-{
-	if (dev->coherent_dma_mask != DMA_BIT_MASK(64))
-		gfp |= GFP_DMA;
-	return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
-}
-
-static void ia64_swiotlb_free_coherent(struct device *dev, size_t size,
-				       void *vaddr, dma_addr_t dma_addr,
-				       unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_addr);
-}
-
-const struct dma_map_ops swiotlb_dma_ops = {
-	.alloc = ia64_swiotlb_alloc_coherent,
-	.free = ia64_swiotlb_free_coherent,
-	.map_page = swiotlb_map_page,
-	.unmap_page = swiotlb_unmap_page,
-	.map_sg = swiotlb_map_sg_attrs,
-	.unmap_sg = swiotlb_unmap_sg_attrs,
-	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
-	.sync_single_for_device = swiotlb_sync_single_for_device,
-	.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
-	.sync_sg_for_device = swiotlb_sync_sg_for_device,
-	.dma_supported = swiotlb_dma_supported,
-	.mapping_error = swiotlb_dma_mapping_error,
-};
-
-void __init swiotlb_dma_init(void)
-{
-	dma_ops = &swiotlb_dma_ops;
-	swiotlb_init(1);
-}
-
-void __init pci_swiotlb_init(void)
-{
-	if (!iommu_detected) {
-#ifdef CONFIG_IA64_GENERIC
-		swiotlb = 1;
-		printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
-		machvec_init("dig");
-		swiotlb_init(1);
-		dma_ops = &swiotlb_dma_ops;
-#else
-		panic("Unable to find Intel IOMMU");
-#endif
-	}
-}
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index 52715a71aede..7d64b30913d1 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -237,9 +237,9 @@ paging_init (void)
 	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
 	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-	max_zone_pfns[ZONE_DMA] = max_dma;
+	max_zone_pfns[ZONE_DMA32] = max_dma;
 #endif
 	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
 
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 9b2d994cddf6..ac46f0d60b66 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -38,7 +38,7 @@ struct early_node_data {
 	struct ia64_node_data *node_data;
 	unsigned long pernode_addr;
 	unsigned long pernode_size;
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
 	unsigned long num_dma_physpages;
 #endif
 	unsigned long min_pfn;
@@ -669,7 +669,7 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
 {
 	unsigned long end = start + len;
 
-#ifdef CONFIG_ZONE_DMA
+#ifdef CONFIG_ZONE_DMA32
 	if (start <= __pa(MAX_DMA_ADDRESS))
 		mem_data[node].num_dma_physpages +=
 			(min(end, __pa(MAX_DMA_ADDRESS)) - start) >>PAGE_SHIFT;
@@ -724,8 +724,8 @@ void __init paging_init(void)
 	}
 
 	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
-#ifdef CONFIG_ZONE_DMA
-	max_zone_pfns[ZONE_DMA] = max_dma;
+#ifdef CONFIG_ZONE_DMA32
+	max_zone_pfns[ZONE_DMA32] = max_dma;
 #endif
 	max_zone_pfns[ZONE_NORMAL] = max_pfn;
 	free_area_init_nodes(max_zone_pfns);
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index 498398d915c1..dd84ee194579 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -19,7 +19,7 @@ config M32R
 	select MODULES_USE_ELF_RELA
 	select HAVE_DEBUG_STACKOVERFLOW
 	select CPU_NO_EFFICIENT_FFS
-	select DMA_NOOP_OPS
+	select DMA_DIRECT_OPS
 	select ARCH_NO_COHERENT_DMA_MMAP if !MMU
 
 config SBUS
diff --git a/arch/m32r/include/asm/Kbuild b/arch/m32r/include/asm/Kbuild
index 7e11b125c35e..ca83fda8177b 100644
--- a/arch/m32r/include/asm/Kbuild
+++ b/arch/m32r/include/asm/Kbuild
@@ -1,5 +1,6 @@
 generic-y += clkdev.h
 generic-y += current.h
+generic-y += dma-mapping.h
 generic-y += exec.h
 generic-y += extable.h
 generic-y += irq_work.h
diff --git a/arch/m32r/include/asm/dma-mapping.h b/arch/m32r/include/asm/dma-mapping.h
deleted file mode 100644
index 336ffe60814b..000000000000
--- a/arch/m32r/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_M32R_DMA_MAPPING_H
-#define _ASM_M32R_DMA_MAPPING_H
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-#include <linux/dma-debug.h>
-#include <linux/io.h>
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return &dma_noop_ops;
-}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return false;
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-#endif /* _ASM_M32R_DMA_MAPPING_H */
diff --git a/arch/m32r/include/asm/io.h b/arch/m32r/include/asm/io.h
index 1b653bb16f9a..a4272d8f0d9c 100644
--- a/arch/m32r/include/asm/io.h
+++ b/arch/m32r/include/asm/io.h
@@ -191,8 +191,6 @@ static inline void _writel(unsigned long l, unsigned long addr)
 
 #define mmiowb()
 
-#define flush_write_buffers() do { } while (0)  /* M32R_FIXME */
-
 static inline void
 memset_io(volatile void __iomem *addr, unsigned char val, int count)
 {
diff --git a/arch/m68k/kernel/dma.c b/arch/m68k/kernel/dma.c
index 87ef73a93856..c01b9b8f97bf 100644
--- a/arch/m68k/kernel/dma.c
+++ b/arch/m68k/kernel/dma.c
@@ -76,8 +76,6 @@ static void *m68k_dma_alloc(struct device *dev, size_t size,
 		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
 	void *ret;
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
 
 	if (dev == NULL || (*dev->dma_mask < 0xffffffff))
 		gfp |= GFP_DMA;
diff --git a/arch/microblaze/include/asm/dma-mapping.h b/arch/microblaze/include/asm/dma-mapping.h
index 6b9ea39405b8..add50c1373bf 100644
--- a/arch/microblaze/include/asm/dma-mapping.h
+++ b/arch/microblaze/include/asm/dma-mapping.h
@@ -18,11 +18,11 @@
 /*
  * Available generic sets of operations
  */
-extern const struct dma_map_ops dma_direct_ops;
+extern const struct dma_map_ops dma_nommu_ops;
 
 static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 {
-	return &dma_direct_ops;
+	return &dma_nommu_ops;
 }
 
 #endif	/* _ASM_MICROBLAZE_DMA_MAPPING_H */
diff --git a/arch/microblaze/kernel/dma.c b/arch/microblaze/kernel/dma.c
index 990bf9ea0ec6..c91e8cef98dd 100644
--- a/arch/microblaze/kernel/dma.c
+++ b/arch/microblaze/kernel/dma.c
@@ -15,42 +15,18 @@
 #include <linux/bug.h>
 #include <asm/cacheflush.h>
 
-#define NOT_COHERENT_CACHE
-
-static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
+static void *dma_nommu_alloc_coherent(struct device *dev, size_t size,
 				       dma_addr_t *dma_handle, gfp_t flag,
 				       unsigned long attrs)
 {
-#ifdef NOT_COHERENT_CACHE
 	return consistent_alloc(flag, size, dma_handle);
-#else
-	void *ret;
-	struct page *page;
-	int node = dev_to_node(dev);
-
-	/* ignore region specifiers */
-	flag  &= ~(__GFP_HIGHMEM);
-
-	page = alloc_pages_node(node, flag, get_order(size));
-	if (page == NULL)
-		return NULL;
-	ret = page_address(page);
-	memset(ret, 0, size);
-	*dma_handle = virt_to_phys(ret);
-
-	return ret;
-#endif
 }
 
-static void dma_direct_free_coherent(struct device *dev, size_t size,
+static void dma_nommu_free_coherent(struct device *dev, size_t size,
 				     void *vaddr, dma_addr_t dma_handle,
 				     unsigned long attrs)
 {
-#ifdef NOT_COHERENT_CACHE
 	consistent_free(size, vaddr);
-#else
-	free_pages((unsigned long)vaddr, get_order(size));
-#endif
 }
 
 static inline void __dma_sync(unsigned long paddr,
@@ -69,7 +45,7 @@ static inline void __dma_sync(unsigned long paddr,
 	}
 }
 
-static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
+static int dma_nommu_map_sg(struct device *dev, struct scatterlist *sgl,
 			     int nents, enum dma_data_direction direction,
 			     unsigned long attrs)
 {
@@ -89,12 +65,7 @@ static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
 	return nents;
 }
 
-static int dma_direct_dma_supported(struct device *dev, u64 mask)
-{
-	return 1;
-}
-
-static inline dma_addr_t dma_direct_map_page(struct device *dev,
+static inline dma_addr_t dma_nommu_map_page(struct device *dev,
 					     struct page *page,
 					     unsigned long offset,
 					     size_t size,
@@ -106,7 +77,7 @@ static inline dma_addr_t dma_direct_map_page(struct device *dev,
 	return page_to_phys(page) + offset;
 }
 
-static inline void dma_direct_unmap_page(struct device *dev,
+static inline void dma_nommu_unmap_page(struct device *dev,
 					 dma_addr_t dma_address,
 					 size_t size,
 					 enum dma_data_direction direction,
@@ -122,7 +93,7 @@ static inline void dma_direct_unmap_page(struct device *dev,
 }
 
 static inline void
-dma_direct_sync_single_for_cpu(struct device *dev,
+dma_nommu_sync_single_for_cpu(struct device *dev,
 			       dma_addr_t dma_handle, size_t size,
 			       enum dma_data_direction direction)
 {
@@ -136,7 +107,7 @@ dma_direct_sync_single_for_cpu(struct device *dev,
 }
 
 static inline void
-dma_direct_sync_single_for_device(struct device *dev,
+dma_nommu_sync_single_for_device(struct device *dev,
 				  dma_addr_t dma_handle, size_t size,
 				  enum dma_data_direction direction)
 {
@@ -150,7 +121,7 @@ dma_direct_sync_single_for_device(struct device *dev,
 }
 
 static inline void
-dma_direct_sync_sg_for_cpu(struct device *dev,
+dma_nommu_sync_sg_for_cpu(struct device *dev,
 			   struct scatterlist *sgl, int nents,
 			   enum dma_data_direction direction)
 {
@@ -164,7 +135,7 @@ dma_direct_sync_sg_for_cpu(struct device *dev,
 }
 
 static inline void
-dma_direct_sync_sg_for_device(struct device *dev,
+dma_nommu_sync_sg_for_device(struct device *dev,
 			      struct scatterlist *sgl, int nents,
 			      enum dma_data_direction direction)
 {
@@ -178,7 +149,7 @@ dma_direct_sync_sg_for_device(struct device *dev,
 }
 
 static
-int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
+int dma_nommu_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
 			     void *cpu_addr, dma_addr_t handle, size_t size,
 			     unsigned long attrs)
 {
@@ -191,12 +162,8 @@ int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
 	if (off >= count || user_count > (count - off))
 		return -ENXIO;
 
-#ifdef NOT_COHERENT_CACHE
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 	pfn = consistent_virt_to_pfn(cpu_addr);
-#else
-	pfn = virt_to_pfn(cpu_addr);
-#endif
 	return remap_pfn_range(vma, vma->vm_start, pfn + off,
 			       vma->vm_end - vma->vm_start, vma->vm_page_prot);
 #else
@@ -204,20 +171,19 @@ int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
 #endif
 }
 
-const struct dma_map_ops dma_direct_ops = {
-	.alloc		= dma_direct_alloc_coherent,
-	.free		= dma_direct_free_coherent,
-	.mmap		= dma_direct_mmap_coherent,
-	.map_sg		= dma_direct_map_sg,
-	.dma_supported	= dma_direct_dma_supported,
-	.map_page	= dma_direct_map_page,
-	.unmap_page	= dma_direct_unmap_page,
-	.sync_single_for_cpu		= dma_direct_sync_single_for_cpu,
-	.sync_single_for_device		= dma_direct_sync_single_for_device,
-	.sync_sg_for_cpu		= dma_direct_sync_sg_for_cpu,
-	.sync_sg_for_device		= dma_direct_sync_sg_for_device,
+const struct dma_map_ops dma_nommu_ops = {
+	.alloc			= dma_nommu_alloc_coherent,
+	.free			= dma_nommu_free_coherent,
+	.mmap			= dma_nommu_mmap_coherent,
+	.map_sg			= dma_nommu_map_sg,
+	.map_page		= dma_nommu_map_page,
+	.unmap_page		= dma_nommu_unmap_page,
+	.sync_single_for_cpu	= dma_nommu_sync_single_for_cpu,
+	.sync_single_for_device	= dma_nommu_sync_single_for_device,
+	.sync_sg_for_cpu	= dma_nommu_sync_sg_for_cpu,
+	.sync_sg_for_device	= dma_nommu_sync_sg_for_device,
 };
-EXPORT_SYMBOL(dma_direct_ops);
+EXPORT_SYMBOL(dma_nommu_ops);
 
 /* Number of entries preallocated for DMA-API debugging */
 #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 8e0b3702f1c0..ab98569994f0 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -431,6 +431,7 @@ config MACH_LOONGSON32
 
 config MACH_LOONGSON64
 	bool "Loongson-2/3 family of machines"
+	select ARCH_HAS_PHYS_TO_DMA
 	select SYS_SUPPORTS_ZBOOT
 	help
 	  This enables the support of Loongson-2/3 family of machines.
@@ -880,6 +881,7 @@ config MIKROTIK_RB532
 config CAVIUM_OCTEON_SOC
 	bool "Cavium Networks Octeon SoC based boards"
 	select CEVT_R4K
+	select ARCH_HAS_PHYS_TO_DMA
 	select ARCH_PHYS_ADDR_T_64BIT
 	select DMA_COHERENT
 	select SYS_SUPPORTS_64BIT_KERNEL
diff --git a/arch/mips/cavium-octeon/Kconfig b/arch/mips/cavium-octeon/Kconfig
index 204a1670fd9b..b5eee1a57d6c 100644
--- a/arch/mips/cavium-octeon/Kconfig
+++ b/arch/mips/cavium-octeon/Kconfig
@@ -75,6 +75,7 @@ config NEED_SG_DMA_LENGTH
 
 config SWIOTLB
 	def_bool y
+	select DMA_DIRECT_OPS
 	select IOMMU_HELPER
 	select NEED_SG_DMA_LENGTH
 
diff --git a/arch/mips/cavium-octeon/dma-octeon.c b/arch/mips/cavium-octeon/dma-octeon.c
index c64bd87f0b6e..c7bb8a407041 100644
--- a/arch/mips/cavium-octeon/dma-octeon.c
+++ b/arch/mips/cavium-octeon/dma-octeon.c
@@ -159,36 +159,13 @@ static void octeon_dma_sync_sg_for_device(struct device *dev,
 static void *octeon_dma_alloc_coherent(struct device *dev, size_t size,
 	dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
-	void *ret;
-
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM);
-
-	if (IS_ENABLED(CONFIG_ZONE_DMA) && dev == NULL)
-		gfp |= __GFP_DMA;
-	else if (IS_ENABLED(CONFIG_ZONE_DMA) &&
-		 dev->coherent_dma_mask <= DMA_BIT_MASK(24))
-		gfp |= __GFP_DMA;
-	else if (IS_ENABLED(CONFIG_ZONE_DMA32) &&
-		 dev->coherent_dma_mask <= DMA_BIT_MASK(32))
-		gfp |= __GFP_DMA32;
-
-	/* Don't invoke OOM killer */
-	gfp |= __GFP_NORETRY;
-
-	ret = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
+	void *ret = swiotlb_alloc(dev, size, dma_handle, gfp, attrs);
 
 	mb();
 
 	return ret;
 }
 
-static void octeon_dma_free_coherent(struct device *dev, size_t size,
-	void *vaddr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_handle);
-}
-
 static dma_addr_t octeon_unity_phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
 	return paddr;
@@ -228,7 +205,7 @@ EXPORT_SYMBOL(dma_to_phys);
 static struct octeon_dma_map_ops octeon_linear_dma_map_ops = {
 	.dma_map_ops = {
 		.alloc = octeon_dma_alloc_coherent,
-		.free = octeon_dma_free_coherent,
+		.free = swiotlb_free,
 		.map_page = octeon_dma_map_page,
 		.unmap_page = swiotlb_unmap_page,
 		.map_sg = octeon_dma_map_sg,
@@ -314,7 +291,7 @@ void __init plat_swiotlb_setup(void)
 static struct octeon_dma_map_ops _octeon_pci_dma_map_ops = {
 	.dma_map_ops = {
 		.alloc = octeon_dma_alloc_coherent,
-		.free = octeon_dma_free_coherent,
+		.free = swiotlb_free,
 		.map_page = octeon_dma_map_page,
 		.unmap_page = swiotlb_unmap_page,
 		.map_sg = octeon_dma_map_sg,
diff --git a/arch/mips/include/asm/dma-direct.h b/arch/mips/include/asm/dma-direct.h
new file mode 100644
index 000000000000..f32f15530aba
--- /dev/null
+++ b/arch/mips/include/asm/dma-direct.h
@@ -0,0 +1 @@
+#include <asm/dma-coherence.h>
diff --git a/arch/mips/include/asm/dma-mapping.h b/arch/mips/include/asm/dma-mapping.h
index 0d9418d264f9..886e75a383f2 100644
--- a/arch/mips/include/asm/dma-mapping.h
+++ b/arch/mips/include/asm/dma-mapping.h
@@ -17,16 +17,6 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return mips_dma_map_ops;
 }
 
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return false;
-
-	return addr + size <= *dev->dma_mask;
-}
-
-static inline void dma_mark_clean(void *addr, size_t size) {}
-
 #define arch_setup_dma_ops arch_setup_dma_ops
 static inline void arch_setup_dma_ops(struct device *dev, u64 dma_base,
 				      u64 size, const struct iommu_ops *iommu,
diff --git a/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h b/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h
index 9110988b92a1..138edf6b5b48 100644
--- a/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h
+++ b/arch/mips/include/asm/mach-cavium-octeon/dma-coherence.h
@@ -61,6 +61,14 @@ static inline void plat_post_dma_flush(struct device *dev)
 {
 }
 
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	if (!dev->dma_mask)
+		return false;
+
+	return addr + size - 1 <= *dev->dma_mask;
+}
+
 dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr);
 phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr);
 
diff --git a/arch/mips/include/asm/mach-generic/dma-coherence.h b/arch/mips/include/asm/mach-generic/dma-coherence.h
index 61addb1677e9..8ad7a40ca786 100644
--- a/arch/mips/include/asm/mach-generic/dma-coherence.h
+++ b/arch/mips/include/asm/mach-generic/dma-coherence.h
@@ -70,16 +70,4 @@ static inline void plat_post_dma_flush(struct device *dev)
 }
 #endif
 
-#ifdef CONFIG_SWIOTLB
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return paddr;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return daddr;
-}
-#endif
-
 #endif /* __ASM_MACH_GENERIC_DMA_COHERENCE_H */
diff --git a/arch/mips/include/asm/mach-loongson64/dma-coherence.h b/arch/mips/include/asm/mach-loongson64/dma-coherence.h
index 1602a9e9e8c2..b1b575f5c6c1 100644
--- a/arch/mips/include/asm/mach-loongson64/dma-coherence.h
+++ b/arch/mips/include/asm/mach-loongson64/dma-coherence.h
@@ -17,6 +17,14 @@
 
 struct device;
 
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	if (!dev->dma_mask)
+		return false;
+
+	return addr + size - 1 <= *dev->dma_mask;
+}
+
 extern dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr);
 extern phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr);
 static inline dma_addr_t plat_map_dma_mem(struct device *dev, void *addr,
diff --git a/arch/mips/include/asm/netlogic/common.h b/arch/mips/include/asm/netlogic/common.h
index a6e6cbebe046..57616649b4f3 100644
--- a/arch/mips/include/asm/netlogic/common.h
+++ b/arch/mips/include/asm/netlogic/common.h
@@ -87,9 +87,6 @@ unsigned int nlm_get_cpu_frequency(void);
 extern const struct plat_smp_ops nlm_smp_ops;
 extern char nlm_reset_entry[], nlm_reset_entry_end[];
 
-/* SWIOTLB */
-extern const struct dma_map_ops nlm_swiotlb_dma_ops;
-
 extern unsigned int nlm_threads_per_core;
 extern cpumask_t nlm_cpumask;
 
diff --git a/arch/mips/loongson64/Kconfig b/arch/mips/loongson64/Kconfig
index 0d249fc3cfe9..6f109bb54cdb 100644
--- a/arch/mips/loongson64/Kconfig
+++ b/arch/mips/loongson64/Kconfig
@@ -136,6 +136,7 @@ config SWIOTLB
 	bool "Soft IOMMU Support for All-Memory DMA"
 	default y
 	depends on CPU_LOONGSON3
+	select DMA_DIRECT_OPS
 	select IOMMU_HELPER
 	select NEED_SG_DMA_LENGTH
 	select NEED_DMA_MAP_STATE
diff --git a/arch/mips/loongson64/common/dma-swiotlb.c b/arch/mips/loongson64/common/dma-swiotlb.c
index ef07740cee61..7bbcf89475f3 100644
--- a/arch/mips/loongson64/common/dma-swiotlb.c
+++ b/arch/mips/loongson64/common/dma-swiotlb.c
@@ -13,32 +13,12 @@
 static void *loongson_dma_alloc_coherent(struct device *dev, size_t size,
 		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
-	void *ret;
+	void *ret = swiotlb_alloc(dev, size, dma_handle, gfp, attrs);
 
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM);
-
-	if ((IS_ENABLED(CONFIG_ISA) && dev == NULL) ||
-	    (IS_ENABLED(CONFIG_ZONE_DMA) &&
-	     dev->coherent_dma_mask < DMA_BIT_MASK(32)))
-		gfp |= __GFP_DMA;
-	else if (IS_ENABLED(CONFIG_ZONE_DMA32) &&
-		 dev->coherent_dma_mask < DMA_BIT_MASK(40))
-		gfp |= __GFP_DMA32;
-
-	gfp |= __GFP_NORETRY;
-
-	ret = swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
 	mb();
 	return ret;
 }
 
-static void loongson_dma_free_coherent(struct device *dev, size_t size,
-		void *vaddr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_handle);
-}
-
 static dma_addr_t loongson_dma_map_page(struct device *dev, struct page *page,
 				unsigned long offset, size_t size,
 				enum dma_data_direction dir,
@@ -109,7 +89,7 @@ phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
 
 static const struct dma_map_ops loongson_dma_map_ops = {
 	.alloc = loongson_dma_alloc_coherent,
-	.free = loongson_dma_free_coherent,
+	.free = swiotlb_free,
 	.map_page = loongson_dma_map_page,
 	.unmap_page = swiotlb_unmap_page,
 	.map_sg = loongson_dma_map_sg,
diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c
index e3e94d05f0fd..237532e89919 100644
--- a/arch/mips/mm/dma-default.c
+++ b/arch/mips/mm/dma-default.c
@@ -93,9 +93,6 @@ static gfp_t massage_gfp_flags(const struct device *dev, gfp_t gfp)
 {
 	gfp_t dma_flag;
 
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM);
-
 #ifdef CONFIG_ISA
 	if (dev == NULL)
 		dma_flag = __GFP_DMA;
diff --git a/arch/mips/netlogic/Kconfig b/arch/mips/netlogic/Kconfig
index 8296b13affd2..7fcfc7fe9f14 100644
--- a/arch/mips/netlogic/Kconfig
+++ b/arch/mips/netlogic/Kconfig
@@ -89,9 +89,4 @@ config IOMMU_HELPER
 config NEED_SG_DMA_LENGTH
 	bool
 
-config SWIOTLB
-	def_bool y
-	select NEED_SG_DMA_LENGTH
-	select IOMMU_HELPER
-
 endif
diff --git a/arch/mips/netlogic/common/Makefile b/arch/mips/netlogic/common/Makefile
index 60d00b5d748e..89f6e3f39fed 100644
--- a/arch/mips/netlogic/common/Makefile
+++ b/arch/mips/netlogic/common/Makefile
@@ -1,6 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
 obj-y				+= irq.o time.o
-obj-y				+= nlm-dma.o
 obj-y				+= reset.o
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o
 obj-$(CONFIG_EARLY_PRINTK)	+= earlycons.o
diff --git a/arch/mips/netlogic/common/nlm-dma.c b/arch/mips/netlogic/common/nlm-dma.c
deleted file mode 100644
index 0ec9d9da6d51..000000000000
--- a/arch/mips/netlogic/common/nlm-dma.c
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
-*  Copyright (C) 2003-2013 Broadcom Corporation
-*  All Rights Reserved
- *
- * This software is available to you under a choice of one of two
- * licenses.  You may choose to be licensed under the terms of the GNU
- * General Public License (GPL) Version 2, available from the file
- * COPYING in the main directory of this source tree, or the Broadcom
- * license below:
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *
- * 1. Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- * 2. Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in
- *    the documentation and/or other materials provided with the
- *    distribution.
- *
- * THIS SOFTWARE IS PROVIDED BY BROADCOM ``AS IS'' AND ANY EXPRESS OR
- * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
- * ARE DISCLAIMED. IN NO EVENT SHALL BROADCOM OR CONTRIBUTORS BE LIABLE
- * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
- * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
- * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
- * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
- * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE
- * OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN
- * IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- */
-#include <linux/dma-mapping.h>
-#include <linux/scatterlist.h>
-#include <linux/bootmem.h>
-#include <linux/export.h>
-#include <linux/swiotlb.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-
-#include <asm/bootinfo.h>
-
-static char *nlm_swiotlb;
-
-static void *nlm_dma_alloc_coherent(struct device *dev, size_t size,
-	dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
-{
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM);
-
-#ifdef CONFIG_ZONE_DMA32
-	if (dev->coherent_dma_mask <= DMA_BIT_MASK(32))
-		gfp |= __GFP_DMA32;
-#endif
-
-	/* Don't invoke OOM killer */
-	gfp |= __GFP_NORETRY;
-
-	return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
-}
-
-static void nlm_dma_free_coherent(struct device *dev, size_t size,
-	void *vaddr, dma_addr_t dma_handle, unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_handle);
-}
-
-const struct dma_map_ops nlm_swiotlb_dma_ops = {
-	.alloc = nlm_dma_alloc_coherent,
-	.free = nlm_dma_free_coherent,
-	.map_page = swiotlb_map_page,
-	.unmap_page = swiotlb_unmap_page,
-	.map_sg = swiotlb_map_sg_attrs,
-	.unmap_sg = swiotlb_unmap_sg_attrs,
-	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
-	.sync_single_for_device = swiotlb_sync_single_for_device,
-	.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
-	.sync_sg_for_device = swiotlb_sync_sg_for_device,
-	.mapping_error = swiotlb_dma_mapping_error,
-	.dma_supported = swiotlb_dma_supported
-};
-
-void __init plat_swiotlb_setup(void)
-{
-	size_t swiotlbsize;
-	unsigned long swiotlb_nslabs;
-
-	swiotlbsize = 1 << 20; /* 1 MB for now */
-	swiotlb_nslabs = swiotlbsize >> IO_TLB_SHIFT;
-	swiotlb_nslabs = ALIGN(swiotlb_nslabs, IO_TLB_SEGSIZE);
-	swiotlbsize = swiotlb_nslabs << IO_TLB_SHIFT;
-
-	nlm_swiotlb = alloc_bootmem_low_pages(swiotlbsize);
-	swiotlb_init_with_tbl(nlm_swiotlb, swiotlb_nslabs, 1);
-}
diff --git a/arch/mn10300/mm/dma-alloc.c b/arch/mn10300/mm/dma-alloc.c
index 86108d2496b3..e3910d4db102 100644
--- a/arch/mn10300/mm/dma-alloc.c
+++ b/arch/mn10300/mm/dma-alloc.c
@@ -37,9 +37,6 @@ static void *mn10300_dma_alloc(struct device *dev, size_t size,
 		goto done;
 	}
 
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
 	if (dev == NULL || dev->coherent_dma_mask < 0xffffffff)
 		gfp |= GFP_DMA;
 
diff --git a/arch/nios2/mm/dma-mapping.c b/arch/nios2/mm/dma-mapping.c
index 7040c1adbb5e..4be815519dd4 100644
--- a/arch/nios2/mm/dma-mapping.c
+++ b/arch/nios2/mm/dma-mapping.c
@@ -63,9 +63,6 @@ static void *nios2_dma_alloc(struct device *dev, size_t size,
 {
 	void *ret;
 
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
 	/* optimized page clearing */
 	gfp |= __GFP_ZERO;
 
diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c
index c0dfd892f70c..91bc0cac03a1 100644
--- a/arch/parisc/kernel/pci-dma.c
+++ b/arch/parisc/kernel/pci-dma.c
@@ -75,11 +75,6 @@ void dump_resmap(void)
 static inline void dump_resmap(void) {;}
 #endif
 
-static int pa11_dma_supported( struct device *dev, u64 mask)
-{
-	return 1;
-}
-
 static inline int map_pte_uncached(pte_t * pte,
 		unsigned long vaddr,
 		unsigned long size, unsigned long *paddr_ptr)
@@ -579,7 +574,6 @@ static void pa11_dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 }
 
 const struct dma_map_ops pcxl_dma_ops = {
-	.dma_supported =	pa11_dma_supported,
 	.alloc =		pa11_dma_alloc,
 	.free =			pa11_dma_free,
 	.map_page =		pa11_dma_map_page,
@@ -616,7 +610,6 @@ static void pcx_dma_free(struct device *dev, size_t size, void *vaddr,
 }
 
 const struct dma_map_ops pcx_dma_ops = {
-	.dma_supported =	pa11_dma_supported,
 	.alloc =		pcx_dma_alloc,
 	.free =			pcx_dma_free,
 	.map_page =		pa11_dma_map_page,
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 2ed525a44734..e92432ae9737 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -139,6 +139,7 @@ config PPC
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
+	select ARCH_HAS_PHYS_TO_DMA
 	select ARCH_HAS_PMEM_API                if PPC64
 	select ARCH_HAS_SCALED_CPUTIME		if VIRT_CPU_ACCOUNTING_NATIVE
 	select ARCH_HAS_SG_CHAIN
diff --git a/arch/powerpc/include/asm/dma-direct.h b/arch/powerpc/include/asm/dma-direct.h
new file mode 100644
index 000000000000..a5b59c765426
--- /dev/null
+++ b/arch/powerpc/include/asm/dma-direct.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef ASM_POWERPC_DMA_DIRECT_H
+#define ASM_POWERPC_DMA_DIRECT_H 1
+
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+#ifdef CONFIG_SWIOTLB
+	struct dev_archdata *sd = &dev->archdata;
+
+	if (sd->max_direct_dma_addr && addr + size > sd->max_direct_dma_addr)
+		return false;
+#endif
+
+	if (!dev->dma_mask)
+		return false;
+
+	return addr + size - 1 <= *dev->dma_mask;
+}
+
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	return paddr + get_dma_offset(dev);
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return daddr - get_dma_offset(dev);
+}
+#endif /* ASM_POWERPC_DMA_DIRECT_H */
diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h
index 5a6cbe11db6f..8fa394520af6 100644
--- a/arch/powerpc/include/asm/dma-mapping.h
+++ b/arch/powerpc/include/asm/dma-mapping.h
@@ -19,13 +19,13 @@
 #include <asm/swiotlb.h>
 
 /* Some dma direct funcs must be visible for use in other dma_ops */
-extern void *__dma_direct_alloc_coherent(struct device *dev, size_t size,
+extern void *__dma_nommu_alloc_coherent(struct device *dev, size_t size,
 					 dma_addr_t *dma_handle, gfp_t flag,
 					 unsigned long attrs);
-extern void __dma_direct_free_coherent(struct device *dev, size_t size,
+extern void __dma_nommu_free_coherent(struct device *dev, size_t size,
 				       void *vaddr, dma_addr_t dma_handle,
 				       unsigned long attrs);
-extern int dma_direct_mmap_coherent(struct device *dev,
+extern int dma_nommu_mmap_coherent(struct device *dev,
 				    struct vm_area_struct *vma,
 				    void *cpu_addr, dma_addr_t handle,
 				    size_t size, unsigned long attrs);
@@ -73,7 +73,7 @@ static inline unsigned long device_to_mask(struct device *dev)
 #ifdef CONFIG_PPC64
 extern struct dma_map_ops dma_iommu_ops;
 #endif
-extern const struct dma_map_ops dma_direct_ops;
+extern const struct dma_map_ops dma_nommu_ops;
 
 static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 {
@@ -107,39 +107,11 @@ static inline void set_dma_offset(struct device *dev, dma_addr_t off)
 		dev->archdata.dma_offset = off;
 }
 
-/* this will be removed soon */
-#define flush_write_buffers()
-
 #define HAVE_ARCH_DMA_SET_MASK 1
 extern int dma_set_mask(struct device *dev, u64 dma_mask);
 
 extern u64 __dma_get_required_mask(struct device *dev);
 
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-#ifdef CONFIG_SWIOTLB
-	struct dev_archdata *sd = &dev->archdata;
-
-	if (sd->max_direct_dma_addr && addr + size > sd->max_direct_dma_addr)
-		return false;
-#endif
-
-	if (!dev->dma_mask)
-		return false;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return paddr + get_dma_offset(dev);
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return daddr - get_dma_offset(dev);
-}
-
 #define ARCH_HAS_DMA_MMAP_COHERENT
 
 #endif /* __KERNEL__ */
diff --git a/arch/powerpc/include/asm/swiotlb.h b/arch/powerpc/include/asm/swiotlb.h
index 01d45a5fd00b..f65ecf57b66c 100644
--- a/arch/powerpc/include/asm/swiotlb.h
+++ b/arch/powerpc/include/asm/swiotlb.h
@@ -13,9 +13,7 @@
 
 #include <linux/swiotlb.h>
 
-extern const struct dma_map_ops swiotlb_dma_ops;
-
-static inline void dma_mark_clean(void *addr, size_t size) {}
+extern const struct dma_map_ops powerpc_swiotlb_dma_ops;
 
 extern unsigned int ppc_swiotlb_enable;
 int __init swiotlb_setup_bus_notifier(void);
diff --git a/arch/powerpc/kernel/dma-iommu.c b/arch/powerpc/kernel/dma-iommu.c
index 66f33e7f8d40..f9fe2080ceb9 100644
--- a/arch/powerpc/kernel/dma-iommu.c
+++ b/arch/powerpc/kernel/dma-iommu.c
@@ -114,7 +114,7 @@ int dma_iommu_mapping_error(struct device *dev, dma_addr_t dma_addr)
 struct dma_map_ops dma_iommu_ops = {
 	.alloc			= dma_iommu_alloc_coherent,
 	.free			= dma_iommu_free_coherent,
-	.mmap			= dma_direct_mmap_coherent,
+	.mmap			= dma_nommu_mmap_coherent,
 	.map_sg			= dma_iommu_map_sg,
 	.unmap_sg		= dma_iommu_unmap_sg,
 	.dma_supported		= dma_iommu_dma_supported,
diff --git a/arch/powerpc/kernel/dma-swiotlb.c b/arch/powerpc/kernel/dma-swiotlb.c
index d0ea7860e02b..88f3963ca30f 100644
--- a/arch/powerpc/kernel/dma-swiotlb.c
+++ b/arch/powerpc/kernel/dma-swiotlb.c
@@ -46,10 +46,10 @@ static u64 swiotlb_powerpc_get_required(struct device *dev)
  * map_page, and unmap_page on highmem, use normal dma_ops
  * for everything else.
  */
-const struct dma_map_ops swiotlb_dma_ops = {
-	.alloc = __dma_direct_alloc_coherent,
-	.free = __dma_direct_free_coherent,
-	.mmap = dma_direct_mmap_coherent,
+const struct dma_map_ops powerpc_swiotlb_dma_ops = {
+	.alloc = __dma_nommu_alloc_coherent,
+	.free = __dma_nommu_free_coherent,
+	.mmap = dma_nommu_mmap_coherent,
 	.map_sg = swiotlb_map_sg_attrs,
 	.unmap_sg = swiotlb_unmap_sg_attrs,
 	.dma_supported = swiotlb_dma_supported,
@@ -89,7 +89,7 @@ static int ppc_swiotlb_bus_notify(struct notifier_block *nb,
 
 	/* May need to bounce if the device can't address all of DRAM */
 	if ((dma_get_mask(dev) + 1) < memblock_end_of_DRAM())
-		set_dma_ops(dev, &swiotlb_dma_ops);
+		set_dma_ops(dev, &powerpc_swiotlb_dma_ops);
 
 	return NOTIFY_DONE;
 }
@@ -121,7 +121,7 @@ static int __init check_swiotlb_enabled(void)
 	if (ppc_swiotlb_enable)
 		swiotlb_print_info();
 	else
-		swiotlb_free();
+		swiotlb_exit();
 
 	return 0;
 }
diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c
index 4194bbbbdb10..da20569de9d4 100644
--- a/arch/powerpc/kernel/dma.c
+++ b/arch/powerpc/kernel/dma.c
@@ -33,14 +33,14 @@ static u64 __maybe_unused get_pfn_limit(struct device *dev)
 	struct dev_archdata __maybe_unused *sd = &dev->archdata;
 
 #ifdef CONFIG_SWIOTLB
-	if (sd->max_direct_dma_addr && dev->dma_ops == &swiotlb_dma_ops)
+	if (sd->max_direct_dma_addr && dev->dma_ops == &powerpc_swiotlb_dma_ops)
 		pfn = min_t(u64, pfn, sd->max_direct_dma_addr >> PAGE_SHIFT);
 #endif
 
 	return pfn;
 }
 
-static int dma_direct_dma_supported(struct device *dev, u64 mask)
+static int dma_nommu_dma_supported(struct device *dev, u64 mask)
 {
 #ifdef CONFIG_PPC64
 	u64 limit = get_dma_offset(dev) + (memblock_end_of_DRAM() - 1);
@@ -62,7 +62,7 @@ static int dma_direct_dma_supported(struct device *dev, u64 mask)
 #endif
 }
 
-void *__dma_direct_alloc_coherent(struct device *dev, size_t size,
+void *__dma_nommu_alloc_coherent(struct device *dev, size_t size,
 				  dma_addr_t *dma_handle, gfp_t flag,
 				  unsigned long attrs)
 {
@@ -105,9 +105,6 @@ void *__dma_direct_alloc_coherent(struct device *dev, size_t size,
 	};
 #endif /* CONFIG_FSL_SOC */
 
-	/* ignore region specifiers */
-	flag  &= ~(__GFP_HIGHMEM);
-
 	page = alloc_pages_node(node, flag, get_order(size));
 	if (page == NULL)
 		return NULL;
@@ -119,7 +116,7 @@ void *__dma_direct_alloc_coherent(struct device *dev, size_t size,
 #endif
 }
 
-void __dma_direct_free_coherent(struct device *dev, size_t size,
+void __dma_nommu_free_coherent(struct device *dev, size_t size,
 				void *vaddr, dma_addr_t dma_handle,
 				unsigned long attrs)
 {
@@ -130,7 +127,7 @@ void __dma_direct_free_coherent(struct device *dev, size_t size,
 #endif
 }
 
-static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
+static void *dma_nommu_alloc_coherent(struct device *dev, size_t size,
 				       dma_addr_t *dma_handle, gfp_t flag,
 				       unsigned long attrs)
 {
@@ -139,8 +136,8 @@ static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
 	/* The coherent mask may be smaller than the real mask, check if
 	 * we can really use the direct ops
 	 */
-	if (dma_direct_dma_supported(dev, dev->coherent_dma_mask))
-		return __dma_direct_alloc_coherent(dev, size, dma_handle,
+	if (dma_nommu_dma_supported(dev, dev->coherent_dma_mask))
+		return __dma_nommu_alloc_coherent(dev, size, dma_handle,
 						   flag, attrs);
 
 	/* Ok we can't ... do we have an iommu ? If not, fail */
@@ -154,15 +151,15 @@ static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
 				    dev_to_node(dev));
 }
 
-static void dma_direct_free_coherent(struct device *dev, size_t size,
+static void dma_nommu_free_coherent(struct device *dev, size_t size,
 				     void *vaddr, dma_addr_t dma_handle,
 				     unsigned long attrs)
 {
 	struct iommu_table *iommu;
 
-	/* See comments in dma_direct_alloc_coherent() */
-	if (dma_direct_dma_supported(dev, dev->coherent_dma_mask))
-		return __dma_direct_free_coherent(dev, size, vaddr, dma_handle,
+	/* See comments in dma_nommu_alloc_coherent() */
+	if (dma_nommu_dma_supported(dev, dev->coherent_dma_mask))
+		return __dma_nommu_free_coherent(dev, size, vaddr, dma_handle,
 						  attrs);
 	/* Maybe we used an iommu ... */
 	iommu = get_iommu_table_base(dev);
@@ -175,7 +172,7 @@ static void dma_direct_free_coherent(struct device *dev, size_t size,
 	iommu_free_coherent(iommu, size, vaddr, dma_handle);
 }
 
-int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
+int dma_nommu_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
 			     void *cpu_addr, dma_addr_t handle, size_t size,
 			     unsigned long attrs)
 {
@@ -193,7 +190,7 @@ int dma_direct_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
 			       vma->vm_page_prot);
 }
 
-static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
+static int dma_nommu_map_sg(struct device *dev, struct scatterlist *sgl,
 			     int nents, enum dma_data_direction direction,
 			     unsigned long attrs)
 {
@@ -213,13 +210,13 @@ static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
 	return nents;
 }
 
-static void dma_direct_unmap_sg(struct device *dev, struct scatterlist *sg,
+static void dma_nommu_unmap_sg(struct device *dev, struct scatterlist *sg,
 				int nents, enum dma_data_direction direction,
 				unsigned long attrs)
 {
 }
 
-static u64 dma_direct_get_required_mask(struct device *dev)
+static u64 dma_nommu_get_required_mask(struct device *dev)
 {
 	u64 end, mask;
 
@@ -231,7 +228,7 @@ static u64 dma_direct_get_required_mask(struct device *dev)
 	return mask;
 }
 
-static inline dma_addr_t dma_direct_map_page(struct device *dev,
+static inline dma_addr_t dma_nommu_map_page(struct device *dev,
 					     struct page *page,
 					     unsigned long offset,
 					     size_t size,
@@ -246,7 +243,7 @@ static inline dma_addr_t dma_direct_map_page(struct device *dev,
 	return page_to_phys(page) + offset + get_dma_offset(dev);
 }
 
-static inline void dma_direct_unmap_page(struct device *dev,
+static inline void dma_nommu_unmap_page(struct device *dev,
 					 dma_addr_t dma_address,
 					 size_t size,
 					 enum dma_data_direction direction,
@@ -255,7 +252,7 @@ static inline void dma_direct_unmap_page(struct device *dev,
 }
 
 #ifdef CONFIG_NOT_COHERENT_CACHE
-static inline void dma_direct_sync_sg(struct device *dev,
+static inline void dma_nommu_sync_sg(struct device *dev,
 		struct scatterlist *sgl, int nents,
 		enum dma_data_direction direction)
 {
@@ -266,7 +263,7 @@ static inline void dma_direct_sync_sg(struct device *dev,
 		__dma_sync_page(sg_page(sg), sg->offset, sg->length, direction);
 }
 
-static inline void dma_direct_sync_single(struct device *dev,
+static inline void dma_nommu_sync_single(struct device *dev,
 					  dma_addr_t dma_handle, size_t size,
 					  enum dma_data_direction direction)
 {
@@ -274,24 +271,24 @@ static inline void dma_direct_sync_single(struct device *dev,
 }
 #endif
 
-const struct dma_map_ops dma_direct_ops = {
-	.alloc				= dma_direct_alloc_coherent,
-	.free				= dma_direct_free_coherent,
-	.mmap				= dma_direct_mmap_coherent,
-	.map_sg				= dma_direct_map_sg,
-	.unmap_sg			= dma_direct_unmap_sg,
-	.dma_supported			= dma_direct_dma_supported,
-	.map_page			= dma_direct_map_page,
-	.unmap_page			= dma_direct_unmap_page,
-	.get_required_mask		= dma_direct_get_required_mask,
+const struct dma_map_ops dma_nommu_ops = {
+	.alloc				= dma_nommu_alloc_coherent,
+	.free				= dma_nommu_free_coherent,
+	.mmap				= dma_nommu_mmap_coherent,
+	.map_sg				= dma_nommu_map_sg,
+	.unmap_sg			= dma_nommu_unmap_sg,
+	.dma_supported			= dma_nommu_dma_supported,
+	.map_page			= dma_nommu_map_page,
+	.unmap_page			= dma_nommu_unmap_page,
+	.get_required_mask		= dma_nommu_get_required_mask,
 #ifdef CONFIG_NOT_COHERENT_CACHE
-	.sync_single_for_cpu 		= dma_direct_sync_single,
-	.sync_single_for_device 	= dma_direct_sync_single,
-	.sync_sg_for_cpu 		= dma_direct_sync_sg,
-	.sync_sg_for_device 		= dma_direct_sync_sg,
+	.sync_single_for_cpu 		= dma_nommu_sync_single,
+	.sync_single_for_device 	= dma_nommu_sync_single,
+	.sync_sg_for_cpu 		= dma_nommu_sync_sg,
+	.sync_sg_for_device 		= dma_nommu_sync_sg,
 #endif
 };
-EXPORT_SYMBOL(dma_direct_ops);
+EXPORT_SYMBOL(dma_nommu_ops);
 
 int dma_set_coherent_mask(struct device *dev, u64 mask)
 {
@@ -302,7 +299,7 @@ int dma_set_coherent_mask(struct device *dev, u64 mask)
 		 * is no dma_op->set_coherent_mask() so we have to do
 		 * things the hard way:
 		 */
-		if (get_dma_ops(dev) != &dma_direct_ops ||
+		if (get_dma_ops(dev) != &dma_nommu_ops ||
 		    get_iommu_table_base(dev) == NULL ||
 		    !dma_iommu_dma_supported(dev, mask))
 			return -EIO;
diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c
index 0ac7aa346c69..590f4d0a6cb1 100644
--- a/arch/powerpc/kernel/pci-common.c
+++ b/arch/powerpc/kernel/pci-common.c
@@ -60,7 +60,7 @@ resource_size_t isa_mem_base;
 EXPORT_SYMBOL(isa_mem_base);
 
 
-static const struct dma_map_ops *pci_dma_ops = &dma_direct_ops;
+static const struct dma_map_ops *pci_dma_ops = &dma_nommu_ops;
 
 void set_pci_dma_ops(const struct dma_map_ops *dma_ops)
 {
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index 8fd3a70047f1..3f33869c6486 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -780,7 +780,7 @@ void arch_setup_pdev_archdata(struct platform_device *pdev)
 {
 	pdev->archdata.dma_mask = DMA_BIT_MASK(32);
 	pdev->dev.dma_mask = &pdev->archdata.dma_mask;
- 	set_dma_ops(&pdev->dev, &dma_direct_ops);
+ 	set_dma_ops(&pdev->dev, &dma_nommu_ops);
 }
 
 static __init void print_system_info(void)
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index 4b91ad08eefd..12352a58072a 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -541,7 +541,7 @@ static struct cbe_iommu *cell_iommu_for_node(int nid)
 	return NULL;
 }
 
-static unsigned long cell_dma_direct_offset;
+static unsigned long cell_dma_nommu_offset;
 
 static unsigned long dma_iommu_fixed_base;
 
@@ -580,7 +580,7 @@ static void *dma_fixed_alloc_coherent(struct device *dev, size_t size,
 					    device_to_mask(dev), flag,
 					    dev_to_node(dev));
 	else
-		return dma_direct_ops.alloc(dev, size, dma_handle, flag,
+		return dma_nommu_ops.alloc(dev, size, dma_handle, flag,
 					    attrs);
 }
 
@@ -592,7 +592,7 @@ static void dma_fixed_free_coherent(struct device *dev, size_t size,
 		iommu_free_coherent(cell_get_iommu_table(dev), size, vaddr,
 				    dma_handle);
 	else
-		dma_direct_ops.free(dev, size, vaddr, dma_handle, attrs);
+		dma_nommu_ops.free(dev, size, vaddr, dma_handle, attrs);
 }
 
 static dma_addr_t dma_fixed_map_page(struct device *dev, struct page *page,
@@ -601,7 +601,7 @@ static dma_addr_t dma_fixed_map_page(struct device *dev, struct page *page,
 				     unsigned long attrs)
 {
 	if (iommu_fixed_is_weak == (attrs & DMA_ATTR_WEAK_ORDERING))
-		return dma_direct_ops.map_page(dev, page, offset, size,
+		return dma_nommu_ops.map_page(dev, page, offset, size,
 					       direction, attrs);
 	else
 		return iommu_map_page(dev, cell_get_iommu_table(dev), page,
@@ -614,7 +614,7 @@ static void dma_fixed_unmap_page(struct device *dev, dma_addr_t dma_addr,
 				 unsigned long attrs)
 {
 	if (iommu_fixed_is_weak == (attrs & DMA_ATTR_WEAK_ORDERING))
-		dma_direct_ops.unmap_page(dev, dma_addr, size, direction,
+		dma_nommu_ops.unmap_page(dev, dma_addr, size, direction,
 					  attrs);
 	else
 		iommu_unmap_page(cell_get_iommu_table(dev), dma_addr, size,
@@ -626,7 +626,7 @@ static int dma_fixed_map_sg(struct device *dev, struct scatterlist *sg,
 			   unsigned long attrs)
 {
 	if (iommu_fixed_is_weak == (attrs & DMA_ATTR_WEAK_ORDERING))
-		return dma_direct_ops.map_sg(dev, sg, nents, direction, attrs);
+		return dma_nommu_ops.map_sg(dev, sg, nents, direction, attrs);
 	else
 		return ppc_iommu_map_sg(dev, cell_get_iommu_table(dev), sg,
 					nents, device_to_mask(dev),
@@ -638,7 +638,7 @@ static void dma_fixed_unmap_sg(struct device *dev, struct scatterlist *sg,
 			       unsigned long attrs)
 {
 	if (iommu_fixed_is_weak == (attrs & DMA_ATTR_WEAK_ORDERING))
-		dma_direct_ops.unmap_sg(dev, sg, nents, direction, attrs);
+		dma_nommu_ops.unmap_sg(dev, sg, nents, direction, attrs);
 	else
 		ppc_iommu_unmap_sg(cell_get_iommu_table(dev), sg, nents,
 				   direction, attrs);
@@ -661,8 +661,8 @@ static void cell_dma_dev_setup(struct device *dev)
 {
 	if (get_pci_dma_ops() == &dma_iommu_ops)
 		set_iommu_table_base(dev, cell_get_iommu_table(dev));
-	else if (get_pci_dma_ops() == &dma_direct_ops)
-		set_dma_offset(dev, cell_dma_direct_offset);
+	else if (get_pci_dma_ops() == &dma_nommu_ops)
+		set_dma_offset(dev, cell_dma_nommu_offset);
 	else
 		BUG();
 }
@@ -810,14 +810,14 @@ static int __init cell_iommu_init_disabled(void)
 	unsigned long base = 0, size;
 
 	/* When no iommu is present, we use direct DMA ops */
-	set_pci_dma_ops(&dma_direct_ops);
+	set_pci_dma_ops(&dma_nommu_ops);
 
 	/* First make sure all IOC translation is turned off */
 	cell_disable_iommus();
 
 	/* If we have no Axon, we set up the spider DMA magic offset */
 	if (of_find_node_by_name(NULL, "axon") == NULL)
-		cell_dma_direct_offset = SPIDER_DMA_OFFSET;
+		cell_dma_nommu_offset = SPIDER_DMA_OFFSET;
 
 	/* Now we need to check to see where the memory is mapped
 	 * in PCI space. We assume that all busses use the same dma
@@ -851,13 +851,13 @@ static int __init cell_iommu_init_disabled(void)
 		return -ENODEV;
 	}
 
-	cell_dma_direct_offset += base;
+	cell_dma_nommu_offset += base;
 
-	if (cell_dma_direct_offset != 0)
+	if (cell_dma_nommu_offset != 0)
 		cell_pci_controller_ops.dma_dev_setup = cell_pci_dma_dev_setup;
 
 	printk("iommu: disabled, direct DMA offset is 0x%lx\n",
-	       cell_dma_direct_offset);
+	       cell_dma_nommu_offset);
 
 	return 0;
 }
diff --git a/arch/powerpc/platforms/pasemi/iommu.c b/arch/powerpc/platforms/pasemi/iommu.c
index 7fec04de27fc..78b80cbd9768 100644
--- a/arch/powerpc/platforms/pasemi/iommu.c
+++ b/arch/powerpc/platforms/pasemi/iommu.c
@@ -186,7 +186,7 @@ static void pci_dma_dev_setup_pasemi(struct pci_dev *dev)
 	 */
 	if (dev->vendor == 0x1959 && dev->device == 0xa007 &&
 	    !firmware_has_feature(FW_FEATURE_LPAR)) {
-		dev->dev.dma_ops = &dma_direct_ops;
+		dev->dev.dma_ops = &dma_nommu_ops;
 		/*
 		 * Set the coherent DMA mask to prevent the iommu
 		 * being used unnecessarily
diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c
index c4a3e93dc324..d0b8ae53660d 100644
--- a/arch/powerpc/platforms/pasemi/setup.c
+++ b/arch/powerpc/platforms/pasemi/setup.c
@@ -363,7 +363,7 @@ static int pcmcia_notify(struct notifier_block *nb, unsigned long action,
 		return 0;
 
 	/* We use the direct ops for localbus */
-	dev->dma_ops = &dma_direct_ops;
+	dev->dma_ops = &dma_nommu_ops;
 
 	return 0;
 }
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 749055553064..9582aeb1fe4c 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -1850,7 +1850,7 @@ static int pnv_pci_ioda_dma_set_mask(struct pci_dev *pdev, u64 dma_mask)
 
 	if (bypass) {
 		dev_info(&pdev->dev, "Using 64-bit DMA iommu bypass\n");
-		set_dma_ops(&pdev->dev, &dma_direct_ops);
+		set_dma_ops(&pdev->dev, &dma_nommu_ops);
 	} else {
 		/*
 		 * If the device can't set the TCE bypass bit but still wants
@@ -1868,7 +1868,7 @@ static int pnv_pci_ioda_dma_set_mask(struct pci_dev *pdev, u64 dma_mask)
 				return rc;
 			/* 4GB offset bypasses 32-bit space */
 			set_dma_offset(&pdev->dev, (1ULL << 32));
-			set_dma_ops(&pdev->dev, &dma_direct_ops);
+			set_dma_ops(&pdev->dev, &dma_nommu_ops);
 		} else if (dma_mask >> 32 && dma_mask != DMA_BIT_MASK(64)) {
 			/*
 			 * Fail the request if a DMA mask between 32 and 64 bits
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 69921f72e2da..eaa11334fc8c 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -1231,7 +1231,7 @@ static int dma_set_mask_pSeriesLP(struct device *dev, u64 dma_mask)
 			if (dma_offset != 0) {
 				dev_info(dev, "Using 64-bit direct DMA at offset %llx\n", dma_offset);
 				set_dma_offset(dev, dma_offset);
-				set_dma_ops(dev, &dma_direct_ops);
+				set_dma_ops(dev, &dma_nommu_ops);
 				ddw_enabled = true;
 			}
 		}
diff --git a/arch/powerpc/platforms/pseries/vio.c b/arch/powerpc/platforms/pseries/vio.c
index d86938260a86..49e04ec19238 100644
--- a/arch/powerpc/platforms/pseries/vio.c
+++ b/arch/powerpc/platforms/pseries/vio.c
@@ -618,7 +618,7 @@ static u64 vio_dma_get_required_mask(struct device *dev)
 static const struct dma_map_ops vio_dma_mapping_ops = {
 	.alloc             = vio_dma_iommu_alloc_coherent,
 	.free              = vio_dma_iommu_free_coherent,
-	.mmap		   = dma_direct_mmap_coherent,
+	.mmap		   = dma_nommu_mmap_coherent,
 	.map_sg            = vio_dma_iommu_map_sg,
 	.unmap_sg          = vio_dma_iommu_unmap_sg,
 	.map_page          = vio_dma_iommu_map_page,
diff --git a/arch/powerpc/sysdev/dart_iommu.c b/arch/powerpc/sysdev/dart_iommu.c
index 3573d54b2770..a6198d4f0f03 100644
--- a/arch/powerpc/sysdev/dart_iommu.c
+++ b/arch/powerpc/sysdev/dart_iommu.c
@@ -402,7 +402,7 @@ static int dart_dma_set_mask(struct device *dev, u64 dma_mask)
 	 */
 	if (dart_device_on_pcie(dev) && dma_mask >= DMA_BIT_MASK(40)) {
 		dev_info(dev, "Using 64-bit DMA iommu bypass\n");
-		set_dma_ops(dev, &dma_direct_ops);
+		set_dma_ops(dev, &dma_nommu_ops);
 	} else {
 		dev_info(dev, "Using 32-bit DMA via iommu\n");
 		set_dma_ops(dev, &dma_iommu_ops);
@@ -446,7 +446,7 @@ void __init iommu_init_early_dart(struct pci_controller_ops *controller_ops)
 	controller_ops->dma_bus_setup = NULL;
 
 	/* Setup pci_dma ops */
-	set_pci_dma_ops(&dma_direct_ops);
+	set_pci_dma_ops(&dma_nommu_ops);
 }
 
 #ifdef CONFIG_PM
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 22d98057f773..61e07c78d64f 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -118,7 +118,7 @@ static void setup_swiotlb_ops(struct pci_controller *hose)
 {
 	if (ppc_swiotlb_enable) {
 		hose->controller_ops.dma_dev_setup = pci_dma_dev_setup_swiotlb;
-		set_pci_dma_ops(&swiotlb_dma_ops);
+		set_pci_dma_ops(&powerpc_swiotlb_dma_ops);
 	}
 }
 #else
@@ -135,7 +135,7 @@ static int fsl_pci_dma_set_mask(struct device *dev, u64 dma_mask)
 	 * mapping that allows addressing any RAM address from across PCI.
 	 */
 	if (dev_is_pci(dev) && dma_mask >= pci64_dma_offset * 2 - 1) {
-		set_dma_ops(dev, &dma_direct_ops);
+		set_dma_ops(dev, &dma_nommu_ops);
 		set_dma_offset(dev, pci64_dma_offset);
 	}
 
diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 2c6adf12713a..865e14f50c14 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -83,7 +83,7 @@ config PGTABLE_LEVELS
 config HAVE_KPROBES
 	def_bool n
 
-config DMA_NOOP_OPS
+config DMA_DIRECT_OPS
 	def_bool y
 
 menu "Platform type"
diff --git a/arch/riscv/include/asm/Kbuild b/arch/riscv/include/asm/Kbuild
index 970460a0b492..197460ccbf21 100644
--- a/arch/riscv/include/asm/Kbuild
+++ b/arch/riscv/include/asm/Kbuild
@@ -7,6 +7,7 @@ generic-y += device.h
 generic-y += div64.h
 generic-y += dma.h
 generic-y += dma-contiguous.h
+generic-y += dma-mapping.h
 generic-y += emergency-restart.h
 generic-y += errno.h
 generic-y += exec.h
diff --git a/arch/riscv/include/asm/dma-mapping.h b/arch/riscv/include/asm/dma-mapping.h
deleted file mode 100644
index 3eec1000196d..000000000000
--- a/arch/riscv/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,38 +0,0 @@
-/*
- * Copyright (C) 2003-2004 Hewlett-Packard Co
- *	David Mosberger-Tang <davidm@hpl.hp.com>
- * Copyright (C) 2012 ARM Ltd.
- * Copyright (C) 2016 SiFive, Inc.
- *
- * 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.
- *
- * This program is distributed in the hope that 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, see <http://www.gnu.org/licenses/>.
- */
-#ifndef __ASM_RISCV_DMA_MAPPING_H
-#define __ASM_RISCV_DMA_MAPPING_H
-
-/* Use ops->dma_mapping_error (if it exists) or assume success */
-// #undef DMA_ERROR_CODE
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return &dma_noop_ops;
-}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return false;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-#endif	/* __ASM_RISCV_DMA_MAPPING_H */
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 829c67986db7..9376637229c9 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -140,7 +140,7 @@ config S390
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_CONTIGUOUS
-	select DMA_NOOP_OPS
+	select DMA_DIRECT_OPS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_DYNAMIC_FTRACE_WITH_REGS
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS
diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild
index 048450869328..dade72be127b 100644
--- a/arch/s390/include/asm/Kbuild
+++ b/arch/s390/include/asm/Kbuild
@@ -4,6 +4,7 @@ generic-y += cacheflush.h
 generic-y += clkdev.h
 generic-y += device.h
 generic-y += dma-contiguous.h
+generic-y += dma-mapping.h
 generic-y += div64.h
 generic-y += emergency-restart.h
 generic-y += export.h
diff --git a/arch/s390/include/asm/dma-mapping.h b/arch/s390/include/asm/dma-mapping.h
deleted file mode 100644
index eaf490f9c5bc..000000000000
--- a/arch/s390/include/asm/dma-mapping.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ASM_S390_DMA_MAPPING_H
-#define _ASM_S390_DMA_MAPPING_H
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-#include <linux/dma-debug.h>
-#include <linux/io.h>
-
-extern const struct dma_map_ops s390_pci_dma_ops;
-
-static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
-{
-	return &dma_noop_ops;
-}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return false;
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-#endif /* _ASM_S390_DMA_MAPPING_H */
diff --git a/arch/s390/include/asm/pci_dma.h b/arch/s390/include/asm/pci_dma.h
index e8d9161fa17a..419fac7a62c0 100644
--- a/arch/s390/include/asm/pci_dma.h
+++ b/arch/s390/include/asm/pci_dma.h
@@ -201,4 +201,7 @@ void dma_cleanup_tables(unsigned long *);
 unsigned long *dma_walk_cpu_trans(unsigned long *rto, dma_addr_t dma_addr);
 void dma_update_cpu_trans(unsigned long *entry, void *page_addr, int flags);
 
+extern const struct dma_map_ops s390_pci_dma_ops;
+
+
 #endif
diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig
index 02f269cfa538..ef9d403cbbe4 100644
--- a/arch/tile/Kconfig
+++ b/arch/tile/Kconfig
@@ -249,7 +249,7 @@ config HIGHMEM
 
 	  If unsure, say "true".
 
-config ZONE_DMA
+config ZONE_DMA32
 	def_bool y
 
 config IOMMU_HELPER
@@ -261,6 +261,7 @@ config NEED_SG_DMA_LENGTH
 config SWIOTLB
 	bool
 	default TILEGX
+	select DMA_DIRECT_OPS
 	select IOMMU_HELPER
 	select NEED_SG_DMA_LENGTH
 	select ARCH_HAS_DMA_SET_COHERENT_MASK
diff --git a/arch/tile/include/asm/dma-mapping.h b/arch/tile/include/asm/dma-mapping.h
index 97ad62878290..d25fce101fc0 100644
--- a/arch/tile/include/asm/dma-mapping.h
+++ b/arch/tile/include/asm/dma-mapping.h
@@ -44,26 +44,6 @@ static inline void set_dma_offset(struct device *dev, dma_addr_t off)
 	dev->archdata.dma_offset = off;
 }
 
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return paddr;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return daddr;
-}
-
-static inline void dma_mark_clean(void *addr, size_t size) {}
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return 0;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
 #define HAVE_ARCH_DMA_SET_MASK 1
 int dma_set_mask(struct device *dev, u64 mask);
 
diff --git a/arch/tile/kernel/pci-dma.c b/arch/tile/kernel/pci-dma.c
index f2abedc8a080..6a1efe5543fa 100644
--- a/arch/tile/kernel/pci-dma.c
+++ b/arch/tile/kernel/pci-dma.c
@@ -54,7 +54,7 @@ static void *tile_dma_alloc_coherent(struct device *dev, size_t size,
 	 * which case we will return NULL.  But such devices are uncommon.
 	 */
 	if (dma_mask <= DMA_BIT_MASK(32)) {
-		gfp |= GFP_DMA;
+		gfp |= GFP_DMA32;
 		node = 0;
 	}
 
@@ -509,39 +509,9 @@ EXPORT_SYMBOL(gx_pci_dma_map_ops);
 /* PCI DMA mapping functions for legacy PCI devices */
 
 #ifdef CONFIG_SWIOTLB
-static void *tile_swiotlb_alloc_coherent(struct device *dev, size_t size,
-					 dma_addr_t *dma_handle, gfp_t gfp,
-					 unsigned long attrs)
-{
-	gfp |= GFP_DMA;
-	return swiotlb_alloc_coherent(dev, size, dma_handle, gfp);
-}
-
-static void tile_swiotlb_free_coherent(struct device *dev, size_t size,
-				       void *vaddr, dma_addr_t dma_addr,
-				       unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_addr);
-}
-
-static const struct dma_map_ops pci_swiotlb_dma_ops = {
-	.alloc = tile_swiotlb_alloc_coherent,
-	.free = tile_swiotlb_free_coherent,
-	.map_page = swiotlb_map_page,
-	.unmap_page = swiotlb_unmap_page,
-	.map_sg = swiotlb_map_sg_attrs,
-	.unmap_sg = swiotlb_unmap_sg_attrs,
-	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
-	.sync_single_for_device = swiotlb_sync_single_for_device,
-	.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
-	.sync_sg_for_device = swiotlb_sync_sg_for_device,
-	.dma_supported = swiotlb_dma_supported,
-	.mapping_error = swiotlb_dma_mapping_error,
-};
-
 static const struct dma_map_ops pci_hybrid_dma_ops = {
-	.alloc = tile_swiotlb_alloc_coherent,
-	.free = tile_swiotlb_free_coherent,
+	.alloc = swiotlb_alloc,
+	.free = swiotlb_free,
 	.map_page = tile_pci_dma_map_page,
 	.unmap_page = tile_pci_dma_unmap_page,
 	.map_sg = tile_pci_dma_map_sg,
@@ -552,7 +522,7 @@ static const struct dma_map_ops pci_hybrid_dma_ops = {
 	.sync_sg_for_device = tile_pci_dma_sync_sg_for_device,
 };
 
-const struct dma_map_ops *gx_legacy_pci_dma_map_ops = &pci_swiotlb_dma_ops;
+const struct dma_map_ops *gx_legacy_pci_dma_map_ops = &swiotlb_dma_ops;
 const struct dma_map_ops *gx_hybrid_pci_dma_map_ops = &pci_hybrid_dma_ops;
 #else
 const struct dma_map_ops *gx_legacy_pci_dma_map_ops;
diff --git a/arch/tile/kernel/setup.c b/arch/tile/kernel/setup.c
index ad83c1e66dbd..eb4e198f6f93 100644
--- a/arch/tile/kernel/setup.c
+++ b/arch/tile/kernel/setup.c
@@ -814,11 +814,11 @@ static void __init zone_sizes_init(void)
 #endif
 
 		if (start < dma_end) {
-			zones_size[ZONE_DMA] = min(zones_size[ZONE_NORMAL],
+			zones_size[ZONE_DMA32] = min(zones_size[ZONE_NORMAL],
 						   dma_end - start);
-			zones_size[ZONE_NORMAL] -= zones_size[ZONE_DMA];
+			zones_size[ZONE_NORMAL] -= zones_size[ZONE_DMA32];
 		} else {
-			zones_size[ZONE_DMA] = 0;
+			zones_size[ZONE_DMA32] = 0;
 		}
 
 		/* Take zone metadata from controller 0 if we're isolnode. */
@@ -830,7 +830,7 @@ static void __init zone_sizes_init(void)
 		       PFN_UP(node_percpu[i]));
 
 		/* Track the type of memory on each node */
-		if (zones_size[ZONE_NORMAL] || zones_size[ZONE_DMA])
+		if (zones_size[ZONE_NORMAL] || zones_size[ZONE_DMA32])
 			node_set_state(i, N_NORMAL_MEMORY);
 #ifdef CONFIG_HIGHMEM
 		if (end != start)
diff --git a/arch/unicore32/include/asm/dma-mapping.h b/arch/unicore32/include/asm/dma-mapping.h
index ac608c2f6af6..790bc2ef4af2 100644
--- a/arch/unicore32/include/asm/dma-mapping.h
+++ b/arch/unicore32/include/asm/dma-mapping.h
@@ -12,38 +12,11 @@
 #ifndef __UNICORE_DMA_MAPPING_H__
 #define __UNICORE_DMA_MAPPING_H__
 
-#ifdef __KERNEL__
-
-#include <linux/mm_types.h>
-#include <linux/scatterlist.h>
 #include <linux/swiotlb.h>
 
-extern const struct dma_map_ops swiotlb_dma_map_ops;
-
 static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 {
-	return &swiotlb_dma_map_ops;
+	return &swiotlb_dma_ops;
 }
 
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (dev && dev->dma_mask)
-		return addr + size - 1 <= *dev->dma_mask;
-
-	return 1;
-}
-
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return paddr;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return daddr;
-}
-
-static inline void dma_mark_clean(void *addr, size_t size) {}
-
-#endif /* __KERNEL__ */
 #endif
diff --git a/arch/unicore32/mm/Kconfig b/arch/unicore32/mm/Kconfig
index c256460cd363..e9154a59d561 100644
--- a/arch/unicore32/mm/Kconfig
+++ b/arch/unicore32/mm/Kconfig
@@ -42,6 +42,7 @@ config CPU_TLB_SINGLE_ENTRY_DISABLE
 
 config SWIOTLB
 	def_bool y
+	select DMA_DIRECT_OPS
 
 config IOMMU_HELPER
 	def_bool SWIOTLB
diff --git a/arch/unicore32/mm/Makefile b/arch/unicore32/mm/Makefile
index 681c0ef5ec9e..8106260583ab 100644
--- a/arch/unicore32/mm/Makefile
+++ b/arch/unicore32/mm/Makefile
@@ -6,8 +6,6 @@
 obj-y				:= extable.o fault.o init.o pgd.o mmu.o
 obj-y				+= flush.o ioremap.o
 
-obj-$(CONFIG_SWIOTLB)		+= dma-swiotlb.o
-
 obj-$(CONFIG_MODULES)		+= proc-syms.o
 
 obj-$(CONFIG_ALIGNMENT_TRAP)	+= alignment.o
diff --git a/arch/unicore32/mm/dma-swiotlb.c b/arch/unicore32/mm/dma-swiotlb.c
deleted file mode 100644
index 525413d6690e..000000000000
--- a/arch/unicore32/mm/dma-swiotlb.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * Contains routines needed to support swiotlb for UniCore32.
- *
- * Copyright (C) 2010 Guan Xuetao
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-#include <linux/pci.h>
-#include <linux/cache.h>
-#include <linux/module.h>
-#include <linux/dma-mapping.h>
-#include <linux/swiotlb.h>
-#include <linux/bootmem.h>
-
-#include <asm/dma.h>
-
-static void *unicore_swiotlb_alloc_coherent(struct device *dev, size_t size,
-					    dma_addr_t *dma_handle, gfp_t flags,
-					    unsigned long attrs)
-{
-	return swiotlb_alloc_coherent(dev, size, dma_handle, flags);
-}
-
-static void unicore_swiotlb_free_coherent(struct device *dev, size_t size,
-					  void *vaddr, dma_addr_t dma_addr,
-					  unsigned long attrs)
-{
-	swiotlb_free_coherent(dev, size, vaddr, dma_addr);
-}
-
-const struct dma_map_ops swiotlb_dma_map_ops = {
-	.alloc = unicore_swiotlb_alloc_coherent,
-	.free = unicore_swiotlb_free_coherent,
-	.map_sg = swiotlb_map_sg_attrs,
-	.unmap_sg = swiotlb_unmap_sg_attrs,
-	.dma_supported = swiotlb_dma_supported,
-	.map_page = swiotlb_map_page,
-	.unmap_page = swiotlb_unmap_page,
-	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
-	.sync_single_for_device = swiotlb_sync_single_for_device,
-	.sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
-	.sync_sg_for_device = swiotlb_sync_sg_for_device,
-	.mapping_error = swiotlb_dma_mapping_error,
-};
-EXPORT_SYMBOL(swiotlb_dma_map_ops);
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 423e4b64e683..dbe5542a6666 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -54,6 +54,7 @@ config X86
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_KCOV			if X86_64
+	select ARCH_HAS_PHYS_TO_DMA
 	select ARCH_HAS_PMEM_API		if X86_64
 	select ARCH_HAS_REFCOUNT
 	select ARCH_HAS_UACCESS_FLUSHCACHE	if X86_64
diff --git a/arch/x86/include/asm/dma-direct.h b/arch/x86/include/asm/dma-direct.h
new file mode 100644
index 000000000000..1295bc622ebe
--- /dev/null
+++ b/arch/x86/include/asm/dma-direct.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef ASM_X86_DMA_DIRECT_H
+#define ASM_X86_DMA_DIRECT_H 1
+
+#include <linux/mem_encrypt.h>
+
+#ifdef CONFIG_X86_DMA_REMAP /* Platform code defines bridge-specific code */
+bool dma_capable(struct device *dev, dma_addr_t addr, size_t size);
+dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr);
+phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr);
+#else
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	if (!dev->dma_mask)
+		return 0;
+
+	return addr + size - 1 <= *dev->dma_mask;
+}
+
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	return __sme_set(paddr);
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
+{
+	return __sme_clr(daddr);
+}
+#endif /* CONFIG_X86_DMA_REMAP */
+#endif /* ASM_X86_DMA_DIRECT_H */
diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h
index 0350d99bb8fd..6277c83c0eb1 100644
--- a/arch/x86/include/asm/dma-mapping.h
+++ b/arch/x86/include/asm/dma-mapping.h
@@ -12,7 +12,6 @@
 #include <asm/io.h>
 #include <asm/swiotlb.h>
 #include <linux/dma-contiguous.h>
-#include <linux/mem_encrypt.h>
 
 #ifdef CONFIG_ISA
 # define ISA_DMA_BIT_MASK DMA_BIT_MASK(24)
@@ -31,6 +30,9 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return dma_ops;
 }
 
+int arch_dma_supported(struct device *dev, u64 mask);
+#define arch_dma_supported arch_dma_supported
+
 bool arch_dma_alloc_attrs(struct device **dev, gfp_t *gfp);
 #define arch_dma_alloc_attrs arch_dma_alloc_attrs
 
@@ -42,31 +44,6 @@ extern void dma_generic_free_coherent(struct device *dev, size_t size,
 				      void *vaddr, dma_addr_t dma_addr,
 				      unsigned long attrs);
 
-#ifdef CONFIG_X86_DMA_REMAP /* Platform code defines bridge-specific code */
-extern bool dma_capable(struct device *dev, dma_addr_t addr, size_t size);
-extern dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr);
-extern phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr);
-#else
-
-static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	if (!dev->dma_mask)
-		return 0;
-
-	return addr + size - 1 <= *dev->dma_mask;
-}
-
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return __sme_set(paddr);
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return __sme_clr(daddr);
-}
-#endif /* CONFIG_X86_DMA_REMAP */
-
 static inline unsigned long dma_alloc_coherent_mask(struct device *dev,
 						    gfp_t gfp)
 {
diff --git a/arch/x86/include/asm/swiotlb.h b/arch/x86/include/asm/swiotlb.h
index bdf9aed40403..1c6a6cb230ff 100644
--- a/arch/x86/include/asm/swiotlb.h
+++ b/arch/x86/include/asm/swiotlb.h
@@ -28,8 +28,6 @@ static inline void pci_swiotlb_late_init(void)
 }
 #endif
 
-static inline void dma_mark_clean(void *addr, size_t size) {}
-
 extern void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 					dma_addr_t *dma_handle, gfp_t flags,
 					unsigned long attrs);
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index cc0e8bc0ea3f..ecd486cb06ab 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -31,6 +31,7 @@
 #include <linux/io.h>
 #include <linux/gfp.h>
 #include <linux/atomic.h>
+#include <linux/dma-direct.h>
 #include <asm/mtrr.h>
 #include <asm/pgtable.h>
 #include <asm/proto.h>
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 599d7462eccc..df7ab02f959f 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -1,5 +1,5 @@
 // SPDX-License-Identifier: GPL-2.0
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/dma-debug.h>
 #include <linux/dmar.h>
 #include <linux/export.h>
@@ -87,7 +87,6 @@ void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 
 	dma_mask = dma_alloc_coherent_mask(dev, flag);
 
-	flag &= ~__GFP_ZERO;
 again:
 	page = NULL;
 	/* CMA can be used only in the context which permits sleeping */
@@ -139,7 +138,6 @@ bool arch_dma_alloc_attrs(struct device **dev, gfp_t *gfp)
 	if (!*dev)
 		*dev = &x86_dma_fallback_dev;
 
-	*gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
 	*gfp = dma_alloc_coherent_gfp_flags(*dev, *gfp);
 
 	if (!is_device_dma_capable(*dev))
@@ -217,7 +215,7 @@ static __init int iommu_setup(char *p)
 }
 early_param("iommu", iommu_setup);
 
-int x86_dma_supported(struct device *dev, u64 mask)
+int arch_dma_supported(struct device *dev, u64 mask)
 {
 #ifdef CONFIG_PCI
 	if (mask > 0xffffffff && forbid_dac > 0) {
@@ -226,12 +224,6 @@ int x86_dma_supported(struct device *dev, u64 mask)
 	}
 #endif
 
-	/* Copied from i386. Doesn't make much sense, because it will
-	   only work for pci_alloc_coherent.
-	   The caller just has to use GFP_DMA in this case. */
-	if (mask < DMA_BIT_MASK(24))
-		return 0;
-
 	/* Tell the device to use SAC when IOMMU force is on.  This
 	   allows the driver to use cheaper accesses in some cases.
 
@@ -251,6 +243,17 @@ int x86_dma_supported(struct device *dev, u64 mask)
 
 	return 1;
 }
+EXPORT_SYMBOL(arch_dma_supported);
+
+int x86_dma_supported(struct device *dev, u64 mask)
+{
+	/* Copied from i386. Doesn't make much sense, because it will
+	   only work for pci_alloc_coherent.
+	   The caller just has to use GFP_DMA in this case. */
+	if (mask < DMA_BIT_MASK(24))
+		return 0;
+	return 1;
+}
 
 static int __init pci_iommu_init(void)
 {
diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index b0caae27e1b7..618285e475c6 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Fallback functions when the main IOMMU code is not compiled in. This
    code is roughly equivalent to i386. */
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/scatterlist.h>
 #include <linux/string.h>
 #include <linux/gfp.h>
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 53bd05ea90d8..0ee0f8f34251 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -6,7 +6,7 @@
 #include <linux/init.h>
 #include <linux/swiotlb.h>
 #include <linux/bootmem.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/mem_encrypt.h>
 
 #include <asm/iommu.h>
@@ -48,7 +48,7 @@ void x86_swiotlb_free_coherent(struct device *dev, size_t size,
 		dma_generic_free_coherent(dev, size, vaddr, dma_addr, attrs);
 }
 
-static const struct dma_map_ops swiotlb_dma_ops = {
+static const struct dma_map_ops x86_swiotlb_dma_ops = {
 	.mapping_error = swiotlb_dma_mapping_error,
 	.alloc = x86_swiotlb_alloc_coherent,
 	.free = x86_swiotlb_free_coherent,
@@ -112,7 +112,7 @@ void __init pci_swiotlb_init(void)
 {
 	if (swiotlb) {
 		swiotlb_init(0);
-		dma_ops = &swiotlb_dma_ops;
+		dma_ops = &x86_swiotlb_dma_ops;
 	}
 }
 
@@ -120,7 +120,7 @@ void __init pci_swiotlb_late_init(void)
 {
 	/* An IOMMU turned us off. */
 	if (!swiotlb)
-		swiotlb_free();
+		swiotlb_exit();
 	else {
 		printk(KERN_INFO "PCI-DMA: "
 		       "Using software bounce buffering for IO (SWIOTLB)\n");
diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
index e1d61e8500f9..1a53071e2e17 100644
--- a/arch/x86/mm/mem_encrypt.c
+++ b/arch/x86/mm/mem_encrypt.c
@@ -15,7 +15,7 @@
 #include <linux/linkage.h>
 #include <linux/init.h>
 #include <linux/mm.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/swiotlb.h>
 #include <linux/mem_encrypt.h>
 
diff --git a/arch/x86/pci/sta2x11-fixup.c b/arch/x86/pci/sta2x11-fixup.c
index 53d600217973..75577c1490c4 100644
--- a/arch/x86/pci/sta2x11-fixup.c
+++ b/arch/x86/pci/sta2x11-fixup.c
@@ -26,6 +26,7 @@
 #include <linux/pci_ids.h>
 #include <linux/export.h>
 #include <linux/list.h>
+#include <linux/dma-direct.h>
 #include <asm/iommu.h>
 
 #define STA2X11_SWIOTLB_SIZE (4*1024*1024)
diff --git a/arch/xtensa/include/asm/dma-mapping.h b/arch/xtensa/include/asm/dma-mapping.h
index 153bf2370988..44098800dad7 100644
--- a/arch/xtensa/include/asm/dma-mapping.h
+++ b/arch/xtensa/include/asm/dma-mapping.h
@@ -23,14 +23,4 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
 	return &xtensa_dma_map_ops;
 }
 
-static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	return (dma_addr_t)paddr;
-}
-
-static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	return (phys_addr_t)daddr;
-}
-
 #endif	/* _XTENSA_DMA_MAPPING_H */
diff --git a/drivers/crypto/marvell/cesa.c b/drivers/crypto/marvell/cesa.c
index 293832488cc9..3a0c40081ffb 100644
--- a/drivers/crypto/marvell/cesa.c
+++ b/drivers/crypto/marvell/cesa.c
@@ -24,6 +24,7 @@
 #include <linux/scatterlist.h>
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/dma-direct.h> /* XXX: drivers shall never use this directly! */
 #include <linux/clk.h>
 #include <linux/of.h>
 #include <linux/of_platform.h>
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 4a2de34895ec..a1373cf34326 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4808,7 +4808,7 @@ int __init intel_iommu_init(void)
 	up_write(&dmar_global_lock);
 	pr_info("Intel(R) Virtualization Technology for Directed I/O\n");
 
-#ifdef CONFIG_SWIOTLB
+#if defined(CONFIG_X86) && defined(CONFIG_SWIOTLB)
 	swiotlb = 0;
 #endif
 	dma_ops = &intel_dma_ops;
diff --git a/drivers/misc/cxl/vphb.c b/drivers/misc/cxl/vphb.c
index 512a4897dbf6..7fd0bdc1436a 100644
--- a/drivers/misc/cxl/vphb.c
+++ b/drivers/misc/cxl/vphb.c
@@ -54,7 +54,7 @@ static bool cxl_pci_enable_device_hook(struct pci_dev *dev)
 		return false;
 	}
 
-	set_dma_ops(&dev->dev, &dma_direct_ops);
+	set_dma_ops(&dev->dev, &dma_nommu_ops);
 	set_dma_offset(&dev->dev, PAGE_OFFSET);
 
 	return _cxl_pci_associate_default_context(dev, afu);
diff --git a/drivers/mtd/nand/qcom_nandc.c b/drivers/mtd/nand/qcom_nandc.c
index 6be555806eca..563b759ffca6 100644
--- a/drivers/mtd/nand/qcom_nandc.c
+++ b/drivers/mtd/nand/qcom_nandc.c
@@ -23,6 +23,7 @@
 #include <linux/of_device.h>
 #include <linux/delay.h>
 #include <linux/dma/qcom_bam_dma.h>
+#include <linux/dma-direct.h> /* XXX: drivers shall never use this directly! */
 
 /* NANDc reg offsets */
 #define	NAND_FLASH_CMD			0x00
diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c
index 82fc54f8eb77..5bb72d3f8337 100644
--- a/drivers/xen/swiotlb-xen.c
+++ b/drivers/xen/swiotlb-xen.c
@@ -36,7 +36,7 @@
 #define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt
 
 #include <linux/bootmem.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/export.h>
 #include <xen/swiotlb-xen.h>
 #include <xen/page.h>
diff --git a/include/asm-generic/dma-mapping.h b/include/asm-generic/dma-mapping.h
new file mode 100644
index 000000000000..880a292d792f
--- /dev/null
+++ b/include/asm-generic/dma-mapping.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_GENERIC_DMA_MAPPING_H
+#define _ASM_GENERIC_DMA_MAPPING_H
+
+static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus)
+{
+	return &dma_direct_ops;
+}
+
+#endif /* _ASM_GENERIC_DMA_MAPPING_H */
diff --git a/include/linux/dma-direct.h b/include/linux/dma-direct.h
new file mode 100644
index 000000000000..bcdb1a3e4b1f
--- /dev/null
+++ b/include/linux/dma-direct.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_DMA_DIRECT_H
+#define _LINUX_DMA_DIRECT_H 1
+
+#include <linux/dma-mapping.h>
+
+#ifdef CONFIG_ARCH_HAS_PHYS_TO_DMA
+#include <asm/dma-direct.h>
+#else
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	dma_addr_t dev_addr = (dma_addr_t)paddr;
+
+	return dev_addr - ((dma_addr_t)dev->dma_pfn_offset << PAGE_SHIFT);
+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
+{
+	phys_addr_t paddr = (phys_addr_t)dev_addr;
+
+	return paddr + ((phys_addr_t)dev->dma_pfn_offset << PAGE_SHIFT);
+}
+
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	if (!dev->dma_mask)
+		return false;
+
+	return addr + size - 1 <= *dev->dma_mask;
+}
+#endif /* !CONFIG_ARCH_HAS_PHYS_TO_DMA */
+
+#ifdef CONFIG_ARCH_HAS_DMA_MARK_CLEAN
+void dma_mark_clean(void *addr, size_t size);
+#else
+static inline void dma_mark_clean(void *addr, size_t size)
+{
+}
+#endif /* CONFIG_ARCH_HAS_DMA_MARK_CLEAN */
+
+void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		gfp_t gfp, unsigned long attrs);
+void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
+		dma_addr_t dma_addr, unsigned long attrs);
+int dma_direct_supported(struct device *dev, u64 mask);
+
+#endif /* _LINUX_DMA_DIRECT_H */
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index 81ed9b2d84dc..34fe8463d10e 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -136,7 +136,7 @@ struct dma_map_ops {
 	int is_phys;
 };
 
-extern const struct dma_map_ops dma_noop_ops;
+extern const struct dma_map_ops dma_direct_ops;
 extern const struct dma_map_ops dma_virt_ops;
 
 #define DMA_BIT_MASK(n)	(((n) == 64) ? ~0ULL : ((1ULL<<(n))-1))
@@ -513,10 +513,18 @@ static inline void *dma_alloc_attrs(struct device *dev, size_t size,
 	void *cpu_addr;
 
 	BUG_ON(!ops);
+	WARN_ON_ONCE(dev && !dev->coherent_dma_mask);
 
 	if (dma_alloc_from_dev_coherent(dev, size, dma_handle, &cpu_addr))
 		return cpu_addr;
 
+	/*
+	 * Let the implementation decide on the zone to allocate from, and
+	 * decide on the way of zeroing the memory given that the memory
+	 * returned should always be zeroed.
+	 */
+	flag &= ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM | __GFP_ZERO);
+
 	if (!arch_dma_alloc_attrs(&dev, &flag))
 		return NULL;
 	if (!ops->alloc)
@@ -568,6 +576,14 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 	return 0;
 }
 
+/*
+ * This is a hack for the legacy x86 forbid_dac and iommu_sac_force. Please
+ * don't use this is new code.
+ */
+#ifndef arch_dma_supported
+#define arch_dma_supported(dev, mask)	(1)
+#endif
+
 static inline void dma_check_mask(struct device *dev, u64 mask)
 {
 	if (sme_active() && (mask < (((u64)sme_get_me_mask() << 1) - 1)))
@@ -580,6 +596,9 @@ static inline int dma_supported(struct device *dev, u64 mask)
 
 	if (!ops)
 		return 0;
+	if (!arch_dma_supported(dev, mask))
+		return 0;
+
 	if (!ops->dma_supported)
 		return 1;
 	return ops->dma_supported(dev, mask);
@@ -692,7 +711,7 @@ static inline int dma_set_seg_boundary(struct device *dev, unsigned long mask)
 #ifndef dma_max_pfn
 static inline unsigned long dma_max_pfn(struct device *dev)
 {
-	return *dev->dma_mask >> PAGE_SHIFT;
+	return (*dev->dma_mask >> PAGE_SHIFT) + dev->dma_pfn_offset;
 }
 #endif
 
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index 24ed817082ee..5b1f2a00491c 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -66,6 +66,12 @@ extern void swiotlb_tbl_sync_single(struct device *hwdev,
 				    enum dma_sync_target target);
 
 /* Accessory functions. */
+
+void *swiotlb_alloc(struct device *hwdev, size_t size, dma_addr_t *dma_handle,
+		gfp_t flags, unsigned long attrs);
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_addr, unsigned long attrs);
+
 extern void
 *swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 			dma_addr_t *dma_handle, gfp_t flags);
@@ -115,10 +121,10 @@ extern int
 swiotlb_dma_supported(struct device *hwdev, u64 mask);
 
 #ifdef CONFIG_SWIOTLB
-extern void __init swiotlb_free(void);
+extern void __init swiotlb_exit(void);
 unsigned int swiotlb_max_segment(void);
 #else
-static inline void swiotlb_free(void) { }
+static inline void swiotlb_exit(void) { }
 static inline unsigned int swiotlb_max_segment(void) { return 0; }
 #endif
 
@@ -126,4 +132,6 @@ extern void swiotlb_print_info(void);
 extern int is_swiotlb_buffer(phys_addr_t paddr);
 extern void swiotlb_set_max_segment(unsigned int);
 
+extern const struct dma_map_ops swiotlb_dma_ops;
+
 #endif /* __LINUX_SWIOTLB_H */
diff --git a/lib/Kconfig b/lib/Kconfig
index 4dd5c11366f9..e96089499371 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -413,7 +413,7 @@ config SGL_ALLOC
 	bool
 	default n
 
-config DMA_NOOP_OPS
+config DMA_DIRECT_OPS
 	bool
 	depends on HAS_DMA && (!64BIT || ARCH_DMA_ADDR_T_64BIT)
 	default n
diff --git a/lib/Makefile b/lib/Makefile
index d11c48ec8ffd..749851abe85a 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -28,7 +28,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \
 
 lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
-lib-$(CONFIG_DMA_NOOP_OPS) += dma-noop.o
+lib-$(CONFIG_DMA_DIRECT_OPS) += dma-direct.o
 lib-$(CONFIG_DMA_VIRT_OPS) += dma-virt.o
 
 lib-y	+= kobject.o klist.o
diff --git a/lib/dma-direct.c b/lib/dma-direct.c
new file mode 100644
index 000000000000..40b1f92f2214
--- /dev/null
+++ b/lib/dma-direct.c
@@ -0,0 +1,156 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DMA operations that map physical memory directly without using an IOMMU or
+ * flushing caches.
+ */
+#include <linux/export.h>
+#include <linux/mm.h>
+#include <linux/dma-direct.h>
+#include <linux/scatterlist.h>
+#include <linux/dma-contiguous.h>
+#include <linux/pfn.h>
+
+#define DIRECT_MAPPING_ERROR		0
+
+/*
+ * Most architectures use ZONE_DMA for the first 16 Megabytes, but
+ * some use it for entirely different regions:
+ */
+#ifndef ARCH_ZONE_DMA_BITS
+#define ARCH_ZONE_DMA_BITS 24
+#endif
+
+static bool
+check_addr(struct device *dev, dma_addr_t dma_addr, size_t size,
+		const char *caller)
+{
+	if (unlikely(dev && !dma_capable(dev, dma_addr, size))) {
+		if (*dev->dma_mask >= DMA_BIT_MASK(32)) {
+			dev_err(dev,
+				"%s: overflow %pad+%zu of device mask %llx\n",
+				caller, &dma_addr, size, *dev->dma_mask);
+		}
+		return false;
+	}
+	return true;
+}
+
+static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
+{
+	return phys_to_dma(dev, phys) + size - 1 <= dev->coherent_dma_mask;
+}
+
+void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		gfp_t gfp, unsigned long attrs)
+{
+	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	int page_order = get_order(size);
+	struct page *page = NULL;
+
+	/* GFP_DMA32 and GFP_DMA are no ops without the corresponding zones: */
+	if (dev->coherent_dma_mask <= DMA_BIT_MASK(ARCH_ZONE_DMA_BITS))
+		gfp |= GFP_DMA;
+	if (dev->coherent_dma_mask <= DMA_BIT_MASK(32) && !(gfp & GFP_DMA))
+		gfp |= GFP_DMA32;
+
+again:
+	/* CMA can be used only in the context which permits sleeping */
+	if (gfpflags_allow_blocking(gfp)) {
+		page = dma_alloc_from_contiguous(dev, count, page_order, gfp);
+		if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
+			dma_release_from_contiguous(dev, page, count);
+			page = NULL;
+		}
+	}
+	if (!page)
+		page = alloc_pages_node(dev_to_node(dev), gfp, page_order);
+
+	if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
+		__free_pages(page, page_order);
+		page = NULL;
+
+		if (dev->coherent_dma_mask < DMA_BIT_MASK(32) &&
+		    !(gfp & GFP_DMA)) {
+			gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
+			goto again;
+		}
+	}
+
+	if (!page)
+		return NULL;
+
+	*dma_handle = phys_to_dma(dev, page_to_phys(page));
+	memset(page_address(page), 0, size);
+	return page_address(page);
+}
+
+void dma_direct_free(struct device *dev, size_t size, void *cpu_addr,
+		dma_addr_t dma_addr, unsigned long attrs)
+{
+	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+
+	if (!dma_release_from_contiguous(dev, virt_to_page(cpu_addr), count))
+		free_pages((unsigned long)cpu_addr, get_order(size));
+}
+
+static dma_addr_t dma_direct_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		unsigned long attrs)
+{
+	dma_addr_t dma_addr = phys_to_dma(dev, page_to_phys(page)) + offset;
+
+	if (!check_addr(dev, dma_addr, size, __func__))
+		return DIRECT_MAPPING_ERROR;
+	return dma_addr;
+}
+
+static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
+		int nents, enum dma_data_direction dir, unsigned long attrs)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sgl, sg, nents, i) {
+		BUG_ON(!sg_page(sg));
+
+		sg_dma_address(sg) = phys_to_dma(dev, sg_phys(sg));
+		if (!check_addr(dev, sg_dma_address(sg), sg->length, __func__))
+			return 0;
+		sg_dma_len(sg) = sg->length;
+	}
+
+	return nents;
+}
+
+int dma_direct_supported(struct device *dev, u64 mask)
+{
+#ifdef CONFIG_ZONE_DMA
+	if (mask < DMA_BIT_MASK(ARCH_ZONE_DMA_BITS))
+		return 0;
+#else
+	/*
+	 * Because 32-bit DMA masks are so common we expect every architecture
+	 * to be able to satisfy them - either by not supporting more physical
+	 * memory, or by providing a ZONE_DMA32.  If neither is the case, the
+	 * architecture needs to use an IOMMU instead of the direct mapping.
+	 */
+	if (mask < DMA_BIT_MASK(32))
+		return 0;
+#endif
+	return 1;
+}
+
+static int dma_direct_mapping_error(struct device *dev, dma_addr_t dma_addr)
+{
+	return dma_addr == DIRECT_MAPPING_ERROR;
+}
+
+const struct dma_map_ops dma_direct_ops = {
+	.alloc			= dma_direct_alloc,
+	.free			= dma_direct_free,
+	.map_page		= dma_direct_map_page,
+	.map_sg			= dma_direct_map_sg,
+	.dma_supported		= dma_direct_supported,
+	.mapping_error		= dma_direct_mapping_error,
+};
+EXPORT_SYMBOL(dma_direct_ops);
diff --git a/lib/dma-noop.c b/lib/dma-noop.c
deleted file mode 100644
index a10185b0c2d4..000000000000
--- a/lib/dma-noop.c
+++ /dev/null
@@ -1,68 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- *	lib/dma-noop.c
- *
- * DMA operations that map to physical addresses without flushing memory.
- */
-#include <linux/export.h>
-#include <linux/mm.h>
-#include <linux/dma-mapping.h>
-#include <linux/scatterlist.h>
-#include <linux/pfn.h>
-
-static void *dma_noop_alloc(struct device *dev, size_t size,
-			    dma_addr_t *dma_handle, gfp_t gfp,
-			    unsigned long attrs)
-{
-	void *ret;
-
-	ret = (void *)__get_free_pages(gfp, get_order(size));
-	if (ret)
-		*dma_handle = virt_to_phys(ret) - PFN_PHYS(dev->dma_pfn_offset);
-
-	return ret;
-}
-
-static void dma_noop_free(struct device *dev, size_t size,
-			  void *cpu_addr, dma_addr_t dma_addr,
-			  unsigned long attrs)
-{
-	free_pages((unsigned long)cpu_addr, get_order(size));
-}
-
-static dma_addr_t dma_noop_map_page(struct device *dev, struct page *page,
-				      unsigned long offset, size_t size,
-				      enum dma_data_direction dir,
-				      unsigned long attrs)
-{
-	return page_to_phys(page) + offset - PFN_PHYS(dev->dma_pfn_offset);
-}
-
-static int dma_noop_map_sg(struct device *dev, struct scatterlist *sgl, int nents,
-			     enum dma_data_direction dir,
-			     unsigned long attrs)
-{
-	int i;
-	struct scatterlist *sg;
-
-	for_each_sg(sgl, sg, nents, i) {
-		dma_addr_t offset = PFN_PHYS(dev->dma_pfn_offset);
-		void *va;
-
-		BUG_ON(!sg_page(sg));
-		va = sg_virt(sg);
-		sg_dma_address(sg) = (dma_addr_t)virt_to_phys(va) - offset;
-		sg_dma_len(sg) = sg->length;
-	}
-
-	return nents;
-}
-
-const struct dma_map_ops dma_noop_ops = {
-	.alloc			= dma_noop_alloc,
-	.free			= dma_noop_free,
-	.map_page		= dma_noop_map_page,
-	.map_sg			= dma_noop_map_sg,
-};
-
-EXPORT_SYMBOL(dma_noop_ops);
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index cea19aaf303c..c43ec2271469 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -18,7 +18,7 @@
  */
 
 #include <linux/cache.h>
-#include <linux/dma-mapping.h>
+#include <linux/dma-direct.h>
 #include <linux/mm.h>
 #include <linux/export.h>
 #include <linux/spinlock.h>
@@ -417,7 +417,7 @@ cleanup2:
 	return -ENOMEM;
 }
 
-void __init swiotlb_free(void)
+void __init swiotlb_exit(void)
 {
 	if (!io_tlb_orig_addr)
 		return;
@@ -586,7 +586,7 @@ phys_addr_t swiotlb_tbl_map_single(struct device *hwdev,
 
 not_found:
 	spin_unlock_irqrestore(&io_tlb_lock, flags);
-	if (printk_ratelimit())
+	if (!(attrs & DMA_ATTR_NO_WARN) && printk_ratelimit())
 		dev_warn(hwdev, "swiotlb buffer is full (sz: %zd bytes)\n", size);
 	return SWIOTLB_MAP_ERROR;
 found:
@@ -605,7 +605,6 @@ found:
 
 	return tlb_addr;
 }
-EXPORT_SYMBOL_GPL(swiotlb_tbl_map_single);
 
 /*
  * Allocates bounce buffer and returns its kernel virtual address.
@@ -675,7 +674,6 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr,
 	}
 	spin_unlock_irqrestore(&io_tlb_lock, flags);
 }
-EXPORT_SYMBOL_GPL(swiotlb_tbl_unmap_single);
 
 void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
 			     size_t size, enum dma_data_direction dir,
@@ -707,92 +705,107 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
 		BUG();
 	}
 }
-EXPORT_SYMBOL_GPL(swiotlb_tbl_sync_single);
+
+static inline bool dma_coherent_ok(struct device *dev, dma_addr_t addr,
+		size_t size)
+{
+	u64 mask = DMA_BIT_MASK(32);
+
+	if (dev && dev->coherent_dma_mask)
+		mask = dev->coherent_dma_mask;
+	return addr + size - 1 <= mask;
+}
+
+static void *
+swiotlb_alloc_buffer(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		unsigned long attrs)
+{
+	phys_addr_t phys_addr;
+
+	if (swiotlb_force == SWIOTLB_NO_FORCE)
+		goto out_warn;
+
+	phys_addr = swiotlb_tbl_map_single(dev,
+			swiotlb_phys_to_dma(dev, io_tlb_start),
+			0, size, DMA_FROM_DEVICE, 0);
+	if (phys_addr == SWIOTLB_MAP_ERROR)
+		goto out_warn;
+
+	*dma_handle = swiotlb_phys_to_dma(dev, phys_addr);
+	if (dma_coherent_ok(dev, *dma_handle, size))
+		goto out_unmap;
+
+	memset(phys_to_virt(phys_addr), 0, size);
+	return phys_to_virt(phys_addr);
+
+out_unmap:
+	dev_warn(dev, "hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n",
+		(unsigned long long)(dev ? dev->coherent_dma_mask : 0),
+		(unsigned long long)*dma_handle);
+
+	/*
+	 * DMA_TO_DEVICE to avoid memcpy in unmap_single.
+	 * DMA_ATTR_SKIP_CPU_SYNC is optional.
+	 */
+	swiotlb_tbl_unmap_single(dev, phys_addr, size, DMA_TO_DEVICE,
+			DMA_ATTR_SKIP_CPU_SYNC);
+out_warn:
+	if ((attrs & DMA_ATTR_NO_WARN) && printk_ratelimit()) {
+		dev_warn(dev,
+			"swiotlb: coherent allocation failed, size=%zu\n",
+			size);
+		dump_stack();
+	}
+	return NULL;
+}
 
 void *
 swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 		       dma_addr_t *dma_handle, gfp_t flags)
 {
-	dma_addr_t dev_addr;
-	void *ret;
 	int order = get_order(size);
-	u64 dma_mask = DMA_BIT_MASK(32);
-
-	if (hwdev && hwdev->coherent_dma_mask)
-		dma_mask = hwdev->coherent_dma_mask;
+	unsigned long attrs = (flags & __GFP_NOWARN) ? DMA_ATTR_NO_WARN : 0;
+	void *ret;
 
 	ret = (void *)__get_free_pages(flags, order);
 	if (ret) {
-		dev_addr = swiotlb_virt_to_bus(hwdev, ret);
-		if (dev_addr + size - 1 > dma_mask) {
-			/*
-			 * The allocated memory isn't reachable by the device.
-			 */
-			free_pages((unsigned long) ret, order);
-			ret = NULL;
+		*dma_handle = swiotlb_virt_to_bus(hwdev, ret);
+		if (dma_coherent_ok(hwdev, *dma_handle, size)) {
+			memset(ret, 0, size);
+			return ret;
 		}
+		free_pages((unsigned long)ret, order);
 	}
-	if (!ret) {
-		/*
-		 * We are either out of memory or the device can't DMA to
-		 * GFP_DMA memory; fall back on map_single(), which
-		 * will grab memory from the lowest available address range.
-		 */
-		phys_addr_t paddr = map_single(hwdev, 0, size,
-					       DMA_FROM_DEVICE, 0);
-		if (paddr == SWIOTLB_MAP_ERROR)
-			goto err_warn;
 
-		ret = phys_to_virt(paddr);
-		dev_addr = swiotlb_phys_to_dma(hwdev, paddr);
-
-		/* Confirm address can be DMA'd by device */
-		if (dev_addr + size - 1 > dma_mask) {
-			printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n",
-			       (unsigned long long)dma_mask,
-			       (unsigned long long)dev_addr);
-
-			/*
-			 * DMA_TO_DEVICE to avoid memcpy in unmap_single.
-			 * The DMA_ATTR_SKIP_CPU_SYNC is optional.
-			 */
-			swiotlb_tbl_unmap_single(hwdev, paddr,
-						 size, DMA_TO_DEVICE,
-						 DMA_ATTR_SKIP_CPU_SYNC);
-			goto err_warn;
-		}
-	}
+	return swiotlb_alloc_buffer(hwdev, size, dma_handle, attrs);
+}
+EXPORT_SYMBOL(swiotlb_alloc_coherent);
 
-	*dma_handle = dev_addr;
-	memset(ret, 0, size);
+static bool swiotlb_free_buffer(struct device *dev, size_t size,
+		dma_addr_t dma_addr)
+{
+	phys_addr_t phys_addr = dma_to_phys(dev, dma_addr);
 
-	return ret;
+	WARN_ON_ONCE(irqs_disabled());
 
-err_warn:
-	pr_warn("swiotlb: coherent allocation failed for device %s size=%zu\n",
-		dev_name(hwdev), size);
-	dump_stack();
+	if (!is_swiotlb_buffer(phys_addr))
+		return false;
 
-	return NULL;
+	/*
+	 * DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single.
+	 * DMA_ATTR_SKIP_CPU_SYNC is optional.
+	 */
+	swiotlb_tbl_unmap_single(dev, phys_addr, size, DMA_TO_DEVICE,
+				 DMA_ATTR_SKIP_CPU_SYNC);
+	return true;
 }
-EXPORT_SYMBOL(swiotlb_alloc_coherent);
 
 void
 swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
 		      dma_addr_t dev_addr)
 {
-	phys_addr_t paddr = dma_to_phys(hwdev, dev_addr);
-
-	WARN_ON(irqs_disabled());
-	if (!is_swiotlb_buffer(paddr))
+	if (!swiotlb_free_buffer(hwdev, size, dev_addr))
 		free_pages((unsigned long)vaddr, get_order(size));
-	else
-		/*
-		 * DMA_TO_DEVICE to avoid memcpy in swiotlb_tbl_unmap_single.
-		 * DMA_ATTR_SKIP_CPU_SYNC is optional.
-		 */
-		swiotlb_tbl_unmap_single(hwdev, paddr, size, DMA_TO_DEVICE,
-					 DMA_ATTR_SKIP_CPU_SYNC);
 }
 EXPORT_SYMBOL(swiotlb_free_coherent);
 
@@ -868,7 +881,6 @@ dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
 
 	return swiotlb_phys_to_dma(dev, io_tlb_overflow_buffer);
 }
-EXPORT_SYMBOL_GPL(swiotlb_map_page);
 
 /*
  * Unmap a single streaming mode DMA translation.  The dma_addr and size must
@@ -909,7 +921,6 @@ void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
 {
 	unmap_single(hwdev, dev_addr, size, dir, attrs);
 }
-EXPORT_SYMBOL_GPL(swiotlb_unmap_page);
 
 /*
  * Make physical memory consistent for a single streaming mode DMA translation
@@ -947,7 +958,6 @@ swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
 {
 	swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_CPU);
 }
-EXPORT_SYMBOL(swiotlb_sync_single_for_cpu);
 
 void
 swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
@@ -955,7 +965,6 @@ swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
 {
 	swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_DEVICE);
 }
-EXPORT_SYMBOL(swiotlb_sync_single_for_device);
 
 /*
  * Map a set of buffers described by scatterlist in streaming mode for DMA.
@@ -1007,7 +1016,6 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems,
 	}
 	return nelems;
 }
-EXPORT_SYMBOL(swiotlb_map_sg_attrs);
 
 /*
  * Unmap a set of streaming mode DMA translations.  Again, cpu read rules
@@ -1027,7 +1035,6 @@ swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 		unmap_single(hwdev, sg->dma_address, sg_dma_len(sg), dir,
 			     attrs);
 }
-EXPORT_SYMBOL(swiotlb_unmap_sg_attrs);
 
 /*
  * Make physical memory consistent for a set of streaming mode DMA translations
@@ -1055,7 +1062,6 @@ swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
 {
 	swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_CPU);
 }
-EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu);
 
 void
 swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
@@ -1063,14 +1069,12 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
 {
 	swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE);
 }
-EXPORT_SYMBOL(swiotlb_sync_sg_for_device);
 
 int
 swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 {
 	return (dma_addr == swiotlb_phys_to_dma(hwdev, io_tlb_overflow_buffer));
 }
-EXPORT_SYMBOL(swiotlb_dma_mapping_error);
 
 /*
  * Return whether the given device DMA address mask can be supported
@@ -1083,4 +1087,49 @@ swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
 	return swiotlb_phys_to_dma(hwdev, io_tlb_end - 1) <= mask;
 }
-EXPORT_SYMBOL(swiotlb_dma_supported);
+
+#ifdef CONFIG_DMA_DIRECT_OPS
+void *swiotlb_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		gfp_t gfp, unsigned long attrs)
+{
+	void *vaddr;
+
+	/* temporary workaround: */
+	if (gfp & __GFP_NOWARN)
+		attrs |= DMA_ATTR_NO_WARN;
+
+	/*
+	 * Don't print a warning when the first allocation attempt fails.
+	 * swiotlb_alloc_coherent() will print a warning when the DMA memory
+	 * allocation ultimately failed.
+	 */
+	gfp |= __GFP_NOWARN;
+
+	vaddr = dma_direct_alloc(dev, size, dma_handle, gfp, attrs);
+	if (!vaddr)
+		vaddr = swiotlb_alloc_buffer(dev, size, dma_handle, attrs);
+	return vaddr;
+}
+
+void swiotlb_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_addr, unsigned long attrs)
+{
+	if (!swiotlb_free_buffer(dev, size, dma_addr))
+		dma_direct_free(dev, size, vaddr, dma_addr, attrs);
+}
+
+const struct dma_map_ops swiotlb_dma_ops = {
+	.mapping_error		= swiotlb_dma_mapping_error,
+	.alloc			= swiotlb_alloc,
+	.free			= swiotlb_free,
+	.sync_single_for_cpu	= swiotlb_sync_single_for_cpu,
+	.sync_single_for_device	= swiotlb_sync_single_for_device,
+	.sync_sg_for_cpu	= swiotlb_sync_sg_for_cpu,
+	.sync_sg_for_device	= swiotlb_sync_sg_for_device,
+	.map_sg			= swiotlb_map_sg_attrs,
+	.unmap_sg		= swiotlb_unmap_sg_attrs,
+	.map_page		= swiotlb_map_page,
+	.unmap_page		= swiotlb_unmap_page,
+	.dma_supported		= swiotlb_dma_supported,
+};
+#endif /* CONFIG_DMA_DIRECT_OPS */