diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 752a3848ffdc..4be85c38250f 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -8,6 +8,7 @@ config ARM
 	select ARCH_HAS_DEBUG_VIRTUAL if MMU
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
 	select ARCH_HAS_DMA_COHERENT_TO_PFN if SWIOTLB
+	select ARCH_HAS_DMA_PREP_COHERENT if IOMMU_DMA
 	select ARCH_HAS_DMA_WRITE_COMBINE if !ARM_DMA_MEM_BUFFERABLE
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_FORTIFY_SOURCE
@@ -20,8 +21,8 @@ config ARM
 	select ARCH_HAS_SET_MEMORY
 	select ARCH_HAS_STRICT_KERNEL_RWX if MMU && !XIP_KERNEL
 	select ARCH_HAS_STRICT_MODULE_RWX if MMU
-	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if SWIOTLB
-	select ARCH_HAS_SYNC_DMA_FOR_CPU if SWIOTLB
+	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if SWIOTLB || IOMMU_DMA
+	select ARCH_HAS_SYNC_DMA_FOR_CPU if SWIOTLB || IOMMU_DMA
 	select ARCH_HAS_TEARDOWN_DMA_OPS if MMU
 	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
 	select ARCH_HAVE_CUSTOM_GPIO_H
@@ -121,6 +122,7 @@ config ARM
 	select REFCOUNT_FULL
 	select RTC_LIB
 	select SYS_SUPPORTS_APM_EMULATION
+	select IOMMU_DMA if IOMMU_SUPPORT
 	# Above selects are sorted alphabetically; please add new ones
 	# according to that.  Thanks.
 	help
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 75b5b4ee22e8..ba23f9059cf4 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -17,6 +17,7 @@
 #include <linux/dma-direct.h>
 #include <linux/dma-mapping.h>
 #include <linux/dma-noncoherent.h>
+#include <linux/dma-iommu.h>
 #include <linux/dma-contiguous.h>
 #include <linux/highmem.h>
 #include <linux/memblock.h>
@@ -2489,8 +2490,9 @@ static void arm_iommu_get_dma_window(struct device *dev, u64 *dma_addr,
 	*dma_size = of_read_number(ranges + naddr, nsize);
 }
 
-static bool arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				    const struct iommu_ops *iommu)
+static bool __maybe_unused
+arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
+			const struct iommu_ops *iommu)
 {
 	struct iommu_group *group;
 	struct iommu_domain *domain;
@@ -2550,8 +2552,9 @@ static void arm_teardown_iommu_dma_ops(struct device *dev)
 }
 #else
 
-static bool arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				    const struct iommu_ops *iommu)
+static bool __maybe_unused
+arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
+			const struct iommu_ops *iommu)
 {
 	return false;
 }
@@ -2566,10 +2569,9 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 			const struct iommu_ops *iommu, bool coherent)
 {
 	const struct dma_map_ops *dma_ops;
-	struct dma_iommu_mapping *mapping;
 
 	dev->archdata.dma_coherent = coherent;
-#ifdef CONFIG_SWIOTLB
+#if defined(CONFIG_SWIOTLB) || defined(CONFIG_IOMMU_DMA)
 	dev->dma_coherent = coherent;
 #endif
 
@@ -2581,14 +2583,13 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	if (dev->dma_ops)
 		return;
 
-	if (arm_setup_iommu_dma_ops(dev, dma_base, size, iommu)) {
-		mapping = to_dma_iommu_mapping(dev);
-		dma_ops = mapping->ops;
-	} else {
-		dma_ops = arm_get_dma_map_ops(coherent);
-	}
+	if (iommu)
+		iommu_setup_dma_ops(dev, dma_base, size);
 
-	set_dma_ops(dev, dma_ops);
+	if (!dev->dma_ops) {
+		dma_ops = arm_get_dma_map_ops(coherent);
+		set_dma_ops(dev, dma_ops);
+	}
 
 #ifdef CONFIG_XEN
 	if (xen_initial_domain())
@@ -2606,7 +2607,7 @@ void arch_teardown_dma_ops(struct device *dev)
 	arm_teardown_iommu_dma_ops(dev);
 }
 
-#ifdef CONFIG_SWIOTLB
+#if defined(CONFIG_SWIOTLB) || defined(CONFIG_IOMMU_DMA)
 void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 		size_t size, enum dma_data_direction dir)
 {
@@ -2620,7 +2621,9 @@ void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
 	__dma_page_dev_to_cpu(phys_to_page(paddr), paddr & (PAGE_SIZE - 1),
 			      size, dir);
 }
+#endif
 
+#ifdef CONFIG_SWIOTLB
 long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
 		dma_addr_t dma_addr)
 {
@@ -2641,3 +2644,10 @@ void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
 	__arm_dma_free(dev, size, cpu_addr, dma_handle, attrs, false);
 }
 #endif /* CONFIG_SWIOTLB */
+
+#ifdef CONFIG_IOMMU_DMA
+void arch_dma_prep_coherent(struct page *page, size_t size)
+{
+	__dma_page_cpu_to_dev(page, 0, size, DMA_BIDIRECTIONAL);
+}
+#endif /* CONFIG_IOMMU_DMA */
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 329f8c15a1e0..3c482f6a978e 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -469,7 +469,6 @@ config ARM_SMMU
 	depends on QCOM_SCM || !QCOM_SCM #if QCOM_SCM=m this can't be =y
 	select IOMMU_API
 	select IOMMU_IO_PGTABLE_LPAE
-	select ARM_DMA_USE_IOMMU if ARM
 	help
 	  Support for implementations of the ARM System MMU architecture
 	  versions 1 and 2.