diff --git a/arch/Kconfig b/arch/Kconfig
index c47b328eada0..e8d19c3cb91f 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -260,6 +260,14 @@ config ARCH_HAS_SET_MEMORY
 config ARCH_HAS_SET_DIRECT_MAP
 	bool
 
+#
+# Select if arch has an uncached kernel segment and provides the
+# uncached_kernel_address / cached_kernel_address symbols to use it
+#
+config ARCH_HAS_UNCACHED_SEGMENT
+	select ARCH_HAS_DMA_PREP_COHERENT
+	bool
+
 # Select if arch init_task must go in the __init_task_data section
 config ARCH_TASK_STRUCT_ON_STACK
        bool
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 1c8137e7247b..8383155c8c82 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -7,6 +7,7 @@ config ARC
 	def_bool y
 	select ARC_TIMERS
 	select ARCH_HAS_DMA_COHERENT_TO_PFN
+	select ARCH_HAS_DMA_PREP_COHERENT
 	select ARCH_HAS_PTE_SPECIAL
 	select ARCH_HAS_SETUP_DMA_OPS
 	select ARCH_HAS_SYNC_DMA_FOR_CPU
@@ -16,6 +17,7 @@ config ARC
 	select BUILDTIME_EXTABLE_SORT
 	select CLONE_BACKWARDS
 	select COMMON_CLK
+	select DMA_DIRECT_REMAP
 	select GENERIC_ATOMIC64 if !ISA_ARCV2 || !(ARC_HAS_LL64 && ARC_HAS_LLSC)
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_FIND_FIRST_BIT
diff --git a/arch/arc/mm/dma.c b/arch/arc/mm/dma.c
index 0bf1468c35a3..62c210e7ee4c 100644
--- a/arch/arc/mm/dma.c
+++ b/arch/arc/mm/dma.c
@@ -8,51 +8,15 @@
 #include <asm/cacheflush.h>
 
 /*
- * ARCH specific callbacks for generic noncoherent DMA ops (dma/noncoherent.c)
+ * ARCH specific callbacks for generic noncoherent DMA ops
  *  - hardware IOC not available (or "dma-coherent" not set for device in DT)
  *  - But still handle both coherent and non-coherent requests from caller
  *
  * For DMA coherent hardware (IOC) generic code suffices
  */
-void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		gfp_t gfp, unsigned long attrs)
+
+void arch_dma_prep_coherent(struct page *page, size_t size)
 {
-	unsigned long order = get_order(size);
-	struct page *page;
-	phys_addr_t paddr;
-	void *kvaddr;
-	bool need_coh = !(attrs & DMA_ATTR_NON_CONSISTENT);
-
-	/*
-	 * __GFP_HIGHMEM flag is cleared by upper layer functions
-	 * (in include/linux/dma-mapping.h) so we should never get a
-	 * __GFP_HIGHMEM here.
-	 */
-	BUG_ON(gfp & __GFP_HIGHMEM);
-
-	page = alloc_pages(gfp | __GFP_ZERO, order);
-	if (!page)
-		return NULL;
-
-	/* This is linear addr (0x8000_0000 based) */
-	paddr = page_to_phys(page);
-
-	*dma_handle = paddr;
-
-	/*
-	 * A coherent buffer needs MMU mapping to enforce non-cachability.
-	 * kvaddr is kernel Virtual address (0x7000_0000 based).
-	 */
-	if (need_coh) {
-		kvaddr = ioremap_nocache(paddr, size);
-		if (kvaddr == NULL) {
-			__free_pages(page, order);
-			return NULL;
-		}
-	} else {
-		kvaddr = (void *)(u32)paddr;
-	}
-
 	/*
 	 * Evict any existing L1 and/or L2 lines for the backing page
 	 * in case it was used earlier as a normal "cached" page.
@@ -63,28 +27,7 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 	 * Currently flush_cache_vmap nukes the L1 cache completely which
 	 * will be optimized as a separate commit
 	 */
-	if (need_coh)
-		dma_cache_wback_inv(paddr, size);
-
-	return kvaddr;
-}
-
-void arch_dma_free(struct device *dev, size_t size, void *vaddr,
-		dma_addr_t dma_handle, unsigned long attrs)
-{
-	phys_addr_t paddr = dma_handle;
-	struct page *page = virt_to_page(paddr);
-
-	if (!(attrs & DMA_ATTR_NON_CONSISTENT))
-		iounmap((void __force __iomem *)vaddr);
-
-	__free_pages(page, get_order(size));
-}
-
-long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
-		dma_addr_t dma_addr)
-{
-	return __phys_to_pfn(dma_addr);
+	dma_cache_wback_inv(page_to_phys(page), size);
 }
 
 /*
@@ -161,3 +104,9 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	dev_info(dev, "use %sncoherent DMA ops\n",
 		 dev->dma_coherent ? "" : "non");
 }
+
+static int __init atomic_pool_init(void)
+{
+	return dma_atomic_pool_init(GFP_KERNEL, pgprot_noncached(PAGE_KERNEL));
+}
+postcore_initcall(atomic_pool_init);
diff --git a/arch/arm/mm/dma-mapping-nommu.c b/arch/arm/mm/dma-mapping-nommu.c
index 1aea01ba1262..52b82559d99b 100644
--- a/arch/arm/mm/dma-mapping-nommu.c
+++ b/arch/arm/mm/dma-mapping-nommu.c
@@ -35,18 +35,7 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
 				 unsigned long attrs)
 
 {
-	void *ret;
-
-	/*
-	 * Try generic allocator first if we are advertised that
-	 * consistency is not required.
-	 */
-
-	if (attrs & DMA_ATTR_NON_CONSISTENT)
-		return dma_direct_alloc_pages(dev, size, dma_handle, gfp,
-				attrs);
-
-	ret = dma_alloc_from_global_coherent(size, dma_handle);
+	void *ret = dma_alloc_from_global_coherent(size, dma_handle);
 
 	/*
 	 * dma_alloc_from_global_coherent() may fail because:
@@ -66,16 +55,9 @@ static void arm_nommu_dma_free(struct device *dev, size_t size,
 			       void *cpu_addr, dma_addr_t dma_addr,
 			       unsigned long attrs)
 {
-	if (attrs & DMA_ATTR_NON_CONSISTENT) {
-		dma_direct_free_pages(dev, size, cpu_addr, dma_addr, attrs);
-	} else {
-		int ret = dma_release_from_global_coherent(get_order(size),
-							   cpu_addr);
+	int ret = dma_release_from_global_coherent(get_order(size), cpu_addr);
 
-		WARN_ON_ONCE(ret == 0);
-	}
-
-	return;
+	WARN_ON_ONCE(ret == 0);
 }
 
 static int arm_nommu_dma_mmap(struct device *dev, struct vm_area_struct *vma,
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 1fb5c0ca1ed8..4789c60a86e3 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -216,25 +216,7 @@ EXPORT_SYMBOL(arm_coherent_dma_ops);
 
 static int __dma_supported(struct device *dev, u64 mask, bool warn)
 {
-	unsigned long max_dma_pfn;
-
-	/*
-	 * If the mask allows for more memory than we can address,
-	 * and we actually have that much memory, then we must
-	 * indicate that DMA to this device is not supported.
-	 */
-	if (sizeof(mask) != sizeof(dma_addr_t) &&
-	    mask > (dma_addr_t)~0 &&
-	    dma_to_pfn(dev, ~0) < max_pfn - 1) {
-		if (warn) {
-			dev_warn(dev, "Coherent DMA mask %#llx is larger than dma_addr_t allows\n",
-				 mask);
-			dev_warn(dev, "Driver did not use or check the return value from dma_set_coherent_mask()?\n");
-		}
-		return 0;
-	}
-
-	max_dma_pfn = min(max_pfn, arm_dma_pfn_limit);
+	unsigned long max_dma_pfn = min(max_pfn, arm_dma_pfn_limit);
 
 	/*
 	 * Translate the device's DMA mask to a PFN limit.  This
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 7957d3457156..d50fafd7bf3a 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1121,6 +1121,7 @@ config DMA_NONCOHERENT
 	bool
 	select ARCH_HAS_DMA_MMAP_PGPROT
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
+	select ARCH_HAS_UNCACHED_SEGMENT
 	select NEED_DMA_MAP_STATE
 	select ARCH_HAS_DMA_COHERENT_TO_PFN
 	select DMA_NONCOHERENT_CACHE_SYNC
diff --git a/arch/mips/include/asm/page.h b/arch/mips/include/asm/page.h
index a25643d258cb..0ba4ce6e2bf3 100644
--- a/arch/mips/include/asm/page.h
+++ b/arch/mips/include/asm/page.h
@@ -258,9 +258,6 @@ extern bool __virt_addr_valid(const volatile void *kaddr);
 	 ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \
 	 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-#define UNCAC_ADDR(addr)	(UNCAC_BASE + __pa(addr))
-#define CAC_ADDR(addr)		((unsigned long)__va((addr) - UNCAC_BASE))
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/getorder.h>
 
diff --git a/arch/mips/jazz/jazzdma.c b/arch/mips/jazz/jazzdma.c
index bedb5047aff3..1804dc9d8136 100644
--- a/arch/mips/jazz/jazzdma.c
+++ b/arch/mips/jazz/jazzdma.c
@@ -575,10 +575,6 @@ static void *jazz_dma_alloc(struct device *dev, size_t size,
 		return NULL;
 	}
 
-	if (!(attrs & DMA_ATTR_NON_CONSISTENT)) {
-		dma_cache_wback_inv((unsigned long)ret, size);
-		ret = (void *)UNCAC_ADDR(ret);
-	}
 	return ret;
 }
 
@@ -586,8 +582,6 @@ static void jazz_dma_free(struct device *dev, size_t size, void *vaddr,
 		dma_addr_t dma_handle, unsigned long attrs)
 {
 	vdma_free(dma_handle);
-	if (!(attrs & DMA_ATTR_NON_CONSISTENT))
-		vaddr = (void *)CAC_ADDR((unsigned long)vaddr);
 	dma_direct_free_pages(dev, size, vaddr, dma_handle, attrs);
 }
 
diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c
index 3da216988672..33b409391ddb 100644
--- a/arch/mips/mm/cache.c
+++ b/arch/mips/mm/cache.c
@@ -62,8 +62,6 @@ void (*_dma_cache_wback_inv)(unsigned long start, unsigned long size);
 void (*_dma_cache_wback)(unsigned long start, unsigned long size);
 void (*_dma_cache_inv)(unsigned long start, unsigned long size);
 
-EXPORT_SYMBOL(_dma_cache_wback_inv);
-
 #endif /* CONFIG_DMA_NONCOHERENT */
 
 /*
diff --git a/arch/mips/mm/dma-noncoherent.c b/arch/mips/mm/dma-noncoherent.c
index f9549d2fbea3..ed56c6fa7be2 100644
--- a/arch/mips/mm/dma-noncoherent.c
+++ b/arch/mips/mm/dma-noncoherent.c
@@ -44,33 +44,25 @@ static inline bool cpu_needs_post_dma_flush(struct device *dev)
 	}
 }
 
-void *arch_dma_alloc(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
+void arch_dma_prep_coherent(struct page *page, size_t size)
 {
-	void *ret;
-
-	ret = dma_direct_alloc_pages(dev, size, dma_handle, gfp, attrs);
-	if (ret && !(attrs & DMA_ATTR_NON_CONSISTENT)) {
-		dma_cache_wback_inv((unsigned long) ret, size);
-		ret = (void *)UNCAC_ADDR(ret);
-	}
-
-	return ret;
+	dma_cache_wback_inv((unsigned long)page_address(page), size);
 }
 
-void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
-		dma_addr_t dma_addr, unsigned long attrs)
+void *uncached_kernel_address(void *addr)
 {
-	if (!(attrs & DMA_ATTR_NON_CONSISTENT))
-		cpu_addr = (void *)CAC_ADDR((unsigned long)cpu_addr);
-	dma_direct_free_pages(dev, size, cpu_addr, dma_addr, attrs);
+	return (void *)(__pa(addr) + UNCAC_BASE);
+}
+
+void *cached_kernel_address(void *addr)
+{
+	return __va(addr) - UNCAC_BASE;
 }
 
 long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
 		dma_addr_t dma_addr)
 {
-	unsigned long addr = CAC_ADDR((unsigned long)cpu_addr);
-	return page_to_pfn(virt_to_page((void *)addr));
+	return page_to_pfn(virt_to_page(cached_kernel_address(cpu_addr)));
 }
 
 pgprot_t arch_dma_mmap_pgprot(struct device *dev, pgprot_t prot,
diff --git a/arch/nds32/Kconfig b/arch/nds32/Kconfig
index fd0d0639454f..fbd68329737f 100644
--- a/arch/nds32/Kconfig
+++ b/arch/nds32/Kconfig
@@ -7,12 +7,14 @@
 config NDS32
 	def_bool y
 	select ARCH_32BIT_OFF_T
+	select ARCH_HAS_DMA_PREP_COHERENT
 	select ARCH_HAS_SYNC_DMA_FOR_CPU
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
 	select ARCH_WANT_FRAME_POINTERS if FTRACE
 	select CLKSRC_MMIO
 	select CLONE_BACKWARDS
 	select COMMON_CLK
+	select DMA_DIRECT_REMAP
 	select GENERIC_ATOMIC64
 	select GENERIC_CPU_DEVICES
 	select GENERIC_CLOCKEVENTS
diff --git a/arch/nds32/kernel/dma.c b/arch/nds32/kernel/dma.c
index d0dbd4fe9645..490e3720d694 100644
--- a/arch/nds32/kernel/dma.c
+++ b/arch/nds32/kernel/dma.c
@@ -3,327 +3,13 @@
 
 #include <linux/types.h>
 #include <linux/mm.h>
-#include <linux/string.h>
 #include <linux/dma-noncoherent.h>
-#include <linux/io.h>
 #include <linux/cache.h>
 #include <linux/highmem.h>
-#include <linux/slab.h>
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
 #include <asm/proc-fns.h>
 
-/*
- * This is the page table (2MB) covering uncached, DMA consistent allocations
- */
-static pte_t *consistent_pte;
-static DEFINE_RAW_SPINLOCK(consistent_lock);
-
-/*
- * VM region handling support.
- *
- * This should become something generic, handling VM region allocations for
- * vmalloc and similar (ioremap, module space, etc).
- *
- * I envisage vmalloc()'s supporting vm_struct becoming:
- *
- *  struct vm_struct {
- *    struct vm_region	region;
- *    unsigned long	flags;
- *    struct page	**pages;
- *    unsigned int	nr_pages;
- *    unsigned long	phys_addr;
- *  };
- *
- * get_vm_area() would then call vm_region_alloc with an appropriate
- * struct vm_region head (eg):
- *
- *  struct vm_region vmalloc_head = {
- *	.vm_list	= LIST_HEAD_INIT(vmalloc_head.vm_list),
- *	.vm_start	= VMALLOC_START,
- *	.vm_end		= VMALLOC_END,
- *  };
- *
- * However, vmalloc_head.vm_start is variable (typically, it is dependent on
- * the amount of RAM found at boot time.)  I would imagine that get_vm_area()
- * would have to initialise this each time prior to calling vm_region_alloc().
- */
-struct arch_vm_region {
-	struct list_head vm_list;
-	unsigned long vm_start;
-	unsigned long vm_end;
-	struct page *vm_pages;
-};
-
-static struct arch_vm_region consistent_head = {
-	.vm_list = LIST_HEAD_INIT(consistent_head.vm_list),
-	.vm_start = CONSISTENT_BASE,
-	.vm_end = CONSISTENT_END,
-};
-
-static struct arch_vm_region *vm_region_alloc(struct arch_vm_region *head,
-					      size_t size, int gfp)
-{
-	unsigned long addr = head->vm_start, end = head->vm_end - size;
-	unsigned long flags;
-	struct arch_vm_region *c, *new;
-
-	new = kmalloc(sizeof(struct arch_vm_region), gfp);
-	if (!new)
-		goto out;
-
-	raw_spin_lock_irqsave(&consistent_lock, flags);
-
-	list_for_each_entry(c, &head->vm_list, vm_list) {
-		if ((addr + size) < addr)
-			goto nospc;
-		if ((addr + size) <= c->vm_start)
-			goto found;
-		addr = c->vm_end;
-		if (addr > end)
-			goto nospc;
-	}
-
-found:
-	/*
-	 * Insert this entry _before_ the one we found.
-	 */
-	list_add_tail(&new->vm_list, &c->vm_list);
-	new->vm_start = addr;
-	new->vm_end = addr + size;
-
-	raw_spin_unlock_irqrestore(&consistent_lock, flags);
-	return new;
-
-nospc:
-	raw_spin_unlock_irqrestore(&consistent_lock, flags);
-	kfree(new);
-out:
-	return NULL;
-}
-
-static struct arch_vm_region *vm_region_find(struct arch_vm_region *head,
-					     unsigned long addr)
-{
-	struct arch_vm_region *c;
-
-	list_for_each_entry(c, &head->vm_list, vm_list) {
-		if (c->vm_start == addr)
-			goto out;
-	}
-	c = NULL;
-out:
-	return c;
-}
-
-void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
-		gfp_t gfp, unsigned long attrs)
-{
-	struct page *page;
-	struct arch_vm_region *c;
-	unsigned long order;
-	u64 mask = ~0ULL, limit;
-	pgprot_t prot = pgprot_noncached(PAGE_KERNEL);
-
-	if (!consistent_pte) {
-		pr_err("%s: not initialized\n", __func__);
-		dump_stack();
-		return NULL;
-	}
-
-	if (dev) {
-		mask = dev->coherent_dma_mask;
-
-		/*
-		 * Sanity check the DMA mask - it must be non-zero, and
-		 * must be able to be satisfied by a DMA allocation.
-		 */
-		if (mask == 0) {
-			dev_warn(dev, "coherent DMA mask is unset\n");
-			goto no_page;
-		}
-
-	}
-
-	/*
-	 * Sanity check the allocation size.
-	 */
-	size = PAGE_ALIGN(size);
-	limit = (mask + 1) & ~mask;
-	if ((limit && size >= limit) ||
-	    size >= (CONSISTENT_END - CONSISTENT_BASE)) {
-		pr_warn("coherent allocation too big "
-			"(requested %#x mask %#llx)\n", size, mask);
-		goto no_page;
-	}
-
-	order = get_order(size);
-
-	if (mask != 0xffffffff)
-		gfp |= GFP_DMA;
-
-	page = alloc_pages(gfp, order);
-	if (!page)
-		goto no_page;
-
-	/*
-	 * Invalidate any data that might be lurking in the
-	 * kernel direct-mapped region for device DMA.
-	 */
-	{
-		unsigned long kaddr = (unsigned long)page_address(page);
-		memset(page_address(page), 0, size);
-		cpu_dma_wbinval_range(kaddr, kaddr + size);
-	}
-
-	/*
-	 * Allocate a virtual address in the consistent mapping region.
-	 */
-	c = vm_region_alloc(&consistent_head, size,
-			    gfp & ~(__GFP_DMA | __GFP_HIGHMEM));
-	if (c) {
-		pte_t *pte = consistent_pte + CONSISTENT_OFFSET(c->vm_start);
-		struct page *end = page + (1 << order);
-
-		c->vm_pages = page;
-
-		/*
-		 * Set the "dma handle"
-		 */
-		*handle = page_to_phys(page);
-
-		do {
-			BUG_ON(!pte_none(*pte));
-
-			/*
-			 * x86 does not mark the pages reserved...
-			 */
-			SetPageReserved(page);
-			set_pte(pte, mk_pte(page, prot));
-			page++;
-			pte++;
-		} while (size -= PAGE_SIZE);
-
-		/*
-		 * Free the otherwise unused pages.
-		 */
-		while (page < end) {
-			__free_page(page);
-			page++;
-		}
-
-		return (void *)c->vm_start;
-	}
-
-	if (page)
-		__free_pages(page, order);
-no_page:
-	*handle = ~0;
-	return NULL;
-}
-
-void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
-		dma_addr_t handle, unsigned long attrs)
-{
-	struct arch_vm_region *c;
-	unsigned long flags, addr;
-	pte_t *ptep;
-
-	size = PAGE_ALIGN(size);
-
-	raw_spin_lock_irqsave(&consistent_lock, flags);
-
-	c = vm_region_find(&consistent_head, (unsigned long)cpu_addr);
-	if (!c)
-		goto no_area;
-
-	if ((c->vm_end - c->vm_start) != size) {
-		pr_err("%s: freeing wrong coherent size (%ld != %d)\n",
-		       __func__, c->vm_end - c->vm_start, size);
-		dump_stack();
-		size = c->vm_end - c->vm_start;
-	}
-
-	ptep = consistent_pte + CONSISTENT_OFFSET(c->vm_start);
-	addr = c->vm_start;
-	do {
-		pte_t pte = ptep_get_and_clear(&init_mm, addr, ptep);
-		unsigned long pfn;
-
-		ptep++;
-		addr += PAGE_SIZE;
-
-		if (!pte_none(pte) && pte_present(pte)) {
-			pfn = pte_pfn(pte);
-
-			if (pfn_valid(pfn)) {
-				struct page *page = pfn_to_page(pfn);
-
-				/*
-				 * x86 does not mark the pages reserved...
-				 */
-				ClearPageReserved(page);
-
-				__free_page(page);
-				continue;
-			}
-		}
-
-		pr_crit("%s: bad page in kernel page table\n", __func__);
-	} while (size -= PAGE_SIZE);
-
-	flush_tlb_kernel_range(c->vm_start, c->vm_end);
-
-	list_del(&c->vm_list);
-
-	raw_spin_unlock_irqrestore(&consistent_lock, flags);
-
-	kfree(c);
-	return;
-
-no_area:
-	raw_spin_unlock_irqrestore(&consistent_lock, flags);
-	pr_err("%s: trying to free invalid coherent area: %p\n",
-	       __func__, cpu_addr);
-	dump_stack();
-}
-
-/*
- * Initialise the consistent memory allocation.
- */
-static int __init consistent_init(void)
-{
-	pgd_t *pgd;
-	pmd_t *pmd;
-	pte_t *pte;
-	int ret = 0;
-
-	do {
-		pgd = pgd_offset(&init_mm, CONSISTENT_BASE);
-		pmd = pmd_alloc(&init_mm, pgd, CONSISTENT_BASE);
-		if (!pmd) {
-			pr_err("%s: no pmd tables\n", __func__);
-			ret = -ENOMEM;
-			break;
-		}
-		/* The first level mapping may be created in somewhere.
-		 * It's not necessary to warn here. */
-		/* WARN_ON(!pmd_none(*pmd)); */
-
-		pte = pte_alloc_kernel(pmd, CONSISTENT_BASE);
-		if (!pte) {
-			ret = -ENOMEM;
-			break;
-		}
-
-		consistent_pte = pte;
-	} while (0);
-
-	return ret;
-}
-
-core_initcall(consistent_init);
-
 static inline void cache_op(phys_addr_t paddr, size_t size,
 		void (*fn)(unsigned long start, unsigned long end))
 {
@@ -389,3 +75,14 @@ void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
 		BUG();
 	}
 }
+
+void arch_dma_prep_coherent(struct page *page, size_t size)
+{
+	cache_op(page_to_phys(page), size, cpu_dma_wbinval_range);
+}
+
+static int __init atomic_pool_init(void)
+{
+	return dma_atomic_pool_init(GFP_KERNEL, pgprot_noncached(PAGE_KERNEL));
+}
+postcore_initcall(atomic_pool_init);
diff --git a/arch/nios2/Kconfig b/arch/nios2/Kconfig
index 26a9c760a98b..44b5da37e8bd 100644
--- a/arch/nios2/Kconfig
+++ b/arch/nios2/Kconfig
@@ -4,6 +4,7 @@ config NIOS2
 	select ARCH_32BIT_OFF_T
 	select ARCH_HAS_SYNC_DMA_FOR_CPU
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
+	select ARCH_HAS_UNCACHED_SEGMENT
 	select ARCH_NO_SWAP
 	select TIMER_OF
 	select GENERIC_ATOMIC64
diff --git a/arch/nios2/include/asm/page.h b/arch/nios2/include/asm/page.h
index f1fbdc47bdaf..79fcac61f6ef 100644
--- a/arch/nios2/include/asm/page.h
+++ b/arch/nios2/include/asm/page.h
@@ -101,12 +101,6 @@ static inline bool pfn_valid(unsigned long pfn)
 # define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
 
-# define UNCAC_ADDR(addr)	\
-	((void *)((unsigned)(addr) | CONFIG_NIOS2_IO_REGION_BASE))
-# define CAC_ADDR(addr)		\
-	((void *)(((unsigned)(addr) & ~CONFIG_NIOS2_IO_REGION_BASE) |	\
-		CONFIG_NIOS2_KERNEL_REGION_BASE))
-
 #include <asm-generic/memory_model.h>
 
 #include <asm-generic/getorder.h>
diff --git a/arch/nios2/mm/dma-mapping.c b/arch/nios2/mm/dma-mapping.c
index 4af9e5b5ba1c..9cb238664584 100644
--- a/arch/nios2/mm/dma-mapping.c
+++ b/arch/nios2/mm/dma-mapping.c
@@ -60,32 +60,28 @@ void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		gfp_t gfp, unsigned long attrs)
+void arch_dma_prep_coherent(struct page *page, size_t size)
 {
-	void *ret;
+	unsigned long start = (unsigned long)page_address(page);
 
-	/* optimized page clearing */
-	gfp |= __GFP_ZERO;
-
-	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
-		gfp |= GFP_DMA;
-
-	ret = (void *) __get_free_pages(gfp, get_order(size));
-	if (ret != NULL) {
-		*dma_handle = virt_to_phys(ret);
-		flush_dcache_range((unsigned long) ret,
-			(unsigned long) ret + size);
-		ret = UNCAC_ADDR(ret);
-	}
-
-	return ret;
+	flush_dcache_range(start, start + size);
 }
 
-void arch_dma_free(struct device *dev, size_t size, void *vaddr,
-		dma_addr_t dma_handle, unsigned long attrs)
+void *uncached_kernel_address(void *ptr)
 {
-	unsigned long addr = (unsigned long) CAC_ADDR((unsigned long) vaddr);
+	unsigned long addr = (unsigned long)ptr;
 
-	free_pages(addr, get_order(size));
+	addr |= CONFIG_NIOS2_IO_REGION_BASE;
+
+	return (void *)ptr;
+}
+
+void *cached_kernel_address(void *ptr)
+{
+	unsigned long addr = (unsigned long)ptr;
+
+	addr &= ~CONFIG_NIOS2_IO_REGION_BASE;
+	addr |= CONFIG_NIOS2_KERNEL_REGION_BASE;
+
+	return (void *)ptr;
 }
diff --git a/arch/openrisc/kernel/dma.c b/arch/openrisc/kernel/dma.c
index 43e340c4cd9c..b41a79fcdbd9 100644
--- a/arch/openrisc/kernel/dma.c
+++ b/arch/openrisc/kernel/dma.c
@@ -94,15 +94,13 @@ arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 
 	va = (unsigned long)page;
 
-	if ((attrs & DMA_ATTR_NON_CONSISTENT) == 0) {
-		/*
-		 * We need to iterate through the pages, clearing the dcache for
-		 * them and setting the cache-inhibit bit.
-		 */
-		if (walk_page_range(va, va + size, &walk)) {
-			free_pages_exact(page, size);
-			return NULL;
-		}
+	/*
+	 * We need to iterate through the pages, clearing the dcache for
+	 * them and setting the cache-inhibit bit.
+	 */
+	if (walk_page_range(va, va + size, &walk)) {
+		free_pages_exact(page, size);
+		return NULL;
 	}
 
 	return (void *)va;
@@ -118,10 +116,8 @@ arch_dma_free(struct device *dev, size_t size, void *vaddr,
 		.mm = &init_mm
 	};
 
-	if ((attrs & DMA_ATTR_NON_CONSISTENT) == 0) {
-		/* walk_page_range shouldn't be able to fail here */
-		WARN_ON(walk_page_range(va, va + size, &walk));
-	}
+	/* walk_page_range shouldn't be able to fail here */
+	WARN_ON(walk_page_range(va, va + size, &walk));
 
 	free_pages_exact(vaddr, size);
 }
diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c
index 239162355b58..ca35d9a76e50 100644
--- a/arch/parisc/kernel/pci-dma.c
+++ b/arch/parisc/kernel/pci-dma.c
@@ -394,17 +394,20 @@ pcxl_dma_init(void)
 
 __initcall(pcxl_dma_init);
 
-static void *pcxl_dma_alloc(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flag, unsigned long attrs)
+void *arch_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
 	unsigned long vaddr;
 	unsigned long paddr;
 	int order;
 
+	if (boot_cpu_data.cpu_type != pcxl2 && boot_cpu_data.cpu_type != pcxl)
+		return NULL;
+
 	order = get_order(size);
 	size = 1 << (order + PAGE_SHIFT);
 	vaddr = pcxl_alloc_range(size);
-	paddr = __get_free_pages(flag | __GFP_ZERO, order);
+	paddr = __get_free_pages(gfp | __GFP_ZERO, order);
 	flush_kernel_dcache_range(paddr, size);
 	paddr = __pa(paddr);
 	map_uncached_pages(vaddr, size, paddr);
@@ -421,44 +424,19 @@ static void *pcxl_dma_alloc(struct device *dev, size_t size,
 	return (void *)vaddr;
 }
 
-static void *pcx_dma_alloc(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t flag, unsigned long attrs)
-{
-	void *addr;
-
-	if ((attrs & DMA_ATTR_NON_CONSISTENT) == 0)
-		return NULL;
-
-	addr = (void *)__get_free_pages(flag | __GFP_ZERO, get_order(size));
-	if (addr)
-		*dma_handle = (dma_addr_t)virt_to_phys(addr);
-
-	return addr;
-}
-
-void *arch_dma_alloc(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
-{
-
-	if (boot_cpu_data.cpu_type == pcxl2 || boot_cpu_data.cpu_type == pcxl)
-		return pcxl_dma_alloc(dev, size, dma_handle, gfp, attrs);
-	else
-		return pcx_dma_alloc(dev, size, dma_handle, gfp, attrs);
-}
-
 void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 		dma_addr_t dma_handle, unsigned long attrs)
 {
 	int order = get_order(size);
 
-	if (boot_cpu_data.cpu_type == pcxl2 || boot_cpu_data.cpu_type == pcxl) {
-		size = 1 << (order + PAGE_SHIFT);
-		unmap_uncached_pages((unsigned long)vaddr, size);
-		pcxl_free_range((unsigned long)vaddr, size);
+	WARN_ON_ONCE(boot_cpu_data.cpu_type != pcxl2 &&
+		     boot_cpu_data.cpu_type != pcxl);
 
-		vaddr = __va(dma_handle);
-	}
-	free_pages((unsigned long)vaddr, get_order(size));
+	size = 1 << (order + PAGE_SHIFT);
+	unmap_uncached_pages((unsigned long)vaddr, size);
+	pcxl_free_range((unsigned long)vaddr, size);
+
+	free_pages((unsigned long)__va(dma_handle), order);
 }
 
 void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
diff --git a/arch/xtensa/kernel/pci-dma.c b/arch/xtensa/kernel/pci-dma.c
index a87f8a308cc1..65f05776d827 100644
--- a/arch/xtensa/kernel/pci-dma.c
+++ b/arch/xtensa/kernel/pci-dma.c
@@ -163,10 +163,6 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
 
 	*handle = phys_to_dma(dev, page_to_phys(page));
 
-	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
-		return page;
-	}
-
 #ifdef CONFIG_MMU
 	if (PageHighMem(page)) {
 		void *p;
@@ -192,9 +188,7 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
 	struct page *page;
 
-	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
-		page = vaddr;
-	} else if (platform_vaddr_uncached(vaddr)) {
+	if (platform_vaddr_uncached(vaddr)) {
 		page = virt_to_page(platform_vaddr_to_cached(vaddr));
 	} else {
 #ifdef CONFIG_MMU
diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
index f802255219d3..a7f9c3edbcb2 100644
--- a/drivers/iommu/dma-iommu.c
+++ b/drivers/iommu/dma-iommu.c
@@ -951,8 +951,8 @@ static void __iommu_dma_free(struct device *dev, size_t size, void *cpu_addr)
 
 	if (pages)
 		__iommu_dma_free_pages(pages, count);
-	if (page && !dma_release_from_contiguous(dev, page, count))
-		__free_pages(page, get_order(alloc_size));
+	if (page)
+		dma_free_contiguous(dev, page, alloc_size);
 }
 
 static void iommu_dma_free(struct device *dev, size_t size, void *cpu_addr,
@@ -970,12 +970,7 @@ static void *iommu_dma_alloc_pages(struct device *dev, size_t size,
 	struct page *page = NULL;
 	void *cpu_addr;
 
-	if (gfpflags_allow_blocking(gfp))
-		page = dma_alloc_from_contiguous(dev, alloc_size >> PAGE_SHIFT,
-						 get_order(alloc_size),
-						 gfp & __GFP_NOWARN);
-	if (!page)
-		page = alloc_pages(gfp, get_order(alloc_size));
+	page = dma_alloc_contiguous(dev, alloc_size, gfp);
 	if (!page)
 		return NULL;
 
@@ -997,8 +992,7 @@ static void *iommu_dma_alloc_pages(struct device *dev, size_t size,
 	memset(cpu_addr, 0, alloc_size);
 	return cpu_addr;
 out_free_pages:
-	if (!dma_release_from_contiguous(dev, page, alloc_size >> PAGE_SHIFT))
-		__free_pages(page, get_order(alloc_size));
+	dma_free_contiguous(dev, page, alloc_size);
 	return NULL;
 }
 
diff --git a/drivers/usb/Kconfig b/drivers/usb/Kconfig
index 94573fb68304..6e59d370ef81 100644
--- a/drivers/usb/Kconfig
+++ b/drivers/usb/Kconfig
@@ -45,6 +45,7 @@ config USB_ARCH_HAS_HCD
 config USB
 	tristate "Support for Host-side USB"
 	depends on USB_ARCH_HAS_HCD
+	select GENERIC_ALLOCATOR
 	select USB_COMMON
 	select NLS  # for UTF-8 strings
 	---help---
diff --git a/drivers/usb/core/buffer.c b/drivers/usb/core/buffer.c
index f641342cdec0..1359b78a624e 100644
--- a/drivers/usb/core/buffer.c
+++ b/drivers/usb/core/buffer.c
@@ -16,6 +16,7 @@
 #include <linux/io.h>
 #include <linux/dma-mapping.h>
 #include <linux/dmapool.h>
+#include <linux/genalloc.h>
 #include <linux/usb.h>
 #include <linux/usb/hcd.h>
 
@@ -67,7 +68,7 @@ int hcd_buffer_create(struct usb_hcd *hcd)
 
 	if (!IS_ENABLED(CONFIG_HAS_DMA) ||
 	    (!is_device_dma_capable(hcd->self.sysdev) &&
-	     !(hcd->driver->flags & HCD_LOCAL_MEM)))
+	     !hcd->localmem_pool))
 		return 0;
 
 	for (i = 0; i < HCD_BUFFER_POOLS; i++) {
@@ -124,10 +125,12 @@ void *hcd_buffer_alloc(
 	if (size == 0)
 		return NULL;
 
+	if (hcd->localmem_pool)
+		return gen_pool_dma_alloc(hcd->localmem_pool, size, dma);
+
 	/* some USB hosts just use PIO */
 	if (!IS_ENABLED(CONFIG_HAS_DMA) ||
-	    (!is_device_dma_capable(bus->sysdev) &&
-	     !(hcd->driver->flags & HCD_LOCAL_MEM))) {
+	    !is_device_dma_capable(bus->sysdev)) {
 		*dma = ~(dma_addr_t) 0;
 		return kmalloc(size, mem_flags);
 	}
@@ -152,9 +155,13 @@ void hcd_buffer_free(
 	if (!addr)
 		return;
 
+	if (hcd->localmem_pool) {
+		gen_pool_free(hcd->localmem_pool, (unsigned long)addr, size);
+		return;
+	}
+
 	if (!IS_ENABLED(CONFIG_HAS_DMA) ||
-	    (!is_device_dma_capable(bus->sysdev) &&
-	     !(hcd->driver->flags & HCD_LOCAL_MEM))) {
+	    !is_device_dma_capable(bus->sysdev)) {
 		kfree(addr);
 		return;
 	}
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 94d22551fc1b..88533938ce19 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -29,6 +29,8 @@
 #include <linux/workqueue.h>
 #include <linux/pm_runtime.h>
 #include <linux/types.h>
+#include <linux/genalloc.h>
+#include <linux/io.h>
 
 #include <linux/phy/phy.h>
 #include <linux/usb.h>
@@ -1345,14 +1347,14 @@ EXPORT_SYMBOL_GPL(usb_hcd_unlink_urb_from_ep);
  * using regular system memory - like pci devices doing bus mastering.
  *
  * To support host controllers with limited dma capabilities we provide dma
- * bounce buffers. This feature can be enabled using the HCD_LOCAL_MEM flag.
+ * bounce buffers. This feature can be enabled by initializing
+ * hcd->localmem_pool using usb_hcd_setup_local_mem().
  * For this to work properly the host controller code must first use the
  * function dma_declare_coherent_memory() to point out which memory area
  * that should be used for dma allocations.
  *
- * The HCD_LOCAL_MEM flag then tells the usb code to allocate all data for
- * dma using dma_alloc_coherent() which in turn allocates from the memory
- * area pointed out with dma_declare_coherent_memory().
+ * The initialized hcd->localmem_pool then tells the usb code to allocate all
+ * data for dma using the genalloc API.
  *
  * So, to summarize...
  *
@@ -1362,9 +1364,6 @@ EXPORT_SYMBOL_GPL(usb_hcd_unlink_urb_from_ep);
  *   (a) "normal" kernel memory is no good, and
  *   (b) there's not enough to share
  *
- * - The only *portable* hook for such stuff in the
- *   DMA framework is dma_declare_coherent_memory()
- *
  * - So we use that, even though the primary requirement
  *   is that the memory be "local" (hence addressable
  *   by that device), not "coherent".
@@ -1531,7 +1530,7 @@ int usb_hcd_map_urb_for_dma(struct usb_hcd *hcd, struct urb *urb,
 						urb->setup_dma))
 				return -EAGAIN;
 			urb->transfer_flags |= URB_SETUP_MAP_SINGLE;
-		} else if (hcd->driver->flags & HCD_LOCAL_MEM) {
+		} else if (hcd->localmem_pool) {
 			ret = hcd_alloc_coherent(
 					urb->dev->bus, mem_flags,
 					&urb->setup_dma,
@@ -1601,7 +1600,7 @@ int usb_hcd_map_urb_for_dma(struct usb_hcd *hcd, struct urb *urb,
 				else
 					urb->transfer_flags |= URB_DMA_MAP_SINGLE;
 			}
-		} else if (hcd->driver->flags & HCD_LOCAL_MEM) {
+		} else if (hcd->localmem_pool) {
 			ret = hcd_alloc_coherent(
 					urb->dev->bus, mem_flags,
 					&urb->transfer_dma,
@@ -3039,6 +3038,40 @@ usb_hcd_platform_shutdown(struct platform_device *dev)
 }
 EXPORT_SYMBOL_GPL(usb_hcd_platform_shutdown);
 
+int usb_hcd_setup_local_mem(struct usb_hcd *hcd, phys_addr_t phys_addr,
+			    dma_addr_t dma, size_t size)
+{
+	int err;
+	void *local_mem;
+
+	hcd->localmem_pool = devm_gen_pool_create(hcd->self.sysdev, 4,
+						  dev_to_node(hcd->self.sysdev),
+						  dev_name(hcd->self.sysdev));
+	if (IS_ERR(hcd->localmem_pool))
+		return PTR_ERR(hcd->localmem_pool);
+
+	local_mem = devm_memremap(hcd->self.sysdev, phys_addr,
+				  size, MEMREMAP_WC);
+	if (!local_mem)
+		return -ENOMEM;
+
+	/*
+	 * Here we pass a dma_addr_t but the arg type is a phys_addr_t.
+	 * It's not backed by system memory and thus there's no kernel mapping
+	 * for it.
+	 */
+	err = gen_pool_add_virt(hcd->localmem_pool, (unsigned long)local_mem,
+				dma, size, dev_to_node(hcd->self.sysdev));
+	if (err < 0) {
+		dev_err(hcd->self.sysdev, "gen_pool_add_virt failed with %d\n",
+			err);
+		return err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(usb_hcd_setup_local_mem);
+
 /*-------------------------------------------------------------------------*/
 
 #if IS_ENABLED(CONFIG_USB_MON)
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index cdafa97f632d..9da7e22848c9 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -559,7 +559,7 @@ static int ehci_init(struct usb_hcd *hcd)
 	ehci->command = temp;
 
 	/* Accept arbitrarily long scatter-gather lists */
-	if (!(hcd->driver->flags & HCD_LOCAL_MEM))
+	if (!hcd->localmem_pool)
 		hcd->self.sg_tablesize = ~0;
 
 	/* Prepare for unlinking active QHs */
diff --git a/drivers/usb/host/fotg210-hcd.c b/drivers/usb/host/fotg210-hcd.c
index e835a22b12af..77cc36efae95 100644
--- a/drivers/usb/host/fotg210-hcd.c
+++ b/drivers/usb/host/fotg210-hcd.c
@@ -4996,7 +4996,7 @@ static int hcd_fotg210_init(struct usb_hcd *hcd)
 	fotg210->command = temp;
 
 	/* Accept arbitrarily long scatter-gather lists */
-	if (!(hcd->driver->flags & HCD_LOCAL_MEM))
+	if (!hcd->localmem_pool)
 		hcd->self.sg_tablesize = ~0;
 	return 0;
 }
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 210181fd98d2..b457fdaff297 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -40,6 +40,7 @@
 #include <linux/dmapool.h>
 #include <linux/workqueue.h>
 #include <linux/debugfs.h>
+#include <linux/genalloc.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
@@ -447,7 +448,7 @@ static int ohci_init (struct ohci_hcd *ohci)
 	struct usb_hcd *hcd = ohci_to_hcd(ohci);
 
 	/* Accept arbitrarily long scatter-gather lists */
-	if (!(hcd->driver->flags & HCD_LOCAL_MEM))
+	if (!hcd->localmem_pool)
 		hcd->self.sg_tablesize = ~0;
 
 	if (distrust_firmware)
@@ -505,8 +506,15 @@ static int ohci_init (struct ohci_hcd *ohci)
 	timer_setup(&ohci->io_watchdog, io_watchdog_func, 0);
 	ohci->prev_frame_no = IO_WATCHDOG_OFF;
 
-	ohci->hcca = dma_alloc_coherent (hcd->self.controller,
-			sizeof(*ohci->hcca), &ohci->hcca_dma, GFP_KERNEL);
+	if (hcd->localmem_pool)
+		ohci->hcca = gen_pool_dma_alloc_align(hcd->localmem_pool,
+						sizeof(*ohci->hcca),
+						&ohci->hcca_dma, 256);
+	else
+		ohci->hcca = dma_alloc_coherent(hcd->self.controller,
+						sizeof(*ohci->hcca),
+						&ohci->hcca_dma,
+						GFP_KERNEL);
 	if (!ohci->hcca)
 		return -ENOMEM;
 
@@ -990,9 +998,14 @@ static void ohci_stop (struct usb_hcd *hcd)
 	remove_debug_files (ohci);
 	ohci_mem_cleanup (ohci);
 	if (ohci->hcca) {
-		dma_free_coherent (hcd->self.controller,
-				sizeof *ohci->hcca,
-				ohci->hcca, ohci->hcca_dma);
+		if (hcd->localmem_pool)
+			gen_pool_free(hcd->localmem_pool,
+				      (unsigned long)ohci->hcca,
+				      sizeof(*ohci->hcca));
+		else
+			dma_free_coherent(hcd->self.controller,
+					  sizeof(*ohci->hcca),
+					  ohci->hcca, ohci->hcca_dma);
 		ohci->hcca = NULL;
 		ohci->hcca_dma = 0;
 	}
diff --git a/drivers/usb/host/ohci-mem.c b/drivers/usb/host/ohci-mem.c
index 3965ac0341eb..1425335c6baf 100644
--- a/drivers/usb/host/ohci-mem.c
+++ b/drivers/usb/host/ohci-mem.c
@@ -36,6 +36,13 @@ static void ohci_hcd_init (struct ohci_hcd *ohci)
 
 static int ohci_mem_init (struct ohci_hcd *ohci)
 {
+	/*
+	 * HCs with local memory allocate from localmem_pool so there's
+	 * no need to create the below dma pools.
+	 */
+	if (ohci_to_hcd(ohci)->localmem_pool)
+		return 0;
+
 	ohci->td_cache = dma_pool_create ("ohci_td",
 		ohci_to_hcd(ohci)->self.controller,
 		sizeof (struct td),
@@ -84,8 +91,13 @@ td_alloc (struct ohci_hcd *hc, gfp_t mem_flags)
 {
 	dma_addr_t	dma;
 	struct td	*td;
+	struct usb_hcd	*hcd = ohci_to_hcd(hc);
 
-	td = dma_pool_zalloc (hc->td_cache, mem_flags, &dma);
+	if (hcd->localmem_pool)
+		td = gen_pool_dma_zalloc_align(hcd->localmem_pool,
+				sizeof(*td), &dma, 32);
+	else
+		td = dma_pool_zalloc(hc->td_cache, mem_flags, &dma);
 	if (td) {
 		/* in case hc fetches it, make it look dead */
 		td->hwNextTD = cpu_to_hc32 (hc, dma);
@@ -99,6 +111,7 @@ static void
 td_free (struct ohci_hcd *hc, struct td *td)
 {
 	struct td	**prev = &hc->td_hash [TD_HASH_FUNC (td->td_dma)];
+	struct usb_hcd	*hcd = ohci_to_hcd(hc);
 
 	while (*prev && *prev != td)
 		prev = &(*prev)->td_hash;
@@ -106,7 +119,12 @@ td_free (struct ohci_hcd *hc, struct td *td)
 		*prev = td->td_hash;
 	else if ((td->hwINFO & cpu_to_hc32(hc, TD_DONE)) != 0)
 		ohci_dbg (hc, "no hash for td %p\n", td);
-	dma_pool_free (hc->td_cache, td, td->td_dma);
+
+	if (hcd->localmem_pool)
+		gen_pool_free(hcd->localmem_pool, (unsigned long)td,
+			      sizeof(*td));
+	else
+		dma_pool_free(hc->td_cache, td, td->td_dma);
 }
 
 /*-------------------------------------------------------------------------*/
@@ -117,8 +135,13 @@ ed_alloc (struct ohci_hcd *hc, gfp_t mem_flags)
 {
 	dma_addr_t	dma;
 	struct ed	*ed;
+	struct usb_hcd	*hcd = ohci_to_hcd(hc);
 
-	ed = dma_pool_zalloc (hc->ed_cache, mem_flags, &dma);
+	if (hcd->localmem_pool)
+		ed = gen_pool_dma_zalloc_align(hcd->localmem_pool,
+				sizeof(*ed), &dma, 16);
+	else
+		ed = dma_pool_zalloc(hc->ed_cache, mem_flags, &dma);
 	if (ed) {
 		INIT_LIST_HEAD (&ed->td_list);
 		ed->dma = dma;
@@ -129,6 +152,12 @@ ed_alloc (struct ohci_hcd *hc, gfp_t mem_flags)
 static void
 ed_free (struct ohci_hcd *hc, struct ed *ed)
 {
-	dma_pool_free (hc->ed_cache, ed, ed->dma);
+	struct usb_hcd	*hcd = ohci_to_hcd(hc);
+
+	if (hcd->localmem_pool)
+		gen_pool_free(hcd->localmem_pool, (unsigned long)ed,
+			      sizeof(*ed));
+	else
+		dma_pool_free(hc->ed_cache, ed, ed->dma);
 }
 
diff --git a/drivers/usb/host/ohci-sm501.c b/drivers/usb/host/ohci-sm501.c
index c26228c25f99..c158cda9e4b9 100644
--- a/drivers/usb/host/ohci-sm501.c
+++ b/drivers/usb/host/ohci-sm501.c
@@ -49,7 +49,7 @@ static const struct hc_driver ohci_sm501_hc_driver = {
 	 * generic hardware linkage
 	 */
 	.irq =			ohci_irq,
-	.flags =		HCD_USB11 | HCD_MEMORY | HCD_LOCAL_MEM,
+	.flags =		HCD_USB11 | HCD_MEMORY,
 
 	/*
 	 * basic lifecycle operations
@@ -110,40 +110,18 @@ static int ohci_hcd_sm501_drv_probe(struct platform_device *pdev)
 		goto err0;
 	}
 
-	/* The sm501 chip is equipped with local memory that may be used
-	 * by on-chip devices such as the video controller and the usb host.
-	 * This driver uses dma_declare_coherent_memory() to make sure
-	 * usb allocations with dma_alloc_coherent() allocate from
-	 * this local memory. The dma_handle returned by dma_alloc_coherent()
-	 * will be an offset starting from 0 for the first local memory byte.
-	 *
-	 * So as long as data is allocated using dma_alloc_coherent() all is
-	 * fine. This is however not always the case - buffers may be allocated
-	 * using kmalloc() - so the usb core needs to be told that it must copy
-	 * data into our local memory if the buffers happen to be placed in
-	 * regular memory. The HCD_LOCAL_MEM flag does just that.
-	 */
-
-	retval = dma_declare_coherent_memory(dev, mem->start,
-					 mem->start - mem->parent->start,
-					 resource_size(mem));
-	if (retval) {
-		dev_err(dev, "cannot declare coherent memory\n");
-		goto err1;
-	}
-
 	/* allocate, reserve and remap resources for registers */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (res == NULL) {
 		dev_err(dev, "no resource definition for registers\n");
 		retval = -ENOENT;
-		goto err2;
+		goto err1;
 	}
 
 	hcd = usb_create_hcd(driver, &pdev->dev, dev_name(&pdev->dev));
 	if (!hcd) {
 		retval = -ENOMEM;
-		goto err2;
+		goto err1;
 	}
 
 	hcd->rsrc_start = res->start;
@@ -164,6 +142,25 @@ static int ohci_hcd_sm501_drv_probe(struct platform_device *pdev)
 
 	ohci_hcd_init(hcd_to_ohci(hcd));
 
+	/* The sm501 chip is equipped with local memory that may be used
+	 * by on-chip devices such as the video controller and the usb host.
+	 * This driver uses genalloc so that usb allocations with
+	 * gen_pool_dma_alloc() allocate from this local memory. The dma_handle
+	 * returned by gen_pool_dma_alloc() will be an offset starting from 0
+	 * for the first local memory byte.
+	 *
+	 * So as long as data is allocated using gen_pool_dma_alloc() all is
+	 * fine. This is however not always the case - buffers may be allocated
+	 * using kmalloc() - so the usb core needs to be told that it must copy
+	 * data into our local memory if the buffers happen to be placed in
+	 * regular memory. A non-null hcd->localmem_pool initialized by the
+	 * the call to usb_hcd_setup_local_mem() below does just that.
+	 */
+
+	if (usb_hcd_setup_local_mem(hcd, mem->start,
+				    mem->start - mem->parent->start,
+				    resource_size(mem)) < 0)
+		goto err5;
 	retval = usb_add_hcd(hcd, irq, IRQF_SHARED);
 	if (retval)
 		goto err5;
@@ -181,8 +178,6 @@ err4:
 	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
 err3:
 	usb_put_hcd(hcd);
-err2:
-	dma_release_declared_memory(dev);
 err1:
 	release_mem_region(mem->start, resource_size(mem));
 err0:
@@ -197,7 +192,6 @@ static int ohci_hcd_sm501_drv_remove(struct platform_device *pdev)
 	usb_remove_hcd(hcd);
 	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
 	usb_put_hcd(hcd);
-	dma_release_declared_memory(&pdev->dev);
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 1);
 	if (mem)
 		release_mem_region(mem->start, resource_size(mem));
diff --git a/drivers/usb/host/ohci-tmio.c b/drivers/usb/host/ohci-tmio.c
index f88a0370659f..d5a293a707b6 100644
--- a/drivers/usb/host/ohci-tmio.c
+++ b/drivers/usb/host/ohci-tmio.c
@@ -153,7 +153,7 @@ static const struct hc_driver ohci_tmio_hc_driver = {
 
 	/* generic hardware linkage */
 	.irq =			ohci_irq,
-	.flags =		HCD_USB11 | HCD_MEMORY | HCD_LOCAL_MEM,
+	.flags =		HCD_USB11 | HCD_MEMORY,
 
 	/* basic lifecycle operations */
 	.start =		ohci_tmio_start,
@@ -224,11 +224,6 @@ static int ohci_hcd_tmio_drv_probe(struct platform_device *dev)
 		goto err_ioremap_regs;
 	}
 
-	ret = dma_declare_coherent_memory(&dev->dev, sram->start, sram->start,
-				resource_size(sram));
-	if (ret)
-		goto err_dma_declare;
-
 	if (cell->enable) {
 		ret = cell->enable(dev);
 		if (ret)
@@ -239,6 +234,11 @@ static int ohci_hcd_tmio_drv_probe(struct platform_device *dev)
 	ohci = hcd_to_ohci(hcd);
 	ohci_hcd_init(ohci);
 
+	ret = usb_hcd_setup_local_mem(hcd, sram->start, sram->start,
+				      resource_size(sram));
+	if (ret < 0)
+		goto err_enable;
+
 	ret = usb_add_hcd(hcd, irq, 0);
 	if (ret)
 		goto err_add_hcd;
@@ -254,8 +254,6 @@ err_add_hcd:
 	if (cell->disable)
 		cell->disable(dev);
 err_enable:
-	dma_release_declared_memory(&dev->dev);
-err_dma_declare:
 	iounmap(hcd->regs);
 err_ioremap_regs:
 	iounmap(tmio->ccr);
@@ -276,7 +274,6 @@ static int ohci_hcd_tmio_drv_remove(struct platform_device *dev)
 	tmio_stop_hc(dev);
 	if (cell->disable)
 		cell->disable(dev);
-	dma_release_declared_memory(&dev->dev);
 	iounmap(hcd->regs);
 	iounmap(tmio->ccr);
 	usb_put_hcd(hcd);
diff --git a/drivers/usb/host/ohci.h b/drivers/usb/host/ohci.h
index ef4813bfc5bf..b015b00774b2 100644
--- a/drivers/usb/host/ohci.h
+++ b/drivers/usb/host/ohci.h
@@ -385,6 +385,8 @@ struct ohci_hcd {
 
 	/*
 	 * memory management for queue data structures
+	 *
+	 * @td_cache and @ed_cache are %NULL if &usb_hcd.localmem_pool is used.
 	 */
 	struct dma_pool		*td_cache;
 	struct dma_pool		*ed_cache;
diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c
index 98deb5f64268..03bc59755123 100644
--- a/drivers/usb/host/uhci-hcd.c
+++ b/drivers/usb/host/uhci-hcd.c
@@ -581,7 +581,7 @@ static int uhci_start(struct usb_hcd *hcd)
 
 	hcd->uses_new_polling = 1;
 	/* Accept arbitrarily long scatter-gather lists */
-	if (!(hcd->driver->flags & HCD_LOCAL_MEM))
+	if (!hcd->localmem_pool)
 		hcd->self.sg_tablesize = ~0;
 
 	spin_lock_init(&uhci->lock);
diff --git a/drivers/video/fbdev/au1100fb.c b/drivers/video/fbdev/au1100fb.c
index 0adf0683cf08..99941ae1f3a1 100644
--- a/drivers/video/fbdev/au1100fb.c
+++ b/drivers/video/fbdev/au1100fb.c
@@ -340,14 +340,12 @@ int au1100fb_fb_pan_display(struct fb_var_screeninfo *var, struct fb_info *fbi)
  */
 int au1100fb_fb_mmap(struct fb_info *fbi, struct vm_area_struct *vma)
 {
-	struct au1100fb_device *fbdev;
+	struct au1100fb_device *fbdev = to_au1100fb_device(fbi);
 
-	fbdev = to_au1100fb_device(fbi);
-
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 	pgprot_val(vma->vm_page_prot) |= (6 << 9); //CCA=6
 
-	return vm_iomap_memory(vma, fbdev->fb_phys, fbdev->fb_len);
+	return dma_mmap_coherent(fbdev->dev, vma, fbdev->fb_mem, fbdev->fb_phys,
+			fbdev->fb_len);
 }
 
 static struct fb_ops au1100fb_ops =
@@ -412,7 +410,6 @@ static int au1100fb_drv_probe(struct platform_device *dev)
 {
 	struct au1100fb_device *fbdev;
 	struct resource *regs_res;
-	unsigned long page;
 	struct clk *c;
 
 	/* Allocate new device private */
@@ -424,6 +421,7 @@ static int au1100fb_drv_probe(struct platform_device *dev)
 		goto failed;
 
 	platform_set_drvdata(dev, (void *)fbdev);
+	fbdev->dev = &dev->dev;
 
 	/* Allocate region for our registers and map them */
 	regs_res = platform_get_resource(dev, IORESOURCE_MEM, 0);
@@ -472,20 +470,6 @@ static int au1100fb_drv_probe(struct platform_device *dev)
 	au1100fb_fix.smem_start = fbdev->fb_phys;
 	au1100fb_fix.smem_len = fbdev->fb_len;
 
-	/*
-	 * Set page reserved so that mmap will work. This is necessary
-	 * since we'll be remapping normal memory.
-	 */
-	for (page = (unsigned long)fbdev->fb_mem;
-	     page < PAGE_ALIGN((unsigned long)fbdev->fb_mem + fbdev->fb_len);
-	     page += PAGE_SIZE) {
-#ifdef CONFIG_DMA_NONCOHERENT
-		SetPageReserved(virt_to_page(CAC_ADDR((void *)page)));
-#else
-		SetPageReserved(virt_to_page(page));
-#endif
-	}
-
 	print_dbg("Framebuffer memory map at %p", fbdev->fb_mem);
 	print_dbg("phys=0x%08x, size=%dK", fbdev->fb_phys, fbdev->fb_len / 1024);
 
diff --git a/drivers/video/fbdev/au1100fb.h b/drivers/video/fbdev/au1100fb.h
index 9af19939a9c6..e7239bceefd3 100644
--- a/drivers/video/fbdev/au1100fb.h
+++ b/drivers/video/fbdev/au1100fb.h
@@ -110,6 +110,7 @@ struct au1100fb_device {
 	dma_addr_t    		fb_phys;
 	int			panel_idx;
 	struct clk		*lcdclk;
+	struct device		*dev;
 };
 
 /********************************************************************/
diff --git a/include/linux/dma-contiguous.h b/include/linux/dma-contiguous.h
index 6665fa03c0d1..c05d4e661489 100644
--- a/include/linux/dma-contiguous.h
+++ b/include/linux/dma-contiguous.h
@@ -50,6 +50,7 @@
 #ifdef __KERNEL__
 
 #include <linux/device.h>
+#include <linux/mm.h>
 
 struct cma;
 struct page;
@@ -111,6 +112,8 @@ struct page *dma_alloc_from_contiguous(struct device *dev, size_t count,
 				       unsigned int order, bool no_warn);
 bool dma_release_from_contiguous(struct device *dev, struct page *pages,
 				 int count);
+struct page *dma_alloc_contiguous(struct device *dev, size_t size, gfp_t gfp);
+void dma_free_contiguous(struct device *dev, struct page *page, size_t size);
 
 #else
 
@@ -153,6 +156,22 @@ bool dma_release_from_contiguous(struct device *dev, struct page *pages,
 	return false;
 }
 
+/* Use fallback alloc() and free() when CONFIG_DMA_CMA=n */
+static inline struct page *dma_alloc_contiguous(struct device *dev, size_t size,
+		gfp_t gfp)
+{
+	int node = dev ? dev_to_node(dev) : NUMA_NO_NODE;
+	size_t align = get_order(PAGE_ALIGN(size));
+
+	return alloc_pages_node(node, gfp, align);
+}
+
+static inline void dma_free_contiguous(struct device *dev, struct page *page,
+		size_t size)
+{
+	__free_pages(page, get_order(size));
+}
+
 #endif
 
 #endif
diff --git a/include/linux/dma-noncoherent.h b/include/linux/dma-noncoherent.h
index 9741767e400f..3813211a9aad 100644
--- a/include/linux/dma-noncoherent.h
+++ b/include/linux/dma-noncoherent.h
@@ -20,6 +20,22 @@ static inline bool dev_is_dma_coherent(struct device *dev)
 }
 #endif /* CONFIG_ARCH_HAS_DMA_COHERENCE_H */
 
+/*
+ * Check if an allocation needs to be marked uncached to be coherent.
+ */
+static __always_inline bool dma_alloc_need_uncached(struct device *dev,
+		unsigned long attrs)
+{
+	if (dev_is_dma_coherent(dev))
+		return false;
+	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING)
+		return false;
+	if (IS_ENABLED(CONFIG_DMA_NONCOHERENT_CACHE_SYNC) &&
+	    (attrs & DMA_ATTR_NON_CONSISTENT))
+		return false;
+	return true;
+}
+
 void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		gfp_t gfp, unsigned long attrs);
 void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
@@ -80,4 +96,7 @@ static inline void arch_dma_prep_coherent(struct page *page, size_t size)
 }
 #endif /* CONFIG_ARCH_HAS_DMA_PREP_COHERENT */
 
+void *uncached_kernel_address(void *addr);
+void *cached_kernel_address(void *addr);
+
 #endif /* _LINUX_DMA_NONCOHERENT_H */
diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h
index 205f62b8d291..4bd583bd6934 100644
--- a/include/linux/genalloc.h
+++ b/include/linux/genalloc.h
@@ -155,6 +155,15 @@ static inline unsigned long gen_pool_alloc(struct gen_pool *pool, size_t size)
 
 extern void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size,
 		dma_addr_t *dma);
+extern void *gen_pool_dma_alloc_algo(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, genpool_algo_t algo, void *data);
+extern void *gen_pool_dma_alloc_align(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, int align);
+extern void *gen_pool_dma_zalloc(struct gen_pool *pool, size_t size, dma_addr_t *dma);
+extern void *gen_pool_dma_zalloc_algo(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, genpool_algo_t algo, void *data);
+extern void *gen_pool_dma_zalloc_align(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, int align);
 extern void gen_pool_free_owner(struct gen_pool *pool, unsigned long addr,
 		size_t size, void **owner);
 static inline void gen_pool_free(struct gen_pool *pool, unsigned long addr,
diff --git a/include/linux/usb/hcd.h b/include/linux/usb/hcd.h
index bb57b5af4700..bab27ccc8ff5 100644
--- a/include/linux/usb/hcd.h
+++ b/include/linux/usb/hcd.h
@@ -216,6 +216,9 @@ struct usb_hcd {
 #define	HC_IS_RUNNING(state) ((state) & __ACTIVE)
 #define	HC_IS_SUSPENDED(state) ((state) & __SUSPEND)
 
+	/* memory pool for HCs having local memory, or %NULL */
+	struct gen_pool         *localmem_pool;
+
 	/* more shared queuing code would be good; it should support
 	 * smarter scheduling, handle transaction translators, etc;
 	 * input size of periodic table to an interrupt scheduler.
@@ -253,7 +256,6 @@ struct hc_driver {
 
 	int	flags;
 #define	HCD_MEMORY	0x0001		/* HC regs use memory (else I/O) */
-#define	HCD_LOCAL_MEM	0x0002		/* HC needs local memory */
 #define	HCD_SHARED	0x0004		/* Two (or more) usb_hcds share HW */
 #define	HCD_USB11	0x0010		/* USB 1.1 */
 #define	HCD_USB2	0x0020		/* USB 2.0 */
@@ -461,6 +463,8 @@ extern int usb_add_hcd(struct usb_hcd *hcd,
 		unsigned int irqnum, unsigned long irqflags);
 extern void usb_remove_hcd(struct usb_hcd *hcd);
 extern int usb_hcd_find_raw_port_number(struct usb_hcd *hcd, int port1);
+int usb_hcd_setup_local_mem(struct usb_hcd *hcd, phys_addr_t phys_addr,
+			    dma_addr_t dma, size_t size);
 
 struct platform_device;
 extern void usb_hcd_platform_shutdown(struct platform_device *dev);
diff --git a/kernel/dma/contiguous.c b/kernel/dma/contiguous.c
index b2a87905846d..bfc0c17f2a3d 100644
--- a/kernel/dma/contiguous.c
+++ b/kernel/dma/contiguous.c
@@ -214,6 +214,62 @@ bool dma_release_from_contiguous(struct device *dev, struct page *pages,
 	return cma_release(dev_get_cma_area(dev), pages, count);
 }
 
+/**
+ * dma_alloc_contiguous() - allocate contiguous pages
+ * @dev:   Pointer to device for which the allocation is performed.
+ * @size:  Requested allocation size.
+ * @gfp:   Allocation flags.
+ *
+ * This function allocates contiguous memory buffer for specified device. It
+ * first tries to use device specific contiguous memory area if available or
+ * the default global one, then tries a fallback allocation of normal pages.
+ *
+ * Note that it byapss one-page size of allocations from the global area as
+ * the addresses within one page are always contiguous, so there is no need
+ * to waste CMA pages for that kind; it also helps reduce fragmentations.
+ */
+struct page *dma_alloc_contiguous(struct device *dev, size_t size, gfp_t gfp)
+{
+	int node = dev ? dev_to_node(dev) : NUMA_NO_NODE;
+	size_t count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	size_t align = get_order(PAGE_ALIGN(size));
+	struct page *page = NULL;
+	struct cma *cma = NULL;
+
+	if (dev && dev->cma_area)
+		cma = dev->cma_area;
+	else if (count > 1)
+		cma = dma_contiguous_default_area;
+
+	/* CMA can be used only in the context which permits sleeping */
+	if (cma && gfpflags_allow_blocking(gfp)) {
+		align = min_t(size_t, align, CONFIG_CMA_ALIGNMENT);
+		page = cma_alloc(cma, count, align, gfp & __GFP_NOWARN);
+	}
+
+	/* Fallback allocation of normal pages */
+	if (!page)
+		page = alloc_pages_node(node, gfp, align);
+	return page;
+}
+
+/**
+ * dma_free_contiguous() - release allocated pages
+ * @dev:   Pointer to device for which the pages were allocated.
+ * @page:  Pointer to the allocated pages.
+ * @size:  Size of allocated pages.
+ *
+ * This function releases memory allocated by dma_alloc_contiguous(). As the
+ * cma_release returns false when provided pages do not belong to contiguous
+ * area and true otherwise, this function then does a fallback __free_pages()
+ * upon a false-return.
+ */
+void dma_free_contiguous(struct device *dev, struct page *page, size_t size)
+{
+	if (!cma_release(dev_get_cma_area(dev), page, size >> PAGE_SHIFT))
+		__free_pages(page, get_order(size));
+}
+
 /*
  * Support for reserved memory regions defined in device tree
  */
diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index 2c2772e9702a..b90e1aede743 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -96,8 +96,6 @@ static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
 struct page *__dma_direct_alloc_pages(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;
 	u64 phys_mask;
 
@@ -109,20 +107,9 @@ struct page *__dma_direct_alloc_pages(struct device *dev, size_t size,
 	gfp |= __dma_direct_optimal_gfp_mask(dev, dev->coherent_dma_mask,
 			&phys_mask);
 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 & __GFP_NOWARN);
-		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);
-
+	page = dma_alloc_contiguous(dev, size, gfp);
 	if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
-		__free_pages(page, page_order);
+		dma_free_contiguous(dev, page, size);
 		page = NULL;
 
 		if (IS_ENABLED(CONFIG_ZONE_DMA32) &&
@@ -151,10 +138,18 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 	if (!page)
 		return NULL;
 
+	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
+		/* remove any dirty cache lines on the kernel alias */
+		if (!PageHighMem(page))
+			arch_dma_prep_coherent(page, size);
+		/* return the page pointer as the opaque cookie */
+		return page;
+	}
+
 	if (PageHighMem(page)) {
 		/*
 		 * Depending on the cma= arguments and per-arch setup
-		 * dma_alloc_from_contiguous could return highmem pages.
+		 * dma_alloc_contiguous could return highmem pages.
 		 * Without remapping there is no way to return them here,
 		 * so log an error and fail.
 		 */
@@ -171,15 +166,19 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 		*dma_handle = phys_to_dma(dev, page_to_phys(page));
 	}
 	memset(ret, 0, size);
+
+	if (IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    dma_alloc_need_uncached(dev, attrs)) {
+		arch_dma_prep_coherent(page, size);
+		ret = uncached_kernel_address(ret);
+	}
+
 	return ret;
 }
 
 void __dma_direct_free_pages(struct device *dev, size_t size, struct page *page)
 {
-	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-
-	if (!dma_release_from_contiguous(dev, page, count))
-		__free_pages(page, get_order(size));
+	dma_free_contiguous(dev, page, size);
 }
 
 void dma_direct_free_pages(struct device *dev, size_t size, void *cpu_addr,
@@ -187,15 +186,26 @@ void dma_direct_free_pages(struct device *dev, size_t size, void *cpu_addr,
 {
 	unsigned int page_order = get_order(size);
 
+	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
+		/* cpu_addr is a struct page cookie, not a kernel address */
+		__dma_direct_free_pages(dev, size, cpu_addr);
+		return;
+	}
+
 	if (force_dma_unencrypted())
 		set_memory_encrypted((unsigned long)cpu_addr, 1 << page_order);
+
+	if (IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    dma_alloc_need_uncached(dev, attrs))
+		cpu_addr = cached_kernel_address(cpu_addr);
 	__dma_direct_free_pages(dev, size, virt_to_page(cpu_addr));
 }
 
 void *dma_direct_alloc(struct device *dev, size_t size,
 		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
-	if (!dev_is_dma_coherent(dev))
+	if (!IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    dma_alloc_need_uncached(dev, attrs))
 		return arch_dma_alloc(dev, size, dma_handle, gfp, attrs);
 	return dma_direct_alloc_pages(dev, size, dma_handle, gfp, attrs);
 }
@@ -203,7 +213,8 @@ void *dma_direct_alloc(struct device *dev, size_t size,
 void dma_direct_free(struct device *dev, size_t size,
 		void *cpu_addr, dma_addr_t dma_addr, unsigned long attrs)
 {
-	if (!dev_is_dma_coherent(dev))
+	if (!IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    dma_alloc_need_uncached(dev, attrs))
 		arch_dma_free(dev, size, cpu_addr, dma_addr, attrs);
 	else
 		dma_direct_free_pages(dev, size, cpu_addr, dma_addr, attrs);
diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index f7afdadb6770..1f628e7ac709 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -317,6 +317,12 @@ void arch_dma_set_mask(struct device *dev, u64 mask);
 
 int dma_set_mask(struct device *dev, u64 mask)
 {
+	/*
+	 * Truncate the mask to the actually supported dma_addr_t width to
+	 * avoid generating unsupportable addresses.
+	 */
+	mask = (dma_addr_t)mask;
+
 	if (!dev->dma_mask || !dma_supported(dev, mask))
 		return -EIO;
 
@@ -330,6 +336,12 @@ EXPORT_SYMBOL(dma_set_mask);
 #ifndef CONFIG_ARCH_HAS_DMA_SET_COHERENT_MASK
 int dma_set_coherent_mask(struct device *dev, u64 mask)
 {
+	/*
+	 * Truncate the mask to the actually supported dma_addr_t width to
+	 * avoid generating unsupportable addresses.
+	 */
+	mask = (dma_addr_t)mask;
+
 	if (!dma_supported(dev, mask))
 		return -EIO;
 
diff --git a/kernel/dma/remap.c b/kernel/dma/remap.c
index 7a723194ecbe..a594aec07882 100644
--- a/kernel/dma/remap.c
+++ b/kernel/dma/remap.c
@@ -158,6 +158,9 @@ out:
 
 bool dma_in_atomic_pool(void *start, size_t size)
 {
+	if (unlikely(!atomic_pool))
+		return false;
+
 	return addr_in_gen_pool(atomic_pool, (unsigned long)start, size);
 }
 
@@ -199,8 +202,7 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 
 	size = PAGE_ALIGN(size);
 
-	if (!gfpflags_allow_blocking(flags) &&
-	    !(attrs & DMA_ATTR_NO_KERNEL_MAPPING)) {
+	if (!gfpflags_allow_blocking(flags)) {
 		ret = dma_alloc_from_pool(size, &page, flags);
 		if (!ret)
 			return NULL;
@@ -214,11 +216,6 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 	/* remove any dirty cache lines on the kernel alias */
 	arch_dma_prep_coherent(page, size);
 
-	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
-		ret = page; /* opaque cookie */
-		goto done;
-	}
-
 	/* create a coherent mapping */
 	ret = dma_common_contiguous_remap(page, size, VM_USERMAP,
 			arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs),
@@ -237,10 +234,7 @@ done:
 void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 		dma_addr_t dma_handle, unsigned long attrs)
 {
-	if (attrs & DMA_ATTR_NO_KERNEL_MAPPING) {
-		/* vaddr is a struct page cookie, not a kernel address */
-		__dma_direct_free_pages(dev, size, vaddr);
-	} else if (!dma_free_from_pool(vaddr, PAGE_ALIGN(size))) {
+	if (!dma_free_from_pool(vaddr, PAGE_ALIGN(size))) {
 		phys_addr_t phys = dma_to_phys(dev, dma_handle);
 		struct page *page = pfn_to_page(__phys_to_pfn(phys));
 
diff --git a/lib/genalloc.c b/lib/genalloc.c
index 5257f74fccf3..9fc31292cfa1 100644
--- a/lib/genalloc.c
+++ b/lib/genalloc.c
@@ -327,21 +327,45 @@ EXPORT_SYMBOL(gen_pool_alloc_algo_owner);
  * gen_pool_dma_alloc - allocate special memory from the pool for DMA usage
  * @pool: pool to allocate from
  * @size: number of bytes to allocate from the pool
- * @dma: dma-view physical address return value.  Use NULL if unneeded.
+ * @dma: dma-view physical address return value.  Use %NULL if unneeded.
  *
  * Allocate the requested number of bytes from the specified pool.
  * Uses the pool allocation function (with first-fit algorithm by default).
  * Can not be used in NMI handler on architectures without
  * NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated memory, or %NULL on failure
  */
 void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
+{
+	return gen_pool_dma_alloc_algo(pool, size, dma, pool->algo, pool->data);
+}
+EXPORT_SYMBOL(gen_pool_dma_alloc);
+
+/**
+ * gen_pool_dma_alloc_algo - allocate special memory from the pool for DMA
+ * usage with the given pool algorithm
+ * @pool: pool to allocate from
+ * @size: number of bytes to allocate from the pool
+ * @dma: DMA-view physical address return value. Use %NULL if unneeded.
+ * @algo: algorithm passed from caller
+ * @data: data passed to algorithm
+ *
+ * Allocate the requested number of bytes from the specified pool. Uses the
+ * given pool allocation function. Can not be used in NMI handler on
+ * architectures without NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated memory, or %NULL on failure
+ */
+void *gen_pool_dma_alloc_algo(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, genpool_algo_t algo, void *data)
 {
 	unsigned long vaddr;
 
 	if (!pool)
 		return NULL;
 
-	vaddr = gen_pool_alloc(pool, size);
+	vaddr = gen_pool_alloc_algo(pool, size, algo, data);
 	if (!vaddr)
 		return NULL;
 
@@ -350,7 +374,102 @@ void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
 
 	return (void *)vaddr;
 }
-EXPORT_SYMBOL(gen_pool_dma_alloc);
+EXPORT_SYMBOL(gen_pool_dma_alloc_algo);
+
+/**
+ * gen_pool_dma_alloc_align - allocate special memory from the pool for DMA
+ * usage with the given alignment
+ * @pool: pool to allocate from
+ * @size: number of bytes to allocate from the pool
+ * @dma: DMA-view physical address return value. Use %NULL if unneeded.
+ * @align: alignment in bytes for starting address
+ *
+ * Allocate the requested number bytes from the specified pool, with the given
+ * alignment restriction. Can not be used in NMI handler on architectures
+ * without NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated memory, or %NULL on failure
+ */
+void *gen_pool_dma_alloc_align(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, int align)
+{
+	struct genpool_data_align data = { .align = align };
+
+	return gen_pool_dma_alloc_algo(pool, size, dma,
+			gen_pool_first_fit_align, &data);
+}
+EXPORT_SYMBOL(gen_pool_dma_alloc_align);
+
+/**
+ * gen_pool_dma_zalloc - allocate special zeroed memory from the pool for
+ * DMA usage
+ * @pool: pool to allocate from
+ * @size: number of bytes to allocate from the pool
+ * @dma: dma-view physical address return value.  Use %NULL if unneeded.
+ *
+ * Allocate the requested number of zeroed bytes from the specified pool.
+ * Uses the pool allocation function (with first-fit algorithm by default).
+ * Can not be used in NMI handler on architectures without
+ * NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated zeroed memory, or %NULL on failure
+ */
+void *gen_pool_dma_zalloc(struct gen_pool *pool, size_t size, dma_addr_t *dma)
+{
+	return gen_pool_dma_zalloc_algo(pool, size, dma, pool->algo, pool->data);
+}
+EXPORT_SYMBOL(gen_pool_dma_zalloc);
+
+/**
+ * gen_pool_dma_zalloc_algo - allocate special zeroed memory from the pool for
+ * DMA usage with the given pool algorithm
+ * @pool: pool to allocate from
+ * @size: number of bytes to allocate from the pool
+ * @dma: DMA-view physical address return value. Use %NULL if unneeded.
+ * @algo: algorithm passed from caller
+ * @data: data passed to algorithm
+ *
+ * Allocate the requested number of zeroed bytes from the specified pool. Uses
+ * the given pool allocation function. Can not be used in NMI handler on
+ * architectures without NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated zeroed memory, or %NULL on failure
+ */
+void *gen_pool_dma_zalloc_algo(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, genpool_algo_t algo, void *data)
+{
+	void *vaddr = gen_pool_dma_alloc_algo(pool, size, dma, algo, data);
+
+	if (vaddr)
+		memset(vaddr, 0, size);
+
+	return vaddr;
+}
+EXPORT_SYMBOL(gen_pool_dma_zalloc_algo);
+
+/**
+ * gen_pool_dma_zalloc_align - allocate special zeroed memory from the pool for
+ * DMA usage with the given alignment
+ * @pool: pool to allocate from
+ * @size: number of bytes to allocate from the pool
+ * @dma: DMA-view physical address return value. Use %NULL if unneeded.
+ * @align: alignment in bytes for starting address
+ *
+ * Allocate the requested number of zeroed bytes from the specified pool,
+ * with the given alignment restriction. Can not be used in NMI handler on
+ * architectures without NMI-safe cmpxchg implementation.
+ *
+ * Return: virtual address of the allocated zeroed memory, or %NULL on failure
+ */
+void *gen_pool_dma_zalloc_align(struct gen_pool *pool, size_t size,
+		dma_addr_t *dma, int align)
+{
+	struct genpool_data_align data = { .align = align };
+
+	return gen_pool_dma_zalloc_algo(pool, size, dma,
+			gen_pool_first_fit_align, &data);
+}
+EXPORT_SYMBOL(gen_pool_dma_zalloc_align);
 
 /**
  * gen_pool_free - free allocated special memory back to the pool