dma-mapping updates for 5.5-rc1

- improve dma-debug scalability (Eric Dumazet)
  - tiny dma-debug cleanup (Dan Carpenter)
  - check for vmap memory in dma_map_single (Kees Cook)
  - check for dma_addr_t overflows in dma-direct when using
    DMA offsets (Nicolas Saenz Julienne)
  - switch the x86 sta2x11 SOC to use more generic DMA code
    (Nicolas Saenz Julienne)
  - fix arm-nommu dma-ranges handling (Vladimir Murzin)
  - use __initdata in CMA (Shyam Saini)
  - replace the bus dma mask with a limit (Nicolas Saenz Julienne)
  - merge the remapping helpers into the main dma-direct flow (me)
  - switch xtensa to the generic dma remap handling (me)
  - various cleanups around dma_capable (me)
  - remove unused dev arguments to various dma-noncoherent helpers (me)
 -----BEGIN PGP SIGNATURE-----
 
 iQI/BAABCgApFiEEgdbnc3r/njty3Iq9D55TZVIEUYMFAl3f+eULHGhjaEBsc3Qu
 ZGUACgkQD55TZVIEUYPyPg/+PVHCrhmepudQQFHu6wfurE5U77iNnoUifvG+b5z5
 5mHmTMkQwyox6rKDe8NuFApAhz1VJDSUgSelPmvTSOIEIGXCvX1p+GqRSVS5YQON
 aLzGvbWKE8hCpaPdDHKYDauD1FZGMM8L2P5oOMF9X9fQ94xxRqfqJM6c8iD16Sgg
 +aOgPNzTnxQHJFF/Dbt/mjJrKXWI+XF+bgUbH+l9yKa7Dd7ibmJR8yl9hs1jmp0H
 1CZ+CizwnAs57rCd1a6Ybc6gj59tySc03NMnnbTko+KDxrcbD3Ee2tpqHVkkCjYz
 Yl0m4FIpbotrpokL/FIS727bVvkjbWgoeM+kiVPoYzmZea3pq/tFDr6tp/BxDhFj
 TZXSFfgQljlYMD3ppSoklFlfjGriVWV0tPO3arPXwuuMF5EX/IMQmvxei05jpc8n
 iELNXOP9iZZkY4tLHy2hn2uWrxBRrS1WQwlLg9hahlNRzyfFSyHeP0zWlVDt+RgF
 5CCbEI+HQcUqg1FApB30lQNWTn1+dJftrpKVBlgNBIyIa/z2rFbt8GdSnItxjfQX
 /XX8EZbFvF6AcXkgURkYFIoKM/EbYShOSLcYA3PTUtcuTnF6Kk5eimySiGWZTVCS
 prruSFDZJOvL3SnOIMIiYVmBdB7lEbDyLI/VYuhoECXEDCJpVmRktNkJNg4q6/E+
 fjQ=
 =e5wO
 -----END PGP SIGNATURE-----

Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux; tag 'dma-mapping-5.5' of git://git.infradead.org/users/hch/dma-mapping

Pull dma-mapping updates from Christoph Hellwig:

 - improve dma-debug scalability (Eric Dumazet)

 - tiny dma-debug cleanup (Dan Carpenter)

 - check for vmap memory in dma_map_single (Kees Cook)

 - check for dma_addr_t overflows in dma-direct when using DMA offsets
   (Nicolas Saenz Julienne)

 - switch the x86 sta2x11 SOC to use more generic DMA code (Nicolas
   Saenz Julienne)

 - fix arm-nommu dma-ranges handling (Vladimir Murzin)

 - use __initdata in CMA (Shyam Saini)

 - replace the bus dma mask with a limit (Nicolas Saenz Julienne)

 - merge the remapping helpers into the main dma-direct flow (me)

 - switch xtensa to the generic dma remap handling (me)

 - various cleanups around dma_capable (me)

 - remove unused dev arguments to various dma-noncoherent helpers (me)

* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux:

* tag 'dma-mapping-5.5' of git://git.infradead.org/users/hch/dma-mapping: (22 commits)
  dma-mapping: treat dev->bus_dma_mask as a DMA limit
  dma-direct: exclude dma_direct_map_resource from the min_low_pfn check
  dma-direct: don't check swiotlb=force in dma_direct_map_resource
  dma-debug: clean up put_hash_bucket()
  powerpc: remove support for NULL dev in __phys_to_dma / __dma_to_phys
  dma-direct: avoid a forward declaration for phys_to_dma
  dma-direct: unify the dma_capable definitions
  dma-mapping: drop the dev argument to arch_sync_dma_for_*
  x86/PCI: sta2x11: use default DMA address translation
  dma-direct: check for overflows on 32 bit DMA addresses
  dma-debug: increase HASH_SIZE
  dma-debug: reorder struct dma_debug_entry fields
  xtensa: use the generic uncached segment support
  dma-mapping: merge the generic remapping helpers into dma-direct
  dma-direct: provide mmap and get_sgtable method overrides
  dma-direct: remove the dma_handle argument to __dma_direct_alloc_pages
  dma-direct: remove __dma_direct_free_pages
  usb: core: Remove redundant vmap checks
  kernel: dma-contiguous: mark CMA parameters __initdata/__initconst
  dma-debug: add a schedule point in debug_dma_dump_mappings()
  ...

arch/arc/Kconfig

@@ -6,7 +6,6 @@
 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

arch/arc/mm/dma.c

@@ -48,8 +48,8 @@ void arch_dma_prep_coherent(struct page *page, size_t size)
  * upper layer functions (in include/linux/dma-mapping.h)
  */
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_TO_DEVICE:
@@ -69,8 +69,8 @@ void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_TO_DEVICE:

arch/arm/Kconfig

@@ -7,7 +7,6 @@ config ARM
 	select ARCH_HAS_BINFMT_FLAT
 	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_WRITE_COMBINE if !ARM_DMA_MEM_BUFFERABLE
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_FORTIFY_SOURCE

arch/arm/include/asm/dma-direct.h

@@ -14,23 +14,4 @@ static inline phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t dev_addr)
 	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 */

arch/arm/mm/dma-mapping-nommu.c

@@ -35,7 +35,7 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
 				 unsigned long attrs)
 
 {
-	void *ret = dma_alloc_from_global_coherent(size, dma_handle);
+	void *ret = dma_alloc_from_global_coherent(dev, size, dma_handle);
 
 	/*
 	 * dma_alloc_from_global_coherent() may fail because:

arch/arm/mm/dma-mapping.c

@@ -2332,26 +2332,20 @@ void arch_teardown_dma_ops(struct device *dev)
 }
 
 #ifdef CONFIG_SWIOTLB
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_page_cpu_to_dev(phys_to_page(paddr), paddr & (PAGE_SIZE - 1),
 			      size, dir);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_page_dev_to_cpu(phys_to_page(paddr), paddr & (PAGE_SIZE - 1),
 			      size, dir);
 }
 
-long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
-		dma_addr_t dma_addr)
-{
-	return dma_to_pfn(dev, dma_addr);
-}
-
 void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		gfp_t gfp, unsigned long attrs)
 {

arch/arm/xen/mm.c

@@ -71,20 +71,20 @@ static void dma_cache_maint(dma_addr_t handle, size_t size, u32 op)
  * pfn_valid returns true the pages is local and we can use the native
  * dma-direct functions, otherwise we call the Xen specific version.
  */
-void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+void xen_dma_sync_for_cpu(dma_addr_t handle, phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	if (pfn_valid(PFN_DOWN(handle)))
-		arch_sync_dma_for_cpu(dev, paddr, size, dir);
+		arch_sync_dma_for_cpu(paddr, size, dir);
 	else if (dir != DMA_TO_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 }
 
-void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+void xen_dma_sync_for_device(dma_addr_t handle, phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	if (pfn_valid(PFN_DOWN(handle)))
-		arch_sync_dma_for_device(dev, paddr, size, dir);
+		arch_sync_dma_for_device(paddr, size, dir);
 	else if (dir == DMA_FROM_DEVICE)
 		dma_cache_maint(handle, size, GNTTAB_CACHE_INVAL);
 	else

arch/arm64/Kconfig

@@ -12,7 +12,6 @@ config ARM64
 	select ARCH_CLOCKSOURCE_DATA
 	select ARCH_HAS_DEBUG_VIRTUAL
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
-	select ARCH_HAS_DMA_COHERENT_TO_PFN
 	select ARCH_HAS_DMA_PREP_COHERENT
 	select ARCH_HAS_ACPI_TABLE_UPGRADE if ACPI
 	select ARCH_HAS_FAST_MULTIPLIER

arch/arm64/mm/dma-mapping.c

@@ -13,14 +13,14 @@
 
 #include <asm/cacheflush.h>
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_map_area(phys_to_virt(paddr), size, dir);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_unmap_area(phys_to_virt(paddr), size, dir);
 }

arch/c6x/mm/dma-coherent.c

@@ -140,7 +140,7 @@ void __init coherent_mem_init(phys_addr_t start, u32 size)
 		      sizeof(long));
 }
 
-static void c6x_dma_sync(struct device *dev, phys_addr_t paddr, size_t size,
+static void c6x_dma_sync(phys_addr_t paddr, size_t size,
 		enum dma_data_direction dir)
 {
 	BUG_ON(!valid_dma_direction(dir));
@@ -160,14 +160,14 @@ static void c6x_dma_sync(struct device *dev, phys_addr_t paddr, size_t size,
 	}
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
-	return c6x_dma_sync(dev, paddr, size, dir);
+	return c6x_dma_sync(paddr, size, dir);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
-	return c6x_dma_sync(dev, paddr, size, dir);
+	return c6x_dma_sync(paddr, size, dir);
 }

arch/csky/mm/dma-mapping.c

@@ -58,8 +58,8 @@ void arch_dma_prep_coherent(struct page *page, size_t size)
 	cache_op(page_to_phys(page), size, dma_wbinv_set_zero_range);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-			      size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_TO_DEVICE:
@@ -74,8 +74,8 @@ void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-			   size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_TO_DEVICE:

arch/hexagon/kernel/dma.c

@@ -55,8 +55,8 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 	gen_pool_free(coherent_pool, (unsigned long) vaddr, size);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	void *addr = phys_to_virt(paddr);
 

arch/ia64/Kconfig

@@ -33,7 +33,7 @@ config IA64
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_MEMBLOCK_NODE_MAP
 	select HAVE_VIRT_CPU_ACCOUNTING
-	select ARCH_HAS_DMA_COHERENT_TO_PFN
+	select DMA_NONCOHERENT_MMAP
 	select ARCH_HAS_SYNC_DMA_FOR_CPU
 	select VIRT_TO_BUS
 	select GENERIC_IRQ_PROBE

arch/ia64/kernel/dma-mapping.c

@@ -19,9 +19,3 @@ void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
 {
 	dma_direct_free_pages(dev, size, cpu_addr, dma_addr, attrs);
 }
-
-long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
-		dma_addr_t dma_addr)
-{
-	return page_to_pfn(virt_to_page(cpu_addr));
-}

arch/ia64/mm/init.c

@@ -73,8 +73,8 @@ __ia64_sync_icache_dcache (pte_t pte)
  * DMA can be marked as "clean" so that lazy_mmu_prot_update() doesn't have to
  * flush them when they get mapped into an executable vm-area.
  */
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	unsigned long pfn = PHYS_PFN(paddr);
 

arch/m68k/kernel/dma.c

@@ -61,8 +61,8 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 
 #endif /* CONFIG_MMU && !CONFIG_COLDFIRE */
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t handle,
+void arch_sync_dma_for_device(phys_addr_t handle, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_BIDIRECTIONAL:

arch/microblaze/Kconfig

@@ -4,7 +4,6 @@ config MICROBLAZE
 	select ARCH_32BIT_OFF_T
 	select ARCH_NO_SWAP
 	select ARCH_HAS_BINFMT_FLAT if !MMU
-	select ARCH_HAS_DMA_COHERENT_TO_PFN if MMU
 	select ARCH_HAS_DMA_PREP_COHERENT
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_SYNC_DMA_FOR_CPU

arch/microblaze/kernel/dma.c

@@ -15,7 +15,7 @@
 #include <linux/bug.h>
 #include <asm/cacheflush.h>
 
-static void __dma_sync(struct device *dev, phys_addr_t paddr, size_t size,
+static void __dma_sync(phys_addr_t paddr, size_t size,
 		enum dma_data_direction direction)
 {
 	switch (direction) {
@@ -31,14 +31,14 @@ static void __dma_sync(struct device *dev, phys_addr_t paddr, size_t size,
 	}
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
-	__dma_sync(dev, paddr, size, dir);
+	__dma_sync(paddr, size, dir);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
-	__dma_sync(dev, paddr, size, dir);
+	__dma_sync(paddr, size, dir);
 }

arch/mips/Kconfig

@@ -1194,9 +1194,9 @@ config DMA_NONCOHERENT
 	select ARCH_HAS_DMA_WRITE_COMBINE
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
 	select ARCH_HAS_UNCACHED_SEGMENT
-	select NEED_DMA_MAP_STATE
+	select DMA_NONCOHERENT_MMAP
-	select ARCH_HAS_DMA_COHERENT_TO_PFN
 	select DMA_NONCOHERENT_CACHE_SYNC
+	select NEED_DMA_MAP_STATE
 
 config SYS_HAS_EARLY_PRINTK
 	bool

arch/mips/bmips/dma.c

@@ -64,7 +64,7 @@ phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t dma_addr)
 	return dma_addr;
 }
 
-void arch_sync_dma_for_cpu_all(struct device *dev)
+void arch_sync_dma_for_cpu_all(void)
 {
 	void __iomem *cbr = BMIPS_GET_CBR();
 	u32 cfg;

arch/mips/include/asm/dma-direct.h

@@ -2,14 +2,6 @@
 #ifndef _MIPS_DMA_DIRECT_H
 #define _MIPS_DMA_DIRECT_H 1
 
-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);
 

arch/mips/jazz/jazzdma.c

@@ -592,7 +592,7 @@ static dma_addr_t jazz_dma_map_page(struct device *dev, struct page *page,
 	phys_addr_t phys = page_to_phys(page) + offset;
 
 	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-		arch_sync_dma_for_device(dev, phys, size, dir);
+		arch_sync_dma_for_device(phys, size, dir);
 	return vdma_alloc(phys, size);
 }
 
@@ -600,7 +600,7 @@ static void jazz_dma_unmap_page(struct device *dev, dma_addr_t dma_addr,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
 	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-		arch_sync_dma_for_cpu(dev, vdma_log2phys(dma_addr), size, dir);
+		arch_sync_dma_for_cpu(vdma_log2phys(dma_addr), size, dir);
 	vdma_free(dma_addr);
 }
 
@@ -612,7 +612,7 @@ static int jazz_dma_map_sg(struct device *dev, struct scatterlist *sglist,
 
 	for_each_sg(sglist, sg, nents, i) {
 		if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-			arch_sync_dma_for_device(dev, sg_phys(sg), sg->length,
+			arch_sync_dma_for_device(sg_phys(sg), sg->length,
 				dir);
 		sg->dma_address = vdma_alloc(sg_phys(sg), sg->length);
 		if (sg->dma_address == DMA_MAPPING_ERROR)
@@ -631,8 +631,7 @@ static void jazz_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
 
 	for_each_sg(sglist, sg, nents, i) {
 		if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-			arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length,
+			arch_sync_dma_for_cpu(sg_phys(sg), sg->length, dir);
-				dir);
 		vdma_free(sg->dma_address);
 	}
 }
@@ -640,13 +639,13 @@ static void jazz_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
 static void jazz_dma_sync_single_for_device(struct device *dev,
 		dma_addr_t addr, size_t size, enum dma_data_direction dir)
 {
-	arch_sync_dma_for_device(dev, vdma_log2phys(addr), size, dir);
+	arch_sync_dma_for_device(vdma_log2phys(addr), size, dir);
 }
 
 static void jazz_dma_sync_single_for_cpu(struct device *dev,
 		dma_addr_t addr, size_t size, enum dma_data_direction dir)
 {
-	arch_sync_dma_for_cpu(dev, vdma_log2phys(addr), size, dir);
+	arch_sync_dma_for_cpu(vdma_log2phys(addr), size, dir);
 }
 
 static void jazz_dma_sync_sg_for_device(struct device *dev,
@@ -656,7 +655,7 @@ static void jazz_dma_sync_sg_for_device(struct device *dev,
 	int i;
 
 	for_each_sg(sgl, sg, nents, i)
-		arch_sync_dma_for_device(dev, sg_phys(sg), sg->length, dir);
+		arch_sync_dma_for_device(sg_phys(sg), sg->length, dir);
 }
 
 static void jazz_dma_sync_sg_for_cpu(struct device *dev,
@@ -666,7 +665,7 @@ static void jazz_dma_sync_sg_for_cpu(struct device *dev,
 	int i;
 
 	for_each_sg(sgl, sg, nents, i)
-		arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length, dir);
+		arch_sync_dma_for_cpu(sg_phys(sg), sg->length, dir);
 }
 
 const struct dma_map_ops jazz_dma_ops = {

arch/mips/mm/dma-noncoherent.c

@@ -27,7 +27,7 @@
  * R10000 and R12000 are used in such systems, the SGI IP28 Indigo² rsp.
  * SGI IP32 aka O2.
  */
-static inline bool cpu_needs_post_dma_flush(struct device *dev)
+static inline bool cpu_needs_post_dma_flush(void)
 {
 	switch (boot_cpu_type()) {
 	case CPU_R10000:
@@ -59,12 +59,6 @@ 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)
-{
-	return page_to_pfn(virt_to_page(cached_kernel_address(cpu_addr)));
-}
-
 static inline void dma_sync_virt(void *addr, size_t size,
 		enum dma_data_direction dir)
 {
@@ -118,17 +112,17 @@ static inline void dma_sync_phys(phys_addr_t paddr, size_t size,
 	} while (left);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	dma_sync_phys(paddr, size, dir);
 }
 
 #ifdef CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
-	if (cpu_needs_post_dma_flush(dev))
+	if (cpu_needs_post_dma_flush())
 		dma_sync_phys(paddr, size, dir);
 }
 #endif

arch/mips/pci/fixup-sb1250.c

@@ -21,22 +21,22 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SIBYTE, PCI_DEVICE_ID_BCM1250_PCI,
 
 /*
  * The BCM1250, etc. PCI host bridge does not support DAC on its 32-bit
- * bus, so we set the bus's DMA mask accordingly.  However the HT link
+ * bus, so we set the bus's DMA limit accordingly.  However the HT link
  * down the artificial PCI-HT bridge supports 40-bit addressing and the
  * SP1011 HT-PCI bridge downstream supports both DAC and a 64-bit bus
  * width, so we record the PCI-HT bridge's secondary and subordinate bus
- * numbers and do not set the mask for devices present in the inclusive
+ * numbers and do not set the limit for devices present in the inclusive
  * range of those.
  */
-struct sb1250_bus_dma_mask_exclude {
+struct sb1250_bus_dma_limit_exclude {
 	bool set;
 	unsigned char start;
 	unsigned char end;
 };
 
-static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
+static int sb1250_bus_dma_limit(struct pci_dev *dev, void *data)
 {
-	struct sb1250_bus_dma_mask_exclude *exclude = data;
+	struct sb1250_bus_dma_limit_exclude *exclude = data;
 	bool exclude_this;
 	bool ht_bridge;
 
@@ -55,7 +55,7 @@ static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
 			exclude->start, exclude->end);
 	} else {
 		dev_dbg(&dev->dev, "disabling DAC for device");
-		dev->dev.bus_dma_mask = DMA_BIT_MASK(32);
+		dev->dev.bus_dma_limit = DMA_BIT_MASK(32);
 	}
 
 	return 0;
@@ -63,9 +63,9 @@ static int sb1250_bus_dma_mask(struct pci_dev *dev, void *data)
 
 static void quirk_sb1250_pci_dac(struct pci_dev *dev)
 {
-	struct sb1250_bus_dma_mask_exclude exclude = { .set = false };
+	struct sb1250_bus_dma_limit_exclude exclude = { .set = false };
 
-	pci_walk_bus(dev->bus, sb1250_bus_dma_mask, &exclude);
+	pci_walk_bus(dev->bus, sb1250_bus_dma_limit, &exclude);
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SIBYTE, PCI_DEVICE_ID_BCM1250_PCI,
 			quirk_sb1250_pci_dac);

arch/nds32/kernel/dma.c

@@ -46,8 +46,8 @@ static inline void cache_op(phys_addr_t paddr, size_t size,
 	} while (left);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_FROM_DEVICE:
@@ -61,8 +61,8 @@ void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_TO_DEVICE:

arch/nios2/mm/dma-mapping.c

@@ -18,8 +18,8 @@
 #include <linux/cache.h>
 #include <asm/cacheflush.h>
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	void *vaddr = phys_to_virt(paddr);
 
@@ -42,8 +42,8 @@ void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	void *vaddr = phys_to_virt(paddr);
 

arch/openrisc/kernel/dma.c

@@ -125,7 +125,7 @@ arch_dma_free(struct device *dev, size_t size, void *vaddr,
 	free_pages_exact(vaddr, size);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t addr, size_t size,
+void arch_sync_dma_for_device(phys_addr_t addr, size_t size,
 		enum dma_data_direction dir)
 {
 	unsigned long cl;

arch/parisc/kernel/pci-dma.c

@@ -439,14 +439,14 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 	free_pages((unsigned long)__va(dma_handle), order);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	flush_kernel_dcache_range((unsigned long)phys_to_virt(paddr), size);
 }

arch/powerpc/include/asm/dma-direct.h

@@ -2,26 +2,13 @@
 #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)
-{
-	if (!dev->dma_mask)
-		return false;
-
-	return addr + size - 1 <=
-		min_not_zero(*dev->dma_mask, dev->bus_dma_mask);
-}
-
 static inline dma_addr_t __phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
-	if (!dev)
-		return paddr + PCI_DRAM_OFFSET;
 	return paddr + dev->archdata.dma_offset;
 }
 
 static inline phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t daddr)
 {
-	if (!dev)
-		return daddr - PCI_DRAM_OFFSET;
 	return daddr - dev->archdata.dma_offset;
 }
 #endif /* ASM_POWERPC_DMA_DIRECT_H */

arch/powerpc/mm/dma-noncoherent.c

@@ -104,14 +104,14 @@ static void __dma_sync_page(phys_addr_t paddr, size_t size, int dir)
 #endif
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_sync_page(paddr, size, dir);
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	__dma_sync_page(paddr, size, dir);
 }

arch/powerpc/platforms/Kconfig.cputype

@@ -459,7 +459,6 @@ config NOT_COHERENT_CACHE
 	bool
 	depends on 4xx || PPC_8xx || E200 || PPC_MPC512x || \
 		GAMECUBE_COMMON || AMIGAONE
-	select ARCH_HAS_DMA_COHERENT_TO_PFN
 	select ARCH_HAS_DMA_PREP_COHERENT
 	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
 	select ARCH_HAS_SYNC_DMA_FOR_CPU

arch/powerpc/sysdev/fsl_pci.c

@@ -115,8 +115,8 @@ static void pci_dma_dev_setup_swiotlb(struct pci_dev *pdev)
 {
 	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
 
-	pdev->dev.bus_dma_mask =
+	pdev->dev.bus_dma_limit =
-		hose->dma_window_base_cur + hose->dma_window_size;
+		hose->dma_window_base_cur + hose->dma_window_size - 1;
 }
 
 static void setup_swiotlb_ops(struct pci_controller *hose)
@@ -135,7 +135,7 @@ static void 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) {
-		dev->bus_dma_mask = 0;
+		dev->bus_dma_limit = 0;
 		dev->archdata.dma_offset = pci64_dma_offset;
 	}
 }

arch/sh/kernel/dma-coherent.c

@@ -25,7 +25,7 @@ void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 	 * Pages from the page allocator may have data present in
 	 * cache. So flush the cache before using uncached memory.
 	 */
-	arch_sync_dma_for_device(dev, virt_to_phys(ret), size,
+	arch_sync_dma_for_device(virt_to_phys(ret), size,
 			DMA_BIDIRECTIONAL);
 
 	ret_nocache = (void __force *)ioremap_nocache(virt_to_phys(ret), size);
@@ -59,8 +59,8 @@ void arch_dma_free(struct device *dev, size_t size, void *vaddr,
 	iounmap(vaddr);
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	void *addr = sh_cacheop_vaddr(phys_to_virt(paddr));
 

arch/sparc/kernel/ioport.c

@@ -366,8 +366,8 @@ void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
 
 /* IIep is write-through, not flushing on cpu to device transfer. */
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	if (dir != PCI_DMA_TODEVICE)
 		dma_make_coherent(paddr, PAGE_ALIGN(size));

arch/x86/Kconfig

@@ -708,7 +708,6 @@ config X86_SUPPORTS_MEMORY_FAILURE
 config STA2X11
 	bool "STA2X11 Companion Chip Support"
 	depends on X86_32_NON_STANDARD && PCI
-	select ARCH_HAS_PHYS_TO_DMA
 	select SWIOTLB
 	select MFD_STA2X11
 	select GPIOLIB

arch/x86/include/asm/device.h

@@ -6,9 +6,6 @@ struct dev_archdata {
 #if defined(CONFIG_INTEL_IOMMU) || defined(CONFIG_AMD_IOMMU)
 	void *iommu; /* hook for IOMMU specific extension */
 #endif
-#ifdef CONFIG_STA2X11
-	bool is_sta2x11;
-#endif
 };
 
 #if defined(CONFIG_X86_DEV_DMA_OPS) && defined(CONFIG_PCI_DOMAINS)

arch/x86/include/asm/dma-direct.h

@@ -1,9 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef ASM_X86_DMA_DIRECT_H
-#define ASM_X86_DMA_DIRECT_H 1
-
-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);
-
-#endif /* ASM_X86_DMA_DIRECT_H */

arch/x86/kernel/amd_gart_64.c

@@ -185,13 +185,13 @@ static void iommu_full(struct device *dev, size_t size, int dir)
 static inline int
 need_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	return force_iommu || !dma_capable(dev, addr, size);
+	return force_iommu || !dma_capable(dev, addr, size, true);
 }
 
 static inline int
 nonforced_iommu(struct device *dev, unsigned long addr, size_t size)
 {
-	return !dma_capable(dev, addr, size);
+	return !dma_capable(dev, addr, size, true);
 }
 
 /* Map a single continuous physical area into the IOMMU.

arch/x86/kernel/pci-dma.c

@@ -140,7 +140,7 @@ rootfs_initcall(pci_iommu_init);
 
 static int via_no_dac_cb(struct pci_dev *pdev, void *data)
 {
-	pdev->dev.bus_dma_mask = DMA_BIT_MASK(32);
+	pdev->dev.bus_dma_limit = DMA_BIT_MASK(32);
 	return 0;
 }
 

arch/x86/mm/mem_encrypt.c

@@ -367,7 +367,7 @@ bool force_dma_unencrypted(struct device *dev)
 	if (sme_active()) {
 		u64 dma_enc_mask = DMA_BIT_MASK(__ffs64(sme_me_mask));
 		u64 dma_dev_mask = min_not_zero(dev->coherent_dma_mask,
-						dev->bus_dma_mask);
+						dev->bus_dma_limit);
 
 		if (dma_dev_mask <= dma_enc_mask)
 			return true;

arch/x86/pci/sta2x11-fixup.c

@@ -30,7 +30,6 @@ struct sta2x11_ahb_regs { /* saved during suspend */
 };
 
 struct sta2x11_mapping {
-	u32 amba_base;
 	int is_suspended;
 	struct sta2x11_ahb_regs regs[STA2X11_NR_FUNCS];
 };
@@ -92,18 +91,6 @@ static int sta2x11_pdev_to_ep(struct pci_dev *pdev)
 	return pdev->bus->number - instance->bus0;
 }
 
-static struct sta2x11_mapping *sta2x11_pdev_to_mapping(struct pci_dev *pdev)
-{
-	struct sta2x11_instance *instance;
-	int ep;
-
-	instance = sta2x11_pdev_to_instance(pdev);
-	if (!instance)
-		return NULL;
-	ep = sta2x11_pdev_to_ep(pdev);
-	return instance->map + ep;
-}
-
 /* This is exported, as some devices need to access the MFD registers */
 struct sta2x11_instance *sta2x11_get_instance(struct pci_dev *pdev)
 {
@@ -111,39 +98,6 @@ struct sta2x11_instance *sta2x11_get_instance(struct pci_dev *pdev)
 }
 EXPORT_SYMBOL(sta2x11_get_instance);
 
-
-/**
- * p2a - Translate physical address to STA2x11 AMBA address,
- *       used for DMA transfers to STA2x11
- * @p: Physical address
- * @pdev: PCI device (must be hosted within the connext)
- */
-static dma_addr_t p2a(dma_addr_t p, struct pci_dev *pdev)
-{
-	struct sta2x11_mapping *map;
-	dma_addr_t a;
-
-	map = sta2x11_pdev_to_mapping(pdev);
-	a = p + map->amba_base;
-	return a;
-}
-
-/**
- * a2p - Translate STA2x11 AMBA address to physical address
- *       used for DMA transfers from STA2x11
- * @a: STA2x11 AMBA address
- * @pdev: PCI device (must be hosted within the connext)
- */
-static dma_addr_t a2p(dma_addr_t a, struct pci_dev *pdev)
-{
-	struct sta2x11_mapping *map;
-	dma_addr_t p;
-
-	map = sta2x11_pdev_to_mapping(pdev);
-	p = a - map->amba_base;
-	return p;
-}
-
 /* At setup time, we use our own ops if the device is a ConneXt one */
 static void sta2x11_setup_pdev(struct pci_dev *pdev)
 {
@@ -151,70 +105,12 @@ static void sta2x11_setup_pdev(struct pci_dev *pdev)
 
 	if (!instance) /* either a sta2x11 bridge or another ST device */
 		return;
-	pci_set_consistent_dma_mask(pdev, STA2X11_AMBA_SIZE - 1);
-	pci_set_dma_mask(pdev, STA2X11_AMBA_SIZE - 1);
-	pdev->dev.archdata.is_sta2x11 = true;
 
 	/* We must enable all devices as master, for audio DMA to work */
 	pci_set_master(pdev);
 }
 DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_STMICRO, PCI_ANY_ID, sta2x11_setup_pdev);
 
-/*
- * The following three functions are exported (used in swiotlb: FIXME)
- */
-/**
- * dma_capable - Check if device can manage DMA transfers (FIXME: kill it)
- * @dev: device for a PCI device
- * @addr: DMA address
- * @size: DMA size
- */
-bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
-{
-	struct sta2x11_mapping *map;
-
-	if (!dev->archdata.is_sta2x11) {
-		if (!dev->dma_mask)
-			return false;
-		return addr + size - 1 <= *dev->dma_mask;
-	}
-
-	map = sta2x11_pdev_to_mapping(to_pci_dev(dev));
-
-	if (!map || (addr < map->amba_base))
-		return false;
-	if (addr + size >= map->amba_base + STA2X11_AMBA_SIZE) {
-		return false;
-	}
-
-	return true;
-}
-
-/**
- * __phys_to_dma - Return the DMA AMBA address used for this STA2x11 device
- * @dev: device for a PCI device
- * @paddr: Physical address
- */
-dma_addr_t __phys_to_dma(struct device *dev, phys_addr_t paddr)
-{
-	if (!dev->archdata.is_sta2x11)
-		return paddr;
-	return p2a(paddr, to_pci_dev(dev));
-}
-
-/**
- * dma_to_phys - Return the physical address used for this STA2x11 DMA address
- * @dev: device for a PCI device
- * @daddr: STA2x11 AMBA DMA address
- */
-phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t daddr)
-{
-	if (!dev->archdata.is_sta2x11)
-		return daddr;
-	return a2p(daddr, to_pci_dev(dev));
-}
-
-
 /*
  * At boot we must set up the mappings for the pcie-to-amba bridge.
  * It involves device access, and the same happens at suspend/resume time
@@ -234,12 +130,22 @@ phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t daddr)
 /* At probe time, enable mapping for each endpoint, using the pdev */
 static void sta2x11_map_ep(struct pci_dev *pdev)
 {
-	struct sta2x11_mapping *map = sta2x11_pdev_to_mapping(pdev);
+	struct sta2x11_instance *instance = sta2x11_pdev_to_instance(pdev);
+	struct device *dev = &pdev->dev;
+	u32 amba_base, max_amba_addr;
 	int i;
 
-	if (!map)
+	if (!instance)
 		return;
-	pci_read_config_dword(pdev, AHB_BASE(0), &map->amba_base);
+
+	pci_read_config_dword(pdev, AHB_BASE(0), &amba_base);
+	max_amba_addr = amba_base + STA2X11_AMBA_SIZE - 1;
+
+	dev->dma_pfn_offset = PFN_DOWN(-amba_base);
+
+	dev->bus_dma_limit = max_amba_addr;
+	pci_set_consistent_dma_mask(pdev, max_amba_addr);
+	pci_set_dma_mask(pdev, max_amba_addr);
 
 	/* Configure AHB mapping */
 	pci_write_config_dword(pdev, AHB_PEXLBASE(0), 0);
@@ -253,13 +159,24 @@ static void sta2x11_map_ep(struct pci_dev *pdev)
 
 	dev_info(&pdev->dev,
 		 "sta2x11: Map EP %i: AMBA address %#8x-%#8x\n",
-		 sta2x11_pdev_to_ep(pdev),  map->amba_base,
+		 sta2x11_pdev_to_ep(pdev), amba_base, max_amba_addr);
-		 map->amba_base + STA2X11_AMBA_SIZE - 1);
 }
 DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_STMICRO, PCI_ANY_ID, sta2x11_map_ep);
 
 #ifdef CONFIG_PM /* Some register values must be saved and restored */
 
+static struct sta2x11_mapping *sta2x11_pdev_to_mapping(struct pci_dev *pdev)
+{
+	struct sta2x11_instance *instance;
+	int ep;
+
+	instance = sta2x11_pdev_to_instance(pdev);
+	if (!instance)
+		return NULL;
+	ep = sta2x11_pdev_to_ep(pdev);
+	return instance->map + ep;
+}
+
 static void suspend_mapping(struct pci_dev *pdev)
 {
 	struct sta2x11_mapping *map = sta2x11_pdev_to_mapping(pdev);

arch/xtensa/Kconfig

@@ -3,8 +3,10 @@ config XTENSA
 	def_bool y
 	select ARCH_32BIT_OFF_T
 	select ARCH_HAS_BINFMT_FLAT if !MMU
-	select ARCH_HAS_SYNC_DMA_FOR_CPU
+	select ARCH_HAS_DMA_PREP_COHERENT if MMU
-	select ARCH_HAS_SYNC_DMA_FOR_DEVICE
+	select ARCH_HAS_SYNC_DMA_FOR_CPU if MMU
+	select ARCH_HAS_SYNC_DMA_FOR_DEVICE if MMU
+	select ARCH_HAS_UNCACHED_SEGMENT if MMU
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_FRAME_POINTERS

arch/xtensa/include/asm/platform.h

@@ -65,31 +65,4 @@ extern void platform_calibrate_ccount (void);
  */
 void cpu_reset(void) __attribute__((noreturn));
 
-/*
- * Memory caching is platform-dependent in noMMU xtensa configurations.
- * The following set of functions should be implemented in platform code
- * in order to enable coherent DMA memory operations when CONFIG_MMU is not
- * enabled. Default implementations do nothing and issue a warning.
- */
-
-/*
- * Check whether p points to a cached memory.
- */
-bool platform_vaddr_cached(const void *p);
-
-/*
- * Check whether p points to an uncached memory.
- */
-bool platform_vaddr_uncached(const void *p);
-
-/*
- * Return pointer to an uncached view of the cached sddress p.
- */
-void *platform_vaddr_to_uncached(void *p);
-
-/*
- * Return pointer to a cached view of the uncached sddress p.
- */
-void *platform_vaddr_to_cached(void *p);
-
 #endif	/* _XTENSA_PLATFORM_H */

arch/xtensa/kernel/Makefile

@@ -5,10 +5,11 @@
 
 extra-y := head.o vmlinux.lds
 
-obj-y := align.o coprocessor.o entry.o irq.o pci-dma.o platform.o process.o \
+obj-y := align.o coprocessor.o entry.o irq.o platform.o process.o \
 	 ptrace.o setup.o signal.o stacktrace.o syscall.o time.o traps.o \
 	 vectors.o
 
+obj-$(CONFIG_MMU) += pci-dma.o
 obj-$(CONFIG_PCI) += pci.o
 obj-$(CONFIG_MODULES) += xtensa_ksyms.o module.o
 obj-$(CONFIG_FUNCTION_TRACER) += mcount.o

arch/xtensa/kernel/pci-dma.c

@@ -44,8 +44,8 @@ static void do_cache_op(phys_addr_t paddr, size_t size,
 		}
 }
 
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_BIDIRECTIONAL:
@@ -62,8 +62,8 @@ void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
 	}
 }
 
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_BIDIRECTIONAL:
@@ -81,122 +81,25 @@ void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
 	}
 }
 
+void arch_dma_prep_coherent(struct page *page, size_t size)
+{
+	__invalidate_dcache_range((unsigned long)page_address(page), size);
+}
+
+/*
+ * Memory caching is platform-dependent in noMMU xtensa configurations.
+ * The following two functions should be implemented in platform code
+ * in order to enable coherent DMA memory operations when CONFIG_MMU is not
+ * enabled.
+ */
 #ifdef CONFIG_MMU
-bool platform_vaddr_cached(const void *p)
+void *uncached_kernel_address(void *p)
-{
-	unsigned long addr = (unsigned long)p;
-
-	return addr >= XCHAL_KSEG_CACHED_VADDR &&
-	       addr - XCHAL_KSEG_CACHED_VADDR < XCHAL_KSEG_SIZE;
-}
-
-bool platform_vaddr_uncached(const void *p)
-{
-	unsigned long addr = (unsigned long)p;
-
-	return addr >= XCHAL_KSEG_BYPASS_VADDR &&
-	       addr - XCHAL_KSEG_BYPASS_VADDR < XCHAL_KSEG_SIZE;
-}
-
-void *platform_vaddr_to_uncached(void *p)
 {
 	return p + XCHAL_KSEG_BYPASS_VADDR - XCHAL_KSEG_CACHED_VADDR;
 }
 
-void *platform_vaddr_to_cached(void *p)
+void *cached_kernel_address(void *p)
 {
 	return p + XCHAL_KSEG_CACHED_VADDR - XCHAL_KSEG_BYPASS_VADDR;
 }
-#else
+#endif /* CONFIG_MMU */
-bool __attribute__((weak)) platform_vaddr_cached(const void *p)
-{
-	WARN_ONCE(1, "Default %s implementation is used\n", __func__);
-	return true;
-}
-
-bool __attribute__((weak)) platform_vaddr_uncached(const void *p)
-{
-	WARN_ONCE(1, "Default %s implementation is used\n", __func__);
-	return false;
-}
-
-void __attribute__((weak)) *platform_vaddr_to_uncached(void *p)
-{
-	WARN_ONCE(1, "Default %s implementation is used\n", __func__);
-	return p;
-}
-
-void __attribute__((weak)) *platform_vaddr_to_cached(void *p)
-{
-	WARN_ONCE(1, "Default %s implementation is used\n", __func__);
-	return p;
-}
-#endif
-
-/*
- * Note: We assume that the full memory space is always mapped to 'kseg'
- *	 Otherwise we have to use page attributes (not implemented).
- */
-
-void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
-		gfp_t flag, unsigned long attrs)
-{
-	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	struct page *page = NULL;
-
-	/* ignore region speicifiers */
-
-	flag &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
-	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
-		flag |= GFP_DMA;
-
-	if (gfpflags_allow_blocking(flag))
-		page = dma_alloc_from_contiguous(dev, count, get_order(size),
-						 flag & __GFP_NOWARN);
-
-	if (!page)
-		page = alloc_pages(flag | __GFP_ZERO, get_order(size));
-
-	if (!page)
-		return NULL;
-
-	*handle = phys_to_dma(dev, page_to_phys(page));
-
-#ifdef CONFIG_MMU
-	if (PageHighMem(page)) {
-		void *p;
-
-		p = dma_common_contiguous_remap(page, size,
-						pgprot_noncached(PAGE_KERNEL),
-						__builtin_return_address(0));
-		if (!p) {
-			if (!dma_release_from_contiguous(dev, page, count))
-				__free_pages(page, get_order(size));
-		}
-		return p;
-	}
-#endif
-	BUG_ON(!platform_vaddr_cached(page_address(page)));
-	__invalidate_dcache_range((unsigned long)page_address(page), size);
-	return platform_vaddr_to_uncached(page_address(page));
-}
-
-void arch_dma_free(struct device *dev, size_t size, void *vaddr,
-		dma_addr_t dma_handle, unsigned long attrs)
-{
-	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	struct page *page;
-
-	if (platform_vaddr_uncached(vaddr)) {
-		page = virt_to_page(platform_vaddr_to_cached(vaddr));
-	} else {
-#ifdef CONFIG_MMU
-		dma_common_free_remap(vaddr, size);
-#endif
-		page = pfn_to_page(PHYS_PFN(dma_to_phys(dev, dma_handle)));
-	}
-
-	if (!dma_release_from_contiguous(dev, page, count))
-		__free_pages(page, get_order(size));
-}

drivers/acpi/arm64/iort.c

@@ -1057,8 +1057,8 @@ static int rc_dma_get_range(struct device *dev, u64 *size)
  */
 void iort_dma_setup(struct device *dev, u64 *dma_addr, u64 *dma_size)
 {
-	u64 mask, dmaaddr = 0, size = 0, offset = 0;
+	u64 end, mask, dmaaddr = 0, size = 0, offset = 0;
-	int ret, msb;
+	int ret;
 
 	/*
 	 * If @dev is expected to be DMA-capable then the bus code that created
@@ -1085,19 +1085,13 @@ void iort_dma_setup(struct device *dev, u64 *dma_addr, u64 *dma_size)
 	}
 
 	if (!ret) {
-		msb = fls64(dmaaddr + size - 1);
 		/*
-		 * Round-up to the power-of-two mask or set
+		 * Limit coherent and dma mask based on size retrieved from
-		 * the mask to the whole 64-bit address space
+		 * firmware.
-		 * in case the DMA region covers the full
-		 * memory window.
 		 */
-		mask = msb == 64 ? U64_MAX : (1ULL << msb) - 1;
+		end = dmaaddr + size - 1;
-		/*
+		mask = DMA_BIT_MASK(ilog2(end) + 1);
-		 * Limit coherent and dma mask based on size
+		dev->bus_dma_limit = end;
-		 * retrieved from firmware.
-		 */
-		dev->bus_dma_mask = mask;
 		dev->coherent_dma_mask = mask;
 		*dev->dma_mask = mask;
 	}

drivers/ata/ahci.c

@@ -910,7 +910,7 @@ static int ahci_configure_dma_masks(struct pci_dev *pdev, int using_dac)
 	 * value, don't extend it here. This happens on STA2X11, for example.
 	 *
 	 * XXX: manipulating the DMA mask from platform code is completely
-	 * bogus, platform code should use dev->bus_dma_mask instead..
+	 * bogus, platform code should use dev->bus_dma_limit instead..
 	 */
 	if (pdev->dma_mask && pdev->dma_mask < DMA_BIT_MASK(32))
 		return 0;

drivers/iommu/dma-iommu.c

@@ -405,8 +405,7 @@ static dma_addr_t iommu_dma_alloc_iova(struct iommu_domain *domain,
 	if (iova_len < (1 << (IOVA_RANGE_CACHE_MAX_SIZE - 1)))
 		iova_len = roundup_pow_of_two(iova_len);
 
-	if (dev->bus_dma_mask)
+	dma_limit = min_not_zero(dma_limit, dev->bus_dma_limit);
-		dma_limit &= dev->bus_dma_mask;
 
 	if (domain->geometry.force_aperture)
 		dma_limit = min(dma_limit, domain->geometry.aperture_end);
@@ -659,7 +658,7 @@ static void iommu_dma_sync_single_for_cpu(struct device *dev,
 		return;
 
 	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
-	arch_sync_dma_for_cpu(dev, phys, size, dir);
+	arch_sync_dma_for_cpu(phys, size, dir);
 }
 
 static void iommu_dma_sync_single_for_device(struct device *dev,
@@ -671,7 +670,7 @@ static void iommu_dma_sync_single_for_device(struct device *dev,
 		return;
 
 	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
-	arch_sync_dma_for_device(dev, phys, size, dir);
+	arch_sync_dma_for_device(phys, size, dir);
 }
 
 static void iommu_dma_sync_sg_for_cpu(struct device *dev,
@@ -685,7 +684,7 @@ static void iommu_dma_sync_sg_for_cpu(struct device *dev,
 		return;
 
 	for_each_sg(sgl, sg, nelems, i)
-		arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length, dir);
+		arch_sync_dma_for_cpu(sg_phys(sg), sg->length, dir);
 }
 
 static void iommu_dma_sync_sg_for_device(struct device *dev,
@@ -699,7 +698,7 @@ static void iommu_dma_sync_sg_for_device(struct device *dev,
 		return;
 
 	for_each_sg(sgl, sg, nelems, i)
-		arch_sync_dma_for_device(dev, sg_phys(sg), sg->length, dir);
+		arch_sync_dma_for_device(sg_phys(sg), sg->length, dir);
 }
 
 static dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
@@ -714,7 +713,7 @@ static dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
 	dma_handle =__iommu_dma_map(dev, phys, size, prot);
 	if (!coherent && !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
 	    dma_handle != DMA_MAPPING_ERROR)
-		arch_sync_dma_for_device(dev, phys, size, dir);
+		arch_sync_dma_for_device(phys, size, dir);
 	return dma_handle;
 }
 

drivers/of/device.c

@@ -93,7 +93,7 @@ int of_dma_configure(struct device *dev, struct device_node *np, bool force_dma)
 	bool coherent;
 	unsigned long offset;
 	const struct iommu_ops *iommu;
-	u64 mask;
+	u64 mask, end;
 
 	ret = of_dma_get_range(np, &dma_addr, &paddr, &size);
 	if (ret < 0) {
@@ -148,12 +148,13 @@ int of_dma_configure(struct device *dev, struct device_node *np, bool force_dma)
 	 * Limit coherent and dma mask based on size and default mask
 	 * set by the driver.
 	 */
-	mask = DMA_BIT_MASK(ilog2(dma_addr + size - 1) + 1);
+	end = dma_addr + size - 1;
+	mask = DMA_BIT_MASK(ilog2(end) + 1);
 	dev->coherent_dma_mask &= mask;
 	*dev->dma_mask &= mask;
-	/* ...but only set bus mask if we found valid dma-ranges earlier */
+	/* ...but only set bus limit if we found valid dma-ranges earlier */
 	if (!ret)
-		dev->bus_dma_mask = mask;
+		dev->bus_dma_limit = end;
 
 	coherent = of_dma_is_coherent(np);
 	dev_dbg(dev, "device is%sdma coherent\n",

drivers/usb/core/hcd.c

@@ -1410,10 +1410,7 @@ int usb_hcd_map_urb_for_dma(struct usb_hcd *hcd, struct urb *urb,
 		if (hcd->self.uses_pio_for_control)
 			return ret;
 		if (hcd_uses_dma(hcd)) {
-			if (is_vmalloc_addr(urb->setup_packet)) {
+			if (object_is_on_stack(urb->setup_packet)) {
-				WARN_ONCE(1, "setup packet is not dma capable\n");
-				return -EAGAIN;
-			} else if (object_is_on_stack(urb->setup_packet)) {
 				WARN_ONCE(1, "setup packet is on stack\n");
 				return -EAGAIN;
 			}
@@ -1479,9 +1476,6 @@ int usb_hcd_map_urb_for_dma(struct usb_hcd *hcd, struct urb *urb,
 					ret = -EAGAIN;
 				else
 					urb->transfer_flags |= URB_DMA_MAP_PAGE;
-			} else if (is_vmalloc_addr(urb->transfer_buffer)) {
-				WARN_ONCE(1, "transfer buffer not dma capable\n");
-				ret = -EAGAIN;
 			} else if (object_is_on_stack(urb->transfer_buffer)) {
 				WARN_ONCE(1, "transfer buffer is on stack\n");
 				ret = -EAGAIN;

drivers/xen/swiotlb-xen.c

@@ -375,7 +375,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 	 * we can safely return the device addr and not worry about bounce
 	 * buffering it.
 	 */
-	if (dma_capable(dev, dev_addr, size) &&
+	if (dma_capable(dev, dev_addr, size, true) &&
 	    !range_straddles_page_boundary(phys, size) &&
 		!xen_arch_need_swiotlb(dev, phys, dev_addr) &&
 		swiotlb_force != SWIOTLB_FORCE)
@@ -397,7 +397,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 	/*
 	 * Ensure that the address returned is DMA'ble
 	 */
-	if (unlikely(!dma_capable(dev, dev_addr, size))) {
+	if (unlikely(!dma_capable(dev, dev_addr, size, true))) {
 		swiotlb_tbl_unmap_single(dev, map, size, size, dir,
 				attrs | DMA_ATTR_SKIP_CPU_SYNC);
 		return DMA_MAPPING_ERROR;
@@ -405,7 +405,7 @@ static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 
 done:
 	if (!dev_is_dma_coherent(dev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-		xen_dma_sync_for_device(dev, dev_addr, phys, size, dir);
+		xen_dma_sync_for_device(dev_addr, phys, size, dir);
 	return dev_addr;
 }
 
@@ -425,7 +425,7 @@ static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
 	BUG_ON(dir == DMA_NONE);
 
 	if (!dev_is_dma_coherent(hwdev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-		xen_dma_sync_for_cpu(hwdev, dev_addr, paddr, size, dir);
+		xen_dma_sync_for_cpu(dev_addr, paddr, size, dir);
 
 	/* NOTE: We use dev_addr here, not paddr! */
 	if (is_xen_swiotlb_buffer(dev_addr))
@@ -439,7 +439,7 @@ xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
 	phys_addr_t paddr = xen_bus_to_phys(dma_addr);
 
 	if (!dev_is_dma_coherent(dev))
-		xen_dma_sync_for_cpu(dev, dma_addr, paddr, size, dir);
+		xen_dma_sync_for_cpu(dma_addr, paddr, size, dir);
 
 	if (is_xen_swiotlb_buffer(dma_addr))
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_CPU);
@@ -455,7 +455,7 @@ xen_swiotlb_sync_single_for_device(struct device *dev, dma_addr_t dma_addr,
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_DEVICE);
 
 	if (!dev_is_dma_coherent(dev))
-		xen_dma_sync_for_device(dev, dma_addr, paddr, size, dir);
+		xen_dma_sync_for_device(dma_addr, paddr, size, dir);
 }
 
 /*

include/linux/device.h

@@ -1213,8 +1213,8 @@ struct dev_links_info {
  * @coherent_dma_mask: Like dma_mask, but for alloc_coherent mapping as not all
  * 		hardware supports 64-bit addresses for consistent allocations
  * 		such descriptors.
- * @bus_dma_mask: Mask of an upstream bridge or bus which imposes a smaller DMA
+ * @bus_dma_limit: Limit of an upstream bridge or bus which imposes a smaller
- *		limit than the device itself supports.
+ *		DMA limit than the device itself supports.
  * @dma_pfn_offset: offset of DMA memory range relatively of RAM
  * @dma_parms:	A low level driver may set these to teach IOMMU code about
  * 		segment limitations.
@@ -1300,7 +1300,7 @@ struct device {
 					     not all hardware supports
 					     64 bit addresses for consistent
 					     allocations such descriptors. */
-	u64		bus_dma_mask;	/* upstream dma_mask constraint */
+	u64		bus_dma_limit;	/* upstream dma constraint */
 	unsigned long	dma_pfn_offset;
 
 	struct device_dma_parameters *dma_parms;

include/linux/dma-direct.h

@@ -3,6 +3,7 @@
 #define _LINUX_DMA_DIRECT_H 1
 
 #include <linux/dma-mapping.h>
+#include <linux/memblock.h> /* for min_low_pfn */
 #include <linux/mem_encrypt.h>
 
 extern unsigned int zone_dma_bits;
@@ -23,15 +24,6 @@ static inline phys_addr_t __dma_to_phys(struct device *dev, dma_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 <=
-		min_not_zero(*dev->dma_mask, dev->bus_dma_mask);
-}
 #endif /* !CONFIG_ARCH_HAS_PHYS_TO_DMA */
 
 #ifdef CONFIG_ARCH_HAS_FORCE_DMA_UNENCRYPTED
@@ -59,6 +51,21 @@ static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t daddr)
 	return __sme_clr(__dma_to_phys(dev, daddr));
 }
 
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size,
+		bool is_ram)
+{
+	dma_addr_t end = addr + size - 1;
+
+	if (!dev->dma_mask)
+		return false;
+
+	if (is_ram && !IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) &&
+	    min(addr, end) < phys_to_dma(dev, PFN_PHYS(min_low_pfn)))
+		return false;
+
+	return end <= min_not_zero(*dev->dma_mask, dev->bus_dma_limit);
+}
+
 u64 dma_direct_get_required_mask(struct device *dev);
 void *dma_direct_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		gfp_t gfp, unsigned long attrs);
@@ -69,7 +76,13 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 void dma_direct_free_pages(struct device *dev, size_t size, void *cpu_addr,
 		dma_addr_t dma_addr, unsigned long attrs);
 struct page *__dma_direct_alloc_pages(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs);
+		gfp_t gfp, unsigned long attrs);
-void __dma_direct_free_pages(struct device *dev, size_t size, struct page *page);
+int dma_direct_get_sgtable(struct device *dev, struct sg_table *sgt,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs);
+bool dma_direct_can_mmap(struct device *dev);
+int dma_direct_mmap(struct device *dev, struct vm_area_struct *vma,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs);
 int dma_direct_supported(struct device *dev, u64 mask);
 #endif /* _LINUX_DMA_DIRECT_H */

include/linux/dma-mapping.h

@@ -159,7 +159,7 @@ int dma_release_from_dev_coherent(struct device *dev, int order, void *vaddr);
 int dma_mmap_from_dev_coherent(struct device *dev, struct vm_area_struct *vma,
 			    void *cpu_addr, size_t size, int *ret);
 
-void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle);
+void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size, dma_addr_t *dma_handle);
 int dma_release_from_global_coherent(int order, void *vaddr);
 int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *cpu_addr,
 				  size_t size, int *ret);
@@ -169,7 +169,7 @@ int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *cpu_addr,
 #define dma_release_from_dev_coherent(dev, order, vaddr) (0)
 #define dma_mmap_from_dev_coherent(dev, vma, vaddr, order, ret) (0)
 
-static inline void *dma_alloc_from_global_coherent(ssize_t size,
+static inline void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size,
 						   dma_addr_t *dma_handle)
 {
 	return NULL;
@@ -580,6 +580,10 @@ static inline unsigned long dma_get_merge_boundary(struct device *dev)
 static inline dma_addr_t dma_map_single_attrs(struct device *dev, void *ptr,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
+	/* DMA must never operate on areas that might be remapped. */
+	if (dev_WARN_ONCE(dev, is_vmalloc_addr(ptr),
+			  "rejecting DMA map of vmalloc memory\n"))
+		return DMA_MAPPING_ERROR;
 	debug_dma_map_single(dev, ptr, size);
 	return dma_map_page_attrs(dev, virt_to_page(ptr), offset_in_page(ptr),
 			size, dir, attrs);
@@ -690,7 +694,7 @@ static inline int dma_coerce_mask_and_coherent(struct device *dev, u64 mask)
  */
 static inline bool dma_addressing_limited(struct device *dev)
 {
-	return min_not_zero(dma_get_mask(dev), dev->bus_dma_mask) <
+	return min_not_zero(dma_get_mask(dev), dev->bus_dma_limit) <
 			    dma_get_required_mask(dev);
 }
 

include/linux/dma-noncoherent.h

@@ -41,8 +41,6 @@ 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,
 		dma_addr_t dma_addr, unsigned long attrs);
-long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
-		dma_addr_t dma_addr);
 
 #ifdef CONFIG_MMU
 /*
@@ -75,29 +73,29 @@ static inline void arch_dma_cache_sync(struct device *dev, void *vaddr,
 #endif /* CONFIG_DMA_NONCOHERENT_CACHE_SYNC */
 
 #ifdef CONFIG_ARCH_HAS_SYNC_DMA_FOR_DEVICE
-void arch_sync_dma_for_device(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir);
+		enum dma_data_direction dir);
 #else
-static inline void arch_sync_dma_for_device(struct device *dev,
+static inline void arch_sync_dma_for_device(phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 }
 #endif /* ARCH_HAS_SYNC_DMA_FOR_DEVICE */
 
 #ifdef CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU
-void arch_sync_dma_for_cpu(struct device *dev, phys_addr_t paddr,
+void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		size_t size, enum dma_data_direction dir);
+		enum dma_data_direction dir);
 #else
-static inline void arch_sync_dma_for_cpu(struct device *dev,
+static inline void arch_sync_dma_for_cpu(phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir)
+		enum dma_data_direction dir)
 {
 }
 #endif /* ARCH_HAS_SYNC_DMA_FOR_CPU */
 
 #ifdef CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU_ALL
-void arch_sync_dma_for_cpu_all(struct device *dev);
+void arch_sync_dma_for_cpu_all(void);
 #else
-static inline void arch_sync_dma_for_cpu_all(struct device *dev)
+static inline void arch_sync_dma_for_cpu_all(void)
 {
 }
 #endif /* CONFIG_ARCH_HAS_SYNC_DMA_FOR_CPU_ALL */

include/xen/swiotlb-xen.h

@@ -4,10 +4,10 @@
 
 #include <linux/swiotlb.h>
 
-void xen_dma_sync_for_cpu(struct device *dev, dma_addr_t handle,
+void xen_dma_sync_for_cpu(dma_addr_t handle, phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+		enum dma_data_direction dir);
-void xen_dma_sync_for_device(struct device *dev, dma_addr_t handle,
+void xen_dma_sync_for_device(dma_addr_t handle, phys_addr_t paddr, size_t size,
-		phys_addr_t paddr, size_t size, enum dma_data_direction dir);
+		enum dma_data_direction dir);
 
 extern int xen_swiotlb_init(int verbose, bool early);
 extern const struct dma_map_ops xen_swiotlb_dma_ops;

kernel/dma/Kconfig

@@ -51,9 +51,6 @@ config ARCH_HAS_SYNC_DMA_FOR_CPU_ALL
 config ARCH_HAS_DMA_PREP_COHERENT
 	bool
 
-config ARCH_HAS_DMA_COHERENT_TO_PFN
-	bool
-
 config ARCH_HAS_FORCE_DMA_UNENCRYPTED
 	bool
 
@@ -68,9 +65,18 @@ config SWIOTLB
 	bool
 	select NEED_DMA_MAP_STATE
 
+#
+# Should be selected if we can mmap non-coherent mappings to userspace.
+# The only thing that is really required is a way to set an uncached bit
+# in the pagetables
+#
+config DMA_NONCOHERENT_MMAP
+	bool
+
 config DMA_REMAP
 	depends on MMU
 	select GENERIC_ALLOCATOR
+	select DMA_NONCOHERENT_MMAP
 	bool
 
 config DMA_DIRECT_REMAP

kernel/dma/coherent.c

@@ -123,8 +123,9 @@ int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
 	return ret;
 }
 
-static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
+static void *__dma_alloc_from_coherent(struct device *dev,
-		ssize_t size, dma_addr_t *dma_handle)
+				       struct dma_coherent_mem *mem,
+				       ssize_t size, dma_addr_t *dma_handle)
 {
 	int order = get_order(size);
 	unsigned long flags;
@@ -143,7 +144,7 @@ static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
 	/*
 	 * Memory was found in the coherent area.
 	 */
-	*dma_handle = mem->device_base + (pageno << PAGE_SHIFT);
+	*dma_handle = dma_get_device_base(dev, mem) + (pageno << PAGE_SHIFT);
 	ret = mem->virt_base + (pageno << PAGE_SHIFT);
 	spin_unlock_irqrestore(&mem->spinlock, flags);
 	memset(ret, 0, size);
@@ -175,17 +176,18 @@ int dma_alloc_from_dev_coherent(struct device *dev, ssize_t size,
 	if (!mem)
 		return 0;
 
-	*ret = __dma_alloc_from_coherent(mem, size, dma_handle);
+	*ret = __dma_alloc_from_coherent(dev, mem, size, dma_handle);
 	return 1;
 }
 
-void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle)
+void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size,
+				     dma_addr_t *dma_handle)
 {
 	if (!dma_coherent_default_memory)
 		return NULL;
 
-	return __dma_alloc_from_coherent(dma_coherent_default_memory, size,
+	return __dma_alloc_from_coherent(dev, dma_coherent_default_memory, size,
-			dma_handle);
+					 dma_handle);
 }
 
 static int __dma_release_from_coherent(struct dma_coherent_mem *mem,

kernel/dma/contiguous.c

@@ -42,10 +42,11 @@ struct cma *dma_contiguous_default_area;
  * Users, who want to set the size of global CMA area for their system
  * should use cma= kernel parameter.
  */
-static const phys_addr_t size_bytes = (phys_addr_t)CMA_SIZE_MBYTES * SZ_1M;
+static const phys_addr_t size_bytes __initconst =
-static phys_addr_t size_cmdline = -1;
+	(phys_addr_t)CMA_SIZE_MBYTES * SZ_1M;
-static phys_addr_t base_cmdline;
+static phys_addr_t  size_cmdline __initdata = -1;
-static phys_addr_t limit_cmdline;
+static phys_addr_t base_cmdline __initdata;
+static phys_addr_t limit_cmdline __initdata;
 
 static int __init early_cma(char *p)
 {

kernel/dma/debug.c

@@ -27,7 +27,7 @@
 
 #include <asm/sections.h>
 
-#define HASH_SIZE       1024ULL
+#define HASH_SIZE       16384ULL
 #define HASH_FN_SHIFT   13
 #define HASH_FN_MASK    (HASH_SIZE - 1)
 
@@ -54,40 +54,40 @@ enum map_err_types {
  * struct dma_debug_entry - track a dma_map* or dma_alloc_coherent mapping
  * @list: node on pre-allocated free_entries list
  * @dev: 'dev' argument to dma_map_{page|single|sg} or dma_alloc_coherent
- * @type: single, page, sg, coherent
- * @pfn: page frame of the start address
- * @offset: offset of mapping relative to pfn
  * @size: length of the mapping
+ * @type: single, page, sg, coherent
  * @direction: enum dma_data_direction
  * @sg_call_ents: 'nents' from dma_map_sg
  * @sg_mapped_ents: 'mapped_ents' from dma_map_sg
+ * @pfn: page frame of the start address
+ * @offset: offset of mapping relative to pfn
  * @map_err_type: track whether dma_mapping_error() was checked
  * @stacktrace: support backtraces when a violation is detected
  */
 struct dma_debug_entry {
 	struct list_head list;
 	struct device    *dev;
-	int              type;
-	unsigned long	 pfn;
-	size_t		 offset;
 	u64              dev_addr;
 	u64              size;
+	int              type;
 	int              direction;
 	int		 sg_call_ents;
 	int		 sg_mapped_ents;
+	unsigned long	 pfn;
+	size_t		 offset;
 	enum map_err_types  map_err_type;
 #ifdef CONFIG_STACKTRACE
 	unsigned int	stack_len;
 	unsigned long	stack_entries[DMA_DEBUG_STACKTRACE_ENTRIES];
 #endif
-};
+} ____cacheline_aligned_in_smp;
 
 typedef bool (*match_fn)(struct dma_debug_entry *, struct dma_debug_entry *);
 
 struct hash_bucket {
 	struct list_head list;
 	spinlock_t lock;
-} ____cacheline_aligned_in_smp;
+};
 
 /* Hash list to save the allocated dma addresses */
 static struct hash_bucket dma_entry_hash[HASH_SIZE];
@@ -255,12 +255,10 @@ static struct hash_bucket *get_hash_bucket(struct dma_debug_entry *entry,
  * Give up exclusive access to the hash bucket
  */
 static void put_hash_bucket(struct hash_bucket *bucket,
-			    unsigned long *flags)
+			    unsigned long flags)
 	__releases(&bucket->lock)
 {
-	unsigned long __flags = *flags;
+	spin_unlock_irqrestore(&bucket->lock, flags);
-
-	spin_unlock_irqrestore(&bucket->lock, __flags);
 }
 
 static bool exact_match(struct dma_debug_entry *a, struct dma_debug_entry *b)
@@ -359,7 +357,7 @@ static struct dma_debug_entry *bucket_find_contain(struct hash_bucket **bucket,
 		/*
 		 * Nothing found, go back a hash bucket
 		 */
-		put_hash_bucket(*bucket, flags);
+		put_hash_bucket(*bucket, *flags);
 		range          += (1 << HASH_FN_SHIFT);
 		index.dev_addr -= (1 << HASH_FN_SHIFT);
 		*bucket = get_hash_bucket(&index, flags);
@@ -420,6 +418,7 @@ void debug_dma_dump_mappings(struct device *dev)
 		}
 
 		spin_unlock_irqrestore(&bucket->lock, flags);
+		cond_resched();
 	}
 }
 
@@ -608,7 +607,7 @@ static void add_dma_entry(struct dma_debug_entry *entry)
 
 	bucket = get_hash_bucket(entry, &flags);
 	hash_bucket_add(bucket, entry);
-	put_hash_bucket(bucket, &flags);
+	put_hash_bucket(bucket, flags);
 
 	rc = active_cacheline_insert(entry);
 	if (rc == -ENOMEM) {
@@ -1001,7 +1000,7 @@ static void check_unmap(struct dma_debug_entry *ref)
 
 	if (!entry) {
 		/* must drop lock before calling dma_mapping_error */
-		put_hash_bucket(bucket, &flags);
+		put_hash_bucket(bucket, flags);
 
 		if (dma_mapping_error(ref->dev, ref->dev_addr)) {
 			err_printk(ref->dev, NULL,
@@ -1083,7 +1082,7 @@ static void check_unmap(struct dma_debug_entry *ref)
 	hash_bucket_del(entry);
 	dma_entry_free(entry);
 
-	put_hash_bucket(bucket, &flags);
+	put_hash_bucket(bucket, flags);
 }
 
 static void check_for_stack(struct device *dev,
@@ -1203,7 +1202,7 @@ static void check_sync(struct device *dev,
 	}
 
 out:
-	put_hash_bucket(bucket, &flags);
+	put_hash_bucket(bucket, flags);
 }
 
 static void check_sg_segment(struct device *dev, struct scatterlist *sg)
@@ -1318,7 +1317,7 @@ void debug_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 		}
 	}
 
-	put_hash_bucket(bucket, &flags);
+	put_hash_bucket(bucket, flags);
 }
 EXPORT_SYMBOL(debug_dma_mapping_error);
 
@@ -1391,7 +1390,7 @@ static int get_nr_mapped_entries(struct device *dev,
 
 	if (entry)
 		mapped_ents = entry->sg_mapped_ents;
-	put_hash_bucket(bucket, &flags);
+	put_hash_bucket(bucket, flags);
 
 	return mapped_ents;
 }

kernel/dma/direct.c

@@ -12,6 +12,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/dma-noncoherent.h>
 #include <linux/pfn.h>
+#include <linux/vmalloc.h>
 #include <linux/set_memory.h>
 #include <linux/swiotlb.h>
 
@@ -26,10 +27,10 @@ static void report_addr(struct device *dev, dma_addr_t dma_addr, size_t size)
 {
 	if (!dev->dma_mask) {
 		dev_err_once(dev, "DMA map on device without dma_mask\n");
-	} else if (*dev->dma_mask >= DMA_BIT_MASK(32) || dev->bus_dma_mask) {
+	} else if (*dev->dma_mask >= DMA_BIT_MASK(32) || dev->bus_dma_limit) {
 		dev_err_once(dev,
-			"overflow %pad+%zu of DMA mask %llx bus mask %llx\n",
+			"overflow %pad+%zu of DMA mask %llx bus limit %llx\n",
-			&dma_addr, size, *dev->dma_mask, dev->bus_dma_mask);
+			&dma_addr, size, *dev->dma_mask, dev->bus_dma_limit);
 	}
 	WARN_ON_ONCE(1);
 }
@@ -42,6 +43,12 @@ static inline dma_addr_t phys_to_dma_direct(struct device *dev,
 	return phys_to_dma(dev, phys);
 }
 
+static inline struct page *dma_direct_to_page(struct device *dev,
+		dma_addr_t dma_addr)
+{
+	return pfn_to_page(PHYS_PFN(dma_to_phys(dev, dma_addr)));
+}
+
 u64 dma_direct_get_required_mask(struct device *dev)
 {
 	u64 max_dma = phys_to_dma_direct(dev, (max_pfn - 1) << PAGE_SHIFT);
@@ -50,15 +57,14 @@ u64 dma_direct_get_required_mask(struct device *dev)
 }
 
 static gfp_t __dma_direct_optimal_gfp_mask(struct device *dev, u64 dma_mask,
-		u64 *phys_mask)
+		u64 *phys_limit)
 {
-	if (dev->bus_dma_mask && dev->bus_dma_mask < dma_mask)
+	u64 dma_limit = min_not_zero(dma_mask, dev->bus_dma_limit);
-		dma_mask = dev->bus_dma_mask;
 
 	if (force_dma_unencrypted(dev))
-		*phys_mask = __dma_to_phys(dev, dma_mask);
+		*phys_limit = __dma_to_phys(dev, dma_limit);
 	else
-		*phys_mask = dma_to_phys(dev, dma_mask);
+		*phys_limit = dma_to_phys(dev, dma_limit);
 
 	/*
 	 * Optimistically try the zone that the physical address mask falls
@@ -68,9 +74,9 @@ static gfp_t __dma_direct_optimal_gfp_mask(struct device *dev, u64 dma_mask,
 	 * Note that GFP_DMA32 and GFP_DMA are no ops without the corresponding
 	 * zones.
 	 */
-	if (*phys_mask <= DMA_BIT_MASK(zone_dma_bits))
+	if (*phys_limit <= DMA_BIT_MASK(zone_dma_bits))
 		return GFP_DMA;
-	if (*phys_mask <= DMA_BIT_MASK(32))
+	if (*phys_limit <= DMA_BIT_MASK(32))
 		return GFP_DMA32;
 	return 0;
 }
@@ -78,16 +84,16 @@ static gfp_t __dma_direct_optimal_gfp_mask(struct device *dev, u64 dma_mask,
 static bool dma_coherent_ok(struct device *dev, phys_addr_t phys, size_t size)
 {
 	return phys_to_dma_direct(dev, phys) + size - 1 <=
-			min_not_zero(dev->coherent_dma_mask, dev->bus_dma_mask);
+			min_not_zero(dev->coherent_dma_mask, dev->bus_dma_limit);
 }
 
 struct page *__dma_direct_alloc_pages(struct device *dev, size_t size,
-		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
+		gfp_t gfp, unsigned long attrs)
 {
 	size_t alloc_size = PAGE_ALIGN(size);
 	int node = dev_to_node(dev);
 	struct page *page = NULL;
-	u64 phys_mask;
+	u64 phys_limit;
 
 	if (attrs & DMA_ATTR_NO_WARN)
 		gfp |= __GFP_NOWARN;
@@ -95,7 +101,7 @@ struct page *__dma_direct_alloc_pages(struct device *dev, size_t size,
 	/* we always manually zero the memory once we are done: */
 	gfp &= ~__GFP_ZERO;
 	gfp |= __dma_direct_optimal_gfp_mask(dev, dev->coherent_dma_mask,
-			&phys_mask);
+			&phys_limit);
 	page = dma_alloc_contiguous(dev, alloc_size, gfp);
 	if (page && !dma_coherent_ok(dev, page_to_phys(page), size)) {
 		dma_free_contiguous(dev, page, alloc_size);
@@ -109,7 +115,7 @@ again:
 		page = NULL;
 
 		if (IS_ENABLED(CONFIG_ZONE_DMA32) &&
-		    phys_mask < DMA_BIT_MASK(64) &&
+		    phys_limit < DMA_BIT_MASK(64) &&
 		    !(gfp & (GFP_DMA32 | GFP_DMA))) {
 			gfp |= GFP_DMA32;
 			goto again;
@@ -130,7 +136,16 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 	struct page *page;
 	void *ret;
 
-	page = __dma_direct_alloc_pages(dev, size, dma_handle, gfp, attrs);
+	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
+	    dma_alloc_need_uncached(dev, attrs) &&
+	    !gfpflags_allow_blocking(gfp)) {
+		ret = dma_alloc_from_pool(PAGE_ALIGN(size), &page, gfp);
+		if (!ret)
+			return NULL;
+		goto done;
+	}
+
+	page = __dma_direct_alloc_pages(dev, size, gfp, attrs);
 	if (!page)
 		return NULL;
 
@@ -139,9 +154,28 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 		/* remove any dirty cache lines on the kernel alias */
 		if (!PageHighMem(page))
 			arch_dma_prep_coherent(page, size);
-		*dma_handle = phys_to_dma(dev, page_to_phys(page));
 		/* return the page pointer as the opaque cookie */
-		return page;
+		ret = page;
+		goto done;
+	}
+
+	if ((IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
+	     dma_alloc_need_uncached(dev, attrs)) ||
+	    (IS_ENABLED(CONFIG_DMA_REMAP) && PageHighMem(page))) {
+		/* remove any dirty cache lines on the kernel alias */
+		arch_dma_prep_coherent(page, PAGE_ALIGN(size));
+
+		/* create a coherent mapping */
+		ret = dma_common_contiguous_remap(page, PAGE_ALIGN(size),
+				dma_pgprot(dev, PAGE_KERNEL, attrs),
+				__builtin_return_address(0));
+		if (!ret) {
+			dma_free_contiguous(dev, page, size);
+			return ret;
+		}
+
+		memset(ret, 0, size);
+		goto done;
 	}
 
 	if (PageHighMem(page)) {
@@ -152,17 +186,14 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 		 * so log an error and fail.
 		 */
 		dev_info(dev, "Rejecting highmem page from CMA.\n");
-		__dma_direct_free_pages(dev, size, page);
+		dma_free_contiguous(dev, page, size);
 		return NULL;
 	}
 
 	ret = page_address(page);
-	if (force_dma_unencrypted(dev)) {
+	if (force_dma_unencrypted(dev))
 		set_memory_decrypted((unsigned long)ret, 1 << get_order(size));
-		*dma_handle = __phys_to_dma(dev, page_to_phys(page));
+
-	} else {
-		*dma_handle = phys_to_dma(dev, page_to_phys(page));
-	}
 	memset(ret, 0, size);
 
 	if (IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
@@ -170,15 +201,14 @@ void *dma_direct_alloc_pages(struct device *dev, size_t size,
 		arch_dma_prep_coherent(page, size);
 		ret = uncached_kernel_address(ret);
 	}
-
+done:
+	if (force_dma_unencrypted(dev))
+		*dma_handle = __phys_to_dma(dev, page_to_phys(page));
+	else
+		*dma_handle = phys_to_dma(dev, page_to_phys(page));
 	return ret;
 }
 
-void __dma_direct_free_pages(struct device *dev, size_t size, struct page *page)
-{
-	dma_free_contiguous(dev, page, size);
-}
-
 void dma_direct_free_pages(struct device *dev, size_t size, void *cpu_addr,
 		dma_addr_t dma_addr, unsigned long attrs)
 {
@@ -187,23 +217,28 @@ void dma_direct_free_pages(struct device *dev, size_t size, void *cpu_addr,
 	if ((attrs & DMA_ATTR_NO_KERNEL_MAPPING) &&
 	    !force_dma_unencrypted(dev)) {
 		/* cpu_addr is a struct page cookie, not a kernel address */
-		__dma_direct_free_pages(dev, size, cpu_addr);
+		dma_free_contiguous(dev, cpu_addr, size);
 		return;
 	}
 
+	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
+	    dma_free_from_pool(cpu_addr, PAGE_ALIGN(size)))
+		return;
+
 	if (force_dma_unencrypted(dev))
 		set_memory_encrypted((unsigned long)cpu_addr, 1 << page_order);
 
-	if (IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr))
-	    dma_alloc_need_uncached(dev, attrs))
+		vunmap(cpu_addr);
-		cpu_addr = cached_kernel_address(cpu_addr);
+
-	__dma_direct_free_pages(dev, size, virt_to_page(cpu_addr));
+	dma_free_contiguous(dev, dma_direct_to_page(dev, dma_addr), size);
 }
 
 void *dma_direct_alloc(struct device *dev, size_t size,
 		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
 {
 	if (!IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    !IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
 	    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);
@@ -213,6 +248,7 @@ void dma_direct_free(struct device *dev, size_t size,
 		void *cpu_addr, dma_addr_t dma_addr, unsigned long attrs)
 {
 	if (!IS_ENABLED(CONFIG_ARCH_HAS_UNCACHED_SEGMENT) &&
+	    !IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
 	    dma_alloc_need_uncached(dev, attrs))
 		arch_dma_free(dev, size, cpu_addr, dma_addr, attrs);
 	else
@@ -230,7 +266,7 @@ void dma_direct_sync_single_for_device(struct device *dev,
 		swiotlb_tbl_sync_single(dev, paddr, size, dir, SYNC_FOR_DEVICE);
 
 	if (!dev_is_dma_coherent(dev))
-		arch_sync_dma_for_device(dev, paddr, size, dir);
+		arch_sync_dma_for_device(paddr, size, dir);
 }
 EXPORT_SYMBOL(dma_direct_sync_single_for_device);
 
@@ -248,7 +284,7 @@ void dma_direct_sync_sg_for_device(struct device *dev,
 					dir, SYNC_FOR_DEVICE);
 
 		if (!dev_is_dma_coherent(dev))
-			arch_sync_dma_for_device(dev, paddr, sg->length,
+			arch_sync_dma_for_device(paddr, sg->length,
 					dir);
 	}
 }
@@ -264,8 +300,8 @@ void dma_direct_sync_single_for_cpu(struct device *dev,
 	phys_addr_t paddr = dma_to_phys(dev, addr);
 
 	if (!dev_is_dma_coherent(dev)) {
-		arch_sync_dma_for_cpu(dev, paddr, size, dir);
+		arch_sync_dma_for_cpu(paddr, size, dir);
-		arch_sync_dma_for_cpu_all(dev);
+		arch_sync_dma_for_cpu_all();
 	}
 
 	if (unlikely(is_swiotlb_buffer(paddr)))
@@ -283,7 +319,7 @@ void dma_direct_sync_sg_for_cpu(struct device *dev,
 		phys_addr_t paddr = dma_to_phys(dev, sg_dma_address(sg));
 
 		if (!dev_is_dma_coherent(dev))
-			arch_sync_dma_for_cpu(dev, paddr, sg->length, dir);
+			arch_sync_dma_for_cpu(paddr, sg->length, dir);
 
 		if (unlikely(is_swiotlb_buffer(paddr)))
 			swiotlb_tbl_sync_single(dev, paddr, sg->length, dir,
@@ -291,7 +327,7 @@ void dma_direct_sync_sg_for_cpu(struct device *dev,
 	}
 
 	if (!dev_is_dma_coherent(dev))
-		arch_sync_dma_for_cpu_all(dev);
+		arch_sync_dma_for_cpu_all();
 }
 EXPORT_SYMBOL(dma_direct_sync_sg_for_cpu);
 
@@ -325,7 +361,7 @@ static inline bool dma_direct_possible(struct device *dev, dma_addr_t dma_addr,
 		size_t size)
 {
 	return swiotlb_force != SWIOTLB_FORCE &&
-		dma_capable(dev, dma_addr, size);
+		dma_capable(dev, dma_addr, size, true);
 }
 
 dma_addr_t dma_direct_map_page(struct device *dev, struct page *page,
@@ -342,7 +378,7 @@ dma_addr_t dma_direct_map_page(struct device *dev, struct page *page,
 	}
 
 	if (!dev_is_dma_coherent(dev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))
-		arch_sync_dma_for_device(dev, phys, size, dir);
+		arch_sync_dma_for_device(phys, size, dir);
 	return dma_addr;
 }
 EXPORT_SYMBOL(dma_direct_map_page);
@@ -374,7 +410,7 @@ dma_addr_t dma_direct_map_resource(struct device *dev, phys_addr_t paddr,
 {
 	dma_addr_t dma_addr = paddr;
 
-	if (unlikely(!dma_direct_possible(dev, dma_addr, size))) {
+	if (unlikely(!dma_capable(dev, dma_addr, size, false))) {
 		report_addr(dev, dma_addr, size);
 		return DMA_MAPPING_ERROR;
 	}
@@ -383,6 +419,59 @@ dma_addr_t dma_direct_map_resource(struct device *dev, phys_addr_t paddr,
 }
 EXPORT_SYMBOL(dma_direct_map_resource);
 
+int dma_direct_get_sgtable(struct device *dev, struct sg_table *sgt,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs)
+{
+	struct page *page = dma_direct_to_page(dev, dma_addr);
+	int ret;
+
+	ret = sg_alloc_table(sgt, 1, GFP_KERNEL);
+	if (!ret)
+		sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0);
+	return ret;
+}
+
+#ifdef CONFIG_MMU
+bool dma_direct_can_mmap(struct device *dev)
+{
+	return dev_is_dma_coherent(dev) ||
+		IS_ENABLED(CONFIG_DMA_NONCOHERENT_MMAP);
+}
+
+int dma_direct_mmap(struct device *dev, struct vm_area_struct *vma,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs)
+{
+	unsigned long user_count = vma_pages(vma);
+	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned long pfn = PHYS_PFN(dma_to_phys(dev, dma_addr));
+	int ret = -ENXIO;
+
+	vma->vm_page_prot = dma_pgprot(dev, vma->vm_page_prot, attrs);
+
+	if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret))
+		return ret;
+
+	if (vma->vm_pgoff >= count || user_count > count - vma->vm_pgoff)
+		return -ENXIO;
+	return remap_pfn_range(vma, vma->vm_start, pfn + vma->vm_pgoff,
+			user_count << PAGE_SHIFT, vma->vm_page_prot);
+}
+#else /* CONFIG_MMU */
+bool dma_direct_can_mmap(struct device *dev)
+{
+	return false;
+}
+
+int dma_direct_mmap(struct device *dev, struct vm_area_struct *vma,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		unsigned long attrs)
+{
+	return -ENXIO;
+}
+#endif /* CONFIG_MMU */
+
 /*
  * 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

kernel/dma/mapping.c

@@ -112,24 +112,9 @@ int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
 		 void *cpu_addr, dma_addr_t dma_addr, size_t size,
 		 unsigned long attrs)
 {
-	struct page *page;
+	struct page *page = virt_to_page(cpu_addr);
 	int ret;
 
-	if (!dev_is_dma_coherent(dev)) {
-		unsigned long pfn;
-
-		if (!IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN))
-			return -ENXIO;
-
-		/* If the PFN is not valid, we do not have a struct page */
-		pfn = arch_dma_coherent_to_pfn(dev, cpu_addr, dma_addr);
-		if (!pfn_valid(pfn))
-			return -ENXIO;
-		page = pfn_to_page(pfn);
-	} else {
-		page = virt_to_page(cpu_addr);
-	}
-
 	ret = sg_alloc_table(sgt, 1, GFP_KERNEL);
 	if (!ret)
 		sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0);
@@ -154,7 +139,7 @@ int dma_get_sgtable_attrs(struct device *dev, struct sg_table *sgt,
 	const struct dma_map_ops *ops = get_dma_ops(dev);
 
 	if (dma_is_direct(ops))
-		return dma_common_get_sgtable(dev, sgt, cpu_addr, dma_addr,
+		return dma_direct_get_sgtable(dev, sgt, cpu_addr, dma_addr,
 				size, attrs);
 	if (!ops->get_sgtable)
 		return -ENXIO;
@@ -192,7 +177,6 @@ int dma_common_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 off = vma->vm_pgoff;
-	unsigned long pfn;
 	int ret = -ENXIO;
 
 	vma->vm_page_prot = dma_pgprot(dev, vma->vm_page_prot, attrs);
@@ -203,19 +187,8 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
 	if (off >= count || user_count > count - off)
 		return -ENXIO;
 
-	if (!dev_is_dma_coherent(dev)) {
+	return remap_pfn_range(vma, vma->vm_start,
-		if (!IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN))
+			page_to_pfn(virt_to_page(cpu_addr)) + vma->vm_pgoff,
-			return -ENXIO;
-
-		/* If the PFN is not valid, we do not have a struct page */
-		pfn = arch_dma_coherent_to_pfn(dev, cpu_addr, dma_addr);
-		if (!pfn_valid(pfn))
-			return -ENXIO;
-	} else {
-		pfn = page_to_pfn(virt_to_page(cpu_addr));
-	}
-
-	return remap_pfn_range(vma, vma->vm_start, pfn + vma->vm_pgoff,
 			user_count << PAGE_SHIFT, vma->vm_page_prot);
 #else
 	return -ENXIO;
@@ -233,12 +206,8 @@ bool dma_can_mmap(struct device *dev)
 {
 	const struct dma_map_ops *ops = get_dma_ops(dev);
 
-	if (dma_is_direct(ops)) {
+	if (dma_is_direct(ops))
-		return IS_ENABLED(CONFIG_MMU) &&
+		return dma_direct_can_mmap(dev);
-		       (dev_is_dma_coherent(dev) ||
-			IS_ENABLED(CONFIG_ARCH_HAS_DMA_COHERENT_TO_PFN));
-	}
-
 	return ops->mmap != NULL;
 }
 EXPORT_SYMBOL_GPL(dma_can_mmap);
@@ -263,7 +232,7 @@ int dma_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
 	const struct dma_map_ops *ops = get_dma_ops(dev);
 
 	if (dma_is_direct(ops))
-		return dma_common_mmap(dev, vma, cpu_addr, dma_addr, size,
+		return dma_direct_mmap(dev, vma, cpu_addr, dma_addr, size,
 				attrs);
 	if (!ops->mmap)
 		return -ENXIO;

kernel/dma/remap.c

@@ -210,59 +210,4 @@ bool dma_free_from_pool(void *start, size_t size)
 	gen_pool_free(atomic_pool, (unsigned long)start, size);
 	return true;
 }
-
-void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		gfp_t flags, unsigned long attrs)
-{
-	struct page *page = NULL;
-	void *ret;
-
-	size = PAGE_ALIGN(size);
-
-	if (!gfpflags_allow_blocking(flags)) {
-		ret = dma_alloc_from_pool(size, &page, flags);
-		if (!ret)
-			return NULL;
-		goto done;
-	}
-
-	page = __dma_direct_alloc_pages(dev, size, dma_handle, flags, attrs);
-	if (!page)
-		return NULL;
-
-	/* remove any dirty cache lines on the kernel alias */
-	arch_dma_prep_coherent(page, size);
-
-	/* create a coherent mapping */
-	ret = dma_common_contiguous_remap(page, size,
-			dma_pgprot(dev, PAGE_KERNEL, attrs),
-			__builtin_return_address(0));
-	if (!ret) {
-		__dma_direct_free_pages(dev, size, page);
-		return ret;
-	}
-
-	memset(ret, 0, size);
-done:
-	*dma_handle = phys_to_dma(dev, page_to_phys(page));
-	return ret;
-}
-
-void arch_dma_free(struct device *dev, size_t size, void *vaddr,
-		dma_addr_t dma_handle, unsigned long attrs)
-{
-	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));
-
-		vunmap(vaddr);
-		__dma_direct_free_pages(dev, size, page);
-	}
-}
-
-long arch_dma_coherent_to_pfn(struct device *dev, void *cpu_addr,
-		dma_addr_t dma_addr)
-{
-	return __phys_to_pfn(dma_to_phys(dev, dma_addr));
-}
 #endif /* CONFIG_DMA_DIRECT_REMAP */

kernel/dma/swiotlb.c

@@ -678,7 +678,7 @@ bool swiotlb_map(struct device *dev, phys_addr_t *phys, dma_addr_t *dma_addr,
 
 	/* Ensure that the address returned is DMA'ble */
 	*dma_addr = __phys_to_dma(dev, *phys);
-	if (unlikely(!dma_capable(dev, *dma_addr, size))) {
+	if (unlikely(!dma_capable(dev, *dma_addr, size, true))) {
 		swiotlb_tbl_unmap_single(dev, *phys, size, size, dir,
 			attrs | DMA_ATTR_SKIP_CPU_SYNC);
 		return false;