Merge branch 'akpm' (patches from Andrew)

Merge third patch-bomb from Andrew Morton:
 "I'm pretty much done for -rc1 now:

   - the rest of MM, basically

   - lib/ updates

   - checkpatch, epoll, hfs, fatfs, ptrace, coredump, exit

   - cpu_mask simplifications

   - kexec, rapidio, MAINTAINERS etc, etc.

   - more dma-mapping cleanups/simplifications from hch"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (109 commits)
  MAINTAINERS: add/fix git URLs for various subsystems
  mm: memcontrol: add "sock" to cgroup2 memory.stat
  mm: memcontrol: basic memory statistics in cgroup2 memory controller
  mm: memcontrol: do not uncharge old page in page cache replacement
  Documentation: cgroup: add memory.swap.{current,max} description
  mm: free swap cache aggressively if memcg swap is full
  mm: vmscan: do not scan anon pages if memcg swap limit is hit
  swap.h: move memcg related stuff to the end of the file
  mm: memcontrol: replace mem_cgroup_lruvec_online with mem_cgroup_online
  mm: vmscan: pass memcg to get_scan_count()
  mm: memcontrol: charge swap to cgroup2
  mm: memcontrol: clean up alloc, online, offline, free functions
  mm: memcontrol: flatten struct cg_proto
  mm: memcontrol: rein in the CONFIG space madness
  net: drop tcp_memcontrol.c
  mm: memcontrol: introduce CONFIG_MEMCG_LEGACY_KMEM
  mm: memcontrol: allow to disable kmem accounting for cgroup2
  mm: memcontrol: account "kmem" consumers in cgroup2 memory controller
  mm: memcontrol: move kmem accounting code to CONFIG_MEMCG
  mm: memcontrol: separate kmem code from legacy tcp accounting code
  ...

CREDITS

@@ -1856,6 +1856,16 @@ S: Korte Heul 95
 S: 1403 ND  BUSSUM
 S: The Netherlands
 
+N: Martin Kepplinger
+E: martink@posteo.de
+E: martin.kepplinger@theobroma-systems.com
+W: http://www.martinkepplinger.com
+D: mma8452 accelerators iio driver
+D: Kernel cleanups
+S: Garnisonstraße 26
+S: 4020 Linz
+S: Austria
+
 N: Karl Keyte
 E: karl@koft.com
 D: Disk usage statistics and modifications to line printer driver

Documentation/DMA-API-HOWTO.txt

@@ -951,16 +951,6 @@ to "Closing".
    alignment constraints (e.g. the alignment constraints about 64-bit
    objects).
 
-3) Supporting multiple types of IOMMUs
-
-   If your architecture needs to support multiple types of IOMMUs, you
-   can use include/linux/asm-generic/dma-mapping-common.h. It's a
-   library to support the DMA API with multiple types of IOMMUs. Lots
-   of architectures (x86, powerpc, sh, alpha, ia64, microblaze and
-   sparc) use it. Choose one to see how it can be used. If you need to
-   support multiple types of IOMMUs in a single system, the example of
-   x86 or powerpc helps.
-
 			   Closing
 
 This document, and the API itself, would not be in its current

Documentation/cgroup-v2.txt

@@ -819,6 +819,78 @@ PAGE_SIZE multiple when read back.
 		the cgroup.  This may not exactly match the number of
 		processes killed but should generally be close.
 
+  memory.stat
+
+	A read-only flat-keyed file which exists on non-root cgroups.
+
+	This breaks down the cgroup's memory footprint into different
+	types of memory, type-specific details, and other information
+	on the state and past events of the memory management system.
+
+	All memory amounts are in bytes.
+
+	The entries are ordered to be human readable, and new entries
+	can show up in the middle. Don't rely on items remaining in a
+	fixed position; use the keys to look up specific values!
+
+	  anon
+
+		Amount of memory used in anonymous mappings such as
+		brk(), sbrk(), and mmap(MAP_ANONYMOUS)
+
+	  file
+
+		Amount of memory used to cache filesystem data,
+		including tmpfs and shared memory.
+
+	  file_mapped
+
+		Amount of cached filesystem data mapped with mmap()
+
+	  file_dirty
+
+		Amount of cached filesystem data that was modified but
+		not yet written back to disk
+
+	  file_writeback
+
+		Amount of cached filesystem data that was modified and
+		is currently being written back to disk
+
+	  inactive_anon
+	  active_anon
+	  inactive_file
+	  active_file
+	  unevictable
+
+		Amount of memory, swap-backed and filesystem-backed,
+		on the internal memory management lists used by the
+		page reclaim algorithm
+
+	  pgfault
+
+		Total number of page faults incurred
+
+	  pgmajfault
+
+		Number of major page faults incurred
+
+  memory.swap.current
+
+	A read-only single value file which exists on non-root
+	cgroups.
+
+	The total amount of swap currently being used by the cgroup
+	and its descendants.
+
+  memory.swap.max
+
+	A read-write single value file which exists on non-root
+	cgroups.  The default is "max".
+
+	Swap usage hard limit.  If a cgroup's swap usage reaches this
+	limit, anonymous meomry of the cgroup will not be swapped out.
+
 
 5-2-2. General Usage
 
@@ -1291,3 +1363,20 @@ allocation from the slack available in other groups or the rest of the
 system than killing the group.  Otherwise, memory.max is there to
 limit this type of spillover and ultimately contain buggy or even
 malicious applications.
+
+The combined memory+swap accounting and limiting is replaced by real
+control over swap space.
+
+The main argument for a combined memory+swap facility in the original
+cgroup design was that global or parental pressure would always be
+able to swap all anonymous memory of a child group, regardless of the
+child's own (possibly untrusted) configuration.  However, untrusted
+groups can sabotage swapping by other means - such as referencing its
+anonymous memory in a tight loop - and an admin can not assume full
+swappability when overcommitting untrusted jobs.
+
+For trusted jobs, on the other hand, a combined counter is not an
+intuitive userspace interface, and it flies in the face of the idea
+that cgroup controllers should account and limit specific physical
+resources.  Swap space is a resource like all others in the system,
+and that's why unified hierarchy allows distributing it separately.

Documentation/features/io/dma_map_attrs/arch-support.txt

-#
-# Feature name:          dma_map_attrs
-#         Kconfig:       HAVE_DMA_ATTRS
-#         description:   arch provides dma_*map*_attrs() APIs
-#
-    -----------------------
-    |         arch |status|
-    -----------------------
-    |       alpha: |  ok  |
-    |         arc: | TODO |
-    |         arm: |  ok  |
-    |       arm64: |  ok  |
-    |       avr32: | TODO |
-    |    blackfin: | TODO |
-    |         c6x: | TODO |
-    |        cris: | TODO |
-    |         frv: | TODO |
-    |       h8300: |  ok  |
-    |     hexagon: |  ok  |
-    |        ia64: |  ok  |
-    |        m32r: | TODO |
-    |        m68k: | TODO |
-    |       metag: | TODO |
-    |  microblaze: |  ok  |
-    |        mips: |  ok  |
-    |     mn10300: | TODO |
-    |       nios2: | TODO |
-    |    openrisc: |  ok  |
-    |      parisc: | TODO |
-    |     powerpc: |  ok  |
-    |        s390: |  ok  |
-    |       score: | TODO |
-    |          sh: |  ok  |
-    |       sparc: |  ok  |
-    |        tile: |  ok  |
-    |          um: | TODO |
-    |   unicore32: |  ok  |
-    |         x86: |  ok  |
-    |      xtensa: | TODO |
-    -----------------------

Documentation/filesystems/vfat.txt

@@ -180,6 +180,16 @@ dos1xfloppy  -- If set, use a fallback default BIOS Parameter Block
 
 <bool>: 0,1,yes,no,true,false
 
+LIMITATION
+---------------------------------------------------------------------
+* The fallocated region of file is discarded at umount/evict time
+  when using fallocate with FALLOC_FL_KEEP_SIZE.
+  So, User should assume that fallocated region can be discarded at
+  last close if there is memory pressure resulting in eviction of
+  the inode from the memory. As a result, for any dependency on
+  the fallocated region, user should make sure to recheck fallocate
+  after reopening the file.
+
 TODO
 ----------------------------------------------------------------------
 * Need to get rid of the raw scanning stuff.  Instead, always use

Documentation/kernel-parameters.txt

@@ -611,6 +611,7 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 	cgroup.memory=	[KNL] Pass options to the cgroup memory controller.
 			Format: <string>
 			nosocket -- Disable socket memory accounting.
+			nokmem -- Disable kernel memory accounting.
 
 	checkreqprot	[SELINUX] Set initial checkreqprot flag value.
 			Format: { "0" | "1" }

Documentation/sysctl/kernel.txt

@@ -825,14 +825,13 @@ via the /proc/sys interface:
        Each write syscall must fully contain the sysctl value to be
        written, and multiple writes on the same sysctl file descriptor
        will rewrite the sysctl value, regardless of file position.
-   0 - (default) Same behavior as above, but warn about processes that
-       perform writes to a sysctl file descriptor when the file position
-       is not 0.
-   1 - Respect file position when writing sysctl strings. Multiple writes
-       will append to the sysctl value buffer. Anything past the max length
-       of the sysctl value buffer will be ignored. Writes to numeric sysctl
-       entries must always be at file position 0 and the value must be
-       fully contained in the buffer sent in the write syscall.
+   0 - Same behavior as above, but warn about processes that perform writes
+       to a sysctl file descriptor when the file position is not 0.
+   1 - (default) Respect file position when writing sysctl strings. Multiple
+       writes will append to the sysctl value buffer. Anything past the max
+       length of the sysctl value buffer will be ignored. Writes to numeric
+       sysctl entries must always be at file position 0 and the value must
+       be fully contained in the buffer sent in the write syscall.
 
 ==============================================================
 

Documentation/ubsan.txt

+Undefined Behavior Sanitizer - UBSAN
+
+Overview
+--------
+
+UBSAN is a runtime undefined behaviour checker.
+
+UBSAN uses compile-time instrumentation to catch undefined behavior (UB).
+Compiler inserts code that perform certain kinds of checks before operations
+that may cause UB. If check fails (i.e. UB detected) __ubsan_handle_*
+function called to print error message.
+
+GCC has that feature since 4.9.x [1] (see -fsanitize=undefined option and
+its suboptions). GCC 5.x has more checkers implemented [2].
+
+Report example
+---------------
+
+	 ================================================================================
+	 UBSAN: Undefined behaviour in ../include/linux/bitops.h:110:33
+	 shift exponent 32 is to large for 32-bit type 'unsigned int'
+	 CPU: 0 PID: 0 Comm: swapper Not tainted 4.4.0-rc1+ #26
+	  0000000000000000 ffffffff82403cc8 ffffffff815e6cd6 0000000000000001
+	  ffffffff82403cf8 ffffffff82403ce0 ffffffff8163a5ed 0000000000000020
+	  ffffffff82403d78 ffffffff8163ac2b ffffffff815f0001 0000000000000002
+	 Call Trace:
+	  [<ffffffff815e6cd6>] dump_stack+0x45/0x5f
+	  [<ffffffff8163a5ed>] ubsan_epilogue+0xd/0x40
+	  [<ffffffff8163ac2b>] __ubsan_handle_shift_out_of_bounds+0xeb/0x130
+	  [<ffffffff815f0001>] ? radix_tree_gang_lookup_slot+0x51/0x150
+	  [<ffffffff8173c586>] _mix_pool_bytes+0x1e6/0x480
+	  [<ffffffff83105653>] ? dmi_walk_early+0x48/0x5c
+	  [<ffffffff8173c881>] add_device_randomness+0x61/0x130
+	  [<ffffffff83105b35>] ? dmi_save_one_device+0xaa/0xaa
+	  [<ffffffff83105653>] dmi_walk_early+0x48/0x5c
+	  [<ffffffff831066ae>] dmi_scan_machine+0x278/0x4b4
+	  [<ffffffff8111d58a>] ? vprintk_default+0x1a/0x20
+	  [<ffffffff830ad120>] ? early_idt_handler_array+0x120/0x120
+	  [<ffffffff830b2240>] setup_arch+0x405/0xc2c
+	  [<ffffffff830ad120>] ? early_idt_handler_array+0x120/0x120
+	  [<ffffffff830ae053>] start_kernel+0x83/0x49a
+	  [<ffffffff830ad120>] ? early_idt_handler_array+0x120/0x120
+	  [<ffffffff830ad386>] x86_64_start_reservations+0x2a/0x2c
+	  [<ffffffff830ad4f3>] x86_64_start_kernel+0x16b/0x17a
+	 ================================================================================
+
+Usage
+-----
+
+To enable UBSAN configure kernel with:
+
+	CONFIG_UBSAN=y
+
+and to check the entire kernel:
+
+        CONFIG_UBSAN_SANITIZE_ALL=y
+
+To enable instrumentation for specific files or directories, add a line
+similar to the following to the respective kernel Makefile:
+
+        For a single file (e.g. main.o):
+                UBSAN_SANITIZE_main.o := y
+
+        For all files in one directory:
+                UBSAN_SANITIZE := y
+
+To exclude files from being instrumented even if
+CONFIG_UBSAN_SANITIZE_ALL=y, use:
+
+                UBSAN_SANITIZE_main.o := n
+        and:
+                UBSAN_SANITIZE := n
+
+Detection of unaligned accesses controlled through the separate option -
+CONFIG_UBSAN_ALIGNMENT. It's off by default on architectures that support
+unaligned accesses (CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y). One could
+still enable it in config, just note that it will produce a lot of UBSAN
+reports.
+
+References
+----------
+
+[1] - https://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Debugging-Options.html
+[2] - https://gcc.gnu.org/onlinedocs/gcc/Debugging-Options.html

Makefile

@@ -411,7 +411,7 @@ export MAKE AWK GENKSYMS INSTALLKERNEL PERL PYTHON UTS_MACHINE
 export HOSTCXX HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
 
 export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS LDFLAGS
-export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE CFLAGS_GCOV CFLAGS_KASAN
+export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE CFLAGS_GCOV CFLAGS_KASAN CFLAGS_UBSAN
 export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE
 export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_LDFLAGS_MODULE
 export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL
@@ -784,6 +784,7 @@ endif
 
 include scripts/Makefile.kasan
 include scripts/Makefile.extrawarn
+include scripts/Makefile.ubsan
 
 # Add any arch overrides and user supplied CPPFLAGS, AFLAGS and CFLAGS as the
 # last assignments

arch/Kconfig

@@ -205,9 +205,6 @@ config HAVE_NMI_WATCHDOG
 config HAVE_ARCH_TRACEHOOK
 	bool
 
-config HAVE_DMA_ATTRS
-	bool
-
 config HAVE_DMA_CONTIGUOUS
 	bool
 
@@ -632,4 +629,7 @@ config OLD_SIGACTION
 config COMPAT_OLD_SIGACTION
 	bool
 
+config ARCH_NO_COHERENT_DMA_MMAP
+	bool
+
 source "kernel/gcov/Kconfig"

arch/alpha/Kconfig

@@ -9,7 +9,6 @@ config ALPHA
 	select HAVE_OPROFILE
 	select HAVE_PCSPKR_PLATFORM
 	select HAVE_PERF_EVENTS
-	select HAVE_DMA_ATTRS
 	select VIRT_TO_BUS
 	select GENERIC_IRQ_PROBE
 	select AUTO_IRQ_AFFINITY if SMP

arch/alpha/include/asm/dma-mapping.h

@@ -10,8 +10,6 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 	return dma_ops;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 #define dma_cache_sync(dev, va, size, dir)		  ((void)0)
 
 #endif	/* _ALPHA_DMA_MAPPING_H */

arch/alpha/include/uapi/asm/mman.h

@@ -47,7 +47,6 @@
 #define MADV_WILLNEED	3		/* will need these pages */
 #define	MADV_SPACEAVAIL	5		/* ensure resources are available */
 #define MADV_DONTNEED	6		/* don't need these pages */
-#define MADV_FREE	7		/* free pages only if memory pressure */
 
 /* common/generic parameters */
 #define MADV_FREE	8		/* free pages only if memory pressure */

arch/arc/include/asm/dma-mapping.h

@@ -11,192 +11,11 @@
 #ifndef ASM_ARC_DMA_MAPPING_H
 #define ASM_ARC_DMA_MAPPING_H
 
-#include <asm-generic/dma-coherent.h>
-#include <asm/cacheflush.h>
+extern struct dma_map_ops arc_dma_ops;
 
-void *dma_alloc_noncoherent(struct device *dev, size_t size,
-			    dma_addr_t *dma_handle, gfp_t gfp);
-
-void dma_free_noncoherent(struct device *dev, size_t size, void *vaddr,
-			  dma_addr_t dma_handle);
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t gfp);
-
-void dma_free_coherent(struct device *dev, size_t size, void *kvaddr,
-		       dma_addr_t dma_handle);
-
-/* drivers/base/dma-mapping.c */
-extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
-			   void *cpu_addr, dma_addr_t dma_addr, size_t size);
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size);
-
-#define dma_mmap_coherent(d, v, c, h, s) dma_common_mmap(d, v, c, h, s)
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
-/*
- * streaming DMA Mapping API...
- * CPU accesses page via normal paddr, thus needs to explicitly made
- * consistent before each use
- */
-
-static inline void __inline_dma_cache_sync(unsigned long paddr, size_t size,
-					   enum dma_data_direction dir)
-{
-	switch (dir) {
-	case DMA_FROM_DEVICE:
-		dma_cache_inv(paddr, size);
-		break;
-	case DMA_TO_DEVICE:
-		dma_cache_wback(paddr, size);
-		break;
-	case DMA_BIDIRECTIONAL:
-		dma_cache_wback_inv(paddr, size);
-		break;
-	default:
-		pr_err("Invalid DMA dir [%d] for OP @ %lx\n", dir, paddr);
-	}
-}
-
-void __arc_dma_cache_sync(unsigned long paddr, size_t size,
-			  enum dma_data_direction dir);
-
-#define _dma_cache_sync(addr, sz, dir)			\
-do {							\
-	if (__builtin_constant_p(dir))			\
-		__inline_dma_cache_sync(addr, sz, dir);	\
-	else						\
-		__arc_dma_cache_sync(addr, sz, dir);	\
-}							\
-while (0);
-
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *cpu_addr, size_t size,
-	       enum dma_data_direction dir)
-{
-	_dma_cache_sync((unsigned long)cpu_addr, size, dir);
-	return (dma_addr_t)cpu_addr;
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
-		 size_t size, enum dma_data_direction dir)
-{
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page,
-	     unsigned long offset, size_t size,
-	     enum dma_data_direction dir)
-{
-	unsigned long paddr = page_to_phys(page) + offset;
-	return dma_map_single(dev, (void *)paddr, size, dir);
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_handle,
-	       size_t size, enum dma_data_direction dir)
-{
-}
-
-static inline int
-dma_map_sg(struct device *dev, struct scatterlist *sg,
-	   int nents, enum dma_data_direction dir)
-{
-	struct scatterlist *s;
-	int i;
-
-	for_each_sg(sg, s, nents, i)
-		s->dma_address = dma_map_page(dev, sg_page(s), s->offset,
-					       s->length, dir);
-
-	return nents;
-}
-
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-	     int nents, enum dma_data_direction dir)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	struct scatterlist *s;
-	int i;
-
-	for_each_sg(sg, s, nents, i)
-		dma_unmap_page(dev, sg_dma_address(s), sg_dma_len(s), dir);
-}
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			size_t size, enum dma_data_direction dir)
-{
-	_dma_cache_sync(dma_handle, size, DMA_FROM_DEVICE);
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
-			   size_t size, enum dma_data_direction dir)
-{
-	_dma_cache_sync(dma_handle, size, DMA_TO_DEVICE);
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			      unsigned long offset, size_t size,
-			      enum dma_data_direction direction)
-{
-	_dma_cache_sync(dma_handle + offset, size, DMA_FROM_DEVICE);
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction direction)
-{
-	_dma_cache_sync(dma_handle + offset, size, DMA_TO_DEVICE);
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist, int nelems,
-		    enum dma_data_direction dir)
-{
-	int i;
-	struct scatterlist *sg;
-
-	for_each_sg(sglist, sg, nelems, i)
-		_dma_cache_sync((unsigned int)sg_virt(sg), sg->length, dir);
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist,
-		       int nelems, enum dma_data_direction dir)
-{
-	int i;
-	struct scatterlist *sg;
-
-	for_each_sg(sglist, sg, nelems, i)
-		_dma_cache_sync((unsigned int)sg_virt(sg), sg->length, dir);
-}
-
-static inline int dma_supported(struct device *dev, u64 dma_mask)
-{
-	/* Support 32 bit DMA mask exclusively */
-	return dma_mask == DMA_BIT_MASK(32);
-}
-
-static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-static inline int dma_set_mask(struct device *dev, u64 dma_mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
-		return -EIO;
-
-	*dev->dma_mask = dma_mask;
-
-	return 0;
+	return &arc_dma_ops;
 }
 
 #endif

arch/arc/mm/dma.c

@@ -17,18 +17,14 @@
  */
 
 #include <linux/dma-mapping.h>
-#include <linux/dma-debug.h>
-#include <linux/export.h>
 #include <asm/cache.h>
 #include <asm/cacheflush.h>
 
-/*
- * Helpers for Coherent DMA API.
- */
-void *dma_alloc_noncoherent(struct device *dev, size_t size,
-			    dma_addr_t *dma_handle, gfp_t gfp)
+
+static void *arc_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
-	void *paddr;
+	void *paddr, *kvaddr;
 
 	/* This is linear addr (0x8000_0000 based) */
 	paddr = alloc_pages_exact(size, gfp);
@@ -38,22 +34,6 @@ void *dma_alloc_noncoherent(struct device *dev, size_t size,
 	/* This is bus address, platform dependent */
 	*dma_handle = (dma_addr_t)paddr;
 
-	return paddr;
-}
-EXPORT_SYMBOL(dma_alloc_noncoherent);
-
-void dma_free_noncoherent(struct device *dev, size_t size, void *vaddr,
-			  dma_addr_t dma_handle)
-{
-	free_pages_exact((void *)dma_handle, size);
-}
-EXPORT_SYMBOL(dma_free_noncoherent);
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t gfp)
-{
-	void *paddr, *kvaddr;
-
 	/*
 	 * IOC relies on all data (even coherent DMA data) being in cache
 	 * Thus allocate normal cached memory
@@ -65,22 +45,15 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	 *   -For coherent data, Read/Write to buffers terminate early in cache
 	 *   (vs. always going to memory - thus are faster)
 	 */
-	if (is_isa_arcv2() && ioc_exists)
-		return dma_alloc_noncoherent(dev, size, dma_handle, gfp);
-
-	/* This is linear addr (0x8000_0000 based) */
-	paddr = alloc_pages_exact(size, gfp);
-	if (!paddr)
-		return NULL;
+	if ((is_isa_arcv2() && ioc_exists) ||
+	    dma_get_attr(DMA_ATTR_NON_CONSISTENT, attrs))
+		return paddr;
 
 	/* This is kernel Virtual address (0x7000_0000 based) */
 	kvaddr = ioremap_nocache((unsigned long)paddr, size);
 	if (kvaddr == NULL)
 		return NULL;
 
-	/* This is bus address, platform dependent */
-	*dma_handle = (dma_addr_t)paddr;
-
 	/*
 	 * Evict any existing L1 and/or L2 lines for the backing page
 	 * in case it was used earlier as a normal "cached" page.
@@ -95,26 +68,111 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 
 	return kvaddr;
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
-void dma_free_coherent(struct device *dev, size_t size, void *kvaddr,
-		       dma_addr_t dma_handle)
+static void arc_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
-	if (is_isa_arcv2() && ioc_exists)
-		return dma_free_noncoherent(dev, size, kvaddr, dma_handle);
-
-	iounmap((void __force __iomem *)kvaddr);
+	if (!dma_get_attr(DMA_ATTR_NON_CONSISTENT, attrs) &&
+	    !(is_isa_arcv2() && ioc_exists))
+		iounmap((void __force __iomem *)vaddr);
 
 	free_pages_exact((void *)dma_handle, size);
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
 /*
- * Helper for streaming DMA...
+ * streaming DMA Mapping API...
+ * CPU accesses page via normal paddr, thus needs to explicitly made
+ * consistent before each use
  */
-void __arc_dma_cache_sync(unsigned long paddr, size_t size,
-			  enum dma_data_direction dir)
+static void _dma_cache_sync(unsigned long paddr, size_t size,
+		enum dma_data_direction dir)
+{
+	switch (dir) {
+	case DMA_FROM_DEVICE:
+		dma_cache_inv(paddr, size);
+		break;
+	case DMA_TO_DEVICE:
+		dma_cache_wback(paddr, size);
+		break;
+	case DMA_BIDIRECTIONAL:
+		dma_cache_wback_inv(paddr, size);
+		break;
+	default:
+		pr_err("Invalid DMA dir [%d] for OP @ %lx\n", dir, paddr);
+	}
+}
+
+static dma_addr_t arc_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
+{
+	unsigned long paddr = page_to_phys(page) + offset;
+	_dma_cache_sync(paddr, size, dir);
+	return (dma_addr_t)paddr;
+}
+
+static int arc_dma_map_sg(struct device *dev, struct scatterlist *sg,
+	   int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
+{
+	struct scatterlist *s;
+	int i;
+
+	for_each_sg(sg, s, nents, i)
+		s->dma_address = dma_map_page(dev, sg_page(s), s->offset,
+					       s->length, dir);
+
+	return nents;
+}
+
+static void arc_dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
+{
+	_dma_cache_sync(dma_handle, size, DMA_FROM_DEVICE);
+}
+
+static void arc_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
 {
-	__inline_dma_cache_sync(paddr, size, dir);
+	_dma_cache_sync(dma_handle, size, DMA_TO_DEVICE);
 }
-EXPORT_SYMBOL(__arc_dma_cache_sync);
+
+static void arc_dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sglist, int nelems,
+		enum dma_data_direction dir)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sglist, sg, nelems, i)
+		_dma_cache_sync((unsigned int)sg_virt(sg), sg->length, dir);
+}
+
+static void arc_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sglist, int nelems,
+		enum dma_data_direction dir)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sglist, sg, nelems, i)
+		_dma_cache_sync((unsigned int)sg_virt(sg), sg->length, dir);
+}
+
+static int arc_dma_supported(struct device *dev, u64 dma_mask)
+{
+	/* Support 32 bit DMA mask exclusively */
+	return dma_mask == DMA_BIT_MASK(32);
+}
+
+struct dma_map_ops arc_dma_ops = {
+	.alloc			= arc_dma_alloc,
+	.free			= arc_dma_free,
+	.map_page		= arc_dma_map_page,
+	.map_sg			= arc_dma_map_sg,
+	.sync_single_for_device	= arc_dma_sync_single_for_device,
+	.sync_single_for_cpu	= arc_dma_sync_single_for_cpu,
+	.sync_sg_for_cpu	= arc_dma_sync_sg_for_cpu,
+	.sync_sg_for_device	= arc_dma_sync_sg_for_device,
+	.dma_supported		= arc_dma_supported,
+};
+EXPORT_SYMBOL(arc_dma_ops);

arch/arm/Kconfig

@@ -47,7 +47,6 @@ config ARM
 	select HAVE_C_RECORDMCOUNT
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_CONTIGUOUS if MMU
 	select HAVE_DYNAMIC_FTRACE if (!XIP_KERNEL) && !CPU_ENDIAN_BE32 && MMU
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && MMU

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

@@ -41,13 +41,6 @@ static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops)
 #define HAVE_ARCH_DMA_SUPPORTED 1
 extern int dma_supported(struct device *dev, u64 mask);
 
-/*
- * Note that while the generic code provides dummy dma_{alloc,free}_noncoherent
- * implementations, we don't provide a dma_cache_sync function so drivers using
- * this API are highlighted with build warnings.
- */
-#include <asm-generic/dma-mapping-common.h>
-
 #ifdef __arch_page_to_dma
 #error Please update to __arch_pfn_to_dma
 #endif

arch/arm64/Kconfig

@@ -64,7 +64,6 @@ config ARM64
 	select HAVE_DEBUG_BUGVERBOSE
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS

arch/arm64/include/asm/dma-mapping.h

@@ -64,8 +64,6 @@ static inline bool is_device_dma_coherent(struct device *dev)
 	return dev->archdata.dma_coherent;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
 {
 	return (dma_addr_t)paddr;

arch/avr32/mm/dma-coherent.c

@@ -9,9 +9,14 @@
 #include <linux/dma-mapping.h>
 #include <linux/gfp.h>
 #include <linux/export.h>
+#include <linux/mm.h>
+#include <linux/device.h>
+#include <linux/scatterlist.h>
 
-#include <asm/addrspace.h>
+#include <asm/processor.h>
 #include <asm/cacheflush.h>
+#include <asm/io.h>
+#include <asm/addrspace.h>
 
 void dma_cache_sync(struct device *dev, void *vaddr, size_t size, int direction)
 {
@@ -93,60 +98,100 @@ static void __dma_free(struct device *dev, size_t size,
 		__free_page(page++);
 }
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *handle, gfp_t gfp)
+static void *avr32_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	struct page *page;
-	void *ret = NULL;
+	dma_addr_t phys;
 
 	page = __dma_alloc(dev, size, handle, gfp);
-	if (page)
-		ret = phys_to_uncached(page_to_phys(page));
+	if (!page)
+		return NULL;
+	phys = page_to_phys(page);
 
-	return ret;
+	if (dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs)) {
+		/* Now, map the page into P3 with write-combining turned on */
+		*handle = phys;
+		return __ioremap(phys, size, _PAGE_BUFFER);
+	} else {
+		return phys_to_uncached(phys);
+	}
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
-void dma_free_coherent(struct device *dev, size_t size,
-		       void *cpu_addr, dma_addr_t handle)
+static void avr32_dma_free(struct device *dev, size_t size,
+		void *cpu_addr, dma_addr_t handle, struct dma_attrs *attrs)
 {
-	void *addr = phys_to_cached(uncached_to_phys(cpu_addr));
 	struct page *page;
 
-	pr_debug("dma_free_coherent addr %p (phys %08lx) size %u\n",
-		 cpu_addr, (unsigned long)handle, (unsigned)size);
-	BUG_ON(!virt_addr_valid(addr));
-	page = virt_to_page(addr);
+	if (dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs)) {
+		iounmap(cpu_addr);
+
+		page = phys_to_page(handle);
+	} else {
+		void *addr = phys_to_cached(uncached_to_phys(cpu_addr));
+
+		pr_debug("avr32_dma_free addr %p (phys %08lx) size %u\n",
+			 cpu_addr, (unsigned long)handle, (unsigned)size);
+
+		BUG_ON(!virt_addr_valid(addr));
+		page = virt_to_page(addr);
+	}
+
 	__dma_free(dev, size, page, handle);
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
-void *dma_alloc_writecombine(struct device *dev, size_t size,
-			     dma_addr_t *handle, gfp_t gfp)
+static dma_addr_t avr32_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
 {
-	struct page *page;
-	dma_addr_t phys;
+	void *cpu_addr = page_address(page) + offset;
 
-	page = __dma_alloc(dev, size, handle, gfp);
-	if (!page)
-		return NULL;
+	dma_cache_sync(dev, cpu_addr, size, direction);
+	return virt_to_bus(cpu_addr);
+}
 
-	phys = page_to_phys(page);
-	*handle = phys;
+static int avr32_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sglist, sg, nents, i) {
+		char *virt;
 
-	/* Now, map the page into P3 with write-combining turned on */
-	return __ioremap(phys, size, _PAGE_BUFFER);
+		sg->dma_address = page_to_bus(sg_page(sg)) + sg->offset;
+		virt = sg_virt(sg);
+		dma_cache_sync(dev, virt, sg->length, direction);
+	}
+
+	return nents;
 }
-EXPORT_SYMBOL(dma_alloc_writecombine);
 
-void dma_free_writecombine(struct device *dev, size_t size,
-			   void *cpu_addr, dma_addr_t handle)
+static void avr32_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
-	struct page *page;
+	dma_cache_sync(dev, bus_to_virt(dma_handle), size, direction);
+}
 
-	iounmap(cpu_addr);
+static void avr32_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sglist, int nents,
+		enum dma_data_direction direction)
+{
+	int i;
+	struct scatterlist *sg;
 
-	page = phys_to_page(handle);
-	__dma_free(dev, size, page, handle);
+	for_each_sg(sglist, sg, nents, i)
+		dma_cache_sync(dev, sg_virt(sg), sg->length, direction);
 }
-EXPORT_SYMBOL(dma_free_writecombine);
+
+struct dma_map_ops avr32_dma_ops = {
+	.alloc			= avr32_dma_alloc,
+	.free			= avr32_dma_free,
+	.map_page		= avr32_dma_map_page,
+	.map_sg			= avr32_dma_map_sg,
+	.sync_single_for_device	= avr32_dma_sync_single_for_device,
+	.sync_sg_for_device	= avr32_dma_sync_sg_for_device,
+};
+EXPORT_SYMBOL(avr32_dma_ops);

arch/blackfin/include/asm/dma-mapping.h

@@ -8,36 +8,6 @@
 #define _BLACKFIN_DMA_MAPPING_H
 
 #include <asm/cacheflush.h>
-struct scatterlist;
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t gfp);
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-		       dma_addr_t dma_handle);
-
-/*
- * Now for the API extensions over the pci_ one
- */
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-#define dma_supported(d, m)         (1)
-
-static inline int
-dma_set_mask(struct device *dev, u64 dma_mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
-		return -EIO;
-
-	*dev->dma_mask = dma_mask;
-
-	return 0;
-}
-
-static inline int
-dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
 
 extern void
 __dma_sync(dma_addr_t addr, size_t size, enum dma_data_direction dir);
@@ -66,102 +36,11 @@ _dma_sync(dma_addr_t addr, size_t size, enum dma_data_direction dir)
 		__dma_sync(addr, size, dir);
 }
 
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *ptr, size_t size,
-	       enum dma_data_direction dir)
-{
-	_dma_sync((dma_addr_t)ptr, size, dir);
-	return (dma_addr_t) ptr;
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page,
-	     unsigned long offset, size_t size,
-	     enum dma_data_direction dir)
-{
-	return dma_map_single(dev, page_address(page) + offset, size, dir);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		 enum dma_data_direction dir)
-{
-	BUG_ON(!valid_dma_direction(dir));
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size,
-	       enum dma_data_direction dir)
-{
-	dma_unmap_single(dev, dma_addr, size, dir);
-}
+extern struct dma_map_ops bfin_dma_ops;
 
-extern int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-		      enum dma_data_direction dir);
-
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-	     int nhwentries, enum dma_data_direction dir)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	BUG_ON(!valid_dma_direction(dir));
+	return &bfin_dma_ops;
 }
 
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t handle,
-			      unsigned long offset, size_t size,
-			      enum dma_data_direction dir)
-{
-	BUG_ON(!valid_dma_direction(dir));
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction dir)
-{
-	_dma_sync(handle + offset, size, dir);
-}
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle, size_t size,
-			enum dma_data_direction dir)
-{
-	dma_sync_single_range_for_cpu(dev, handle, 0, size, dir);
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t handle, size_t size,
-			   enum dma_data_direction dir)
-{
-	dma_sync_single_range_for_device(dev, handle, 0, size, dir);
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nents,
-		    enum dma_data_direction dir)
-{
-	BUG_ON(!valid_dma_direction(dir));
-}
-
-extern void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-		       int nents, enum dma_data_direction dir);
-
-static inline void
-dma_cache_sync(struct device *dev, void *vaddr, size_t size,
-	       enum dma_data_direction dir)
-{
-	_dma_sync((dma_addr_t)vaddr, size, dir);
-}
-
-/* drivers/base/dma-mapping.c */
-extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
-			   void *cpu_addr, dma_addr_t dma_addr, size_t size);
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size);
-
-#define dma_mmap_coherent(d, v, c, h, s) dma_common_mmap(d, v, c, h, s)
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
 #endif				/* _BLACKFIN_DMA_MAPPING_H */

arch/blackfin/kernel/dma-mapping.c

@@ -78,8 +78,8 @@ static void __free_dma_pages(unsigned long addr, unsigned int pages)
 	spin_unlock_irqrestore(&dma_page_lock, flags);
 }
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t gfp)
+static void *bfin_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	void *ret;
 
@@ -92,15 +92,12 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 
 	return ret;
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
-void
-dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-		  dma_addr_t dma_handle)
+static void bfin_dma_free(struct device *dev, size_t size, void *vaddr,
+		  dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	__free_dma_pages((unsigned long)vaddr, get_pages(size));
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
 /*
  * Streaming DMA mappings
@@ -112,9 +109,9 @@ void __dma_sync(dma_addr_t addr, size_t size,
 }
 EXPORT_SYMBOL(__dma_sync);
 
-int
-dma_map_sg(struct device *dev, struct scatterlist *sg_list, int nents,
-	   enum dma_data_direction direction)
+static int bfin_dma_map_sg(struct device *dev, struct scatterlist *sg_list,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	struct scatterlist *sg;
 	int i;
@@ -126,10 +123,10 @@ dma_map_sg(struct device *dev, struct scatterlist *sg_list, int nents,
 
 	return nents;
 }
-EXPORT_SYMBOL(dma_map_sg);
 
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg_list,
-			    int nelems, enum dma_data_direction direction)
+static void bfin_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sg_list, int nelems,
+		enum dma_data_direction direction)
 {
 	struct scatterlist *sg;
 	int i;
@@ -139,4 +136,31 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg_list,
 		__dma_sync(sg_dma_address(sg), sg_dma_len(sg), direction);
 	}
 }
-EXPORT_SYMBOL(dma_sync_sg_for_device);
+
+static dma_addr_t bfin_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
+{
+	dma_addr_t handle = (dma_addr_t)(page_address(page) + offset);
+
+	_dma_sync(handle, size, dir);
+	return handle;
+}
+
+static inline void bfin_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
+{
+	_dma_sync(handle, size, dir);
+}
+
+struct dma_map_ops bfin_dma_ops = {
+	.alloc			= bfin_dma_alloc,
+	.free			= bfin_dma_free,
+
+	.map_page		= bfin_dma_map_page,
+	.map_sg			= bfin_dma_map_sg,
+
+	.sync_single_for_device	= bfin_dma_sync_single_for_device,
+	.sync_sg_for_device	= bfin_dma_sync_sg_for_device,
+};
+EXPORT_SYMBOL(bfin_dma_ops);

arch/c6x/Kconfig

@@ -17,6 +17,7 @@ config C6X
 	select OF_EARLY_FLATTREE
 	select GENERIC_CLOCKEVENTS
 	select MODULES_USE_ELF_RELA
+	select ARCH_NO_COHERENT_DMA_MMAP
 
 config MMU
 	def_bool n

arch/c6x/include/asm/dma-mapping.h

@@ -12,104 +12,22 @@
 #ifndef _ASM_C6X_DMA_MAPPING_H
 #define _ASM_C6X_DMA_MAPPING_H
 
-#include <linux/dma-debug.h>
-#include <asm-generic/dma-coherent.h>
-
-#define dma_supported(d, m)	1
-
-static inline void dma_sync_single_range_for_device(struct device *dev,
-						    dma_addr_t addr,
-						    unsigned long offset,
-						    size_t size,
-						    enum dma_data_direction dir)
-{
-}
-
-static inline int dma_set_mask(struct device *dev, u64 dma_mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
-		return -EIO;
-
-	*dev->dma_mask = dma_mask;
-
-	return 0;
-}
-
 /*
  * DMA errors are defined by all-bits-set in the DMA address.
  */
-static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	debug_dma_mapping_error(dev, dma_addr);
-	return dma_addr == ~0;
-}
-
-extern dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
-				 size_t size, enum dma_data_direction dir);
+#define DMA_ERROR_CODE ~0
 
-extern void dma_unmap_single(struct device *dev, dma_addr_t handle,
-			     size_t size, enum dma_data_direction dir);
+extern struct dma_map_ops c6x_dma_ops;
 
-extern int dma_map_sg(struct device *dev, struct scatterlist *sglist,
-		      int nents, enum dma_data_direction direction);
-
-extern void dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
-			 int nents, enum dma_data_direction direction);
-
-static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
-				      unsigned long offset, size_t size,
-				      enum dma_data_direction dir)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	dma_addr_t handle;
-
-	handle = dma_map_single(dev, page_address(page) + offset, size, dir);
-
-	debug_dma_map_page(dev, page, offset, size, dir, handle, false);
-
-	return handle;
-}
-
-static inline void dma_unmap_page(struct device *dev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir)
-{
-	dma_unmap_single(dev, handle, size, dir);
-
-	debug_dma_unmap_page(dev, handle, size, dir, false);
+	return &c6x_dma_ops;
 }
 
-extern void dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle,
-				    size_t size, enum dma_data_direction dir);
-
-extern void dma_sync_single_for_device(struct device *dev, dma_addr_t handle,
-				       size_t size,
-				       enum dma_data_direction dir);
-
-extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-				int nents, enum dma_data_direction dir);
-
-extern void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-				   int nents, enum dma_data_direction dir);
-
 extern void coherent_mem_init(u32 start, u32 size);
-extern void *dma_alloc_coherent(struct device *, size_t, dma_addr_t *, gfp_t);
-extern void dma_free_coherent(struct device *, size_t, void *, dma_addr_t);
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent((d), (s), (h), (f))
-#define dma_free_noncoherent(d, s, v, h)  dma_free_coherent((d), (s), (v), (h))
-
-/* Not supported for now */
-static inline int dma_mmap_coherent(struct device *dev,
-				    struct vm_area_struct *vma, void *cpu_addr,
-				    dma_addr_t dma_addr, size_t size)
-{
-	return -EINVAL;
-}
-
-static inline int dma_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size)
-{
-	return -EINVAL;
-}
+void *c6x_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t gfp, struct dma_attrs *attrs);
+void c6x_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs);
 
 #endif	/* _ASM_C6X_DMA_MAPPING_H */

arch/c6x/kernel/dma.c

@@ -36,110 +36,101 @@ static void c6x_dma_sync(dma_addr_t handle, size_t size,
 	}
 }
 
-dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
-			  enum dma_data_direction dir)
+static dma_addr_t c6x_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
 {
-	dma_addr_t addr = virt_to_phys(ptr);
+	dma_addr_t handle = virt_to_phys(page_address(page) + offset);
 
-	c6x_dma_sync(addr, size, dir);
-
-	debug_dma_map_page(dev, virt_to_page(ptr),
-			   (unsigned long)ptr & ~PAGE_MASK, size,
-			   dir, addr, true);
-	return addr;
+	c6x_dma_sync(handle, size, dir);
+	return handle;
 }
-EXPORT_SYMBOL(dma_map_single);
-
 
-void dma_unmap_single(struct device *dev, dma_addr_t handle,
-		      size_t size, enum dma_data_direction dir)
+static void c6x_dma_unmap_page(struct device *dev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir, struct dma_attrs *attrs)
 {
 	c6x_dma_sync(handle, size, dir);
-
-	debug_dma_unmap_page(dev, handle, size, dir, true);
 }
-EXPORT_SYMBOL(dma_unmap_single);
-
 
-int dma_map_sg(struct device *dev, struct scatterlist *sglist,
-	       int nents, enum dma_data_direction dir)
+static int c6x_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
 {
 	struct scatterlist *sg;
 	int i;
 
-	for_each_sg(sglist, sg, nents, i)
-		sg->dma_address = dma_map_single(dev, sg_virt(sg), sg->length,
-						 dir);
-
-	debug_dma_map_sg(dev, sglist, nents, nents, dir);
+	for_each_sg(sglist, sg, nents, i) {
+		sg->dma_address = sg_phys(sg);
+		c6x_dma_sync(sg->dma_address, sg->length, dir);
+	}
 
 	return nents;
 }
-EXPORT_SYMBOL(dma_map_sg);
-
 
-void dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
-		  int nents, enum dma_data_direction dir)
+static void c6x_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
+		  int nents, enum dma_data_direction dir,
+		  struct dma_attrs *attrs)
 {
 	struct scatterlist *sg;
 	int i;
 
 	for_each_sg(sglist, sg, nents, i)
-		dma_unmap_single(dev, sg_dma_address(sg), sg->length, dir);
+		c6x_dma_sync(sg_dma_address(sg), sg->length, dir);
 
-	debug_dma_unmap_sg(dev, sglist,	nents, dir);
 }
-EXPORT_SYMBOL(dma_unmap_sg);
 
-void dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle,
-			     size_t size, enum dma_data_direction dir)
+static void c6x_dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle,
+		size_t size, enum dma_data_direction dir)
 {
 	c6x_dma_sync(handle, size, dir);
 
-	debug_dma_sync_single_for_cpu(dev, handle, size, dir);
 }
-EXPORT_SYMBOL(dma_sync_single_for_cpu);
 
-
-void dma_sync_single_for_device(struct device *dev, dma_addr_t handle,
-				size_t size, enum dma_data_direction dir)
+static void c6x_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
 	c6x_dma_sync(handle, size, dir);
 
-	debug_dma_sync_single_for_device(dev, handle, size, dir);
 }
-EXPORT_SYMBOL(dma_sync_single_for_device);
-
 
-void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist,
-			 int nents, enum dma_data_direction dir)
+static void c6x_dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sglist, int nents,
+		enum dma_data_direction dir)
 {
 	struct scatterlist *sg;
 	int i;
 
 	for_each_sg(sglist, sg, nents, i)
-		dma_sync_single_for_cpu(dev, sg_dma_address(sg),
+		c6x_dma_sync_single_for_cpu(dev, sg_dma_address(sg),
 					sg->length, dir);
 
-	debug_dma_sync_sg_for_cpu(dev, sglist, nents, dir);
 }
-EXPORT_SYMBOL(dma_sync_sg_for_cpu);
-
 
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist,
-			    int nents, enum dma_data_direction dir)
+static void c6x_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sglist, int nents,
+		enum dma_data_direction dir)
 {
 	struct scatterlist *sg;
 	int i;
 
 	for_each_sg(sglist, sg, nents, i)
-		dma_sync_single_for_device(dev, sg_dma_address(sg),
+		c6x_dma_sync_single_for_device(dev, sg_dma_address(sg),
 					   sg->length, dir);
 
-	debug_dma_sync_sg_for_device(dev, sglist, nents, dir);
 }
-EXPORT_SYMBOL(dma_sync_sg_for_device);
 
+struct dma_map_ops c6x_dma_ops = {
+	.alloc			= c6x_dma_alloc,
+	.free			= c6x_dma_free,
+	.map_page		= c6x_dma_map_page,
+	.unmap_page		= c6x_dma_unmap_page,
+	.map_sg			= c6x_dma_map_sg,
+	.unmap_sg		= c6x_dma_unmap_sg,
+	.sync_single_for_device	= c6x_dma_sync_single_for_device,
+	.sync_single_for_cpu	= c6x_dma_sync_single_for_cpu,
+	.sync_sg_for_device	= c6x_dma_sync_sg_for_device,
+	.sync_sg_for_cpu	= c6x_dma_sync_sg_for_cpu,
+};
+EXPORT_SYMBOL(c6x_dma_ops);
 
 /* Number of entries preallocated for DMA-API debugging */
 #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)

arch/c6x/mm/dma-coherent.c

@@ -73,8 +73,8 @@ static void __free_dma_pages(u32 addr, int order)
  * Allocate DMA coherent memory space and return both the kernel
  * virtual and DMA address for that space.
  */
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *handle, gfp_t gfp)
+void *c6x_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t gfp, struct dma_attrs *attrs)
 {
 	u32 paddr;
 	int order;
@@ -94,13 +94,12 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 
 	return phys_to_virt(paddr);
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
 /*
  * Free DMA coherent memory as defined by the above mapping.
  */
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-		       dma_addr_t dma_handle)
+void c6x_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	int order;
 
@@ -111,7 +110,6 @@ void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
 
 	__free_dma_pages(virt_to_phys(vaddr), order);
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
 /*
  * Initialise the coherent DMA memory allocator using the given uncached region.

arch/cris/arch-v32/drivers/pci/dma.c

@@ -16,21 +16,18 @@
 #include <linux/gfp.h>
 #include <asm/io.h>
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t gfp)
+static void *v32_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp,  struct dma_attrs *attrs)
 {
 	void *ret;
-	int order = get_order(size);
+
 	/* ignore region specifiers */
 	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
 
-	if (dma_alloc_from_coherent(dev, size, dma_handle, &ret))
-		return ret;
-
 	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
 		gfp |= GFP_DMA;
 
-	ret = (void *)__get_free_pages(gfp, order);
+	ret = (void *)__get_free_pages(gfp,  get_order(size));
 
 	if (ret != NULL) {
 		memset(ret, 0, size);
@@ -39,12 +36,45 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	return ret;
 }
 
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
+static void v32_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
+{
+	free_pages((unsigned long)vaddr, get_order(size));
+}
+
+static inline dma_addr_t v32_dma_map_page(struct device *dev,
+		struct page *page, unsigned long offset, size_t size,
+		enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
-	int order = get_order(size);
+	return page_to_phys(page) + offset;
+}
 
-	if (!dma_release_from_coherent(dev, order, vaddr))
-		free_pages((unsigned long)vaddr, order);
+static inline int v32_dma_map_sg(struct device *dev, struct scatterlist *sg,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	printk("Map sg\n");
+	return nents;
+}
+
+static inline int v32_dma_supported(struct device *dev, u64 mask)
+{
+        /*
+         * we fall back to GFP_DMA when the mask isn't all 1s,
+         * so we can't guarantee allocations that must be
+         * within a tighter range than GFP_DMA..
+         */
+        if (mask < 0x00ffffff)
+                return 0;
+	return 1;
 }
 
+struct dma_map_ops v32_dma_ops = {
+	.alloc			= v32_dma_alloc,
+	.free			= v32_dma_free,
+	.map_page		= v32_dma_map_page,
+	.map_sg                 = v32_dma_map_sg,
+	.dma_supported		= v32_dma_supported,
+};
+EXPORT_SYMBOL(v32_dma_ops);

arch/cris/include/asm/dma-mapping.h

-/* DMA mapping. Nothing tricky here, just virt_to_phys */
-
 #ifndef _ASM_CRIS_DMA_MAPPING_H
 #define _ASM_CRIS_DMA_MAPPING_H
 
-#include <linux/mm.h>
-#include <linux/kernel.h>
-#include <linux/scatterlist.h>
-
-#include <asm/cache.h>
-#include <asm/io.h>
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
 #ifdef CONFIG_PCI
-#include <asm-generic/dma-coherent.h>
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t flag);
+extern struct dma_map_ops v32_dma_ops;
 
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle);
-#else
-static inline void *
-dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-                   gfp_t flag)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-        BUG();
-        return NULL;
+	return &v32_dma_ops;
 }
-
-static inline void
-dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
-                    dma_addr_t dma_handle)
+#else
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-        BUG();
+	BUG();
+	return NULL;
 }
 #endif
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *ptr, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	return virt_to_phys(ptr);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		 enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-static inline int
-dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-	   enum dma_data_direction direction)
-{
-	printk("Map sg\n");
-	return nents;
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-	     size_t size, enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	return page_to_phys(page) + offset;
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-	     enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
-			enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
-			enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			      unsigned long offset, size_t size,
-			      enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-		    enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
-		    enum dma_data_direction direction)
-{
-}
-
-static inline int
-dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-static inline int
-dma_supported(struct device *dev, u64 mask)
-{
-        /*
-         * we fall back to GFP_DMA when the mask isn't all 1s,
-         * so we can't guarantee allocations that must be
-         * within a tighter range than GFP_DMA..
-         */
-        if(mask < 0x00ffffff)
-                return 0;
-
-	return 1;
-}
-
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
-	if(!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
-}
 
 static inline void
 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
@@ -158,15 +22,4 @@ dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 {
 }
 
-/* drivers/base/dma-mapping.c */
-extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
-			   void *cpu_addr, dma_addr_t dma_addr, size_t size);
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size);
-
-#define dma_mmap_coherent(d, v, c, h, s) dma_common_mmap(d, v, c, h, s)
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
-
 #endif

arch/frv/Kconfig

@@ -15,6 +15,7 @@ config FRV
 	select OLD_SIGSUSPEND3
 	select OLD_SIGACTION
 	select HAVE_DEBUG_STACKOVERFLOW
+	select ARCH_NO_COHERENT_DMA_MMAP
 
 config ZONE_DMA
 	bool

arch/frv/include/asm/dma-mapping.h

 #ifndef _ASM_DMA_MAPPING_H
 #define _ASM_DMA_MAPPING_H
 
-#include <linux/device.h>
-#include <linux/scatterlist.h>
 #include <asm/cache.h>
 #include <asm/cacheflush.h>
-#include <asm/io.h>
-
-/*
- * See Documentation/DMA-API.txt for the description of how the
- * following DMA API should work.
- */
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
 
 extern unsigned long __nongprelbss dma_coherent_mem_start;
 extern unsigned long __nongprelbss dma_coherent_mem_end;
 
-void *dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t gfp);
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle);
-
-extern dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
-				 enum dma_data_direction direction);
-
-static inline
-void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		      enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-extern int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-		      enum dma_data_direction direction);
-
-static inline
-void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-	     enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-extern
-dma_addr_t dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-			size_t size, enum dma_data_direction direction);
-
-static inline
-void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-		    enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-
-static inline
-void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
-			     enum dma_data_direction direction)
-{
-}
-
-static inline
-void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
-				enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline
-void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-				   unsigned long offset, size_t size,
-				   enum dma_data_direction direction)
-{
-}
-
-static inline
-void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				      unsigned long offset, size_t size,
-				      enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline
-void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-			 enum dma_data_direction direction)
-{
-}
-
-static inline
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
-			    enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline
-int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-static inline
-int dma_supported(struct device *dev, u64 mask)
-{
-        /*
-         * we fall back to GFP_DMA when the mask isn't all 1s,
-         * so we can't guarantee allocations that must be
-         * within a tighter range than GFP_DMA..
-         */
-        if (mask < 0x00ffffff)
-                return 0;
-
-	return 1;
-}
+extern struct dma_map_ops frv_dma_ops;
 
-static inline
-int dma_set_mask(struct device *dev, u64 mask)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
+	return &frv_dma_ops;
 }
 
 static inline
@@ -132,19 +21,4 @@ void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	flush_write_buffers();
 }
 
-/* Not supported for now */
-static inline int dma_mmap_coherent(struct device *dev,
-				    struct vm_area_struct *vma, void *cpu_addr,
-				    dma_addr_t dma_addr, size_t size)
-{
-	return -EINVAL;
-}
-
-static inline int dma_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size)
-{
-	return -EINVAL;
-}
-
 #endif  /* _ASM_DMA_MAPPING_H */

arch/frv/include/asm/io.h

@@ -43,9 +43,20 @@ static inline unsigned long _swapl(unsigned long v)
 //#define __iormb() asm volatile("membar")
 //#define __iowmb() asm volatile("membar")
 
-#define __raw_readb __builtin_read8
-#define __raw_readw __builtin_read16
-#define __raw_readl __builtin_read32
+static inline u8 __raw_readb(const volatile void __iomem *addr)
+{
+	return __builtin_read8((volatile void __iomem *)addr);
+}
+
+static inline u16 __raw_readw(const volatile void __iomem *addr)
+{
+	return __builtin_read16((volatile void __iomem *)addr);
+}
+
+static inline u32 __raw_readl(const volatile void __iomem *addr)
+{
+	return __builtin_read32((volatile void __iomem *)addr);
+}
 
 #define __raw_writeb(datum, addr) __builtin_write8(addr, datum)
 #define __raw_writew(datum, addr) __builtin_write16(addr, datum)

arch/frv/mb93090-mb00/pci-dma-nommu.c

@@ -34,7 +34,8 @@ struct dma_alloc_record {
 static DEFINE_SPINLOCK(dma_alloc_lock);
 static LIST_HEAD(dma_alloc_list);
 
-void *dma_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t gfp)
+static void *frv_dma_alloc(struct device *hwdev, size_t size, dma_addr_t *dma_handle,
+		gfp_t gfp, struct dma_attrs *attrs)
 {
 	struct dma_alloc_record *new;
 	struct list_head *this = &dma_alloc_list;
@@ -84,9 +85,8 @@ void *dma_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_hand
 	return NULL;
 }
 
-EXPORT_SYMBOL(dma_alloc_coherent);
-
-void dma_free_coherent(struct device *hwdev, size_t size, void *vaddr, dma_addr_t dma_handle)
+static void frv_dma_free(struct device *hwdev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	struct dma_alloc_record *rec;
 	unsigned long flags;
@@ -105,22 +105,9 @@ void dma_free_coherent(struct device *hwdev, size_t size, void *vaddr, dma_addr_
 	BUG();
 }
 
-EXPORT_SYMBOL(dma_free_coherent);
-
-dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
-			  enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-
-	frv_cache_wback_inv((unsigned long) ptr, (unsigned long) ptr + size);
-
-	return virt_to_bus(ptr);
-}
-
-EXPORT_SYMBOL(dma_map_single);
-
-int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	       enum dma_data_direction direction)
+static int frv_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	int i;
 	struct scatterlist *sg;
@@ -135,14 +122,49 @@ int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
 	return nents;
 }
 
-EXPORT_SYMBOL(dma_map_sg);
-
-dma_addr_t dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-			size_t size, enum dma_data_direction direction)
+static dma_addr_t frv_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
 {
 	BUG_ON(direction == DMA_NONE);
 	flush_dcache_page(page);
 	return (dma_addr_t) page_to_phys(page) + offset;
 }
 
-EXPORT_SYMBOL(dma_map_page);
+static void frv_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	flush_write_buffers();
+}
+
+static void frv_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sg, int nelems,
+		enum dma_data_direction direction)
+{
+	flush_write_buffers();
+}
+
+
+static int frv_dma_supported(struct device *dev, u64 mask)
+{
+        /*
+         * we fall back to GFP_DMA when the mask isn't all 1s,
+         * so we can't guarantee allocations that must be
+         * within a tighter range than GFP_DMA..
+         */
+        if (mask < 0x00ffffff)
+                return 0;
+	return 1;
+}
+
+struct dma_map_ops frv_dma_ops = {
+	.alloc			= frv_dma_alloc,
+	.free			= frv_dma_free,
+	.map_page		= frv_dma_map_page,
+	.map_sg			= frv_dma_map_sg,
+	.sync_single_for_device	= frv_dma_sync_single_for_device,
+	.sync_sg_for_device	= frv_dma_sync_sg_for_device,
+	.dma_supported		= frv_dma_supported,
+};
+EXPORT_SYMBOL(frv_dma_ops);

arch/frv/mb93090-mb00/pci-dma.c

@@ -18,7 +18,9 @@
 #include <linux/scatterlist.h>
 #include <asm/io.h>
 
-void *dma_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t gfp)
+static void *frv_dma_alloc(struct device *hwdev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp,
+		struct dma_attrs *attrs)
 {
 	void *ret;
 
@@ -29,29 +31,15 @@ void *dma_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_hand
 	return ret;
 }
 
-EXPORT_SYMBOL(dma_alloc_coherent);
-
-void dma_free_coherent(struct device *hwdev, size_t size, void *vaddr, dma_addr_t dma_handle)
+static void frv_dma_free(struct device *hwdev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	consistent_free(vaddr);
 }
 
-EXPORT_SYMBOL(dma_free_coherent);
-
-dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
-			  enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-
-	frv_cache_wback_inv((unsigned long) ptr, (unsigned long) ptr + size);
-
-	return virt_to_bus(ptr);
-}
-
-EXPORT_SYMBOL(dma_map_single);
-
-int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	       enum dma_data_direction direction)
+static int frv_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	unsigned long dampr2;
 	void *vaddr;
@@ -79,14 +67,48 @@ int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
 	return nents;
 }
 
-EXPORT_SYMBOL(dma_map_sg);
-
-dma_addr_t dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-			size_t size, enum dma_data_direction direction)
+static dma_addr_t frv_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
 {
-	BUG_ON(direction == DMA_NONE);
 	flush_dcache_page(page);
 	return (dma_addr_t) page_to_phys(page) + offset;
 }
 
-EXPORT_SYMBOL(dma_map_page);
+static void frv_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	flush_write_buffers();
+}
+
+static void frv_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sg, int nelems,
+		enum dma_data_direction direction)
+{
+	flush_write_buffers();
+}
+
+
+static int frv_dma_supported(struct device *dev, u64 mask)
+{
+        /*
+         * we fall back to GFP_DMA when the mask isn't all 1s,
+         * so we can't guarantee allocations that must be
+         * within a tighter range than GFP_DMA..
+         */
+        if (mask < 0x00ffffff)
+                return 0;
+	return 1;
+}
+
+struct dma_map_ops frv_dma_ops = {
+	.alloc			= frv_dma_alloc,
+	.free			= frv_dma_free,
+	.map_page		= frv_dma_map_page,
+	.map_sg			= frv_dma_map_sg,
+	.sync_single_for_device	= frv_dma_sync_single_for_device,
+	.sync_sg_for_device	= frv_dma_sync_sg_for_device,
+	.dma_supported		= frv_dma_supported,
+};
+EXPORT_SYMBOL(frv_dma_ops);

arch/h8300/Kconfig

@@ -15,7 +15,6 @@ config H8300
 	select OF_IRQ
 	select OF_EARLY_FLATTREE
 	select HAVE_MEMBLOCK
-	select HAVE_DMA_ATTRS
 	select CLKSRC_OF
 	select H8300_TMR8
 	select HAVE_KERNEL_GZIP

arch/h8300/include/asm/dma-mapping.h

@@ -8,6 +8,4 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 	return &h8300_dma_map_ops;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 #endif

arch/hexagon/Kconfig

@@ -27,7 +27,6 @@ config HEXAGON
 	select GENERIC_CLOCKEVENTS_BROADCAST
 	select MODULES_USE_ELF_RELA
 	select GENERIC_CPU_DEVICES
-	select HAVE_DMA_ATTRS
 	---help---
 	  Qualcomm Hexagon is a processor architecture designed for high
 	  performance and low power across a wide variety of applications.

arch/hexagon/include/asm/dma-mapping.h

@@ -49,8 +49,6 @@ extern int dma_is_consistent(struct device *dev, dma_addr_t dma_handle);
 extern void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 			   enum dma_data_direction direction);
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 {
 	if (!dev->dma_mask)

arch/ia64/Kconfig

@@ -25,7 +25,6 @@ config IA64
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_DYNAMIC_FTRACE if (!ITANIUM)
 	select HAVE_FUNCTION_TRACER
-	select HAVE_DMA_ATTRS
 	select TTY
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_DMA_API_DEBUG

arch/ia64/include/asm/dma-mapping.h

@@ -25,8 +25,6 @@ extern void machvec_dma_sync_sg(struct device *, struct scatterlist *, int,
 
 #define get_dma_ops(dev) platform_dma_get_ops(dev)
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 {
 	if (!dev->dma_mask)

arch/m68k/include/asm/dma-mapping.h

 #ifndef _M68K_DMA_MAPPING_H
 #define _M68K_DMA_MAPPING_H
 
-#include <asm/cache.h>
+extern struct dma_map_ops m68k_dma_ops;
 
-struct scatterlist;
-
-static inline int dma_supported(struct device *dev, u64 mask)
-{
-	return 1;
-}
-
-static inline int dma_set_mask(struct device *dev, u64 mask)
-{
-	return 0;
-}
-
-extern void *dma_alloc_coherent(struct device *, size_t,
-				dma_addr_t *, gfp_t);
-extern void dma_free_coherent(struct device *, size_t,
-			      void *, dma_addr_t);
-
-static inline void *dma_alloc_attrs(struct device *dev, size_t size,
-				    dma_addr_t *dma_handle, gfp_t flag,
-				    struct dma_attrs *attrs)
-{
-	/* attrs is not supported and ignored */
-	return dma_alloc_coherent(dev, size, dma_handle, flag);
-}
-
-static inline void dma_free_attrs(struct device *dev, size_t size,
-				  void *cpu_addr, dma_addr_t dma_handle,
-				  struct dma_attrs *attrs)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	/* attrs is not supported and ignored */
-	dma_free_coherent(dev, size, cpu_addr, dma_handle);
+        return &m68k_dma_ops;
 }
 
-static inline void *dma_alloc_noncoherent(struct device *dev, size_t size,
-					  dma_addr_t *handle, gfp_t flag)
-{
-	return dma_alloc_coherent(dev, size, handle, flag);
-}
-static inline void dma_free_noncoherent(struct device *dev, size_t size,
-					void *addr, dma_addr_t handle)
-{
-	dma_free_coherent(dev, size, addr, handle);
-}
 static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 				  enum dma_data_direction dir)
 {
 	/* we use coherent allocation, so not much to do here. */
 }
 
-extern dma_addr_t dma_map_single(struct device *, void *, size_t,
-				 enum dma_data_direction);
-static inline void dma_unmap_single(struct device *dev, dma_addr_t addr,
-				    size_t size, enum dma_data_direction dir)
-{
-}
-
-extern dma_addr_t dma_map_page(struct device *, struct page *,
-			       unsigned long, size_t size,
-			       enum dma_data_direction);
-static inline void dma_unmap_page(struct device *dev, dma_addr_t address,
-				  size_t size, enum dma_data_direction dir)
-{
-}
-
-extern int dma_map_sg(struct device *, struct scatterlist *, int,
-		      enum dma_data_direction);
-static inline void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-				int nhwentries, enum dma_data_direction dir)
-{
-}
-
-extern void dma_sync_single_for_device(struct device *, dma_addr_t, size_t,
-				       enum dma_data_direction);
-extern void dma_sync_sg_for_device(struct device *, struct scatterlist *, int,
-				   enum dma_data_direction);
-
-static inline void dma_sync_single_range_for_device(struct device *dev,
-		dma_addr_t dma_handle, unsigned long offset, size_t size,
-		enum dma_data_direction direction)
-{
-	/* just sync everything for now */
-	dma_sync_single_for_device(dev, dma_handle, offset + size, direction);
-}
-
-static inline void dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle,
-					   size_t size, enum dma_data_direction dir)
-{
-}
-
-static inline void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-				       int nents, enum dma_data_direction dir)
-{
-}
-
-static inline void dma_sync_single_range_for_cpu(struct device *dev,
-		dma_addr_t dma_handle, unsigned long offset, size_t size,
-		enum dma_data_direction direction)
-{
-	/* just sync everything for now */
-	dma_sync_single_for_cpu(dev, dma_handle, offset + size, direction);
-}
-
-static inline int dma_mapping_error(struct device *dev, dma_addr_t handle)
-{
-	return 0;
-}
-
-/* drivers/base/dma-mapping.c */
-extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
-			   void *cpu_addr, dma_addr_t dma_addr, size_t size);
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size);
-
-#define dma_mmap_coherent(d, v, c, h, s) dma_common_mmap(d, v, c, h, s)
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
 #endif  /* _M68K_DMA_MAPPING_H */

arch/m68k/kernel/dma.c

@@ -18,8 +18,8 @@
 
 #if defined(CONFIG_MMU) && !defined(CONFIG_COLDFIRE)
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *handle, gfp_t flag)
+static void *m68k_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
+		gfp_t flag, struct dma_attrs *attrs)
 {
 	struct page *page, **map;
 	pgprot_t pgprot;
@@ -61,8 +61,8 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	return addr;
 }
 
-void dma_free_coherent(struct device *dev, size_t size,
-		       void *addr, dma_addr_t handle)
+static void m68k_dma_free(struct device *dev, size_t size, void *addr,
+		dma_addr_t handle, struct dma_attrs *attrs)
 {
 	pr_debug("dma_free_coherent: %p, %x\n", addr, handle);
 	vfree(addr);
@@ -72,8 +72,8 @@ void dma_free_coherent(struct device *dev, size_t size,
 
 #include <asm/cacheflush.h>
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t gfp)
+static void *m68k_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	void *ret;
 	/* ignore region specifiers */
@@ -90,19 +90,16 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	return ret;
 }
 
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
+static void m68k_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	free_pages((unsigned long)vaddr, get_order(size));
 }
 
 #endif /* CONFIG_MMU && !CONFIG_COLDFIRE */
 
-EXPORT_SYMBOL(dma_alloc_coherent);
-EXPORT_SYMBOL(dma_free_coherent);
-
-void dma_sync_single_for_device(struct device *dev, dma_addr_t handle,
-				size_t size, enum dma_data_direction dir)
+static void m68k_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t handle, size_t size, enum dma_data_direction dir)
 {
 	switch (dir) {
 	case DMA_BIDIRECTIONAL:
@@ -118,10 +115,9 @@ void dma_sync_single_for_device(struct device *dev, dma_addr_t handle,
 		break;
 	}
 }
-EXPORT_SYMBOL(dma_sync_single_for_device);
 
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist,
-			    int nents, enum dma_data_direction dir)
+static void m68k_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sglist, int nents, enum dma_data_direction dir)
 {
 	int i;
 	struct scatterlist *sg;
@@ -131,31 +127,19 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist,
 					   dir);
 	}
 }
-EXPORT_SYMBOL(dma_sync_sg_for_device);
-
-dma_addr_t dma_map_single(struct device *dev, void *addr, size_t size,
-			  enum dma_data_direction dir)
-{
-	dma_addr_t handle = virt_to_bus(addr);
-
-	dma_sync_single_for_device(dev, handle, size, dir);
-	return handle;
-}
-EXPORT_SYMBOL(dma_map_single);
 
-dma_addr_t dma_map_page(struct device *dev, struct page *page,
-			unsigned long offset, size_t size,
-			enum dma_data_direction dir)
+static dma_addr_t m68k_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size, enum dma_data_direction dir,
+		struct dma_attrs *attrs)
 {
 	dma_addr_t handle = page_to_phys(page) + offset;
 
 	dma_sync_single_for_device(dev, handle, size, dir);
 	return handle;
 }
-EXPORT_SYMBOL(dma_map_page);
 
-int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	       enum dma_data_direction dir)
+static int m68k_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
 {
 	int i;
 	struct scatterlist *sg;
@@ -167,4 +151,13 @@ int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
 	}
 	return nents;
 }
-EXPORT_SYMBOL(dma_map_sg);
+
+struct dma_map_ops m68k_dma_ops = {
+	.alloc			= m68k_dma_alloc,
+	.free			= m68k_dma_free,
+	.map_page		= m68k_dma_map_page,
+	.map_sg			= m68k_dma_map_sg,
+	.sync_single_for_device	= m68k_dma_sync_single_for_device,
+	.sync_sg_for_device	= m68k_dma_sync_sg_for_device,
+};
+EXPORT_SYMBOL(m68k_dma_ops);

arch/metag/include/asm/dma-mapping.h

 #ifndef _ASM_METAG_DMA_MAPPING_H
 #define _ASM_METAG_DMA_MAPPING_H
 
-#include <linux/mm.h>
+extern struct dma_map_ops metag_dma_ops;
 
-#include <asm/cache.h>
-#include <asm/io.h>
-#include <linux/scatterlist.h>
-#include <asm/bug.h>
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, gfp_t flag);
-
-void dma_free_coherent(struct device *dev, size_t size,
-		       void *vaddr, dma_addr_t dma_handle);
-
-void dma_sync_for_device(void *vaddr, size_t size, int dma_direction);
-void dma_sync_for_cpu(void *vaddr, size_t size, int dma_direction);
-
-int dma_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
-		      void *cpu_addr, dma_addr_t dma_addr, size_t size);
-
-int dma_mmap_writecombine(struct device *dev, struct vm_area_struct *vma,
-			  void *cpu_addr, dma_addr_t dma_addr, size_t size);
-
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *ptr, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(size == 0);
-	dma_sync_for_device(ptr, size, direction);
-	return virt_to_phys(ptr);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		 enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	dma_sync_for_cpu(phys_to_virt(dma_addr), size, direction);
-}
-
-static inline int
-dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	   enum dma_data_direction direction)
-{
-	struct scatterlist *sg;
-	int i;
-
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(nents == 0 || sglist[0].length == 0);
-
-	for_each_sg(sglist, sg, nents, i) {
-		BUG_ON(!sg_page(sg));
-
-		sg->dma_address = sg_phys(sg);
-		dma_sync_for_device(sg_virt(sg), sg->length, direction);
-	}
-
-	return nents;
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-	     size_t size, enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	dma_sync_for_device((void *)(page_to_phys(page) + offset), size,
-			    direction);
-	return page_to_phys(page) + offset;
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	dma_sync_for_cpu(phys_to_virt(dma_address), size, direction);
-}
-
-
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sglist, int nhwentries,
-	     enum dma_data_direction direction)
-{
-	struct scatterlist *sg;
-	int i;
-
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(nhwentries == 0 || sglist[0].length == 0);
-
-	for_each_sg(sglist, sg, nhwentries, i) {
-		BUG_ON(!sg_page(sg));
-
-		sg->dma_address = sg_phys(sg);
-		dma_sync_for_cpu(sg_virt(sg), sg->length, direction);
-	}
-}
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
-			enum dma_data_direction direction)
-{
-	dma_sync_for_cpu(phys_to_virt(dma_handle), size, direction);
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
-			   size_t size, enum dma_data_direction direction)
-{
-	dma_sync_for_device(phys_to_virt(dma_handle), size, direction);
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			      unsigned long offset, size_t size,
-			      enum dma_data_direction direction)
-{
-	dma_sync_for_cpu(phys_to_virt(dma_handle)+offset, size,
-			 direction);
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction direction)
-{
-	dma_sync_for_device(phys_to_virt(dma_handle)+offset, size,
-			    direction);
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist, int nelems,
-		    enum dma_data_direction direction)
-{
-	int i;
-	struct scatterlist *sg;
-
-	for_each_sg(sglist, sg, nelems, i)
-		dma_sync_for_cpu(sg_virt(sg), sg->length, direction);
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sglist,
-		       int nelems, enum dma_data_direction direction)
-{
-	int i;
-	struct scatterlist *sg;
-
-	for_each_sg(sglist, sg, nelems, i)
-		dma_sync_for_device(sg_virt(sg), sg->length, direction);
-}
-
-static inline int
-dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	return 0;
-}
-
-#define dma_supported(dev, mask)        (1)
-
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
+	return &metag_dma_ops;
 }
 
 /*
@@ -184,11 +18,4 @@ dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 {
 }
 
-/* drivers/base/dma-mapping.c */
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size);
-
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
 #endif

arch/metag/kernel/dma.c

@@ -171,8 +171,8 @@ static struct metag_vm_region *metag_vm_region_find(struct metag_vm_region
  * Allocate DMA-coherent memory space and return both the kernel remapped
  * virtual and bus address for that space.
  */
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *handle, gfp_t gfp)
+static void *metag_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	struct page *page;
 	struct metag_vm_region *c;
@@ -263,13 +263,12 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 no_page:
 	return NULL;
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
 /*
  * free a page as defined by the above mapping.
  */
-void dma_free_coherent(struct device *dev, size_t size,
-		       void *vaddr, dma_addr_t dma_handle)
+static void metag_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	struct metag_vm_region *c;
 	unsigned long flags, addr;
@@ -329,16 +328,19 @@ void dma_free_coherent(struct device *dev, size_t size,
 	       __func__, vaddr);
 	dump_stack();
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
-
-static int dma_mmap(struct device *dev, struct vm_area_struct *vma,
-		    void *cpu_addr, dma_addr_t dma_addr, size_t size)
+static int metag_dma_mmap(struct device *dev, struct vm_area_struct *vma,
+		void *cpu_addr, dma_addr_t dma_addr, size_t size,
+		struct dma_attrs *attrs)
 {
-	int ret = -ENXIO;
-
 	unsigned long flags, user_size, kern_size;
 	struct metag_vm_region *c;
+	int ret = -ENXIO;
+
+	if (dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs))
+		vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
+	else
+		vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 
 	user_size = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
 
@@ -364,25 +366,6 @@ static int dma_mmap(struct device *dev, struct vm_area_struct *vma,
 	return ret;
 }
 
-int dma_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
-		      void *cpu_addr, dma_addr_t dma_addr, size_t size)
-{
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
-	return dma_mmap(dev, vma, cpu_addr, dma_addr, size);
-}
-EXPORT_SYMBOL(dma_mmap_coherent);
-
-int dma_mmap_writecombine(struct device *dev, struct vm_area_struct *vma,
-			  void *cpu_addr, dma_addr_t dma_addr, size_t size)
-{
-	vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
-	return dma_mmap(dev, vma, cpu_addr, dma_addr, size);
-}
-EXPORT_SYMBOL(dma_mmap_writecombine);
-
-
-
-
 /*
  * Initialise the consistent memory allocation.
  */
@@ -423,7 +406,7 @@ early_initcall(dma_alloc_init);
 /*
  * make an area consistent to devices.
  */
-void dma_sync_for_device(void *vaddr, size_t size, int dma_direction)
+static void dma_sync_for_device(void *vaddr, size_t size, int dma_direction)
 {
 	/*
 	 * Ensure any writes get through the write combiner. This is necessary
@@ -465,12 +448,11 @@ void dma_sync_for_device(void *vaddr, size_t size, int dma_direction)
 
 	wmb();
 }
-EXPORT_SYMBOL(dma_sync_for_device);
 
 /*
  * make an area consistent to the core.
  */
-void dma_sync_for_cpu(void *vaddr, size_t size, int dma_direction)
+static void dma_sync_for_cpu(void *vaddr, size_t size, int dma_direction)
 {
 	/*
 	 * Hardware L2 cache prefetch doesn't occur across 4K physical
@@ -497,4 +479,100 @@ void dma_sync_for_cpu(void *vaddr, size_t size, int dma_direction)
 
 	rmb();
 }
-EXPORT_SYMBOL(dma_sync_for_cpu);
+
+static dma_addr_t metag_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
+{
+	dma_sync_for_device((void *)(page_to_phys(page) + offset), size,
+			    direction);
+	return page_to_phys(page) + offset;
+}
+
+static void metag_dma_unmap_page(struct device *dev, dma_addr_t dma_address,
+		size_t size, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	dma_sync_for_cpu(phys_to_virt(dma_address), size, direction);
+}
+
+static int metag_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	struct scatterlist *sg;
+	int i;
+
+	for_each_sg(sglist, sg, nents, i) {
+		BUG_ON(!sg_page(sg));
+
+		sg->dma_address = sg_phys(sg);
+		dma_sync_for_device(sg_virt(sg), sg->length, direction);
+	}
+
+	return nents;
+}
+
+
+static void metag_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
+		int nhwentries, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	struct scatterlist *sg;
+	int i;
+
+	for_each_sg(sglist, sg, nhwentries, i) {
+		BUG_ON(!sg_page(sg));
+
+		sg->dma_address = sg_phys(sg);
+		dma_sync_for_cpu(sg_virt(sg), sg->length, direction);
+	}
+}
+
+static void metag_dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	dma_sync_for_cpu(phys_to_virt(dma_handle), size, direction);
+}
+
+static void metag_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
+{
+	dma_sync_for_device(phys_to_virt(dma_handle), size, direction);
+}
+
+static void metag_dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sglist, int nelems,
+		enum dma_data_direction direction)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sglist, sg, nelems, i)
+		dma_sync_for_cpu(sg_virt(sg), sg->length, direction);
+}
+
+static void metag_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sglist, int nelems,
+		enum dma_data_direction direction)
+{
+	int i;
+	struct scatterlist *sg;
+
+	for_each_sg(sglist, sg, nelems, i)
+		dma_sync_for_device(sg_virt(sg), sg->length, direction);
+}
+
+struct dma_map_ops metag_dma_ops = {
+	.alloc			= metag_dma_alloc,
+	.free			= metag_dma_free,
+	.map_page		= metag_dma_map_page,
+	.map_sg			= metag_dma_map_sg,
+	.sync_single_for_device	= metag_dma_sync_single_for_device,
+	.sync_single_for_cpu	= metag_dma_sync_single_for_cpu,
+	.sync_sg_for_cpu	= metag_dma_sync_sg_for_cpu,
+	.mmap			= metag_dma_mmap,
+};
+EXPORT_SYMBOL(metag_dma_ops);

arch/microblaze/Kconfig

@@ -19,7 +19,6 @@ config MICROBLAZE
 	select HAVE_ARCH_KGDB
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_FUNCTION_GRAPH_TRACER

arch/microblaze/include/asm/dma-mapping.h

@@ -44,8 +44,6 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 	return &dma_direct_ops;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline void __dma_sync(unsigned long paddr,
 			      size_t size, enum dma_data_direction direction)
 {

arch/mips/Kconfig

@@ -31,7 +31,6 @@ config MIPS
 	select RTC_LIB if !MACH_LOONGSON64
 	select GENERIC_ATOMIC64 if !64BIT
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DMA_API_DEBUG
 	select GENERIC_IRQ_PROBE

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

@@ -29,8 +29,6 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 
 static inline void dma_mark_clean(void *addr, size_t size) {}
 
-#include <asm-generic/dma-mapping-common.h>
-
 extern void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	       enum dma_data_direction direction);
 

arch/mips/include/uapi/asm/mman.h

@@ -73,7 +73,6 @@
 #define MADV_SEQUENTIAL 2		/* expect sequential page references */
 #define MADV_WILLNEED	3		/* will need these pages */
 #define MADV_DONTNEED	4		/* don't need these pages */
-#define MADV_FREE	5		/* free pages only if memory pressure */
 
 /* common parameters: try to keep these consistent across architectures */
 #define MADV_FREE	8		/* free pages only if memory pressure */

arch/mn10300/Kconfig

@@ -14,6 +14,7 @@ config MN10300
 	select OLD_SIGSUSPEND3
 	select OLD_SIGACTION
 	select HAVE_DEBUG_STACKOVERFLOW
+	select ARCH_NO_COHERENT_DMA_MMAP
 
 config AM33_2
 	def_bool n

arch/mn10300/include/asm/dma-mapping.h

@@ -11,154 +11,14 @@
 #ifndef _ASM_DMA_MAPPING_H
 #define _ASM_DMA_MAPPING_H
 
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-
 #include <asm/cache.h>
 #include <asm/io.h>
 
-/*
- * See Documentation/DMA-API.txt for the description of how the
- * following DMA API should work.
- */
-
-extern void *dma_alloc_coherent(struct device *dev, size_t size,
-				dma_addr_t *dma_handle, int flag);
-
-extern void dma_free_coherent(struct device *dev, size_t size,
-			      void *vaddr, dma_addr_t dma_handle);
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent((d), (s), (h), (f))
-#define dma_free_noncoherent(d, s, v, h)  dma_free_coherent((d), (s), (v), (h))
-
-static inline
-dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
-			  enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	mn10300_dcache_flush_inv();
-	return virt_to_bus(ptr);
-}
-
-static inline
-void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		      enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-static inline
-int dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	       enum dma_data_direction direction)
-{
-	struct scatterlist *sg;
-	int i;
-
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(nents == 0 || sglist[0].length == 0);
-
-	for_each_sg(sglist, sg, nents, i) {
-		BUG_ON(!sg_page(sg));
-
-		sg->dma_address = sg_phys(sg);
-	}
+extern struct dma_map_ops mn10300_dma_ops;
 
-	mn10300_dcache_flush_inv();
-	return nents;
-}
-
-static inline
-void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-		  enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-}
-
-static inline
-dma_addr_t dma_map_page(struct device *dev, struct page *page,
-			unsigned long offset, size_t size,
-			enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-	return page_to_bus(page) + offset;
-}
-
-static inline
-void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-		    enum dma_data_direction direction)
-{
-	BUG_ON(direction == DMA_NONE);
-}
-
-static inline
-void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			     size_t size, enum dma_data_direction direction)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-}
-
-static inline
-void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
-				size_t size, enum dma_data_direction direction)
-{
-	mn10300_dcache_flush_inv();
-}
-
-static inline
-void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-				   unsigned long offset, size_t size,
-				   enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction direction)
-{
-	mn10300_dcache_flush_inv();
-}
-
-
-static inline
-void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-			 int nelems, enum dma_data_direction direction)
-{
-}
-
-static inline
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-			    int nelems, enum dma_data_direction direction)
-{
-	mn10300_dcache_flush_inv();
-}
-
-static inline
-int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-static inline
-int dma_supported(struct device *dev, u64 mask)
-{
-	/*
-	 * we fall back to GFP_DMA when the mask isn't all 1s, so we can't
-	 * guarantee allocations that must be within a tighter range than
-	 * GFP_DMA
-	 */
-	if (mask < 0x00ffffff)
-		return 0;
-	return 1;
-}
-
-static inline
-int dma_set_mask(struct device *dev, u64 mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-	return 0;
+	return &mn10300_dma_ops;
 }
 
 static inline
@@ -168,19 +28,4 @@ void dma_cache_sync(void *vaddr, size_t size,
 	mn10300_dcache_flush_inv();
 }
 
-/* Not supported for now */
-static inline int dma_mmap_coherent(struct device *dev,
-				    struct vm_area_struct *vma, void *cpu_addr,
-				    dma_addr_t dma_addr, size_t size)
-{
-	return -EINVAL;
-}
-
-static inline int dma_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size)
-{
-	return -EINVAL;
-}
-
 #endif

arch/mn10300/mm/dma-alloc.c

@@ -20,8 +20,8 @@
 
 static unsigned long pci_sram_allocated = 0xbc000000;
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			 dma_addr_t *dma_handle, int gfp)
+static void *mn10300_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	unsigned long addr;
 	void *ret;
@@ -61,10 +61,9 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 	printk("dma_alloc_coherent() = %p [%x]\n", ret, *dma_handle);
 	return ret;
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-		       dma_addr_t dma_handle)
+static void mn10300_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	unsigned long addr = (unsigned long) vaddr & ~0x20000000;
 
@@ -73,4 +72,60 @@ void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
 
 	free_pages(addr, get_order(size));
 }
-EXPORT_SYMBOL(dma_free_coherent);
+
+static int mn10300_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	struct scatterlist *sg;
+	int i;
+
+	for_each_sg(sglist, sg, nents, i) {
+		BUG_ON(!sg_page(sg));
+
+		sg->dma_address = sg_phys(sg);
+	}
+
+	mn10300_dcache_flush_inv();
+	return nents;
+}
+
+static dma_addr_t mn10300_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
+{
+	return page_to_bus(page) + offset;
+}
+
+static void mn10300_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
+				size_t size, enum dma_data_direction direction)
+{
+	mn10300_dcache_flush_inv();
+}
+
+static void mn10300_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
+			    int nelems, enum dma_data_direction direction)
+{
+	mn10300_dcache_flush_inv();
+}
+
+static int mn10300_dma_supported(struct device *dev, u64 mask)
+{
+	/*
+	 * we fall back to GFP_DMA when the mask isn't all 1s, so we can't
+	 * guarantee allocations that must be within a tighter range than
+	 * GFP_DMA
+	 */
+	if (mask < 0x00ffffff)
+		return 0;
+	return 1;
+}
+
+struct dma_map_ops mn10300_dma_ops = {
+	.alloc			= mn10300_dma_alloc,
+	.free			= mn10300_dma_free,
+	.map_page		= mn10300_dma_map_page,
+	.map_sg			= mn10300_dma_map_sg,
+	.sync_single_for_device	= mn10300_dma_sync_single_for_device,
+	.sync_sg_for_device	= mn10300_dma_sync_sg_for_device,
+};

arch/nios2/include/asm/dma-mapping.h

@@ -10,131 +10,20 @@
 #ifndef _ASM_NIOS2_DMA_MAPPING_H
 #define _ASM_NIOS2_DMA_MAPPING_H
 
-#include <linux/scatterlist.h>
-#include <linux/cache.h>
-#include <asm/cacheflush.h>
+extern struct dma_map_ops nios2_dma_ops;
 
-static inline void __dma_sync_for_device(void *vaddr, size_t size,
-			      enum dma_data_direction direction)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	switch (direction) {
-	case DMA_FROM_DEVICE:
-		invalidate_dcache_range((unsigned long)vaddr,
-			(unsigned long)(vaddr + size));
-		break;
-	case DMA_TO_DEVICE:
-		/*
-		 * We just need to flush the caches here , but Nios2 flush
-		 * instruction will do both writeback and invalidate.
-		 */
-	case DMA_BIDIRECTIONAL: /* flush and invalidate */
-		flush_dcache_range((unsigned long)vaddr,
-			(unsigned long)(vaddr + size));
-		break;
-	default:
-		BUG();
-	}
-}
-
-static inline void __dma_sync_for_cpu(void *vaddr, size_t size,
-			      enum dma_data_direction direction)
-{
-	switch (direction) {
-	case DMA_BIDIRECTIONAL:
-	case DMA_FROM_DEVICE:
-		invalidate_dcache_range((unsigned long)vaddr,
-			(unsigned long)(vaddr + size));
-		break;
-	case DMA_TO_DEVICE:
-		break;
-	default:
-		BUG();
-	}
-}
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t flag);
-
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle);
-
-static inline dma_addr_t dma_map_single(struct device *dev, void *ptr,
-					size_t size,
-					enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	__dma_sync_for_device(ptr, size, direction);
-	return virt_to_phys(ptr);
-}
-
-static inline void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
-				size_t size, enum dma_data_direction direction)
-{
-}
-
-extern int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-	enum dma_data_direction direction);
-extern dma_addr_t dma_map_page(struct device *dev, struct page *page,
-	unsigned long offset, size_t size, enum dma_data_direction direction);
-extern void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
-	size_t size, enum dma_data_direction direction);
-extern void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
-	int nhwentries, enum dma_data_direction direction);
-extern void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-	size_t size, enum dma_data_direction direction);
-extern void dma_sync_single_for_device(struct device *dev,
-	dma_addr_t dma_handle, size_t size, enum dma_data_direction direction);
-extern void dma_sync_single_range_for_cpu(struct device *dev,
-	dma_addr_t dma_handle, unsigned long offset, size_t size,
-	enum dma_data_direction direction);
-extern void dma_sync_single_range_for_device(struct device *dev,
-	dma_addr_t dma_handle, unsigned long offset, size_t size,
-	enum dma_data_direction direction);
-extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
-	int nelems, enum dma_data_direction direction);
-extern void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-	int nelems, enum dma_data_direction direction);
-
-static inline int dma_supported(struct device *dev, u64 mask)
-{
-	return 1;
-}
-
-static inline int dma_set_mask(struct device *dev, u64 mask)
-{
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
-}
-
-static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
-{
-	return 0;
+	return &nios2_dma_ops;
 }
 
 /*
-* dma_alloc_noncoherent() returns non-cacheable memory, so there's no need to
-* do any flushing here.
-*/
+ * dma_alloc_noncoherent() returns non-cacheable memory, so there's no need to
+ * do any flushing here.
+ */
 static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 				  enum dma_data_direction direction)
 {
 }
 
-/* drivers/base/dma-mapping.c */
-extern int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
-		void *cpu_addr, dma_addr_t dma_addr, size_t size);
-extern int dma_common_get_sgtable(struct device *dev, struct sg_table *sgt,
-		void *cpu_addr, dma_addr_t dma_addr,
-		size_t size);
-
-#define dma_mmap_coherent(d, v, c, h, s) dma_common_mmap(d, v, c, h, s)
-#define dma_get_sgtable(d, t, v, h, s) dma_common_get_sgtable(d, t, v, h, s)
-
 #endif /* _ASM_NIOS2_DMA_MAPPING_H */

arch/nios2/mm/dma-mapping.c

@@ -20,9 +20,46 @@
 #include <linux/cache.h>
 #include <asm/cacheflush.h>
 
+static inline void __dma_sync_for_device(void *vaddr, size_t size,
+			      enum dma_data_direction direction)
+{
+	switch (direction) {
+	case DMA_FROM_DEVICE:
+		invalidate_dcache_range((unsigned long)vaddr,
+			(unsigned long)(vaddr + size));
+		break;
+	case DMA_TO_DEVICE:
+		/*
+		 * We just need to flush the caches here , but Nios2 flush
+		 * instruction will do both writeback and invalidate.
+		 */
+	case DMA_BIDIRECTIONAL: /* flush and invalidate */
+		flush_dcache_range((unsigned long)vaddr,
+			(unsigned long)(vaddr + size));
+		break;
+	default:
+		BUG();
+	}
+}
 
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			    dma_addr_t *dma_handle, gfp_t gfp)
+static inline void __dma_sync_for_cpu(void *vaddr, size_t size,
+			      enum dma_data_direction direction)
+{
+	switch (direction) {
+	case DMA_BIDIRECTIONAL:
+	case DMA_FROM_DEVICE:
+		invalidate_dcache_range((unsigned long)vaddr,
+			(unsigned long)(vaddr + size));
+		break;
+	case DMA_TO_DEVICE:
+		break;
+	default:
+		BUG();
+	}
+}
+
+static void *nios2_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t gfp, struct dma_attrs *attrs)
 {
 	void *ret;
 
@@ -45,24 +82,21 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 
 	return ret;
 }
-EXPORT_SYMBOL(dma_alloc_coherent);
 
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-			dma_addr_t dma_handle)
+static void nios2_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	unsigned long addr = (unsigned long) CAC_ADDR((unsigned long) vaddr);
 
 	free_pages(addr, get_order(size));
 }
-EXPORT_SYMBOL(dma_free_coherent);
 
-int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-		enum dma_data_direction direction)
+static int nios2_dma_map_sg(struct device *dev, struct scatterlist *sg,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	int i;
 
-	BUG_ON(!valid_dma_direction(direction));
-
 	for_each_sg(sg, sg, nents, i) {
 		void *addr;
 
@@ -75,40 +109,32 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 
 	return nents;
 }
-EXPORT_SYMBOL(dma_map_sg);
 
-dma_addr_t dma_map_page(struct device *dev, struct page *page,
+static dma_addr_t nios2_dma_map_page(struct device *dev, struct page *page,
 			unsigned long offset, size_t size,
-			enum dma_data_direction direction)
+			enum dma_data_direction direction,
+			struct dma_attrs *attrs)
 {
-	void *addr;
-
-	BUG_ON(!valid_dma_direction(direction));
+	void *addr = page_address(page) + offset;
 
-	addr = page_address(page) + offset;
 	__dma_sync_for_device(addr, size, direction);
-
 	return page_to_phys(page) + offset;
 }
-EXPORT_SYMBOL(dma_map_page);
 
-void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-		    enum dma_data_direction direction)
+static void nios2_dma_unmap_page(struct device *dev, dma_addr_t dma_address,
+		size_t size, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
-	BUG_ON(!valid_dma_direction(direction));
-
 	__dma_sync_for_cpu(phys_to_virt(dma_address), size, direction);
 }
-EXPORT_SYMBOL(dma_unmap_page);
 
-void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-		  enum dma_data_direction direction)
+static void nios2_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+		int nhwentries, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	void *addr;
 	int i;
 
-	BUG_ON(!valid_dma_direction(direction));
-
 	if (direction == DMA_TO_DEVICE)
 		return;
 
@@ -118,69 +144,54 @@ void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
 			__dma_sync_for_cpu(addr, sg->length, direction);
 	}
 }
-EXPORT_SYMBOL(dma_unmap_sg);
-
-void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			     size_t size, enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
 
-	__dma_sync_for_cpu(phys_to_virt(dma_handle), size, direction);
-}
-EXPORT_SYMBOL(dma_sync_single_for_cpu);
-
-void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
-				size_t size, enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-
-	__dma_sync_for_device(phys_to_virt(dma_handle), size, direction);
-}
-EXPORT_SYMBOL(dma_sync_single_for_device);
-
-void dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-					unsigned long offset, size_t size,
-					enum dma_data_direction direction)
+static void nios2_dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
-	BUG_ON(!valid_dma_direction(direction));
-
 	__dma_sync_for_cpu(phys_to_virt(dma_handle), size, direction);
 }
-EXPORT_SYMBOL(dma_sync_single_range_for_cpu);
 
-void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-					unsigned long offset, size_t size,
-					enum dma_data_direction direction)
+static void nios2_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
-	BUG_ON(!valid_dma_direction(direction));
-
 	__dma_sync_for_device(phys_to_virt(dma_handle), size, direction);
 }
-EXPORT_SYMBOL(dma_sync_single_range_for_device);
 
-void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-			 enum dma_data_direction direction)
+static void nios2_dma_sync_sg_for_cpu(struct device *dev,
+		struct scatterlist *sg, int nelems,
+		enum dma_data_direction direction)
 {
 	int i;
 
-	BUG_ON(!valid_dma_direction(direction));
-
 	/* Make sure that gcc doesn't leave the empty loop body.  */
 	for_each_sg(sg, sg, nelems, i)
 		__dma_sync_for_cpu(sg_virt(sg), sg->length, direction);
 }
-EXPORT_SYMBOL(dma_sync_sg_for_cpu);
 
-void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
-				int nelems, enum dma_data_direction direction)
+static void nios2_dma_sync_sg_for_device(struct device *dev,
+		struct scatterlist *sg, int nelems,
+		enum dma_data_direction direction)
 {
 	int i;
 
-	BUG_ON(!valid_dma_direction(direction));
-
 	/* Make sure that gcc doesn't leave the empty loop body.  */
 	for_each_sg(sg, sg, nelems, i)
 		__dma_sync_for_device(sg_virt(sg), sg->length, direction);
 
 }
-EXPORT_SYMBOL(dma_sync_sg_for_device);
+
+struct dma_map_ops nios2_dma_ops = {
+	.alloc			= nios2_dma_alloc,
+	.free			= nios2_dma_free,
+	.map_page		= nios2_dma_map_page,
+	.unmap_page		= nios2_dma_unmap_page,
+	.map_sg			= nios2_dma_map_sg,
+	.unmap_sg		= nios2_dma_unmap_sg,
+	.sync_single_for_device	= nios2_dma_sync_single_for_device,
+	.sync_single_for_cpu	= nios2_dma_sync_single_for_cpu,
+	.sync_sg_for_cpu	= nios2_dma_sync_sg_for_cpu,
+	.sync_sg_for_device	= nios2_dma_sync_sg_for_device,
+};
+EXPORT_SYMBOL(nios2_dma_ops);

arch/openrisc/Kconfig

@@ -29,9 +29,6 @@ config OPENRISC
 config MMU
 	def_bool y
 
-config HAVE_DMA_ATTRS
-	def_bool y
-
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y
 

arch/openrisc/include/asm/dma-mapping.h

@@ -42,6 +42,4 @@ static inline int dma_supported(struct device *dev, u64 dma_mask)
 	return dma_mask == DMA_BIT_MASK(32);
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 #endif	/* __ASM_OPENRISC_DMA_MAPPING_H */

arch/parisc/Kconfig

@@ -29,6 +29,7 @@ config PARISC
 	select TTY # Needed for pdc_cons.c
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_ARCH_AUDITSYSCALL
+	select ARCH_NO_COHERENT_DMA_MMAP
 
 	help
 	  The PA-RISC microprocessor is designed by Hewlett-Packard and used

arch/parisc/include/asm/dma-mapping.h

 #ifndef _PARISC_DMA_MAPPING_H
 #define _PARISC_DMA_MAPPING_H
 
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
 #include <asm/cacheflush.h>
 
-/* See Documentation/DMA-API-HOWTO.txt */
-struct hppa_dma_ops {
-	int  (*dma_supported)(struct device *dev, u64 mask);
-	void *(*alloc_consistent)(struct device *dev, size_t size, dma_addr_t *iova, gfp_t flag);
-	void *(*alloc_noncoherent)(struct device *dev, size_t size, dma_addr_t *iova, gfp_t flag);
-	void (*free_consistent)(struct device *dev, size_t size, void *vaddr, dma_addr_t iova);
-	dma_addr_t (*map_single)(struct device *dev, void *addr, size_t size, enum dma_data_direction direction);
-	void (*unmap_single)(struct device *dev, dma_addr_t iova, size_t size, enum dma_data_direction direction);
-	int  (*map_sg)(struct device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction);
-	void (*unmap_sg)(struct device *dev, struct scatterlist *sg, int nhwents, enum dma_data_direction direction);
-	void (*dma_sync_single_for_cpu)(struct device *dev, dma_addr_t iova, unsigned long offset, size_t size, enum dma_data_direction direction);
-	void (*dma_sync_single_for_device)(struct device *dev, dma_addr_t iova, unsigned long offset, size_t size, enum dma_data_direction direction);
-	void (*dma_sync_sg_for_cpu)(struct device *dev, struct scatterlist *sg, int nelems, enum dma_data_direction direction);
-	void (*dma_sync_sg_for_device)(struct device *dev, struct scatterlist *sg, int nelems, enum dma_data_direction direction);
-};
-
 /*
-** We could live without the hppa_dma_ops indirection if we didn't want
-** to support 4 different coherent dma models with one binary (they will
-** someday be loadable modules):
+** We need to support 4 different coherent dma models with one binary:
+**
 **     I/O MMU        consistent method           dma_sync behavior
 **  =============   ======================       =======================
 **  a) PA-7x00LC    uncachable host memory          flush/purge
@@ -40,158 +21,22 @@ struct hppa_dma_ops {
 */
 
 #ifdef CONFIG_PA11
-extern struct hppa_dma_ops pcxl_dma_ops;
-extern struct hppa_dma_ops pcx_dma_ops;
+extern struct dma_map_ops pcxl_dma_ops;
+extern struct dma_map_ops pcx_dma_ops;
 #endif
 
-extern struct hppa_dma_ops *hppa_dma_ops;
-
-#define dma_alloc_attrs(d, s, h, f, a) dma_alloc_coherent(d, s, h, f)
-#define dma_free_attrs(d, s, h, f, a) dma_free_coherent(d, s, h, f)
-
-static inline void *
-dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		   gfp_t flag)
-{
-	return hppa_dma_ops->alloc_consistent(dev, size, dma_handle, flag);
-}
-
-static inline void *
-dma_alloc_noncoherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		      gfp_t flag)
-{
-	return hppa_dma_ops->alloc_noncoherent(dev, size, dma_handle, flag);
-}
-
-static inline void
-dma_free_coherent(struct device *dev, size_t size, 
-		    void *vaddr, dma_addr_t dma_handle)
-{
-	hppa_dma_ops->free_consistent(dev, size, vaddr, dma_handle);
-}
-
-static inline void
-dma_free_noncoherent(struct device *dev, size_t size, 
-		    void *vaddr, dma_addr_t dma_handle)
-{
-	hppa_dma_ops->free_consistent(dev, size, vaddr, dma_handle);
-}
-
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *ptr, size_t size,
-	       enum dma_data_direction direction)
-{
-	return hppa_dma_ops->map_single(dev, ptr, size, direction);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		 enum dma_data_direction direction)
-{
-	hppa_dma_ops->unmap_single(dev, dma_addr, size, direction);
-}
-
-static inline int
-dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-	   enum dma_data_direction direction)
-{
-	return hppa_dma_ops->map_sg(dev, sg, nents, direction);
-}
+extern struct dma_map_ops *hppa_dma_ops;
 
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-	     enum dma_data_direction direction)
-{
-	hppa_dma_ops->unmap_sg(dev, sg, nhwentries, direction);
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-	     size_t size, enum dma_data_direction direction)
-{
-	return dma_map_single(dev, (page_address(page) + (offset)), size, direction);
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-	       enum dma_data_direction direction)
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 {
-	dma_unmap_single(dev, dma_address, size, direction);
-}
-
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
-		enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_single_for_cpu)
-		hppa_dma_ops->dma_sync_single_for_cpu(dev, dma_handle, 0, size, direction);
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
-		enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_single_for_device)
-		hppa_dma_ops->dma_sync_single_for_device(dev, dma_handle, 0, size, direction);
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-		      unsigned long offset, size_t size,
-		      enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_single_for_cpu)
-		hppa_dma_ops->dma_sync_single_for_cpu(dev, dma_handle, offset, size, direction);
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-		      unsigned long offset, size_t size,
-		      enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_single_for_device)
-		hppa_dma_ops->dma_sync_single_for_device(dev, dma_handle, offset, size, direction);
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-		 enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_sg_for_cpu)
-		hppa_dma_ops->dma_sync_sg_for_cpu(dev, sg, nelems, direction);
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
-		 enum dma_data_direction direction)
-{
-	if(hppa_dma_ops->dma_sync_sg_for_device)
-		hppa_dma_ops->dma_sync_sg_for_device(dev, sg, nelems, direction);
-}
-
-static inline int
-dma_supported(struct device *dev, u64 mask)
-{
-	return hppa_dma_ops->dma_supported(dev, mask);
-}
-
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
-	if(!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
+	return hppa_dma_ops;
 }
 
 static inline void
 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	       enum dma_data_direction direction)
 {
-	if(hppa_dma_ops->dma_sync_single_for_cpu)
+	if (hppa_dma_ops->sync_single_for_cpu)
 		flush_kernel_dcache_range((unsigned long)vaddr, size);
 }
 
@@ -238,22 +83,4 @@ struct parisc_device;
 void * sba_get_iommu(struct parisc_device *dev);
 #endif
 
-/* At the moment, we panic on error for IOMMU resource exaustion */
-#define dma_mapping_error(dev, x)	0
-
-/* This API cannot be supported on PA-RISC */
-static inline int dma_mmap_coherent(struct device *dev,
-				    struct vm_area_struct *vma, void *cpu_addr,
-				    dma_addr_t dma_addr, size_t size)
-{
-	return -EINVAL;
-}
-
-static inline int dma_get_sgtable(struct device *dev, struct sg_table *sgt,
-				  void *cpu_addr, dma_addr_t dma_addr,
-				  size_t size)
-{
-	return -EINVAL;
-}
-
 #endif

arch/parisc/include/uapi/asm/mman.h

@@ -43,7 +43,6 @@
 #define MADV_SPACEAVAIL 5               /* insure that resources are reserved */
 #define MADV_VPS_PURGE  6               /* Purge pages from VM page cache */
 #define MADV_VPS_INHERIT 7              /* Inherit parents page size */
-#define MADV_FREE	8		/* free pages only if memory pressure */
 
 /* common/generic parameters */
 #define MADV_FREE	8		/* free pages only if memory pressure */

arch/parisc/kernel/drivers.c

@@ -40,7 +40,7 @@
 #include <asm/parisc-device.h>
 
 /* See comments in include/asm-parisc/pci.h */
-struct hppa_dma_ops *hppa_dma_ops __read_mostly;
+struct dma_map_ops *hppa_dma_ops __read_mostly;
 EXPORT_SYMBOL(hppa_dma_ops);
 
 static struct device root = {

arch/parisc/kernel/pci-dma.c

@@ -413,7 +413,8 @@ pcxl_dma_init(void)
 
 __initcall(pcxl_dma_init);
 
-static void * pa11_dma_alloc_consistent (struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag)
+static void *pa11_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flag, struct dma_attrs *attrs)
 {
 	unsigned long vaddr;
 	unsigned long paddr;
@@ -439,7 +440,8 @@ static void * pa11_dma_alloc_consistent (struct device *dev, size_t size, dma_ad
 	return (void *)vaddr;
 }
 
-static void pa11_dma_free_consistent (struct device *dev, size_t size, void *vaddr, dma_addr_t dma_handle)
+static void pa11_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t dma_handle, struct dma_attrs *attrs)
 {
 	int order;
 
@@ -450,15 +452,20 @@ static void pa11_dma_free_consistent (struct device *dev, size_t size, void *vad
 	free_pages((unsigned long)__va(dma_handle), order);
 }
 
-static dma_addr_t pa11_dma_map_single(struct device *dev, void *addr, size_t size, enum dma_data_direction direction)
+static dma_addr_t pa11_dma_map_page(struct device *dev, struct page *page,
+		unsigned long offset, size_t size,
+		enum dma_data_direction direction, struct dma_attrs *attrs)
 {
+	void *addr = page_address(page) + offset;
 	BUG_ON(direction == DMA_NONE);
 
 	flush_kernel_dcache_range((unsigned long) addr, size);
 	return virt_to_phys(addr);
 }
 
-static void pa11_dma_unmap_single(struct device *dev, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction)
+static void pa11_dma_unmap_page(struct device *dev, dma_addr_t dma_handle,
+		size_t size, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	BUG_ON(direction == DMA_NONE);
 
@@ -475,7 +482,9 @@ static void pa11_dma_unmap_single(struct device *dev, dma_addr_t dma_handle, siz
 	return;
 }
 
-static int pa11_dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction)
+static int pa11_dma_map_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	int i;
 	struct scatterlist *sg;
@@ -492,7 +501,9 @@ static int pa11_dma_map_sg(struct device *dev, struct scatterlist *sglist, int n
 	return nents;
 }
 
-static void pa11_dma_unmap_sg(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction)
+static void pa11_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
+		int nents, enum dma_data_direction direction,
+		struct dma_attrs *attrs)
 {
 	int i;
 	struct scatterlist *sg;
@@ -509,18 +520,24 @@ static void pa11_dma_unmap_sg(struct device *dev, struct scatterlist *sglist, in
 	return;
 }
 
-static void pa11_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction)
+static void pa11_dma_sync_single_for_cpu(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
 	BUG_ON(direction == DMA_NONE);
 
-	flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle) + offset, size);
+	flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle),
+			size);
 }
 
-static void pa11_dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, unsigned long offset, size_t size, enum dma_data_direction direction)
+static void pa11_dma_sync_single_for_device(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction)
 {
 	BUG_ON(direction == DMA_NONE);
 
-	flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle) + offset, size);
+	flush_kernel_dcache_range((unsigned long) phys_to_virt(dma_handle),
+			size);
 }
 
 static void pa11_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sglist, int nents, enum dma_data_direction direction)
@@ -545,32 +562,28 @@ static void pa11_dma_sync_sg_for_device(struct device *dev, struct scatterlist *
 		flush_kernel_vmap_range(sg_virt(sg), sg->length);
 }
 
-struct hppa_dma_ops pcxl_dma_ops = {
+struct dma_map_ops pcxl_dma_ops = {
 	.dma_supported =	pa11_dma_supported,
-	.alloc_consistent =	pa11_dma_alloc_consistent,
-	.alloc_noncoherent =	pa11_dma_alloc_consistent,
-	.free_consistent =	pa11_dma_free_consistent,
-	.map_single =		pa11_dma_map_single,
-	.unmap_single =		pa11_dma_unmap_single,
+	.alloc =		pa11_dma_alloc,
+	.free =			pa11_dma_free,
+	.map_page =		pa11_dma_map_page,
+	.unmap_page =		pa11_dma_unmap_page,
 	.map_sg =		pa11_dma_map_sg,
 	.unmap_sg =		pa11_dma_unmap_sg,
-	.dma_sync_single_for_cpu = pa11_dma_sync_single_for_cpu,
-	.dma_sync_single_for_device = pa11_dma_sync_single_for_device,
-	.dma_sync_sg_for_cpu = pa11_dma_sync_sg_for_cpu,
-	.dma_sync_sg_for_device = pa11_dma_sync_sg_for_device,
+	.sync_single_for_cpu =	pa11_dma_sync_single_for_cpu,
+	.sync_single_for_device = pa11_dma_sync_single_for_device,
+	.sync_sg_for_cpu =	pa11_dma_sync_sg_for_cpu,
+	.sync_sg_for_device =	pa11_dma_sync_sg_for_device,
 };
 
-static void *fail_alloc_consistent(struct device *dev, size_t size,
-				   dma_addr_t *dma_handle, gfp_t flag)
-{
-	return NULL;
-}
-
-static void *pa11_dma_alloc_noncoherent(struct device *dev, size_t size,
-					  dma_addr_t *dma_handle, gfp_t flag)
+static void *pcx_dma_alloc(struct device *dev, size_t size,
+		dma_addr_t *dma_handle, gfp_t flag, struct dma_attrs *attrs)
 {
 	void *addr;
 
+	if (!dma_get_attr(DMA_ATTR_NON_CONSISTENT, attrs))
+		return NULL;
+
 	addr = (void *)__get_free_pages(flag, get_order(size));
 	if (addr)
 		*dma_handle = (dma_addr_t)virt_to_phys(addr);
@@ -578,24 +591,23 @@ static void *pa11_dma_alloc_noncoherent(struct device *dev, size_t size,
 	return addr;
 }
 
-static void pa11_dma_free_noncoherent(struct device *dev, size_t size,
-					void *vaddr, dma_addr_t iova)
+static void pcx_dma_free(struct device *dev, size_t size, void *vaddr,
+		dma_addr_t iova, struct dma_attrs *attrs)
 {
 	free_pages((unsigned long)vaddr, get_order(size));
 	return;
 }
 
-struct hppa_dma_ops pcx_dma_ops = {
+struct dma_map_ops pcx_dma_ops = {
 	.dma_supported =	pa11_dma_supported,
-	.alloc_consistent =	fail_alloc_consistent,
-	.alloc_noncoherent =	pa11_dma_alloc_noncoherent,
-	.free_consistent =	pa11_dma_free_noncoherent,
-	.map_single =		pa11_dma_map_single,
-	.unmap_single =		pa11_dma_unmap_single,
+	.alloc =		pcx_dma_alloc,
+	.free =			pcx_dma_free,
+	.map_page =		pa11_dma_map_page,
+	.unmap_page =		pa11_dma_unmap_page,
 	.map_sg =		pa11_dma_map_sg,
 	.unmap_sg =		pa11_dma_unmap_sg,
-	.dma_sync_single_for_cpu =	pa11_dma_sync_single_for_cpu,
-	.dma_sync_single_for_device =	pa11_dma_sync_single_for_device,
-	.dma_sync_sg_for_cpu =		pa11_dma_sync_sg_for_cpu,
-	.dma_sync_sg_for_device =	pa11_dma_sync_sg_for_device,
+	.sync_single_for_cpu =	pa11_dma_sync_single_for_cpu,
+	.sync_single_for_device = pa11_dma_sync_single_for_device,
+	.sync_sg_for_cpu =	pa11_dma_sync_sg_for_cpu,
+	.sync_sg_for_device =	pa11_dma_sync_sg_for_device,
 };

arch/powerpc/Kconfig

@@ -108,7 +108,6 @@ config PPC
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_MEMBLOCK
 	select HAVE_MEMBLOCK_NODE_MAP
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select HAVE_OPROFILE
 	select HAVE_DEBUG_KMEMLEAK
@@ -158,6 +157,7 @@ config PPC
 	select ARCH_HAS_DMA_SET_COHERENT_MASK
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
 	select HAVE_ARCH_SECCOMP_FILTER
+	select ARCH_HAS_UBSAN_SANITIZE_ALL
 
 config GENERIC_CSUM
 	def_bool CPU_LITTLE_ENDIAN

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

@@ -125,8 +125,6 @@ static inline void set_dma_offset(struct device *dev, dma_addr_t off)
 #define HAVE_ARCH_DMA_SET_MASK 1
 extern int dma_set_mask(struct device *dev, u64 dma_mask);
 
-#include <asm-generic/dma-mapping-common.h>
-
 extern int __dma_set_mask(struct device *dev, u64 dma_mask);
 extern u64 __dma_get_required_mask(struct device *dev);
 

arch/powerpc/include/asm/fadump.h

@@ -191,7 +191,7 @@ struct fadump_crash_info_header {
 	u64		elfcorehdr_addr;
 	u32		crashing_cpu;
 	struct pt_regs	regs;
-	struct cpumask	cpu_online_mask;
+	struct cpumask	online_mask;
 };
 
 /* Crash memory ranges */

arch/powerpc/kernel/Makefile

@@ -136,12 +136,18 @@ endif
 obj-$(CONFIG_EPAPR_PARAVIRT)	+= epapr_paravirt.o epapr_hcalls.o
 obj-$(CONFIG_KVM_GUEST)		+= kvm.o kvm_emul.o
 
-# Disable GCOV in odd or sensitive code
+# Disable GCOV & sanitizers in odd or sensitive code
 GCOV_PROFILE_prom_init.o := n
+UBSAN_SANITIZE_prom_init.o := n
 GCOV_PROFILE_ftrace.o := n
+UBSAN_SANITIZE_ftrace.o := n
 GCOV_PROFILE_machine_kexec_64.o := n
+UBSAN_SANITIZE_machine_kexec_64.o := n
 GCOV_PROFILE_machine_kexec_32.o := n
+UBSAN_SANITIZE_machine_kexec_32.o := n
 GCOV_PROFILE_kprobes.o := n
+UBSAN_SANITIZE_kprobes.o := n
+UBSAN_SANITIZE_vdso.o := n
 
 extra-$(CONFIG_PPC_FPU)		+= fpu.o
 extra-$(CONFIG_ALTIVEC)		+= vector.o

arch/powerpc/kernel/fadump.c

@@ -415,7 +415,7 @@ void crash_fadump(struct pt_regs *regs, const char *str)
 	else
 		ppc_save_regs(&fdh->regs);
 
-	fdh->cpu_online_mask = *cpu_online_mask;
+	fdh->online_mask = *cpu_online_mask;
 
 	/* Call ibm,os-term rtas call to trigger firmware assisted dump */
 	rtas_os_term((char *)str);
@@ -646,7 +646,7 @@ static int __init fadump_build_cpu_notes(const struct fadump_mem_struct *fdm)
 		}
 		/* Lower 4 bytes of reg_value contains logical cpu id */
 		cpu = be64_to_cpu(reg_entry->reg_value) & FADUMP_CPU_ID_MASK;
-		if (fdh && !cpumask_test_cpu(cpu, &fdh->cpu_online_mask)) {
+		if (fdh && !cpumask_test_cpu(cpu, &fdh->online_mask)) {
 			SKIP_TO_NEXT_CPU(reg_entry);
 			continue;
 		}

arch/powerpc/kernel/vdso32/Makefile

@@ -15,6 +15,7 @@ targets := $(obj-vdso32) vdso32.so vdso32.so.dbg
 obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32))
 
 GCOV_PROFILE := n
+UBSAN_SANITIZE := n
 
 ccflags-y := -shared -fno-common -fno-builtin
 ccflags-y += -nostdlib -Wl,-soname=linux-vdso32.so.1 \

arch/powerpc/kernel/vdso64/Makefile

@@ -8,6 +8,7 @@ targets := $(obj-vdso64) vdso64.so vdso64.so.dbg
 obj-vdso64 := $(addprefix $(obj)/, $(obj-vdso64))
 
 GCOV_PROFILE := n
+UBSAN_SANITIZE := n
 
 ccflags-y := -shared -fno-common -fno-builtin
 ccflags-y += -nostdlib -Wl,-soname=linux-vdso64.so.1 \

arch/powerpc/xmon/Makefile

@@ -3,6 +3,7 @@
 subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror
 
 GCOV_PROFILE := n
+UBSAN_SANITIZE := n
 
 ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
 

arch/s390/Kconfig

@@ -579,7 +579,6 @@ config QDIO
 
 menuconfig PCI
 	bool "PCI support"
-	select HAVE_DMA_ATTRS
 	select PCI_MSI
 	select IOMMU_SUPPORT
 	help

arch/s390/include/asm/dma-mapping.h

@@ -23,8 +23,6 @@ static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 {
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 {
 	if (!dev->dma_mask)

arch/sh/Kconfig

@@ -11,7 +11,6 @@ config SUPERH
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_PERF_EVENTS
 	select HAVE_DEBUG_BUGVERBOSE
 	select ARCH_HAVE_CUSTOM_GPIO_H

arch/sh/include/asm/dma-mapping.h

@@ -11,8 +11,6 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 
 #define DMA_ERROR_CODE 0
 
-#include <asm-generic/dma-mapping-common.h>
-
 void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 		    enum dma_data_direction dir);
 

arch/sparc/Kconfig

@@ -26,7 +26,6 @@ config SPARC
 	select RTC_CLASS
 	select RTC_DRV_M48T59
 	select RTC_SYSTOHC
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select HAVE_ARCH_JUMP_LABEL if SPARC64
 	select GENERIC_IRQ_SHOW

arch/sparc/include/asm/dma-mapping.h

@@ -37,21 +37,4 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 	return dma_ops;
 }
 
-#define HAVE_ARCH_DMA_SET_MASK 1
-
-static inline int dma_set_mask(struct device *dev, u64 mask)
-{
-#ifdef CONFIG_PCI
-	if (dev->bus == &pci_bus_type) {
-		if (!dev->dma_mask || !dma_supported(dev, mask))
-			return -EINVAL;
-		*dev->dma_mask = mask;
-		return 0;
-	}
-#endif
-	return -EINVAL;
-}
-
-#include <asm-generic/dma-mapping-common.h>
-
 #endif

arch/tile/Kconfig

@@ -5,7 +5,6 @@ config TILE
 	def_bool y
 	select HAVE_PERF_EVENTS
 	select USE_PMC if PERF_EVENTS
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select HAVE_KVM if !TILEGX
 	select GENERIC_FIND_FIRST_BIT

arch/tile/include/asm/dma-mapping.h

@@ -73,37 +73,7 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 }
 
 #define HAVE_ARCH_DMA_SET_MASK 1
-
-#include <asm-generic/dma-mapping-common.h>
-
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
-	struct dma_map_ops *dma_ops = get_dma_ops(dev);
-
-	/*
-	 * For PCI devices with 64-bit DMA addressing capability, promote
-	 * the dma_ops to hybrid, with the consistent memory DMA space limited
-	 * to 32-bit. For 32-bit capable devices, limit the streaming DMA
-	 * address range to max_direct_dma_addr.
-	 */
-	if (dma_ops == gx_pci_dma_map_ops ||
-	    dma_ops == gx_hybrid_pci_dma_map_ops ||
-	    dma_ops == gx_legacy_pci_dma_map_ops) {
-		if (mask == DMA_BIT_MASK(64) &&
-		    dma_ops == gx_legacy_pci_dma_map_ops)
-			set_dma_ops(dev, gx_hybrid_pci_dma_map_ops);
-		else if (mask > dev->archdata.max_direct_dma_addr)
-			mask = dev->archdata.max_direct_dma_addr;
-	}
-
-	if (!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
-}
+int dma_set_mask(struct device *dev, u64 mask);
 
 /*
  * dma_alloc_noncoherent() is #defined to return coherent memory,

arch/tile/kernel/pci-dma.c

@@ -583,6 +583,35 @@ struct dma_map_ops *gx_hybrid_pci_dma_map_ops;
 EXPORT_SYMBOL(gx_legacy_pci_dma_map_ops);
 EXPORT_SYMBOL(gx_hybrid_pci_dma_map_ops);
 
+int dma_set_mask(struct device *dev, u64 mask)
+{
+	struct dma_map_ops *dma_ops = get_dma_ops(dev);
+
+	/*
+	 * For PCI devices with 64-bit DMA addressing capability, promote
+	 * the dma_ops to hybrid, with the consistent memory DMA space limited
+	 * to 32-bit. For 32-bit capable devices, limit the streaming DMA
+	 * address range to max_direct_dma_addr.
+	 */
+	if (dma_ops == gx_pci_dma_map_ops ||
+	    dma_ops == gx_hybrid_pci_dma_map_ops ||
+	    dma_ops == gx_legacy_pci_dma_map_ops) {
+		if (mask == DMA_BIT_MASK(64) &&
+		    dma_ops == gx_legacy_pci_dma_map_ops)
+			set_dma_ops(dev, gx_hybrid_pci_dma_map_ops);
+		else if (mask > dev->archdata.max_direct_dma_addr)
+			mask = dev->archdata.max_direct_dma_addr;
+	}
+
+	if (!dev->dma_mask || !dma_supported(dev, mask))
+		return -EIO;
+
+	*dev->dma_mask = mask;
+
+	return 0;
+}
+EXPORT_SYMBOL(dma_set_mask);
+
 #ifdef CONFIG_ARCH_HAS_DMA_SET_COHERENT_MASK
 int dma_set_coherent_mask(struct device *dev, u64 mask)
 {

arch/unicore32/Kconfig

@@ -5,7 +5,6 @@ config UNICORE32
 	select ARCH_MIGHT_HAVE_PC_SERIO
 	select HAVE_MEMBLOCK
 	select HAVE_GENERIC_DMA_COHERENT
-	select HAVE_DMA_ATTRS
 	select HAVE_KERNEL_GZIP
 	select HAVE_KERNEL_BZIP2
 	select GENERIC_ATOMIC64

arch/unicore32/include/asm/dma-mapping.h

@@ -28,8 +28,6 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 	return &swiotlb_dma_map_ops;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 {
 	if (dev && dev->dma_mask)

arch/x86/Kconfig

@@ -31,6 +31,7 @@ config X86
 	select ARCH_HAS_PMEM_API		if X86_64
 	select ARCH_HAS_MMIO_FLUSH
 	select ARCH_HAS_SG_CHAIN
+	select ARCH_HAS_UBSAN_SANITIZE_ALL
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	select ARCH_MIGHT_HAVE_ACPI_PDC		if ACPI
 	select ARCH_MIGHT_HAVE_PC_PARPORT
@@ -99,7 +100,6 @@ config X86
 	select HAVE_DEBUG_KMEMLEAK
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_DMA_CONTIGUOUS
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_DYNAMIC_FTRACE_WITH_REGS

arch/x86/boot/Makefile

@@ -60,6 +60,7 @@ clean-files += cpustr.h
 KBUILD_CFLAGS	:= $(USERINCLUDE) $(REALMODE_CFLAGS) -D_SETUP
 KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
 GCOV_PROFILE := n
+UBSAN_SANITIZE := n
 
 $(obj)/bzImage: asflags-y  := $(SVGA_MODE)
 

arch/x86/boot/compressed/Makefile

@@ -33,6 +33,7 @@ KBUILD_CFLAGS += $(call cc-option,-fno-stack-protector)
 
 KBUILD_AFLAGS  := $(KBUILD_CFLAGS) -D__ASSEMBLY__
 GCOV_PROFILE := n
+UBSAN_SANITIZE :=n
 
 LDFLAGS := -m elf_$(UTS_MACHINE)
 LDFLAGS_vmlinux := -T

arch/x86/entry/vdso/Makefile

@@ -4,6 +4,7 @@
 
 KBUILD_CFLAGS += $(DISABLE_LTO)
 KASAN_SANITIZE := n
+UBSAN_SANITIZE := n
 
 VDSO64-$(CONFIG_X86_64)		:= y
 VDSOX32-$(CONFIG_X86_X32_ABI)	:= y

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

@@ -46,8 +46,6 @@ bool arch_dma_alloc_attrs(struct device **dev, gfp_t *gfp);
 #define HAVE_ARCH_DMA_SUPPORTED 1
 extern int dma_supported(struct device *hwdev, u64 mask);
 
-#include <asm-generic/dma-mapping-common.h>
-
 extern void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 					dma_addr_t *dma_addr, gfp_t flag,
 					struct dma_attrs *attrs);

arch/x86/kernel/machine_kexec_64.c

@@ -385,6 +385,7 @@ int arch_kimage_file_post_load_cleanup(struct kimage *image)
 	return image->fops->cleanup(image->image_loader_data);
 }
 
+#ifdef CONFIG_KEXEC_VERIFY_SIG
 int arch_kexec_kernel_verify_sig(struct kimage *image, void *kernel,
 				 unsigned long kernel_len)
 {
@@ -395,6 +396,7 @@ int arch_kexec_kernel_verify_sig(struct kimage *image, void *kernel,
 
 	return image->fops->verify_sig(kernel, kernel_len);
 }
+#endif
 
 /*
  * Apply purgatory relocations.

arch/x86/realmode/rm/Makefile

@@ -70,3 +70,4 @@ KBUILD_CFLAGS	:= $(LINUXINCLUDE) $(REALMODE_CFLAGS) -D_SETUP -D_WAKEUP \
 		   -I$(srctree)/arch/x86/boot
 KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
 GCOV_PROFILE := n
+UBSAN_SANITIZE := n

arch/xtensa/Kconfig

@@ -15,7 +15,6 @@ config XTENSA
 	select GENERIC_PCI_IOMAP
 	select GENERIC_SCHED_CLOCK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_DMA_ATTRS
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUTEX_CMPXCHG if !MMU
 	select HAVE_IRQ_TIME_ACCOUNTING

arch/xtensa/include/asm/dma-mapping.h

@@ -13,8 +13,6 @@
 #include <asm/cache.h>
 #include <asm/io.h>
 
-#include <asm-generic/dma-coherent.h>
-
 #include <linux/mm.h>
 #include <linux/scatterlist.h>
 
@@ -30,8 +28,6 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 		return &xtensa_dma_map_ops;
 }
 
-#include <asm-generic/dma-mapping-common.h>
-
 void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 		    enum dma_data_direction direction);
 

arch/xtensa/include/uapi/asm/mman.h

@@ -86,7 +86,6 @@
 #define MADV_SEQUENTIAL	2		/* expect sequential page references */
 #define MADV_WILLNEED	3		/* will need these pages */
 #define MADV_DONTNEED	4		/* don't need these pages */
-#define MADV_FREE	5		/* free pages only if memory pressure */
 
 /* common parameters: try to keep these consistent across architectures */
 #define MADV_FREE	8		/* free pages only if memory pressure */

drivers/base/cpu.c

@@ -200,7 +200,7 @@ static const struct attribute_group *hotplugable_cpu_attr_groups[] = {
 
 struct cpu_attr {
 	struct device_attribute attr;
-	const struct cpumask *const * const map;
+	const struct cpumask *const map;
 };
 
 static ssize_t show_cpus_attr(struct device *dev,
@@ -209,7 +209,7 @@ static ssize_t show_cpus_attr(struct device *dev,
 {
 	struct cpu_attr *ca = container_of(attr, struct cpu_attr, attr);
 
-	return cpumap_print_to_pagebuf(true, buf, *ca->map);
+	return cpumap_print_to_pagebuf(true, buf, ca->map);
 }
 
 #define _CPU_ATTR(name, map) \
@@ -217,9 +217,9 @@ static ssize_t show_cpus_attr(struct device *dev,
 
 /* Keep in sync with cpu_subsys_attrs */
 static struct cpu_attr cpu_attrs[] = {
-	_CPU_ATTR(online, &cpu_online_mask),
-	_CPU_ATTR(possible, &cpu_possible_mask),
-	_CPU_ATTR(present, &cpu_present_mask),
+	_CPU_ATTR(online, &__cpu_online_mask),
+	_CPU_ATTR(possible, &__cpu_possible_mask),
+	_CPU_ATTR(present, &__cpu_present_mask),
 };
 
 /*

drivers/base/dma-mapping.c

@@ -12,7 +12,6 @@
 #include <linux/gfp.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
-#include <asm-generic/dma-coherent.h>
 
 /*
  * Managed DMA API
@@ -167,7 +166,7 @@ void dmam_free_noncoherent(struct device *dev, size_t size, void *vaddr,
 }
 EXPORT_SYMBOL(dmam_free_noncoherent);
 
-#ifdef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
+#ifdef CONFIG_HAVE_GENERIC_DMA_COHERENT
 
 static void dmam_coherent_decl_release(struct device *dev, void *res)
 {
@@ -247,7 +246,7 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
 		    void *cpu_addr, dma_addr_t dma_addr, size_t size)
 {
 	int ret = -ENXIO;
-#ifdef CONFIG_MMU
+#if defined(CONFIG_MMU) && !defined(CONFIG_ARCH_NO_COHERENT_DMA_MMAP)
 	unsigned long user_count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
 	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
 	unsigned long pfn = page_to_pfn(virt_to_page(cpu_addr));
@@ -264,7 +263,7 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
 				      user_count << PAGE_SHIFT,
 				      vma->vm_page_prot);
 	}
-#endif	/* CONFIG_MMU */
+#endif	/* CONFIG_MMU && !CONFIG_ARCH_NO_COHERENT_DMA_MMAP */
 
 	return ret;
 }

drivers/firmware/broadcom/bcm47xx_nvram.c

@@ -56,9 +56,7 @@ static u32 find_nvram_size(void __iomem *end)
 static int nvram_find_and_copy(void __iomem *iobase, u32 lim)
 {
 	struct nvram_header __iomem *header;
-	int i;
 	u32 off;
-	u32 *src, *dst;
 	u32 size;
 
 	if (nvram_len) {
@@ -95,10 +93,7 @@ static int nvram_find_and_copy(void __iomem *iobase, u32 lim)
 	return -ENXIO;
 
 found:
-	src = (u32 *)header;
-	dst = (u32 *)nvram_buf;
-	for (i = 0; i < sizeof(struct nvram_header); i += 4)
-		*dst++ = __raw_readl(src++);
+	__ioread32_copy(nvram_buf, header, sizeof(*header) / 4);
 	header = (struct nvram_header *)nvram_buf;
 	nvram_len = header->len;
 	if (nvram_len > size) {
@@ -111,8 +106,8 @@ static int nvram_find_and_copy(void __iomem *iobase, u32 lim)
 		nvram_len = NVRAM_SPACE - 1;
 	}
 	/* proceed reading data after header */
-	for (; i < nvram_len; i += 4)
-		*dst++ = readl(src++);
+	__ioread32_copy(nvram_buf + sizeof(*header), header + 1,
+			DIV_ROUND_UP(nvram_len, 4));
 	nvram_buf[NVRAM_SPACE - 1] = '\0';
 
 	return 0;

drivers/firmware/efi/libstub/Makefile

@@ -22,6 +22,7 @@ KBUILD_CFLAGS			:= $(cflags-y) -DDISABLE_BRANCH_PROFILING \
 
 GCOV_PROFILE			:= n
 KASAN_SANITIZE			:= n
+UBSAN_SANITIZE			:= n
 
 lib-y				:= efi-stub-helper.o
 

drivers/gpu/drm/Kconfig

@@ -82,13 +82,13 @@ config DRM_TTM
 
 config DRM_GEM_CMA_HELPER
 	bool
-	depends on DRM && HAVE_DMA_ATTRS
+	depends on DRM
 	help
 	  Choose this if you need the GEM CMA helper functions
 
 config DRM_KMS_CMA_HELPER
 	bool
-	depends on DRM && HAVE_DMA_ATTRS
+	depends on DRM
 	select DRM_GEM_CMA_HELPER
 	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT

drivers/gpu/drm/imx/Kconfig

@@ -5,7 +5,7 @@ config DRM_IMX
 	select VIDEOMODE_HELPERS
 	select DRM_GEM_CMA_HELPER
 	select DRM_KMS_CMA_HELPER
-	depends on DRM && (ARCH_MXC || ARCH_MULTIPLATFORM) && HAVE_DMA_ATTRS
+	depends on DRM && (ARCH_MXC || ARCH_MULTIPLATFORM)
 	depends on IMX_IPUV3_CORE
 	help
 	  enable i.MX graphics support

drivers/gpu/drm/rcar-du/Kconfig

 config DRM_RCAR_DU
 	tristate "DRM Support for R-Car Display Unit"
-	depends on DRM && ARM && HAVE_DMA_ATTRS && OF
+	depends on DRM && ARM && OF
 	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER

drivers/gpu/drm/shmobile/Kconfig

 config DRM_SHMOBILE
 	tristate "DRM Support for SH Mobile"
-	depends on DRM && ARM && HAVE_DMA_ATTRS
+	depends on DRM && ARM
 	depends on ARCH_SHMOBILE || COMPILE_TEST
 	depends on FB_SH_MOBILE_MERAM || !FB_SH_MOBILE_MERAM
 	select BACKLIGHT_CLASS_DEVICE

drivers/gpu/drm/sti/Kconfig

 config DRM_STI
 	tristate "DRM Support for STMicroelectronics SoC stiH41x Series"
-	depends on DRM && (SOC_STIH415 || SOC_STIH416 || ARCH_MULTIPLATFORM) && HAVE_DMA_ATTRS
+	depends on DRM && (SOC_STIH415 || SOC_STIH416 || ARCH_MULTIPLATFORM)
 	select RESET_CONTROLLER
 	select DRM_KMS_HELPER
 	select DRM_GEM_CMA_HELPER