diff mbox

[RFC,3/3] iommu: dma-iommu: use common implementation also on ARM architecture

Message ID 1455870164-25337-4-git-send-email-m.szyprowski@samsung.com (mailing list archive)
State New, archived
Headers show

Commit Message

Marek Szyprowski Feb. 19, 2016, 8:22 a.m. UTC
This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
side-effect of this change is a switch from bitmap-based IO address space
management to tree-based code. There should be no functional changes
for drivers, which rely on initialization from generic arch_setup_dna_ops()
interface. Code, which used old arm_iommu_* functions must be updated to
new interface.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
---
 arch/arm/Kconfig                   |   22 +-
 arch/arm/include/asm/device.h      |    9 -
 arch/arm/include/asm/dma-iommu.h   |   37 --
 arch/arm/include/asm/dma-mapping.h |   59 +-
 arch/arm/mm/dma-mapping.c          | 1158 +-----------------------------------
 drivers/gpu/drm/rockchip/Kconfig   |    1 +
 drivers/iommu/Kconfig              |    1 +
 drivers/media/platform/Kconfig     |    1 +
 8 files changed, 82 insertions(+), 1206 deletions(-)
 delete mode 100644 arch/arm/include/asm/dma-iommu.h

Comments

Arnd Bergmann Feb. 19, 2016, 10:30 a.m. UTC | #1
On Friday 19 February 2016 09:22:44 Marek Szyprowski wrote:
> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
> side-effect of this change is a switch from bitmap-based IO address space
> management to tree-based code. There should be no functional changes
> for drivers, which rely on initialization from generic arch_setup_dna_ops()
> interface. Code, which used old arm_iommu_* functions must be updated to
> new interface.
> 
> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>

I like the overall idea. However, this interface from the iommu
subsystem into architecture specific code:

> +/*
> + * The DMA API is built upon the notion of "buffer ownership".  A buffer
> + * is either exclusively owned by the CPU (and therefore may be accessed
> + * by it) or exclusively owned by the DMA device.  These helper functions
> + * represent the transitions between these two ownership states.
> + *
> + * Note, however, that on later ARMs, this notion does not work due to
> + * speculative prefetches.  We model our approach on the assumption that
> + * the CPU does do speculative prefetches, which means we clean caches
> + * before transfers and delay cache invalidation until transfer completion.
> + *
> + */
> +extern void __dma_page_cpu_to_dev(struct page *, unsigned long, size_t,
> +				  enum dma_data_direction);
> +extern void __dma_page_dev_to_cpu(struct page *, unsigned long, size_t,
> +				  enum dma_data_direction);
> +
> +static inline void arch_flush_page(struct device *dev, const void *virt,
> +			    phys_addr_t phys)
> +{
> +	dmac_flush_range(virt, virt + PAGE_SIZE);
> +	outer_flush_range(phys, phys + PAGE_SIZE);
> +}
> +
> +static inline void arch_dma_map_area(phys_addr_t phys, size_t size,
> +				     enum dma_data_direction dir)
> +{
> +	unsigned int offset = phys & ~PAGE_MASK;
> +	__dma_page_cpu_to_dev(phys_to_page(phys & PAGE_MASK), offset, size, dir);
> +}
> +
> +static inline void arch_dma_unmap_area(phys_addr_t phys, size_t size,
> +				       enum dma_data_direction dir)
> +{
> +	unsigned int offset = phys & ~PAGE_MASK;
> +	__dma_page_dev_to_cpu(phys_to_page(phys & PAGE_MASK), offset, size, dir);
> +}
> +
> +static inline pgprot_t arch_get_dma_pgprot(struct dma_attrs *attrs,
> +					pgprot_t prot, bool coherent)
> +{
> +	if (coherent)
> +		return prot;
> +
> +	prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ?
> +			    pgprot_writecombine(prot) :
> +			    pgprot_dmacoherent(prot);
> +	return prot;
> +}
> +
> +extern void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page,
> +					 gfp_t flags);
> +extern bool arch_in_atomic_pool(void *start, size_t size);
> +extern int arch_free_from_atomic_pool(void *start, size_t size);
> +
> +

doesn't feel completely right yet. In particular the arch_flush_page()
interface is probably still too specific to ARM/ARM64 and won't work
that way on other architectures.

I think it would be better to do this either more generic, or less generic:

a) leave the iommu_dma_map_ops definition in the architecture specific
   code, but make it call helper functions in the drivers/iommu to do all
   of the really generic parts.

b) clarify that this is only applicable to arch/arm and arch/arm64, and
   unify things further between these two, as they have very similar
   requirements in the CPU architecture.

	Arnd
Marek Szyprowski Feb. 25, 2016, 12:26 p.m. UTC | #2
Hello,

On 2016-02-19 11:30, Arnd Bergmann wrote:
> On Friday 19 February 2016 09:22:44 Marek Szyprowski wrote:
>> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
>> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
>> side-effect of this change is a switch from bitmap-based IO address space
>> management to tree-based code. There should be no functional changes
>> for drivers, which rely on initialization from generic arch_setup_dna_ops()
>> interface. Code, which used old arm_iommu_* functions must be updated to
>> new interface.
>>
>> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
> I like the overall idea. However, this interface from the iommu
> subsystem into architecture specific code:
>
>> +/*
>> + * The DMA API is built upon the notion of "buffer ownership".  A buffer
>> + * is either exclusively owned by the CPU (and therefore may be accessed
>> + * by it) or exclusively owned by the DMA device.  These helper functions
>> + * represent the transitions between these two ownership states.
>> + *
>> + * Note, however, that on later ARMs, this notion does not work due to
>> + * speculative prefetches.  We model our approach on the assumption that
>> + * the CPU does do speculative prefetches, which means we clean caches
>> + * before transfers and delay cache invalidation until transfer completion.
>> + *
>> + */
>> +extern void __dma_page_cpu_to_dev(struct page *, unsigned long, size_t,
>> +				  enum dma_data_direction);
>> +extern void __dma_page_dev_to_cpu(struct page *, unsigned long, size_t,
>> +				  enum dma_data_direction);
>> +
>> +static inline void arch_flush_page(struct device *dev, const void *virt,
>> +			    phys_addr_t phys)
>> +{
>> +	dmac_flush_range(virt, virt + PAGE_SIZE);
>> +	outer_flush_range(phys, phys + PAGE_SIZE);
>> +}
>> +
>> +static inline void arch_dma_map_area(phys_addr_t phys, size_t size,
>> +				     enum dma_data_direction dir)
>> +{
>> +	unsigned int offset = phys & ~PAGE_MASK;
>> +	__dma_page_cpu_to_dev(phys_to_page(phys & PAGE_MASK), offset, size, dir);
>> +}
>> +
>> +static inline void arch_dma_unmap_area(phys_addr_t phys, size_t size,
>> +				       enum dma_data_direction dir)
>> +{
>> +	unsigned int offset = phys & ~PAGE_MASK;
>> +	__dma_page_dev_to_cpu(phys_to_page(phys & PAGE_MASK), offset, size, dir);
>> +}
>> +
>> +static inline pgprot_t arch_get_dma_pgprot(struct dma_attrs *attrs,
>> +					pgprot_t prot, bool coherent)
>> +{
>> +	if (coherent)
>> +		return prot;
>> +
>> +	prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ?
>> +			    pgprot_writecombine(prot) :
>> +			    pgprot_dmacoherent(prot);
>> +	return prot;
>> +}
>> +
>> +extern void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page,
>> +					 gfp_t flags);
>> +extern bool arch_in_atomic_pool(void *start, size_t size);
>> +extern int arch_free_from_atomic_pool(void *start, size_t size);
>> +
>> +
> doesn't feel completely right yet. In particular the arch_flush_page()
> interface is probably still too specific to ARM/ARM64 and won't work
> that way on other architectures.
>
> I think it would be better to do this either more generic, or less generic:
>
> a) leave the iommu_dma_map_ops definition in the architecture specific
>     code, but make it call helper functions in the drivers/iommu to do all
>     of the really generic parts.
>
> b) clarify that this is only applicable to arch/arm and arch/arm64, and
>     unify things further between these two, as they have very similar
>     requirements in the CPU architecture.

Some really generic parts are already in iommu/dma-iommu.c and one can build
it's own, non-ARM CPU architecture based IOMMU/DMA-mapping code. Initially I
also wanted to use that generic code on both ARM and ARM64, but it 
turned out
that both archs, ARM and ARM64 will duplicate 99% of code, which use this
'generic' functions. This was the reason why I dedided to move all that
common code from arch/{arm,arm64}/mm/dma-mapping.c to
drivers/iommu/dma-iommu-ops.c

I'm not sure if I can design all the changes that need to be made to
drivers/iommu/dma-iommu-ops.c to make it more generic. Maybe when one will
try to use that code with other, non-ARM architecture based arch glue code,
a better abstraction can be developed. For now I would like to keep all this
code in a common place so both arm and arm64 will benefit from improvements
done there.

Best regards
Arnd Bergmann Feb. 25, 2016, 2:44 p.m. UTC | #3
On Thursday 25 February 2016 13:26:17 Marek Szyprowski wrote:
> >> +}
> >> +
> >> +extern void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page,
> >> +                                     gfp_t flags);
> >> +extern bool arch_in_atomic_pool(void *start, size_t size);
> >> +extern int arch_free_from_atomic_pool(void *start, size_t size);
> >> +
> >> +
> > doesn't feel completely right yet. In particular the arch_flush_page()
> > interface is probably still too specific to ARM/ARM64 and won't work
> > that way on other architectures.
> >
> > I think it would be better to do this either more generic, or less generic:
> >
> > a) leave the iommu_dma_map_ops definition in the architecture specific
> >     code, but make it call helper functions in the drivers/iommu to do all
> >     of the really generic parts.
> >
> > b) clarify that this is only applicable to arch/arm and arch/arm64, and
> >     unify things further between these two, as they have very similar
> >     requirements in the CPU architecture.
> 
> Some really generic parts are already in iommu/dma-iommu.c and one can build
> it's own, non-ARM CPU architecture based IOMMU/DMA-mapping code. Initially I
> also wanted to use that generic code on both ARM and ARM64, but it 
> turned out
> that both archs, ARM and ARM64 will duplicate 99% of code, which use this
> 'generic' functions. This was the reason why I dedided to move all that
> common code from arch/{arm,arm64}/mm/dma-mapping.c to
> drivers/iommu/dma-iommu-ops.c
> 
> I'm not sure if I can design all the changes that need to be made to
> drivers/iommu/dma-iommu-ops.c to make it more generic. Maybe when one will
> try to use that code with other, non-ARM architecture based arch glue code,
> a better abstraction can be developed. For now I would like to keep all this
> code in a common place so both arm and arm64 will benefit from improvements
> done there.

Fair enough. Let's stay with your approach then.

	Arnd
Magnus Damm March 15, 2016, 11:18 a.m. UTC | #4
Hi Marek,

On Fri, Feb 19, 2016 at 5:22 PM, Marek Szyprowski
<m.szyprowski@samsung.com> wrote:
> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
> side-effect of this change is a switch from bitmap-based IO address space
> management to tree-based code. There should be no functional changes
> for drivers, which rely on initialization from generic arch_setup_dna_ops()
> interface. Code, which used old arm_iommu_* functions must be updated to
> new interface.
>
> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
> ---

Thanks for your efforts and my apologies for late comments. Just FYI
I'll try your patch (and this series) with the ipmmu-vmsa.c driver on
32-bit ARM and see how it goes. Nice not to have to support multiple
interfaces depending on architecture!

One question that comes to mind is how to handle features.

For instance, the 32-bit ARM code supports DMA_ATTR_FORCE_CONTIGUOUS
while the shared code in drivers/iommu/dma-iommu.c does not. I assume
existing users may rely on such features so from my point of view it
probably makes sense to carry over features from the 32-bit ARM code
into the shared code before pulling the plug.

I also wonder if it is possible to do a step-by-step migration and
support both old and new interfaces in the same binary? That may make
things easier for multiplatform enablement. So far I've managed to
make one IOMMU driver support both 32-bit ARM and 64-bit ARM with some
ugly magic, so adjusting 32-bit ARM dma-mapping code to coexist with
the shared code in drivers/iommu/dma-iommu.c may also be possible. And
probably involving even more ugly magic. =)

Cheers,

/ magnus
Marek Szyprowski March 15, 2016, 12:03 p.m. UTC | #5
Hello,

On 2016-03-15 12:18, Magnus Damm wrote:
> Hi Marek,
>
> On Fri, Feb 19, 2016 at 5:22 PM, Marek Szyprowski
> <m.szyprowski@samsung.com> wrote:
>> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
>> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
>> side-effect of this change is a switch from bitmap-based IO address space
>> management to tree-based code. There should be no functional changes
>> for drivers, which rely on initialization from generic arch_setup_dna_ops()
>> interface. Code, which used old arm_iommu_* functions must be updated to
>> new interface.
>>
>> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
>> ---
> Thanks for your efforts and my apologies for late comments. Just FYI
> I'll try your patch (and this series) with the ipmmu-vmsa.c driver on
> 32-bit ARM and see how it goes. Nice not to have to support multiple
> interfaces depending on architecture!

Thanks for testing!

> One question that comes to mind is how to handle features.
>
> For instance, the 32-bit ARM code supports DMA_ATTR_FORCE_CONTIGUOUS
> while the shared code in drivers/iommu/dma-iommu.c does not. I assume
> existing users may rely on such features so from my point of view it
> probably makes sense to carry over features from the 32-bit ARM code
> into the shared code before pulling the plug.

Right, this has to be added to common code before merging.

> I also wonder if it is possible to do a step-by-step migration and
> support both old and new interfaces in the same binary? That may make
> things easier for multiplatform enablement. So far I've managed to
> make one IOMMU driver support both 32-bit ARM and 64-bit ARM with some
> ugly magic, so adjusting 32-bit ARM dma-mapping code to coexist with
> the shared code in drivers/iommu/dma-iommu.c may also be possible. And
> probably involving even more ugly magic. =)

Having one IOMMU driver for both 32-bit and 64-bit ARM archs is quite easy
IF you rely on the iommu core to setup everything. See exynos-iommu driver
- after my last patches it now works fine on both archs (using arch
specific interfaces). Most of the magic is done automatically by
arch_setup_dma_ops().

The real problem is the fact that there are drivers (like DRM) which rely
on specific dma-mapping functions from ARM architecture, which need to be
rewritten.

Best regards
Robin Murphy March 15, 2016, 12:33 p.m. UTC | #6
Hi Marek, Arnd,

On 19/02/16 10:30, Arnd Bergmann wrote:
> On Friday 19 February 2016 09:22:44 Marek Szyprowski wrote:
>> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
>> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
>> side-effect of this change is a switch from bitmap-based IO address space
>> management to tree-based code. There should be no functional changes
>> for drivers, which rely on initialization from generic arch_setup_dna_ops()
>> interface. Code, which used old arm_iommu_* functions must be updated to
>> new interface.
>>
>> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
>
> I like the overall idea. However, this interface from the iommu
> subsystem into architecture specific code:
>
>> +/*
>> + * The DMA API is built upon the notion of "buffer ownership".  A buffer
>> + * is either exclusively owned by the CPU (and therefore may be accessed
>> + * by it) or exclusively owned by the DMA device.  These helper functions
>> + * represent the transitions between these two ownership states.
>> + *
>> + * Note, however, that on later ARMs, this notion does not work due to
>> + * speculative prefetches.  We model our approach on the assumption that
>> + * the CPU does do speculative prefetches, which means we clean caches
>> + * before transfers and delay cache invalidation until transfer completion.
>> + *
>> + */
>> +extern void __dma_page_cpu_to_dev(struct page *, unsigned long, size_t,
>> +				  enum dma_data_direction);
>> +extern void __dma_page_dev_to_cpu(struct page *, unsigned long, size_t,
>> +				  enum dma_data_direction);
>> +
>> +static inline void arch_flush_page(struct device *dev, const void *virt,
>> +			    phys_addr_t phys)
>> +{
>> +	dmac_flush_range(virt, virt + PAGE_SIZE);
>> +	outer_flush_range(phys, phys + PAGE_SIZE);
>> +}
>> +
>> +static inline void arch_dma_map_area(phys_addr_t phys, size_t size,
>> +				     enum dma_data_direction dir)
>> +{
>> +	unsigned int offset = phys & ~PAGE_MASK;
>> +	__dma_page_cpu_to_dev(phys_to_page(phys & PAGE_MASK), offset, size, dir);
>> +}
>> +
>> +static inline void arch_dma_unmap_area(phys_addr_t phys, size_t size,
>> +				       enum dma_data_direction dir)
>> +{
>> +	unsigned int offset = phys & ~PAGE_MASK;
>> +	__dma_page_dev_to_cpu(phys_to_page(phys & PAGE_MASK), offset, size, dir);
>> +}
>> +
>> +static inline pgprot_t arch_get_dma_pgprot(struct dma_attrs *attrs,
>> +					pgprot_t prot, bool coherent)
>> +{
>> +	if (coherent)
>> +		return prot;
>> +
>> +	prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ?
>> +			    pgprot_writecombine(prot) :
>> +			    pgprot_dmacoherent(prot);
>> +	return prot;
>> +}
>> +
>> +extern void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page,
>> +					 gfp_t flags);
>> +extern bool arch_in_atomic_pool(void *start, size_t size);
>> +extern int arch_free_from_atomic_pool(void *start, size_t size);
>> +
>> +
>
> doesn't feel completely right yet. In particular the arch_flush_page()
> interface is probably still too specific to ARM/ARM64 and won't work
> that way on other architectures.
>
> I think it would be better to do this either more generic, or less generic:
>
> a) leave the iommu_dma_map_ops definition in the architecture specific
>     code, but make it call helper functions in the drivers/iommu to do all
>     of the really generic parts.

This was certainly the original intent of the arm64 code. The division 
of responsibility there is a conscious decision - IOMMU-API-wrangling 
goes in the common code, cache maintenance and actual dma_map_ops stay 
hidden in architecture-private code, safe from abuse. It's very much 
modelled on SWIOTLB.

Given all the work Russell did last year getting rid of direct uses of 
the dmac_* cache maintenance functions by ARM drivers, I don't think 
bringing all of that back is a good way to go - Personally I'd much 
rather see several dozen lines of very similar looking (other than 
highmem and outer cache stuff) arch-private code if it maintains a 
robust and clearly-defined abstraction (and avoids yet another level of 
indirection). It does also seem a little odd to factor out only half the 
file on the grounds of architectural similarity, when that argument 
applies equally to the other (non-IOMMU) half too. I think the recent 
tree-wide conversion to generic dma_map_ops was in part motivated by the 
thought of common implementations, so I'm sure that's something we can 
revisit in due course.

Robin.

>
> b) clarify that this is only applicable to arch/arm and arch/arm64, and
>     unify things further between these two, as they have very similar
>     requirements in the CPU architecture.
>
> 	Arnd
> _______________________________________________
> iommu mailing list
> iommu@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/iommu
>
yao mark April 18, 2016, 2:20 a.m. UTC | #7
On 2016?02?19? 16:22, Marek Szyprowski wrote:
> This patch replaces ARM-specific IOMMU-based DMA-mapping implementation
> with generic IOMMU DMA-mapping code shared with ARM64 architecture. The
> side-effect of this change is a switch from bitmap-based IO address space
> management to tree-based code. There should be no functional changes
> for drivers, which rely on initialization from generic arch_setup_dna_ops()
> interface. Code, which used old arm_iommu_* functions must be updated to
> new interface.
>
> Signed-off-by: Marek Szyprowski<m.szyprowski@samsung.com>
IOMMU works good on drm/rockchip ARM64 platform with this patch. So

Tested-by: Mark Yao <mark.yao@rock-chips.com>

Thanks.
diff mbox

Patch

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 4f799e567fc8..ed45f0d63cee 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -107,27 +107,7 @@  config ARM_DMA_USE_IOMMU
 	bool
 	select ARM_HAS_SG_CHAIN
 	select NEED_SG_DMA_LENGTH
-
-if ARM_DMA_USE_IOMMU
-
-config ARM_DMA_IOMMU_ALIGNMENT
-	int "Maximum PAGE_SIZE order of alignment for DMA IOMMU buffers"
-	range 4 9
-	default 8
-	help
-	  DMA mapping framework by default aligns all buffers to the smallest
-	  PAGE_SIZE order which is greater than or equal to the requested buffer
-	  size. This works well for buffers up to a few hundreds kilobytes, but
-	  for larger buffers it just a waste of address space. Drivers which has
-	  relatively small addressing window (like 64Mib) might run out of
-	  virtual space with just a few allocations.
-
-	  With this parameter you can specify the maximum PAGE_SIZE order for
-	  DMA IOMMU buffers. Larger buffers will be aligned only to this
-	  specified order. The order is expressed as a power of two multiplied
-	  by the PAGE_SIZE.
-
-endif
+	select IOMMU_DMA
 
 config MIGHT_HAVE_PCI
 	bool
diff --git a/arch/arm/include/asm/device.h b/arch/arm/include/asm/device.h
index 4111592f0130..6ea939c39cd4 100644
--- a/arch/arm/include/asm/device.h
+++ b/arch/arm/include/asm/device.h
@@ -14,9 +14,6 @@  struct dev_archdata {
 #ifdef CONFIG_IOMMU_API
 	void *iommu; /* private IOMMU data */
 #endif
-#ifdef CONFIG_ARM_DMA_USE_IOMMU
-	struct dma_iommu_mapping	*mapping;
-#endif
 	bool dma_coherent;
 };
 
@@ -28,10 +25,4 @@  struct pdev_archdata {
 #endif
 };
 
-#ifdef CONFIG_ARM_DMA_USE_IOMMU
-#define to_dma_iommu_mapping(dev) ((dev)->archdata.mapping)
-#else
-#define to_dma_iommu_mapping(dev) NULL
-#endif
-
 #endif
diff --git a/arch/arm/include/asm/dma-iommu.h b/arch/arm/include/asm/dma-iommu.h
deleted file mode 100644
index 2ef282f96651..000000000000
--- a/arch/arm/include/asm/dma-iommu.h
+++ /dev/null
@@ -1,37 +0,0 @@ 
-#ifndef ASMARM_DMA_IOMMU_H
-#define ASMARM_DMA_IOMMU_H
-
-#ifdef __KERNEL__
-
-#include <linux/mm_types.h>
-#include <linux/scatterlist.h>
-#include <linux/dma-debug.h>
-#include <linux/kmemcheck.h>
-#include <linux/kref.h>
-
-struct dma_iommu_mapping {
-	/* iommu specific data */
-	struct iommu_domain	*domain;
-
-	unsigned long		**bitmaps;	/* array of bitmaps */
-	unsigned int		nr_bitmaps;	/* nr of elements in array */
-	unsigned int		extensions;
-	size_t			bitmap_size;	/* size of a single bitmap */
-	size_t			bits;		/* per bitmap */
-	dma_addr_t		base;
-
-	spinlock_t		lock;
-	struct kref		kref;
-};
-
-struct dma_iommu_mapping *
-arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, u64 size);
-
-void arm_iommu_release_mapping(struct dma_iommu_mapping *mapping);
-
-int arm_iommu_attach_device(struct device *dev,
-					struct dma_iommu_mapping *mapping);
-void arm_iommu_detach_device(struct device *dev);
-
-#endif /* __KERNEL__ */
-#endif
diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index 6ad1ceda62a5..08bedb0c02c6 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h
@@ -8,6 +8,7 @@ 
 #include <linux/dma-attrs.h>
 #include <linux/dma-debug.h>
 
+#include <asm/cacheflush.h>
 #include <asm/memory.h>
 
 #include <xen/xen.h>
@@ -32,7 +33,7 @@  static inline struct dma_map_ops *get_dma_ops(struct device *dev)
 		return __generic_dma_ops(dev);
 }
 
-static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops)
+static inline void arch_set_dma_ops(struct device *dev, struct dma_map_ops *ops)
 {
 	BUG_ON(!dev);
 	dev->archdata.dma_ops = ops;
@@ -275,5 +276,61 @@  extern int arm_dma_get_sgtable(struct device *dev, struct sg_table *sgt,
 		void *cpu_addr, dma_addr_t dma_addr, size_t size,
 		struct dma_attrs *attrs);
 
+/*
+ * The DMA API is built upon the notion of "buffer ownership".  A buffer
+ * is either exclusively owned by the CPU (and therefore may be accessed
+ * by it) or exclusively owned by the DMA device.  These helper functions
+ * represent the transitions between these two ownership states.
+ *
+ * Note, however, that on later ARMs, this notion does not work due to
+ * speculative prefetches.  We model our approach on the assumption that
+ * the CPU does do speculative prefetches, which means we clean caches
+ * before transfers and delay cache invalidation until transfer completion.
+ *
+ */
+extern void __dma_page_cpu_to_dev(struct page *, unsigned long, size_t,
+				  enum dma_data_direction);
+extern void __dma_page_dev_to_cpu(struct page *, unsigned long, size_t,
+				  enum dma_data_direction);
+
+static inline void arch_flush_page(struct device *dev, const void *virt,
+			    phys_addr_t phys)
+{
+	dmac_flush_range(virt, virt + PAGE_SIZE);
+	outer_flush_range(phys, phys + PAGE_SIZE);
+}
+
+static inline void arch_dma_map_area(phys_addr_t phys, size_t size,
+				     enum dma_data_direction dir)
+{
+	unsigned int offset = phys & ~PAGE_MASK;
+	__dma_page_cpu_to_dev(phys_to_page(phys & PAGE_MASK), offset, size, dir);
+}
+
+static inline void arch_dma_unmap_area(phys_addr_t phys, size_t size,
+				       enum dma_data_direction dir)
+{
+	unsigned int offset = phys & ~PAGE_MASK;
+	__dma_page_dev_to_cpu(phys_to_page(phys & PAGE_MASK), offset, size, dir);
+}
+
+static inline pgprot_t arch_get_dma_pgprot(struct dma_attrs *attrs,
+					pgprot_t prot, bool coherent)
+{
+	if (coherent)
+		return prot;
+
+	prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ?
+			    pgprot_writecombine(prot) :
+			    pgprot_dmacoherent(prot);
+	return prot;
+}
+
+extern void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page,
+					 gfp_t flags);
+extern bool arch_in_atomic_pool(void *start, size_t size);
+extern int arch_free_from_atomic_pool(void *start, size_t size);
+
+
 #endif /* __KERNEL__ */
 #endif
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 0eca3812527e..5d497f3c5924 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -20,6 +20,7 @@ 
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
 #include <linux/dma-contiguous.h>
+#include <linux/dma-iommu.h>
 #include <linux/highmem.h>
 #include <linux/memblock.h>
 #include <linux/slab.h>
@@ -34,7 +35,6 @@ 
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
 #include <asm/mach/arch.h>
-#include <asm/dma-iommu.h>
 #include <asm/mach/map.h>
 #include <asm/system_info.h>
 #include <asm/dma-contiguous.h>
@@ -42,23 +42,6 @@ 
 #include "dma.h"
 #include "mm.h"
 
-/*
- * The DMA API is built upon the notion of "buffer ownership".  A buffer
- * is either exclusively owned by the CPU (and therefore may be accessed
- * by it) or exclusively owned by the DMA device.  These helper functions
- * represent the transitions between these two ownership states.
- *
- * Note, however, that on later ARMs, this notion does not work due to
- * speculative prefetches.  We model our approach on the assumption that
- * the CPU does do speculative prefetches, which means we clean caches
- * before transfers and delay cache invalidation until transfer completion.
- *
- */
-static void __dma_page_cpu_to_dev(struct page *, unsigned long,
-		size_t, enum dma_data_direction);
-static void __dma_page_dev_to_cpu(struct page *, unsigned long,
-		size_t, enum dma_data_direction);
-
 /**
  * arm_dma_map_page - map a portion of a page for streaming DMA
  * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
@@ -492,7 +475,7 @@  static void *__alloc_remap_buffer(struct device *dev, size_t size, gfp_t gfp,
 	return ptr;
 }
 
-static void *__alloc_from_pool(size_t size, struct page **ret_page)
+void *arch_alloc_from_atomic_pool(size_t size, struct page **ret_page, gfp_t gfp)
 {
 	unsigned long val;
 	void *ptr = NULL;
@@ -513,14 +496,14 @@  static void *__alloc_from_pool(size_t size, struct page **ret_page)
 	return ptr;
 }
 
-static bool __in_atomic_pool(void *start, size_t size)
+bool arch_in_atomic_pool(void *start, size_t size)
 {
 	return addr_in_gen_pool(atomic_pool, (unsigned long)start, size);
 }
 
-static int __free_from_pool(void *start, size_t size)
+int arch_free_from_atomic_pool(void *start, size_t size)
 {
-	if (!__in_atomic_pool(start, size))
+	if (!arch_in_atomic_pool(start, size))
 		return 0;
 
 	gen_pool_free(atomic_pool, (unsigned long)start, size);
@@ -574,25 +557,21 @@  static void __free_from_contiguous(struct device *dev, struct page *page,
 	dma_release_from_contiguous(dev, page, size >> PAGE_SHIFT);
 }
 
-static inline pgprot_t __get_dma_pgprot(struct dma_attrs *attrs, pgprot_t prot)
-{
-	prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ?
-			    pgprot_writecombine(prot) :
-			    pgprot_dmacoherent(prot);
-	return prot;
-}
-
 #define nommu() 0
 
+#define __alloc_from_pool(size, ret, gfp) arch_alloc_from_atomic_pool(size, ret, gfp)
+#define __free_from_pool(addr, size) arch_free_from_atomic_pool(addr, size)
+#define __get_dma_pgprot(attrs, prot, coherent) arch_get_dma_pgprot(attrs, prot, coherent)
+
 #else	/* !CONFIG_MMU */
 
 #define nommu() 1
 
-#define __get_dma_pgprot(attrs, prot)				__pgprot(0)
+#define __get_dma_pgprot(attrs, prot, coherent)				__pgprot(0)
 #define __alloc_remap_buffer(dev, size, gfp, prot, ret, c, wv)	NULL
-#define __alloc_from_pool(size, ret_page)			NULL
+#define __alloc_from_pool(size, ret_page, gfp)			NULL
 #define __alloc_from_contiguous(dev, size, prot, ret, c, wv)	NULL
-#define __free_from_pool(cpu_addr, size)			0
+#define __free_from_atomic_pool(cpu_addr, size)			0
 #define __free_from_contiguous(dev, page, cpu_addr, size, wv)	do { } while (0)
 #define __dma_free_remap(cpu_addr, size)			do { } while (0)
 
@@ -657,7 +636,7 @@  static void *__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
 	else if (is_coherent)
 		addr = __alloc_simple_buffer(dev, size, gfp, &page);
 	else if (!gfpflags_allow_blocking(gfp))
-		addr = __alloc_from_pool(size, &page);
+		addr = __alloc_from_pool(size, &page, gfp);
 	else
 		addr = __alloc_remap_buffer(dev, size, gfp, prot, &page,
 					    caller, want_vaddr);
@@ -675,7 +654,7 @@  static void *__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
 void *arm_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle,
 		    gfp_t gfp, struct dma_attrs *attrs)
 {
-	pgprot_t prot = __get_dma_pgprot(attrs, PAGE_KERNEL);
+	pgprot_t prot = __get_dma_pgprot(attrs, PAGE_KERNEL, false);
 
 	return __dma_alloc(dev, size, handle, gfp, prot, false,
 			   attrs, __builtin_return_address(0));
@@ -728,7 +707,7 @@  int arm_dma_mmap(struct device *dev, struct vm_area_struct *vma,
 		 struct dma_attrs *attrs)
 {
 #ifdef CONFIG_MMU
-	vma->vm_page_prot = __get_dma_pgprot(attrs, vma->vm_page_prot);
+	vma->vm_page_prot = __get_dma_pgprot(attrs, vma->vm_page_prot, false);
 #endif	/* CONFIG_MMU */
 	return __arm_dma_mmap(dev, vma, cpu_addr, dma_addr, size, attrs);
 }
@@ -842,7 +821,7 @@  static void dma_cache_maint_page(struct page *page, unsigned long offset,
  * platforms with CONFIG_DMABOUNCE.
  * Use the driver DMA support - see dma-mapping.h (dma_sync_*)
  */
-static void __dma_page_cpu_to_dev(struct page *page, unsigned long off,
+void __dma_page_cpu_to_dev(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
 	phys_addr_t paddr;
@@ -858,7 +837,7 @@  static void __dma_page_cpu_to_dev(struct page *page, unsigned long off,
 	/* FIXME: non-speculating: flush on bidirectional mappings? */
 }
 
-static void __dma_page_dev_to_cpu(struct page *page, unsigned long off,
+void __dma_page_dev_to_cpu(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
 	phys_addr_t paddr = page_to_phys(page) + off;
@@ -1023,1098 +1002,6 @@  static int __init dma_debug_do_init(void)
 }
 fs_initcall(dma_debug_do_init);
 
-#ifdef CONFIG_ARM_DMA_USE_IOMMU
-
-/* IOMMU */
-
-static int extend_iommu_mapping(struct dma_iommu_mapping *mapping);
-
-static inline dma_addr_t __alloc_iova(struct dma_iommu_mapping *mapping,
-				      size_t size)
-{
-	unsigned int order = get_order(size);
-	unsigned int align = 0;
-	unsigned int count, start;
-	size_t mapping_size = mapping->bits << PAGE_SHIFT;
-	unsigned long flags;
-	dma_addr_t iova;
-	int i;
-
-	if (order > CONFIG_ARM_DMA_IOMMU_ALIGNMENT)
-		order = CONFIG_ARM_DMA_IOMMU_ALIGNMENT;
-
-	count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	align = (1 << order) - 1;
-
-	spin_lock_irqsave(&mapping->lock, flags);
-	for (i = 0; i < mapping->nr_bitmaps; i++) {
-		start = bitmap_find_next_zero_area(mapping->bitmaps[i],
-				mapping->bits, 0, count, align);
-
-		if (start > mapping->bits)
-			continue;
-
-		bitmap_set(mapping->bitmaps[i], start, count);
-		break;
-	}
-
-	/*
-	 * No unused range found. Try to extend the existing mapping
-	 * and perform a second attempt to reserve an IO virtual
-	 * address range of size bytes.
-	 */
-	if (i == mapping->nr_bitmaps) {
-		if (extend_iommu_mapping(mapping)) {
-			spin_unlock_irqrestore(&mapping->lock, flags);
-			return DMA_ERROR_CODE;
-		}
-
-		start = bitmap_find_next_zero_area(mapping->bitmaps[i],
-				mapping->bits, 0, count, align);
-
-		if (start > mapping->bits) {
-			spin_unlock_irqrestore(&mapping->lock, flags);
-			return DMA_ERROR_CODE;
-		}
-
-		bitmap_set(mapping->bitmaps[i], start, count);
-	}
-	spin_unlock_irqrestore(&mapping->lock, flags);
-
-	iova = mapping->base + (mapping_size * i);
-	iova += start << PAGE_SHIFT;
-
-	return iova;
-}
-
-static inline void __free_iova(struct dma_iommu_mapping *mapping,
-			       dma_addr_t addr, size_t size)
-{
-	unsigned int start, count;
-	size_t mapping_size = mapping->bits << PAGE_SHIFT;
-	unsigned long flags;
-	dma_addr_t bitmap_base;
-	u32 bitmap_index;
-
-	if (!size)
-		return;
-
-	bitmap_index = (u32) (addr - mapping->base) / (u32) mapping_size;
-	BUG_ON(addr < mapping->base || bitmap_index > mapping->extensions);
-
-	bitmap_base = mapping->base + mapping_size * bitmap_index;
-
-	start = (addr - bitmap_base) >>	PAGE_SHIFT;
-
-	if (addr + size > bitmap_base + mapping_size) {
-		/*
-		 * The address range to be freed reaches into the iova
-		 * range of the next bitmap. This should not happen as
-		 * we don't allow this in __alloc_iova (at the
-		 * moment).
-		 */
-		BUG();
-	} else
-		count = size >> PAGE_SHIFT;
-
-	spin_lock_irqsave(&mapping->lock, flags);
-	bitmap_clear(mapping->bitmaps[bitmap_index], start, count);
-	spin_unlock_irqrestore(&mapping->lock, flags);
-}
-
-static struct page **__iommu_alloc_buffer(struct device *dev, size_t size,
-					  gfp_t gfp, struct dma_attrs *attrs)
-{
-	struct page **pages;
-	int count = size >> PAGE_SHIFT;
-	int array_size = count * sizeof(struct page *);
-	int i = 0;
-
-	if (array_size <= PAGE_SIZE)
-		pages = kzalloc(array_size, GFP_KERNEL);
-	else
-		pages = vzalloc(array_size);
-	if (!pages)
-		return NULL;
-
-	if (dma_get_attr(DMA_ATTR_FORCE_CONTIGUOUS, attrs))
-	{
-		unsigned long order = get_order(size);
-		struct page *page;
-
-		page = dma_alloc_from_contiguous(dev, count, order);
-		if (!page)
-			goto error;
-
-		__dma_clear_buffer(page, size);
-
-		for (i = 0; i < count; i++)
-			pages[i] = page + i;
-
-		return pages;
-	}
-
-	/*
-	 * IOMMU can map any pages, so himem can also be used here
-	 */
-	gfp |= __GFP_NOWARN | __GFP_HIGHMEM;
-
-	while (count) {
-		int j, order;
-
-		for (order = __fls(count); order > 0; --order) {
-			/*
-			 * We do not want OOM killer to be invoked as long
-			 * as we can fall back to single pages, so we force
-			 * __GFP_NORETRY for orders higher than zero.
-			 */
-			pages[i] = alloc_pages(gfp | __GFP_NORETRY, order);
-			if (pages[i])
-				break;
-		}
-
-		if (!pages[i]) {
-			/*
-			 * Fall back to single page allocation.
-			 * Might invoke OOM killer as last resort.
-			 */
-			pages[i] = alloc_pages(gfp, 0);
-			if (!pages[i])
-				goto error;
-		}
-
-		if (order) {
-			split_page(pages[i], order);
-			j = 1 << order;
-			while (--j)
-				pages[i + j] = pages[i] + j;
-		}
-
-		__dma_clear_buffer(pages[i], PAGE_SIZE << order);
-		i += 1 << order;
-		count -= 1 << order;
-	}
-
-	return pages;
-error:
-	while (i--)
-		if (pages[i])
-			__free_pages(pages[i], 0);
-	kvfree(pages);
-	return NULL;
-}
-
-static int __iommu_free_buffer(struct device *dev, struct page **pages,
-			       size_t size, struct dma_attrs *attrs)
-{
-	int count = size >> PAGE_SHIFT;
-	int i;
-
-	if (dma_get_attr(DMA_ATTR_FORCE_CONTIGUOUS, attrs)) {
-		dma_release_from_contiguous(dev, pages[0], count);
-	} else {
-		for (i = 0; i < count; i++)
-			if (pages[i])
-				__free_pages(pages[i], 0);
-	}
-
-	kvfree(pages);
-	return 0;
-}
-
-/*
- * Create a CPU mapping for a specified pages
- */
-static void *
-__iommu_alloc_remap(struct page **pages, size_t size, gfp_t gfp, pgprot_t prot,
-		    const void *caller)
-{
-	return dma_common_pages_remap(pages, size,
-			VM_ARM_DMA_CONSISTENT | VM_USERMAP, prot, caller);
-}
-
-/*
- * Create a mapping in device IO address space for specified pages
- */
-static dma_addr_t
-__iommu_create_mapping(struct device *dev, struct page **pages, size_t size)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	dma_addr_t dma_addr, iova;
-	int i;
-
-	dma_addr = __alloc_iova(mapping, size);
-	if (dma_addr == DMA_ERROR_CODE)
-		return dma_addr;
-
-	iova = dma_addr;
-	for (i = 0; i < count; ) {
-		int ret;
-
-		unsigned int next_pfn = page_to_pfn(pages[i]) + 1;
-		phys_addr_t phys = page_to_phys(pages[i]);
-		unsigned int len, j;
-
-		for (j = i + 1; j < count; j++, next_pfn++)
-			if (page_to_pfn(pages[j]) != next_pfn)
-				break;
-
-		len = (j - i) << PAGE_SHIFT;
-		ret = iommu_map(mapping->domain, iova, phys, len,
-				IOMMU_READ|IOMMU_WRITE);
-		if (ret < 0)
-			goto fail;
-		iova += len;
-		i = j;
-	}
-	return dma_addr;
-fail:
-	iommu_unmap(mapping->domain, dma_addr, iova-dma_addr);
-	__free_iova(mapping, dma_addr, size);
-	return DMA_ERROR_CODE;
-}
-
-static int __iommu_remove_mapping(struct device *dev, dma_addr_t iova, size_t size)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-
-	/*
-	 * add optional in-page offset from iova to size and align
-	 * result to page size
-	 */
-	size = PAGE_ALIGN((iova & ~PAGE_MASK) + size);
-	iova &= PAGE_MASK;
-
-	iommu_unmap(mapping->domain, iova, size);
-	__free_iova(mapping, iova, size);
-	return 0;
-}
-
-static struct page **__atomic_get_pages(void *addr)
-{
-	struct page *page;
-	phys_addr_t phys;
-
-	phys = gen_pool_virt_to_phys(atomic_pool, (unsigned long)addr);
-	page = phys_to_page(phys);
-
-	return (struct page **)page;
-}
-
-static struct page **__iommu_get_pages(void *cpu_addr, struct dma_attrs *attrs)
-{
-	struct vm_struct *area;
-
-	if (__in_atomic_pool(cpu_addr, PAGE_SIZE))
-		return __atomic_get_pages(cpu_addr);
-
-	if (dma_get_attr(DMA_ATTR_NO_KERNEL_MAPPING, attrs))
-		return cpu_addr;
-
-	area = find_vm_area(cpu_addr);
-	if (area && (area->flags & VM_ARM_DMA_CONSISTENT))
-		return area->pages;
-	return NULL;
-}
-
-static void *__iommu_alloc_atomic(struct device *dev, size_t size,
-				  dma_addr_t *handle)
-{
-	struct page *page;
-	void *addr;
-
-	addr = __alloc_from_pool(size, &page);
-	if (!addr)
-		return NULL;
-
-	*handle = __iommu_create_mapping(dev, &page, size);
-	if (*handle == DMA_ERROR_CODE)
-		goto err_mapping;
-
-	return addr;
-
-err_mapping:
-	__free_from_pool(addr, size);
-	return NULL;
-}
-
-static void __iommu_free_atomic(struct device *dev, void *cpu_addr,
-				dma_addr_t handle, size_t size)
-{
-	__iommu_remove_mapping(dev, handle, size);
-	__free_from_pool(cpu_addr, size);
-}
-
-static void *arm_iommu_alloc_attrs(struct device *dev, size_t size,
-	    dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs)
-{
-	pgprot_t prot = __get_dma_pgprot(attrs, PAGE_KERNEL);
-	struct page **pages;
-	void *addr = NULL;
-
-	*handle = DMA_ERROR_CODE;
-	size = PAGE_ALIGN(size);
-
-	if (!gfpflags_allow_blocking(gfp))
-		return __iommu_alloc_atomic(dev, size, handle);
-
-	/*
-	 * Following is a work-around (a.k.a. hack) to prevent pages
-	 * with __GFP_COMP being passed to split_page() which cannot
-	 * handle them.  The real problem is that this flag probably
-	 * should be 0 on ARM as it is not supported on this
-	 * platform; see CONFIG_HUGETLBFS.
-	 */
-	gfp &= ~(__GFP_COMP);
-
-	pages = __iommu_alloc_buffer(dev, size, gfp, attrs);
-	if (!pages)
-		return NULL;
-
-	*handle = __iommu_create_mapping(dev, pages, size);
-	if (*handle == DMA_ERROR_CODE)
-		goto err_buffer;
-
-	if (dma_get_attr(DMA_ATTR_NO_KERNEL_MAPPING, attrs))
-		return pages;
-
-	addr = __iommu_alloc_remap(pages, size, gfp, prot,
-				   __builtin_return_address(0));
-	if (!addr)
-		goto err_mapping;
-
-	return addr;
-
-err_mapping:
-	__iommu_remove_mapping(dev, *handle, size);
-err_buffer:
-	__iommu_free_buffer(dev, pages, size, attrs);
-	return NULL;
-}
-
-static int arm_iommu_mmap_attrs(struct device *dev, struct vm_area_struct *vma,
-		    void *cpu_addr, dma_addr_t dma_addr, size_t size,
-		    struct dma_attrs *attrs)
-{
-	unsigned long uaddr = vma->vm_start;
-	unsigned long usize = vma->vm_end - vma->vm_start;
-	struct page **pages = __iommu_get_pages(cpu_addr, attrs);
-	unsigned long nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	unsigned long off = vma->vm_pgoff;
-
-	vma->vm_page_prot = __get_dma_pgprot(attrs, vma->vm_page_prot);
-
-	if (!pages)
-		return -ENXIO;
-
-	if (off >= nr_pages || (usize >> PAGE_SHIFT) > nr_pages - off)
-		return -ENXIO;
-
-	pages += off;
-
-	do {
-		int ret = vm_insert_page(vma, uaddr, *pages++);
-		if (ret) {
-			pr_err("Remapping memory failed: %d\n", ret);
-			return ret;
-		}
-		uaddr += PAGE_SIZE;
-		usize -= PAGE_SIZE;
-	} while (usize > 0);
-
-	return 0;
-}
-
-/*
- * free a page as defined by the above mapping.
- * Must not be called with IRQs disabled.
- */
-void arm_iommu_free_attrs(struct device *dev, size_t size, void *cpu_addr,
-			  dma_addr_t handle, struct dma_attrs *attrs)
-{
-	struct page **pages;
-	size = PAGE_ALIGN(size);
-
-	if (__in_atomic_pool(cpu_addr, size)) {
-		__iommu_free_atomic(dev, cpu_addr, handle, size);
-		return;
-	}
-
-	pages = __iommu_get_pages(cpu_addr, attrs);
-	if (!pages) {
-		WARN(1, "trying to free invalid coherent area: %p\n", cpu_addr);
-		return;
-	}
-
-	if (!dma_get_attr(DMA_ATTR_NO_KERNEL_MAPPING, attrs)) {
-		dma_common_free_remap(cpu_addr, size,
-			VM_ARM_DMA_CONSISTENT | VM_USERMAP);
-	}
-
-	__iommu_remove_mapping(dev, handle, size);
-	__iommu_free_buffer(dev, pages, size, attrs);
-}
-
-static int arm_iommu_get_sgtable(struct device *dev, struct sg_table *sgt,
-				 void *cpu_addr, dma_addr_t dma_addr,
-				 size_t size, struct dma_attrs *attrs)
-{
-	unsigned int count = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	struct page **pages = __iommu_get_pages(cpu_addr, attrs);
-
-	if (!pages)
-		return -ENXIO;
-
-	return sg_alloc_table_from_pages(sgt, pages, count, 0, size,
-					 GFP_KERNEL);
-}
-
-static int __dma_direction_to_prot(enum dma_data_direction dir)
-{
-	int prot;
-
-	switch (dir) {
-	case DMA_BIDIRECTIONAL:
-		prot = IOMMU_READ | IOMMU_WRITE;
-		break;
-	case DMA_TO_DEVICE:
-		prot = IOMMU_READ;
-		break;
-	case DMA_FROM_DEVICE:
-		prot = IOMMU_WRITE;
-		break;
-	default:
-		prot = 0;
-	}
-
-	return prot;
-}
-
-/*
- * Map a part of the scatter-gather list into contiguous io address space
- */
-static int __map_sg_chunk(struct device *dev, struct scatterlist *sg,
-			  size_t size, dma_addr_t *handle,
-			  enum dma_data_direction dir, struct dma_attrs *attrs,
-			  bool is_coherent)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t iova, iova_base;
-	int ret = 0;
-	unsigned int count;
-	struct scatterlist *s;
-	int prot;
-
-	size = PAGE_ALIGN(size);
-	*handle = DMA_ERROR_CODE;
-
-	iova_base = iova = __alloc_iova(mapping, size);
-	if (iova == DMA_ERROR_CODE)
-		return -ENOMEM;
-
-	for (count = 0, s = sg; count < (size >> PAGE_SHIFT); s = sg_next(s)) {
-		phys_addr_t phys = page_to_phys(sg_page(s));
-		unsigned int len = PAGE_ALIGN(s->offset + s->length);
-
-		if (!is_coherent &&
-			!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs))
-			__dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir);
-
-		prot = __dma_direction_to_prot(dir);
-
-		ret = iommu_map(mapping->domain, iova, phys, len, prot);
-		if (ret < 0)
-			goto fail;
-		count += len >> PAGE_SHIFT;
-		iova += len;
-	}
-	*handle = iova_base;
-
-	return 0;
-fail:
-	iommu_unmap(mapping->domain, iova_base, count * PAGE_SIZE);
-	__free_iova(mapping, iova_base, size);
-	return ret;
-}
-
-static int __iommu_map_sg(struct device *dev, struct scatterlist *sg, int nents,
-		     enum dma_data_direction dir, struct dma_attrs *attrs,
-		     bool is_coherent)
-{
-	struct scatterlist *s = sg, *dma = sg, *start = sg;
-	int i, count = 0;
-	unsigned int offset = s->offset;
-	unsigned int size = s->offset + s->length;
-	unsigned int max = dma_get_max_seg_size(dev);
-
-	for (i = 1; i < nents; i++) {
-		s = sg_next(s);
-
-		s->dma_address = DMA_ERROR_CODE;
-		s->dma_length = 0;
-
-		if (s->offset || (size & ~PAGE_MASK) || size + s->length > max) {
-			if (__map_sg_chunk(dev, start, size, &dma->dma_address,
-			    dir, attrs, is_coherent) < 0)
-				goto bad_mapping;
-
-			dma->dma_address += offset;
-			dma->dma_length = size - offset;
-
-			size = offset = s->offset;
-			start = s;
-			dma = sg_next(dma);
-			count += 1;
-		}
-		size += s->length;
-	}
-	if (__map_sg_chunk(dev, start, size, &dma->dma_address, dir, attrs,
-		is_coherent) < 0)
-		goto bad_mapping;
-
-	dma->dma_address += offset;
-	dma->dma_length = size - offset;
-
-	return count+1;
-
-bad_mapping:
-	for_each_sg(sg, s, count, i)
-		__iommu_remove_mapping(dev, sg_dma_address(s), sg_dma_len(s));
-	return 0;
-}
-
-/**
- * arm_coherent_iommu_map_sg - map a set of SG buffers for streaming mode DMA
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to map
- * @dir: DMA transfer direction
- *
- * Map a set of i/o coherent buffers described by scatterlist in streaming
- * mode for DMA. The scatter gather list elements are merged together (if
- * possible) and tagged with the appropriate dma address and length. They are
- * obtained via sg_dma_{address,length}.
- */
-int arm_coherent_iommu_map_sg(struct device *dev, struct scatterlist *sg,
-		int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
-{
-	return __iommu_map_sg(dev, sg, nents, dir, attrs, true);
-}
-
-/**
- * arm_iommu_map_sg - map a set of SG buffers for streaming mode DMA
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to map
- * @dir: DMA transfer direction
- *
- * Map a set of buffers described by scatterlist in streaming mode for DMA.
- * The scatter gather list elements are merged together (if possible) and
- * tagged with the appropriate dma address and length. They are obtained via
- * sg_dma_{address,length}.
- */
-int arm_iommu_map_sg(struct device *dev, struct scatterlist *sg,
-		int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
-{
-	return __iommu_map_sg(dev, sg, nents, dir, attrs, false);
-}
-
-static void __iommu_unmap_sg(struct device *dev, struct scatterlist *sg,
-		int nents, enum dma_data_direction dir, struct dma_attrs *attrs,
-		bool is_coherent)
-{
-	struct scatterlist *s;
-	int i;
-
-	for_each_sg(sg, s, nents, i) {
-		if (sg_dma_len(s))
-			__iommu_remove_mapping(dev, sg_dma_address(s),
-					       sg_dma_len(s));
-		if (!is_coherent &&
-		    !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs))
-			__dma_page_dev_to_cpu(sg_page(s), s->offset,
-					      s->length, dir);
-	}
-}
-
-/**
- * arm_coherent_iommu_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to unmap (same as was passed to dma_map_sg)
- * @dir: DMA transfer direction (same as was passed to dma_map_sg)
- *
- * Unmap a set of streaming mode DMA translations.  Again, CPU access
- * rules concerning calls here are the same as for dma_unmap_single().
- */
-void arm_coherent_iommu_unmap_sg(struct device *dev, struct scatterlist *sg,
-		int nents, enum dma_data_direction dir, struct dma_attrs *attrs)
-{
-	__iommu_unmap_sg(dev, sg, nents, dir, attrs, true);
-}
-
-/**
- * arm_iommu_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to unmap (same as was passed to dma_map_sg)
- * @dir: DMA transfer direction (same as was passed to dma_map_sg)
- *
- * Unmap a set of streaming mode DMA translations.  Again, CPU access
- * rules concerning calls here are the same as for dma_unmap_single().
- */
-void arm_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, int nents,
-			enum dma_data_direction dir, struct dma_attrs *attrs)
-{
-	__iommu_unmap_sg(dev, sg, nents, dir, attrs, false);
-}
-
-/**
- * arm_iommu_sync_sg_for_cpu
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to map (returned from dma_map_sg)
- * @dir: DMA transfer direction (same as was passed to dma_map_sg)
- */
-void arm_iommu_sync_sg_for_cpu(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)
-		__dma_page_dev_to_cpu(sg_page(s), s->offset, s->length, dir);
-
-}
-
-/**
- * arm_iommu_sync_sg_for_device
- * @dev: valid struct device pointer
- * @sg: list of buffers
- * @nents: number of buffers to map (returned from dma_map_sg)
- * @dir: DMA transfer direction (same as was passed to dma_map_sg)
- */
-void arm_iommu_sync_sg_for_device(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)
-		__dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir);
-}
-
-
-/**
- * arm_coherent_iommu_map_page
- * @dev: valid struct device pointer
- * @page: page that buffer resides in
- * @offset: offset into page for start of buffer
- * @size: size of buffer to map
- * @dir: DMA transfer direction
- *
- * Coherent IOMMU aware version of arm_dma_map_page()
- */
-static dma_addr_t arm_coherent_iommu_map_page(struct device *dev, struct page *page,
-	     unsigned long offset, size_t size, enum dma_data_direction dir,
-	     struct dma_attrs *attrs)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t dma_addr;
-	int ret, prot, len = PAGE_ALIGN(size + offset);
-
-	dma_addr = __alloc_iova(mapping, len);
-	if (dma_addr == DMA_ERROR_CODE)
-		return dma_addr;
-
-	prot = __dma_direction_to_prot(dir);
-
-	ret = iommu_map(mapping->domain, dma_addr, page_to_phys(page), len, prot);
-	if (ret < 0)
-		goto fail;
-
-	return dma_addr + offset;
-fail:
-	__free_iova(mapping, dma_addr, len);
-	return DMA_ERROR_CODE;
-}
-
-/**
- * arm_iommu_map_page
- * @dev: valid struct device pointer
- * @page: page that buffer resides in
- * @offset: offset into page for start of buffer
- * @size: size of buffer to map
- * @dir: DMA transfer direction
- *
- * IOMMU aware version of arm_dma_map_page()
- */
-static dma_addr_t arm_iommu_map_page(struct device *dev, struct page *page,
-	     unsigned long offset, size_t size, enum dma_data_direction dir,
-	     struct dma_attrs *attrs)
-{
-	if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs))
-		__dma_page_cpu_to_dev(page, offset, size, dir);
-
-	return arm_coherent_iommu_map_page(dev, page, offset, size, dir, attrs);
-}
-
-/**
- * arm_coherent_iommu_unmap_page
- * @dev: valid struct device pointer
- * @handle: DMA address of buffer
- * @size: size of buffer (same as passed to dma_map_page)
- * @dir: DMA transfer direction (same as passed to dma_map_page)
- *
- * Coherent IOMMU aware version of arm_dma_unmap_page()
- */
-static void arm_coherent_iommu_unmap_page(struct device *dev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		struct dma_attrs *attrs)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t iova = handle & PAGE_MASK;
-	int offset = handle & ~PAGE_MASK;
-	int len = PAGE_ALIGN(size + offset);
-
-	if (!iova)
-		return;
-
-	iommu_unmap(mapping->domain, iova, len);
-	__free_iova(mapping, iova, len);
-}
-
-/**
- * arm_iommu_unmap_page
- * @dev: valid struct device pointer
- * @handle: DMA address of buffer
- * @size: size of buffer (same as passed to dma_map_page)
- * @dir: DMA transfer direction (same as passed to dma_map_page)
- *
- * IOMMU aware version of arm_dma_unmap_page()
- */
-static void arm_iommu_unmap_page(struct device *dev, dma_addr_t handle,
-		size_t size, enum dma_data_direction dir,
-		struct dma_attrs *attrs)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t iova = handle & PAGE_MASK;
-	struct page *page = phys_to_page(iommu_iova_to_phys(mapping->domain, iova));
-	int offset = handle & ~PAGE_MASK;
-	int len = PAGE_ALIGN(size + offset);
-
-	if (!iova)
-		return;
-
-	if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs))
-		__dma_page_dev_to_cpu(page, offset, size, dir);
-
-	iommu_unmap(mapping->domain, iova, len);
-	__free_iova(mapping, iova, len);
-}
-
-static void arm_iommu_sync_single_for_cpu(struct device *dev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t iova = handle & PAGE_MASK;
-	struct page *page = phys_to_page(iommu_iova_to_phys(mapping->domain, iova));
-	unsigned int offset = handle & ~PAGE_MASK;
-
-	if (!iova)
-		return;
-
-	__dma_page_dev_to_cpu(page, offset, size, dir);
-}
-
-static void arm_iommu_sync_single_for_device(struct device *dev,
-		dma_addr_t handle, size_t size, enum dma_data_direction dir)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-	dma_addr_t iova = handle & PAGE_MASK;
-	struct page *page = phys_to_page(iommu_iova_to_phys(mapping->domain, iova));
-	unsigned int offset = handle & ~PAGE_MASK;
-
-	if (!iova)
-		return;
-
-	__dma_page_cpu_to_dev(page, offset, size, dir);
-}
-
-struct dma_map_ops iommu_ops = {
-	.alloc		= arm_iommu_alloc_attrs,
-	.free		= arm_iommu_free_attrs,
-	.mmap		= arm_iommu_mmap_attrs,
-	.get_sgtable	= arm_iommu_get_sgtable,
-
-	.map_page		= arm_iommu_map_page,
-	.unmap_page		= arm_iommu_unmap_page,
-	.sync_single_for_cpu	= arm_iommu_sync_single_for_cpu,
-	.sync_single_for_device	= arm_iommu_sync_single_for_device,
-
-	.map_sg			= arm_iommu_map_sg,
-	.unmap_sg		= arm_iommu_unmap_sg,
-	.sync_sg_for_cpu	= arm_iommu_sync_sg_for_cpu,
-	.sync_sg_for_device	= arm_iommu_sync_sg_for_device,
-
-	.set_dma_mask		= arm_dma_set_mask,
-};
-
-struct dma_map_ops iommu_coherent_ops = {
-	.alloc		= arm_iommu_alloc_attrs,
-	.free		= arm_iommu_free_attrs,
-	.mmap		= arm_iommu_mmap_attrs,
-	.get_sgtable	= arm_iommu_get_sgtable,
-
-	.map_page	= arm_coherent_iommu_map_page,
-	.unmap_page	= arm_coherent_iommu_unmap_page,
-
-	.map_sg		= arm_coherent_iommu_map_sg,
-	.unmap_sg	= arm_coherent_iommu_unmap_sg,
-
-	.set_dma_mask	= arm_dma_set_mask,
-};
-
-/**
- * arm_iommu_create_mapping
- * @bus: pointer to the bus holding the client device (for IOMMU calls)
- * @base: start address of the valid IO address space
- * @size: maximum size of the valid IO address space
- *
- * Creates a mapping structure which holds information about used/unused
- * IO address ranges, which is required to perform memory allocation and
- * mapping with IOMMU aware functions.
- *
- * The client device need to be attached to the mapping with
- * arm_iommu_attach_device function.
- */
-struct dma_iommu_mapping *
-arm_iommu_create_mapping(struct bus_type *bus, dma_addr_t base, u64 size)
-{
-	unsigned int bits = size >> PAGE_SHIFT;
-	unsigned int bitmap_size = BITS_TO_LONGS(bits) * sizeof(long);
-	struct dma_iommu_mapping *mapping;
-	int extensions = 1;
-	int err = -ENOMEM;
-
-	/* currently only 32-bit DMA address space is supported */
-	if (size > DMA_BIT_MASK(32) + 1)
-		return ERR_PTR(-ERANGE);
-
-	if (!bitmap_size)
-		return ERR_PTR(-EINVAL);
-
-	if (bitmap_size > PAGE_SIZE) {
-		extensions = bitmap_size / PAGE_SIZE;
-		bitmap_size = PAGE_SIZE;
-	}
-
-	mapping = kzalloc(sizeof(struct dma_iommu_mapping), GFP_KERNEL);
-	if (!mapping)
-		goto err;
-
-	mapping->bitmap_size = bitmap_size;
-	mapping->bitmaps = kzalloc(extensions * sizeof(unsigned long *),
-				GFP_KERNEL);
-	if (!mapping->bitmaps)
-		goto err2;
-
-	mapping->bitmaps[0] = kzalloc(bitmap_size, GFP_KERNEL);
-	if (!mapping->bitmaps[0])
-		goto err3;
-
-	mapping->nr_bitmaps = 1;
-	mapping->extensions = extensions;
-	mapping->base = base;
-	mapping->bits = BITS_PER_BYTE * bitmap_size;
-
-	spin_lock_init(&mapping->lock);
-
-	mapping->domain = iommu_domain_alloc(bus);
-	if (!mapping->domain)
-		goto err4;
-
-	kref_init(&mapping->kref);
-	return mapping;
-err4:
-	kfree(mapping->bitmaps[0]);
-err3:
-	kfree(mapping->bitmaps);
-err2:
-	kfree(mapping);
-err:
-	return ERR_PTR(err);
-}
-EXPORT_SYMBOL_GPL(arm_iommu_create_mapping);
-
-static void release_iommu_mapping(struct kref *kref)
-{
-	int i;
-	struct dma_iommu_mapping *mapping =
-		container_of(kref, struct dma_iommu_mapping, kref);
-
-	iommu_domain_free(mapping->domain);
-	for (i = 0; i < mapping->nr_bitmaps; i++)
-		kfree(mapping->bitmaps[i]);
-	kfree(mapping->bitmaps);
-	kfree(mapping);
-}
-
-static int extend_iommu_mapping(struct dma_iommu_mapping *mapping)
-{
-	int next_bitmap;
-
-	if (mapping->nr_bitmaps >= mapping->extensions)
-		return -EINVAL;
-
-	next_bitmap = mapping->nr_bitmaps;
-	mapping->bitmaps[next_bitmap] = kzalloc(mapping->bitmap_size,
-						GFP_ATOMIC);
-	if (!mapping->bitmaps[next_bitmap])
-		return -ENOMEM;
-
-	mapping->nr_bitmaps++;
-
-	return 0;
-}
-
-void arm_iommu_release_mapping(struct dma_iommu_mapping *mapping)
-{
-	if (mapping)
-		kref_put(&mapping->kref, release_iommu_mapping);
-}
-EXPORT_SYMBOL_GPL(arm_iommu_release_mapping);
-
-static int __arm_iommu_attach_device(struct device *dev,
-				     struct dma_iommu_mapping *mapping)
-{
-	int err;
-
-	err = iommu_attach_device(mapping->domain, dev);
-	if (err)
-		return err;
-
-	kref_get(&mapping->kref);
-	to_dma_iommu_mapping(dev) = mapping;
-
-	pr_debug("Attached IOMMU controller to %s device.\n", dev_name(dev));
-	return 0;
-}
-
-/**
- * arm_iommu_attach_device
- * @dev: valid struct device pointer
- * @mapping: io address space mapping structure (returned from
- *	arm_iommu_create_mapping)
- *
- * Attaches specified io address space mapping to the provided device.
- * This replaces the dma operations (dma_map_ops pointer) with the
- * IOMMU aware version.
- *
- * More than one client might be attached to the same io address space
- * mapping.
- */
-int arm_iommu_attach_device(struct device *dev,
-			    struct dma_iommu_mapping *mapping)
-{
-	int err;
-
-	err = __arm_iommu_attach_device(dev, mapping);
-	if (err)
-		return err;
-
-	set_dma_ops(dev, &iommu_ops);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(arm_iommu_attach_device);
-
-static void __arm_iommu_detach_device(struct device *dev)
-{
-	struct dma_iommu_mapping *mapping;
-
-	mapping = to_dma_iommu_mapping(dev);
-	if (!mapping) {
-		dev_warn(dev, "Not attached\n");
-		return;
-	}
-
-	iommu_detach_device(mapping->domain, dev);
-	kref_put(&mapping->kref, release_iommu_mapping);
-	to_dma_iommu_mapping(dev) = NULL;
-
-	pr_debug("Detached IOMMU controller from %s device.\n", dev_name(dev));
-}
-
-/**
- * arm_iommu_detach_device
- * @dev: valid struct device pointer
- *
- * Detaches the provided device from a previously attached map.
- * This voids the dma operations (dma_map_ops pointer)
- */
-void arm_iommu_detach_device(struct device *dev)
-{
-	__arm_iommu_detach_device(dev);
-	set_dma_ops(dev, NULL);
-}
-EXPORT_SYMBOL_GPL(arm_iommu_detach_device);
-
-static struct dma_map_ops *arm_get_iommu_dma_map_ops(bool coherent)
-{
-	return coherent ? &iommu_coherent_ops : &iommu_ops;
-}
-
-static bool arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				    struct iommu_ops *iommu)
-{
-	struct dma_iommu_mapping *mapping;
-
-	if (!iommu)
-		return false;
-
-	mapping = arm_iommu_create_mapping(dev->bus, dma_base, size);
-	if (IS_ERR(mapping)) {
-		pr_warn("Failed to create %llu-byte IOMMU mapping for device %s\n",
-				size, dev_name(dev));
-		return false;
-	}
-
-	if (__arm_iommu_attach_device(dev, mapping)) {
-		pr_warn("Failed to attached device %s to IOMMU_mapping\n",
-				dev_name(dev));
-		arm_iommu_release_mapping(mapping);
-		return false;
-	}
-
-	return true;
-}
-
-static void arm_teardown_iommu_dma_ops(struct device *dev)
-{
-	struct dma_iommu_mapping *mapping = to_dma_iommu_mapping(dev);
-
-	if (!mapping)
-		return;
-
-	__arm_iommu_detach_device(dev);
-	arm_iommu_release_mapping(mapping);
-}
-
-#else
-
-static bool arm_setup_iommu_dma_ops(struct device *dev, u64 dma_base, u64 size,
-				    struct iommu_ops *iommu)
-{
-	return false;
-}
-
-static void arm_teardown_iommu_dma_ops(struct device *dev) { }
-
-#define arm_get_iommu_dma_map_ops arm_get_dma_map_ops
-
-#endif	/* CONFIG_ARM_DMA_USE_IOMMU */
-
 static struct dma_map_ops *arm_get_dma_map_ops(bool coherent)
 {
 	return coherent ? &arm_coherent_dma_ops : &arm_dma_ops;
@@ -2123,18 +1010,13 @@  static struct dma_map_ops *arm_get_dma_map_ops(bool coherent)
 void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 			struct iommu_ops *iommu, bool coherent)
 {
-	struct dma_map_ops *dma_ops;
-
 	dev->archdata.dma_coherent = coherent;
-	if (arm_setup_iommu_dma_ops(dev, dma_base, size, iommu))
-		dma_ops = arm_get_iommu_dma_map_ops(coherent);
-	else
-		dma_ops = arm_get_dma_map_ops(coherent);
 
-	set_dma_ops(dev, dma_ops);
+	if (!common_iommu_setup_dma_ops(dev, dma_base, size, iommu))
+		arch_set_dma_ops(dev, arm_get_dma_map_ops(coherent));
 }
 
 void arch_teardown_dma_ops(struct device *dev)
 {
-	arm_teardown_iommu_dma_ops(dev);
+	common_iommu_teardown_dma_ops(dev);
 }
diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index 85739859dffc..7bdb5cf64ba3 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -1,5 +1,6 @@ 
 config DRM_ROCKCHIP
 	tristate "DRM Support for Rockchip"
+	depends on BROKEN
 	depends on DRM && ROCKCHIP_IOMMU
 	depends on RESET_CONTROLLER
 	select DRM_KMS_HELPER
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 1674de1cfed0..8a99210f1cbc 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -265,6 +265,7 @@  config EXYNOS_IOMMU_DEBUG
 
 config IPMMU_VMSA
 	bool "Renesas VMSA-compatible IPMMU"
+	depends on BROKEN
 	depends on ARM_LPAE
 	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select IOMMU_API
diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
index 8b89ebe16d94..fb8bb372e489 100644
--- a/drivers/media/platform/Kconfig
+++ b/drivers/media/platform/Kconfig
@@ -88,6 +88,7 @@  config VIDEO_OMAP3
 	depends on VIDEO_V4L2 && I2C && VIDEO_V4L2_SUBDEV_API && ARCH_OMAP3
 	depends on HAS_DMA && OF
 	depends on OMAP_IOMMU
+	depends on BROKEN
 	select ARM_DMA_USE_IOMMU
 	select VIDEOBUF2_DMA_CONTIG
 	select MFD_SYSCON