[PATCH v8 01/17] arm: make SWIOTLB available

On Thu, 10 Oct 2013, Will Deacon wrote:

On Wed, Oct 09, 2013 at 09:48:14PM +0100, Stefano Stabellini wrote:

quoted

IOMMU_HELPER is needed because SWIOTLB calls iommu_is_span_boundary,
provided by lib/iommu_helper.c.

Signed-off-by: Stefano Stabellini <redacted>
Reviewed-by: Konrad Rzeszutek Wilk <redacted>
CC: will.deacon at arm.com
CC: linux at arm.linux.org.uk

[...]

quoted

diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index 5b579b9..9aa5384 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h

@@ -10,6 +10,7 @@
 
 #include <asm-generic/dma-coherent.h>
 #include <asm/memory.h>
+#include <asm/cacheflush.h>
 
 #define DMA_ERROR_CODE	(~0)
 extern struct dma_map_ops arm_dma_ops;

@@ -86,6 +87,42 @@ static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
 }
 #endif
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr)
+{
+	unsigned int offset = paddr & ~PAGE_MASK;
+	return pfn_to_dma(dev, paddr >> PAGE_SHIFT) + offset;

__phys_to_pfn instead of the explicit shift?

quoted

+}
+
+static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
+{
+	unsigned int offset = dev_addr & ~PAGE_MASK;
+	return (dma_to_pfn(dev, dev_addr) << PAGE_SHIFT) + offset;i

then __pfn_to_phys here.

good point

quoted

+}
+
+static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
+{
+	u64 limit, mask;
+	
+	if (dev->dma_mask)
+		mask = *dev->dma_mask;
+	else 
+		mask = dev->coherent_dma_mask;
+
+	if (mask == 0)
+		return 0;
+
+	limit = (mask + 1) & ~mask;

This looks like homebrew alignment to me. Can you use __ALIGN_KERNEL_MASK or
one of thoese guys from kernel.h?

It was suggested by Russell here:

http://marc.info/?l=linux-arm-kernel&m=137535542901235&w=2

I don't think that __ALIGN_KERNEL_MASK would be useful in this case.
I could add a comment to explain the purpose of the check though.