* [PATCH] ARM: dma-mapping: Add maximum alignment order for dma iommu buffers
@ 2013-02-06 2:06 Seung-Woo Kim
2013-02-06 4:21 ` [RESEND][PATCH] " Seung-Woo Kim
0 siblings, 1 reply; 2+ messages in thread
From: Seung-Woo Kim @ 2013-02-06 2:06 UTC (permalink / raw)
To: linux-arm-kernel
Alignment order for a dma iommu buffer is set by buffer size. For
large buffer, it is a waste of iommu address space. So configurable
parameter to limit maximum alignment order can reduce the waste.
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Kyungmin.park <kyungmin.park@samsung.com>
---
arch/arm/Kconfig | 21 +++++++++++++++++++++
arch/arm/mm/dma-mapping.c | 3 +++
2 files changed, 24 insertions(+), 0 deletions(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 67874b8..4e89112 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -75,6 +75,27 @@ config ARM_DMA_USE_IOMMU
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
+
config HAVE_PWM
bool
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 076c26d..4cce854 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -1002,6 +1002,9 @@ static inline dma_addr_t __alloc_iova(struct dma_iommu_mapping *mapping,
unsigned int count, start;
unsigned long flags;
+ if (order > ARM_DMA_IOMMU_ALIGNMENT)
+ order = ARM_DMA_IOMMU_ALIGNMENT;
+
count = ((PAGE_ALIGN(size) >> PAGE_SHIFT) +
(1 << mapping->order) - 1) >> mapping->order;
--
1.7.4.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* [RESEND][PATCH] ARM: dma-mapping: Add maximum alignment order for dma iommu buffers
2013-02-06 2:06 [PATCH] ARM: dma-mapping: Add maximum alignment order for dma iommu buffers Seung-Woo Kim
@ 2013-02-06 4:21 ` Seung-Woo Kim
0 siblings, 0 replies; 2+ messages in thread
From: Seung-Woo Kim @ 2013-02-06 4:21 UTC (permalink / raw)
To: linux-arm-kernel
Alignment order for a dma iommu buffer is set by buffer size. For
large buffer, it is a waste of iommu address space. So configurable
parameter to limit maximum alignment order can reduce the waste.
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Signed-off-by: Kyungmin.park <kyungmin.park@samsung.com>
---
There was stupid misses in previous patch, so I resend fixed version.
arch/arm/Kconfig | 21 +++++++++++++++++++++
arch/arm/mm/dma-mapping.c | 3 +++
2 files changed, 24 insertions(+), 0 deletions(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 67874b8..4e89112 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -75,6 +75,27 @@ config ARM_DMA_USE_IOMMU
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
+
config HAVE_PWM
bool
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 076c26d..4cce854 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -1002,6 +1002,9 @@ static inline dma_addr_t __alloc_iova(struct dma_iommu_mapping *mapping,
unsigned int count, start;
unsigned long flags;
+ if (order > CONFIG_ARM_DMA_IOMMU_ALIGNMENT)
+ order = CONFIG_ARM_DMA_IOMMU_ALIGNMENT;
+
count = ((PAGE_ALIGN(size) >> PAGE_SHIFT) +
(1 << mapping->order) - 1) >> mapping->order;
--
1.7.4.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
end of thread, other threads:[~2013-02-06 4:21 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-02-06 2:06 [PATCH] ARM: dma-mapping: Add maximum alignment order for dma iommu buffers Seung-Woo Kim
2013-02-06 4:21 ` [RESEND][PATCH] " Seung-Woo Kim
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.