diff options
| author | Manoj Chourasia <mchourasia@nvidia.com> | 2012-06-04 17:25:43 +0530 |
|---|---|---|
| committer | Rohan Somvanshi <rsomvanshi@nvidia.com> | 2012-06-13 02:48:30 -0700 |
| commit | 017d8bd2c4bd03afe04721476dd26388a4bfe7f6 (patch) | |
| tree | 7b09ed282288a302c0a9f43cd43115bc03989ff6 /arch/arm/mm/Kconfig | |
| parent | 6a42ac50db6ff2832ff616f586fe7217b885df14 (diff) | |
Avoid aliasing mappings in DMA coherent allocator
Avoid multiple mappings with DMA coherent/writecombine allocator by pre-
allocating the mappings, and removing that memory from the system memory
mapping. (See previous discussions on linux-arm-kernel as to why this
is bad.)
NB1: By default, we preallocate 2MB for DMA coherent, and 2MB for write
combine memory, rather than 1MB for each in case 1MB is not sufficient
for existing platform usage. Platforms have the option of shrinking
this down to 1MB DMA / 1MB WC (or even 2MB DMA / 0MB WC) if they so wish.
The DMA memory must be a multiple of 1MB, the write combine memory must
also be a multiple of 1MB, and the two together must be a multiple of
2MB.
NB2: On ARMv6/7 where we use 'normal uncacheable' memory for both DMA
and WC, the two pools are combined into one, as was the case with the
previous implementation.
The down side to this change is that the memory is permanently set aside
for DMA purposes, but I believe that to be unavoidable if we are to
avoid the possibility of the cache getting in the way on VIPT CPUs.
This removes the last known offender (at this time) from the kernel.
Given that DMA memory is fully coherent by this patch, cache
invalidation/clean is not required and so, we skip cache related
activities for the memory managed by the DMA layer. The bus
address -> virtual address conversion normally used in the calling
path and the fact that we remove kernel static mapping corresponding
to the DMA buffers leads to exceptions otherwise.
bug 876019
bug 965047
bug 987589
Change-Id: I72beb386605aafe1a301494a95a67d094ea6b2e4
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Manoj Chourasia <mchourasia@nvidia.com>
Reviewed-on: http://git-master/r/106212
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-by: Krishna Reddy <vdumpa@nvidia.com>
Reviewed-by: Venkat Moganty <vmoganty@nvidia.com>
Diffstat (limited to 'arch/arm/mm/Kconfig')
| -rw-r--r-- | arch/arm/mm/Kconfig | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index aaea6d487bae..e23ad5ebc330 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -885,6 +885,17 @@ config ARM_DMA_MEM_BUFFERABLE You are recommended say 'Y' here and debug any affected drivers. +config NON_ALIASED_COHERENT_MEM + bool "Avoid aliasing mappings in DMA coherent allocator" if CPU_V7 + help + Avoid multiple mappings with DMA coherent/writecombine allocator by + pre-allocating the mappings, and removing that memory from the + system memory mapping. + + This will make sure DMA memory is fully coherent, so cache + dma_sync APIs will not be necessary to maintain cache cohereny + for DMA coherent region. + config ARCH_HAS_BARRIERS bool help |