[PATCH 0/9][RFC] stackable dma_ops for x86

[Joerg Roedel <joerg.roedel@amd.com>]

Hi,

this patch series implements stackable dma_ops on x86. This is useful to
be able to fall back to a different dma_ops implementation if one can
not handle a particular device (as necessary for example with
paravirtualized device passthrough or if a hardware IOMMU only handles a
subset of available devices).
The patch series is split to change the different iommu implementations
in different patches. This should make it easier for the specific
maintainers to review the part that changes their code.


Thanks for review and comments,

Joerg

diffstat:

 arch/x86/kernel/amd_iommu.c      |    5 ++-
 arch/x86/kernel/pci-calgary_64.c |   21 ++++++------
 arch/x86/kernel/pci-dma.c        |   63 +++++++++++++++++++++++++++++++++++++-
 arch/x86/kernel/pci-gart_64.c    |    9 +++++
 arch/x86/kernel/pci-nommu.c      |   12 +++++++
 arch/x86/kernel/pci-swiotlb_64.c |   16 +++++++++-
 drivers/pci/intel-iommu.c        |    8 +++++
 include/asm-x86/device.h         |    6 ++--
 include/asm-x86/dma-mapping.h    |   43 +++++++++++++++++++++----
 include/asm-x86/swiotlb.h        |    1 +
 10 files changed, 160 insertions(+), 24 deletions(-)