[PATCH v7 00/22] Generic DT bindings for PCI IOMMUs and ARM SMMU
Auger Eric
eric.auger at redhat.com
Tue Sep 13 05:57:59 PDT 2016
Hi Robin,
On 13/09/2016 14:40, Robin Murphy wrote:
> Hi Eric,
>
> On 13/09/16 13:14, Auger Eric wrote:
>> Hi Robin
>>
>> On 12/09/2016 18:13, Robin Murphy wrote:
>>> Hi all,
>>>
>>> To any more confusing fixups and crazily numbered extra patches, here's
>>> a quick v7 with everything rebased into the right order. The significant
>>> change this time is to implement iommu_fwspec properly from the start,
>>> which ends up being far simpler and more robust than faffing about
>>> introducing it somewhere 'less intrusive' to move toward core code later.
>>>
>>> New branch in the logical place:
>>>
>>> git://linux-arm.org/linux-rm iommu/generic-v7
>> I just tested your branch on AMD overdrive *without* updating the device
>> tree description according to the new syntax and I get a kernel oops.
>> See logs attached. Continuing my investigations ...
>
> Looking at that backtrace, it seems the offending commit is actually in
> Will's devel branch _underneath_ this series; what's blowing up there is
> the short-descriptor io-pgtable selftests, which you should be able to
> reproduce on anything back to 4.6-rc1 with
> CONFIG_IOMMU_IO_PGTABLE_ARMV7S_SELFTEST=y.
I confirm that when disabling the option, I don't get the oops anymore.
Thanks!
Eric
>
> The short-descriptor code is never going to work on Seattle due to the
> lack of 32-bit addressable memory - in normal use it would fail
> gracefully because it couldn't allocate anything, but since the
> selftests bypass the DMA API and corresponding checks, you end up with
> nastiness happening via truncated addresses. A while back I did start
> looking into generalising the selftests to remove all the "if
> (!selftest_running)" special-casing; might be time to pick that up again.
>
> Robin.
>
>>
>> Best Regards
>>
>> Eric
>>>
>>> Robin.
>>>
>>> Mark Rutland (1):
>>> Docs: dt: add PCI IOMMU map bindings
>>>
>>> Robin Murphy (21):
>>> of/irq: Break out msi-map lookup (again)
>>> iommu/of: Handle iommu-map property for PCI
>>> iommu: Introduce iommu_fwspec
>>> Docs: dt: document ARM SMMUv3 generic binding usage
>>> iommu/arm-smmu: Fall back to global bypass
>>> iommu/arm-smmu: Implement of_xlate() for SMMUv3
>>> iommu/arm-smmu: Support non-PCI devices with SMMUv3
>>> iommu/arm-smmu: Set PRIVCFG in stage 1 STEs
>>> iommu/arm-smmu: Handle stream IDs more dynamically
>>> iommu/arm-smmu: Consolidate stream map entry state
>>> iommu/arm-smmu: Keep track of S2CR state
>>> iommu/arm-smmu: Refactor mmu-masters handling
>>> iommu/arm-smmu: Streamline SMMU data lookups
>>> iommu/arm-smmu: Add a stream map entry iterator
>>> iommu/arm-smmu: Intelligent SMR allocation
>>> iommu/arm-smmu: Convert to iommu_fwspec
>>> Docs: dt: document ARM SMMU generic binding usage
>>> iommu/arm-smmu: Wire up generic configuration support
>>> iommu/arm-smmu: Set domain geometry
>>> iommu/dma: Add support for mapping MSIs
>>> iommu/dma: Avoid PCI host bridge windows
>>>
>>> .../devicetree/bindings/iommu/arm,smmu-v3.txt | 8 +-
>>> .../devicetree/bindings/iommu/arm,smmu.txt | 63 +-
>>> .../devicetree/bindings/pci/pci-iommu.txt | 171 ++++
>>> arch/arm64/mm/dma-mapping.c | 2 +-
>>> drivers/gpu/drm/exynos/exynos_drm_iommu.h | 2 +-
>>> drivers/iommu/Kconfig | 2 +-
>>> drivers/iommu/arm-smmu-v3.c | 386 +++++----
>>> drivers/iommu/arm-smmu.c | 962 ++++++++++-----------
>>> drivers/iommu/dma-iommu.c | 161 +++-
>>> drivers/iommu/iommu.c | 56 ++
>>> drivers/iommu/of_iommu.c | 52 +-
>>> drivers/irqchip/irq-gic-v2m.c | 3 +
>>> drivers/irqchip/irq-gic-v3-its.c | 3 +
>>> drivers/of/irq.c | 78 +-
>>> drivers/of/of_pci.c | 102 +++
>>> include/linux/device.h | 3 +
>>> include/linux/dma-iommu.h | 12 +-
>>> include/linux/iommu.h | 38 +
>>> include/linux/of_pci.h | 10 +
>>> 19 files changed, 1323 insertions(+), 791 deletions(-)
>>> create mode 100644 Documentation/devicetree/bindings/pci/pci-iommu.txt
>>>
>>
>>
>>
>
More information about the linux-arm-kernel
mailing list