Re: [RFC v2 2/4] iommu/arm-smmu-v3: Add tlbi_on_map option

From: Auger Eric
Date: Thu Aug 17 2017 - 13:47:51 EST


Hi Will,

On 17/08/2017 18:34, Will Deacon wrote:
> On Fri, Aug 11, 2017 at 03:45:28PM +0200, Eric Auger wrote:
>> When running a virtual SMMU on a guest we sometimes need to trap
>> all changes to the translation structures. This is especially useful
>> to integrate with VFIO. This patch adds a new option that forces
>> the IO_PGTABLE_QUIRK_TLBI_ON_MAP to be applied on LPAE page tables.
>>
>> TLBI commands then can be trapped.
>>
>> Signed-off-by: Eric Auger <eric.auger@xxxxxxxxxx>
>>
>> ---
>> v1 -> v2:
>> - rebase on v4.13-rc2
>> ---
>> Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt | 4 ++++
>> drivers/iommu/arm-smmu-v3.c | 5 +++++
>> 2 files changed, 9 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt b/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt
>> index c9abbf3..ebb85e9 100644
>> --- a/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt
>> +++ b/Documentation/devicetree/bindings/iommu/arm,smmu-v3.txt
>> @@ -52,6 +52,10 @@ the PCIe specification.
>> devicetree/bindings/interrupt-controller/msi.txt
>> for a description of the msi-parent property.
>>
>> +- tlbi-on-map : invalidate caches whenever there is an update of
>> + any remapping structure (updates to not-present or
>> + present entries).
>> +
>
> My position on this hasn't changed, so NAK for this patch. If you want to
> emulate something outside of the SMMUv3 architecture, please do so, but
> don't pretend that it's an SMMUv3.
OK understood. I wanted to go down the road and enable DPDK use case
using the same solution as Intel.

So to me the approach is not adapted to SMMUv3 because
- SMMUv3 does not expose anything similar to the Intel Caching Mode
(when set, indicates the OS needs to invalidate TLB on map)
- SMMUv3 does not expose any IOVA range TLB invalidation command.

Do you agree with this conclusion? ACPI enablement was not a showstopper
I think.

I will see with Peter and other potential users in the community whether
it is worth to pursue the efforts on upstreaming the QEMU vSMMUv3
device, considering the VFIO/VHOST integration is made impossible.

Thanks

Eric

>
> Will
>