From: David Stevens <[email protected]>
Fall back to domain selective flush if the target address is not aligned
to the mask being used for invalidation. This is necessary because page
selective invalidation masks out the lower order bits of the target
address based on the mask value, so if a non-aligned address is targeted
for psi, then mappings at the end of [pfn, pfn+pages) may not properly
be flushed from the iotlb.
This is not normally an issue because iova.c always allocates iovas that
are aligned to their size. However, iovas which come from other sources
(e.g. userspace via VFIO) may not be aligned.
Signed-off-by: David Stevens <[email protected]>
---
drivers/iommu/intel/iommu.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index 5b196cfe9ed2..c122686e0a5c 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -1735,7 +1735,8 @@ static void iommu_flush_iotlb_psi(struct intel_iommu *iommu,
* and the base address is naturally aligned to the size.
*/
if (!cap_pgsel_inv(iommu->cap) ||
- mask > cap_max_amask_val(iommu->cap))
+ mask > cap_max_amask_val(iommu->cap) ||
+ unlikely(((1 << mask) - 1) & pfn))
iommu->flush.flush_iotlb(iommu, did, 0, 0,
DMA_TLB_DSI_FLUSH);
else
--
2.35.1.723.g4982287a31-goog
> -----Original Message-----
> From: iommu <[email protected]> On Behalf Of
> David Stevens
> Sent: Wednesday, March 16, 2022 1:07 PM
> To: Lu Baolu <[email protected]>
> Cc: [email protected]; [email protected]; David
> Stevens <[email protected]>
> Subject: [PATCH] iommu/vt-d: check alignment before using psi
>
> From: David Stevens <[email protected]>
>
> Fall back to domain selective flush if the target address is not aligned to the
> mask being used for invalidation. This is necessary because page selective
> invalidation masks out the lower order bits of the target address based on
> the mask value, so if a non-aligned address is targeted for psi, then mappings
> at the end of [pfn, pfn+pages) may not properly be flushed from the iotlb.
>
> This is not normally an issue because iova.c always allocates iovas that are
> aligned to their size. However, iovas which come from other sources (e.g.
> userspace via VFIO) may not be aligned.
Indeed, the dma_map/unmap always use alloca_iova(..., true) to allocate iova, which means size_aligned. But the iova used by userspace might not be size aligned. And it seems we cannot expect userspace can always use size_aligned iova base address.
BR,
Tina
>
> Signed-off-by: David Stevens <[email protected]>
> ---
> drivers/iommu/intel/iommu.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
> index 5b196cfe9ed2..c122686e0a5c 100644
> --- a/drivers/iommu/intel/iommu.c
> +++ b/drivers/iommu/intel/iommu.c
> @@ -1735,7 +1735,8 @@ static void iommu_flush_iotlb_psi(struct
> intel_iommu *iommu,
> * and the base address is naturally aligned to the size.
> */
> if (!cap_pgsel_inv(iommu->cap) ||
> - mask > cap_max_amask_val(iommu->cap))
> + mask > cap_max_amask_val(iommu->cap) ||
> + unlikely(((1 << mask) - 1) & pfn))
> iommu->flush.flush_iotlb(iommu, did, 0, 0,
>
> DMA_TLB_DSI_FLUSH);
> else
> --
> 2.35.1.723.g4982287a31-goog
>
> _______________________________________________
> iommu mailing list
> [email protected]
> https://lists.linuxfoundation.org/mailman/listinfo/iommu
> From: David Stevens <[email protected]>
> Sent: Wednesday, March 16, 2022 1:07 PM
>
> From: David Stevens <[email protected]>
>
> Fall back to domain selective flush if the target address is not aligned
> to the mask being used for invalidation. This is necessary because page
using domain selective flush is a bit conservative. What about trying
mask+1 first?
> selective invalidation masks out the lower order bits of the target
> address based on the mask value, so if a non-aligned address is targeted
> for psi, then mappings at the end of [pfn, pfn+pages) may not properly
> be flushed from the iotlb.
>
> This is not normally an issue because iova.c always allocates iovas that
> are aligned to their size. However, iovas which come from other sources
> (e.g. userspace via VFIO) may not be aligned.
>
> Signed-off-by: David Stevens <[email protected]>
> ---
> drivers/iommu/intel/iommu.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
> index 5b196cfe9ed2..c122686e0a5c 100644
> --- a/drivers/iommu/intel/iommu.c
> +++ b/drivers/iommu/intel/iommu.c
> @@ -1735,7 +1735,8 @@ static void iommu_flush_iotlb_psi(struct
> intel_iommu *iommu,
> * and the base address is naturally aligned to the size.
> */
> if (!cap_pgsel_inv(iommu->cap) ||
> - mask > cap_max_amask_val(iommu->cap))
> + mask > cap_max_amask_val(iommu->cap) ||
> + unlikely(((1 << mask) - 1) & pfn))
> iommu->flush.flush_iotlb(iommu, did, 0, 0,
>
> DMA_TLB_DSI_FLUSH);
> else
> --
> 2.35.1.723.g4982287a31-goog