Consolidating the flush queue logic also meant that the "iommu.strict"
option started taking effect on x86 as well. Make sure we document that.
Fixes: a250c23f15c2 ("iommu: remove DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE")
Signed-off-by: Robin Murphy <[email protected]>
---
Documentation/admin-guide/kernel-parameters.txt | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index cb89dbdedc46..20a32de990ed 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -1987,7 +1987,7 @@
forcing Dual Address Cycle for PCI cards supporting
greater than 32-bit addressing.
- iommu.strict= [ARM64] Configure TLB invalidation behaviour
+ iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
Format: { "0" | "1" }
0 - Lazy mode.
Request that DMA unmap operations use deferred
@@ -1998,6 +1998,10 @@
1 - Strict mode (default).
DMA unmap operations invalidate IOMMU hardware TLBs
synchronously.
+ Note: on x86, the default behaviour depends on the
+ equivalent driver-specific parameters, but a strict
+ mode explicitly specified by either method takes
+ precedence.
iommu.passthrough=
[ARM64, X86] Configure DMA to bypass the IOMMU by default.
--
2.25.1
On 14/06/2021 15:57, Robin Murphy wrote:
> Consolidating the flush queue logic also meant that the "iommu.strict"
> option started taking effect on x86 as well. Make sure we document that.
>
> Fixes: a250c23f15c2 ("iommu: remove DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE")
> Signed-off-by: Robin Murphy <[email protected]>
I assumed that all bets were off when we use a cmdline param which is
not "officially" supported.
Anyway, FWIW:
Reviewed-by: John Garry <[email protected]>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index cb89dbdedc46..20a32de990ed 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -1987,7 +1987,7 @@
> forcing Dual Address Cycle for PCI cards supporting
> greater than 32-bit addressing.
>
> - iommu.strict= [ARM64] Configure TLB invalidation behaviour
> + iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
> Format: { "0" | "1" }
> 0 - Lazy mode.
> Request that DMA unmap operations use deferred
On 6/14/21 10:57 PM, Robin Murphy wrote:
> Consolidating the flush queue logic also meant that the "iommu.strict"
> option started taking effect on x86 as well. Make sure we document that.
>
> Fixes: a250c23f15c2 ("iommu: remove DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE")
> Signed-off-by: Robin Murphy <[email protected]>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index cb89dbdedc46..20a32de990ed 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -1987,7 +1987,7 @@
> forcing Dual Address Cycle for PCI cards supporting
> greater than 32-bit addressing.
>
> - iommu.strict= [ARM64] Configure TLB invalidation behaviour
> + iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour
> Format: { "0" | "1" }
> 0 - Lazy mode.
> Request that DMA unmap operations use deferred
> @@ -1998,6 +1998,10 @@
> 1 - Strict mode (default).
> DMA unmap operations invalidate IOMMU hardware TLBs
> synchronously.
> + Note: on x86, the default behaviour depends on the
> + equivalent driver-specific parameters, but a strict
> + mode explicitly specified by either method takes
> + precedence.
>
> iommu.passthrough=
> [ARM64, X86] Configure DMA to bypass the IOMMU by default.
>
Reviewed-by: Lu Baolu <[email protected]>
Best regards,
baolu
On Mon, Jun 14, 2021 at 03:57:26PM +0100, Robin Murphy wrote:
> Consolidating the flush queue logic also meant that the "iommu.strict"
> option started taking effect on x86 as well. Make sure we document that.
>
> Fixes: a250c23f15c2 ("iommu: remove DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE")
> Signed-off-by: Robin Murphy <[email protected]>
> ---
> Documentation/admin-guide/kernel-parameters.txt | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
Applied, thanks.