[PATCH v8 0/3] iommu/arm-smmu-v3: Tegra264 invalidation workaround

[PATCH v8 0/3] iommu/arm-smmu-v3: Tegra264 invalidation workaround

[Ashish Mhetre]

Nvidia Tegra264 SMMUs are affected by an erratum where a TLB entry can
survive an invalidation that races with concurrent traffic targeting
the same entry. The hardware-recommended software workaround is to
issue every CFGI/TLBI command (each followed by CMD_SYNC) twice.
The second issue must execute only after the first issue's CMD_SYNC
has completed, giving the sequence:

    TLBI/CFGI ... CMD_SYNC TLBI/CFGI ... CMD_SYNC

ATC_INV is not affected and must not be doubled.

The erratum is not flagged by any SMMUv3 IDR/IIDR register, so it
cannot be detected from hardware ID. Tegra264 is device-tree-only
(no ACPI/IORT support), so detection is purely by compatible string.

This series is structured as a small refactor + infrastructure + enable
sequence so that each step is reviewable in isolation:

 1/3 Pure refactor (no functional change): lift the existing
     force-sync conditions out of arm_smmu_cmdq_batch_add_cmd_p()
     into a new arm_smmu_cmdq_batch_force_sync() helper, so that
     adding another condition (in patch 2) is a one-line addition.
     Authored by Nicolin Chen.

 2/3 Add the workaround infrastructure without enabling it. Defines
     the file-local arm_smmu_erratum_repeat_tlbi_cfgi_key static key
     with an inline erratum description, the shared
     arm_smmu_erratum_cmd_needs_repeating() predicate, the
     arm_smmu_cmdq_issue_cmdlist() wrapper that re-issues matching
     cmdlists for host-internal callers, and the batch-helper
     force-sync condition. The iommufd user-invalidation path keeps
     issuing user commands exactly once via the raw
     __arm_smmu_cmdq_issue_cmdlist().

 3/3 Enable the workaround for the existing "nvidia,tegra264-smmu"
     compatible, document the erratum in silicon-errata.rst, and
     report it to user space via a new
     IOMMU_HW_INFO_ARM_SMMUV3_ERRATA_REPEAT_TLBI_CFGI hw_info flag so
     a VMM/guest can apply the workaround to its own invalidations.

The series applies cleanly on linux-next/master (base-commit below).

Changes since v7:
 - Rework the nesting/iommufd handling to report the erratum to user
   space and stop repeating user-issued invalidations in the host.
 - 2/3: drop arm_vsmmu_can_batch_cmd() and the iommufd batching split
   it added. Expose __arm_smmu_cmdq_issue_cmdlist() and make
   arm_vsmmu_cache_invalidate() call it directly.
 - 3/3: add the uapi flag + enum, report it from arm_smmu_hw_info()
   via a new arm_smmu_erratum_repeat_tlbi_cfgi() accessor.
 - Carry Jason Gunthorpe's Reviewed-by on 1/3 and 2/3, and Nicolin
   Chen's Reviewed-by on 2/3. Drop earlier tags from 3/3 since it
   gained the uapi/hw_info change; re-review appreciated.

Changes since v6:
 - Add #include <linux/jump_label.h> now that the static key is
   defined in arm-smmu-v3.c.
 - Drop the unused smmu parameter from arm_vsmmu_can_batch_cmd().
 - Expand the arm_smmu_cmdq_batch_force_sync() comment to note that
   batches never mix CFGI/TLBI with other commands, so checking
   cmds[0] alone is enough.
 - Note in 3/3 that a guest kernel enabling CMDQV on Tegra264 must
   also apply this workaround, since guest-level VCMDQs issue
   commands directly to the hardware.
 - Carry Reviewed-by: Nicolin Chen on 2/3 and 3/3.

Changes since v5:
 - Move arm_smmu_erratum_cmd_needs_repeating() into arm-smmu-v3.c
   and leave a declaration-only stub in arm-smmu-v3.h. Make
   arm_smmu_erratum_repeat_tlbi_cfgi_key file-local static.
 - Add an inline erratum/workaround description at the static key,
   referenced from arm_smmu_cmdq_batch_force_sync().
 - Fix (rather than drop) the misleading !n comment above
   arm_smmu_cmdq_issue_cmdlist(); keep the defensive !n guard.
 - Remove the unused smmu parameter from the predicate.
 - Tweak 2/3 commit-message wording ("commit" vs "patch").

Changes since v4:
 - Drop ARM_SMMU_OPT_REPEAT_TLBI_CFGI entirely: the option bit was
   set and read on the exact same "nvidia,tegra264-smmu" compatible
   as the static key, so it added no per-instance signal that the
   static key did not already carry. The predicate now gates purely
   on arm_smmu_erratum_repeat_tlbi_cfgi_key.
 - Reorder the series so the compatible-string detection lands
   last, once all the infrastructure exists:
     1/3  factor out force_sync helper (unchanged)
     2/3  add static key + WAR functions (no functional change)
     3/3  enable the key on nvidia,tegra264-smmu + silicon-errata
   Split the old v4 "Detect" and "Issue twice" patches accordingly.
 - Update the /* See ARM_SMMU_OPT_REPEAT_TLBI_CFGI */ comment inside
   arm_smmu_cmdq_batch_force_sync() to reference the static key
   description instead.

Changes since v3:
 - Drop the cmds->num == 0 early-return so the refactor is
   truly "no functional change".
 - Rename ARM_SMMU_OPT_TLBI_TWICE -> ARM_SMMU_OPT_REPEAT_TLBI_CFGI
   and rephrase its kdoc to be hardware-agnostic.
 - Rename arm_smmu_cmd_needs_tlbi_twice() ->
   arm_smmu_erratum_cmd_needs_repeating() and drop the kdoc
   above it.
 - Replace the explicit opcode switch with a single range check
   opcode >= CMDQ_OP_CFGI_STE && opcode < CMDQ_OP_ATC_INV.
 - Introduce arm_smmu_erratum_repeat_tlbi_cfgi_key static key:
   the predicate gates on it first so unaffected kernels pay
   only a single static_branch_unlikely() check.
 - Drop the verbose Tegra264-specific comments above
   arm_vsmmu_can_batch_cmd() and inside the batch helper.
 - Document the erratum in
   Documentation/arch/arm64/silicon-errata.rst.
 - Guard the repeat path in arm_smmu_cmdq_issue_cmdlist() with
   an n > 0 check so cmds[0] is never inspected on an empty
   cmdlist.
 - Drop the carried Reviewed-by tags now that the patch
   shape has changed; re-review appreciated.

Changes since v2:
 - Split into a 3-patch series (refactor / detect / apply) to keep
   each step small and bisectable.
 - Move the classifier to arm-smmu-v3.h as static inline so the
   iommufd file can share it.
 - Add arm_vsmmu_can_batch_cmd() to split iommufd batches at
   "needs repeating" transitions so the per-batch decision based
   on the first command stays correct under mixed user input.
 - Spell out in the commit message why detection is via DT and
   not via IIDR/ACPI.

Changes since v1:
 - Detect the erratum from the existing "nvidia,tegra264-smmu"
   compatible instead of adding a new property.
 - Centralise the doubling at the CMDQ submission layer and only
   apply it to CFGI/TLBI (not ATC_INV).
 - Drop the binding/dtsi patches accordingly.

Ashish Mhetre (2):
  iommu/arm-smmu-v3: Introduce CFGI/TLBI-repeat workaround
    infrastructure
  iommu/arm-smmu-v3: Enable CFGI/TLBI-repeat workaround on Tegra264

Nicolin Chen (1):
  iommu/arm-smmu-v3: Factor out CMDQ batch force-sync conditions

 Documentation/arch/arm64/silicon-errata.rst   |  2 +
 .../arm/arm-smmu-v3/arm-smmu-v3-iommufd.c     |  7 +-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 89 +++++++---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  6 ++
 include/uapi/linux/iommufd.h                  | 12 ++-
 5 files changed, 102 insertions(+), 14 deletions(-)


base-commit: 0718283ab28bc3907e10b61a6b4be6fefa1cbb2f
-- 
2.50.1

[PATCH v8 1/3] iommu/arm-smmu-v3: Factor out CMDQ batch force-sync conditions

[Ashish Mhetre]

From: Nicolin Chen <nicolinc@nvidia.com>

arm_smmu_cmdq_batch_add_cmd_p() carries two distinct reasons for
flushing the current batch with a CMD_SYNC before appending the
new command:

  - The batch's pre-assigned cmdq does not support the new command.
  - The Arm erratum 2812531 workaround (ARM_SMMU_OPT_CMDQ_FORCE_SYNC)
    forces a SYNC at one entry before the batch is full.

Lift those checks into a new arm_smmu_cmdq_batch_force_sync() helper
so that adding another force-sync condition becomes a one-line
addition. No functional change.

Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Ashish Mhetre <amhetre@nvidia.com>
---
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 23 +++++++++++++++------
 1 file changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 57b750ebcd3d..dd7475c50afc 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -847,16 +847,27 @@ static void arm_smmu_cmdq_batch_init_cmd(struct arm_smmu_device *smmu,
 	cmds->cmdq = arm_smmu_get_cmdq(smmu, cmd);
 }
 
+static bool arm_smmu_cmdq_batch_force_sync(struct arm_smmu_device *smmu,
+					   struct arm_smmu_cmdq_batch *cmds,
+					   struct arm_smmu_cmd *cmd)
+{
+	/* The batch's pre-assigned cmdq doesn't support the new command */
+	if (!arm_smmu_cmdq_supports_cmd(cmds->cmdq, cmd))
+		return true;
+
+	/* Arm erratum 2812531 */
+	if (cmds->num == CMDQ_BATCH_ENTRIES - 1 &&
+	    (smmu->options & ARM_SMMU_OPT_CMDQ_FORCE_SYNC))
+		return true;
+
+	return false;
+}
+
 static void arm_smmu_cmdq_batch_add_cmd_p(struct arm_smmu_device *smmu,
 					  struct arm_smmu_cmdq_batch *cmds,
 					  struct arm_smmu_cmd *cmd)
 {
-	bool force_sync = (cmds->num == CMDQ_BATCH_ENTRIES - 1) &&
-			  (smmu->options & ARM_SMMU_OPT_CMDQ_FORCE_SYNC);
-	bool unsupported_cmd;
-
-	unsupported_cmd = !arm_smmu_cmdq_supports_cmd(cmds->cmdq, cmd);
-	if (force_sync || unsupported_cmd) {
+	if (arm_smmu_cmdq_batch_force_sync(smmu, cmds, cmd)) {
 		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmdq, cmds->cmds,
 					    cmds->num, true);
 		arm_smmu_cmdq_batch_init_cmd(smmu, cmds, cmd);
-- 
2.50.1

[PATCH v8 2/3] iommu/arm-smmu-v3: Introduce CFGI/TLBI-repeat workaround infrastructure

[Ashish Mhetre]

Tegra264 SMMU instances need every CFGI/TLBI command sequence issued
twice, with the second issue executing only after the first issue's
CMD_SYNC has completed:

    TLBI/CFGI ... CMD_SYNC TLBI/CFGI ... CMD_SYNC

ATC_INV is not affected and must never be doubled.

Add arm_smmu_erratum_repeat_tlbi_cfgi_key and an
arm_smmu_erratum_cmd_needs_repeating() helper that gates on the static
key first and then range-checks the opcode (CFGI_STE .. ATC_INV), so
subsequent changes wiring the workaround into the CMDQ submission path
can share a single predicate.

Rename the existing arm_smmu_cmdq_issue_cmdlist() to
__arm_smmu_cmdq_issue_cmdlist() and add a thin wrapper of the original
name that re-issues the same cmdlist a second time when the predicate
fires. Host-internal callers go through the wrapper and get the repeat;
register the new condition with arm_smmu_cmdq_batch_force_sync() too.

The iommufd user-invalidation path (arm_vsmmu_cache_invalidate()) keeps
issuing user commands exactly once by calling the raw
__arm_smmu_cmdq_issue_cmdlist(). The erratum is reported to user space
in a subsequent change so an affected guest doubles its own CFGI/TLBI,
doubling them again on the host would be wrong.

No callers enable the static key yet, so there is no functional change.
A subsequent change will enable the key on affected instances.

Suggested-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Ashish Mhetre <amhetre@nvidia.com>
---
 .../arm/arm-smmu-v3/arm-smmu-v3-iommufd.c     |  4 +-
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   | 57 +++++++++++++++++--
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  5 ++
 3 files changed, 60 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
index 85ebfdb3d2a7..032d0c272f3f 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
@@ -400,8 +400,8 @@ int arm_vsmmu_cache_invalidate(struct iommufd_viommu *viommu,
 			continue;
 
 		/* FIXME always uses the main cmdq rather than trying to group by type */
-		ret = arm_smmu_cmdq_issue_cmdlist(smmu, &smmu->cmdq, &last->cmd,
-						  cur - last, true);
+		ret = __arm_smmu_cmdq_issue_cmdlist(smmu, &smmu->cmdq, &last->cmd,
+						    cur - last, true);
 		if (ret) {
 			cur--;
 			goto out;
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index dd7475c50afc..f0f8b2daf8d5 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -18,6 +18,7 @@
 #include <linux/interrupt.h>
 #include <linux/io-pgtable.h>
 #include <linux/iopoll.h>
+#include <linux/jump_label.h>
 #include <linux/module.h>
 #include <linux/msi.h>
 #include <linux/of.h>
@@ -42,6 +43,14 @@ MODULE_PARM_DESC(disable_msipolling,
 static const struct iommu_ops arm_smmu_ops;
 static struct iommu_dirty_ops arm_smmu_dirty_ops;
 
+/*
+ * Repeat every {CFGI,TLBI};CMD_SYNC command sequence so that the second
+ * issue executes only after the first issue's CMD_SYNC has completed.
+ * Does not apply to ATC_INV. The key is global and is enabled from DT
+ * probe on affected hardware (currently Tegra264 only).
+ */
+static DEFINE_STATIC_KEY_FALSE(arm_smmu_erratum_repeat_tlbi_cfgi_key);
+
 enum arm_smmu_msi_index {
 	EVTQ_MSI_INDEX,
 	GERROR_MSI_INDEX,
@@ -698,10 +707,10 @@ static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq,
  *   insert their own list of commands then all of the commands from one
  *   CPU will appear before any of the commands from the other CPU.
  */
-int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
-				struct arm_smmu_cmdq *cmdq,
-				struct arm_smmu_cmd *cmds, int n,
-				bool sync)
+int __arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
+				  struct arm_smmu_cmdq *cmdq,
+				  struct arm_smmu_cmd *cmds, int n,
+				  bool sync)
 {
 	struct arm_smmu_cmd cmd_sync;
 	u32 prod;
@@ -820,6 +829,38 @@ int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 	return ret;
 }
 
+bool arm_smmu_erratum_cmd_needs_repeating(struct arm_smmu_cmd *cmd)
+{
+	u8 opcode;
+
+	if (!static_branch_unlikely(&arm_smmu_erratum_repeat_tlbi_cfgi_key))
+		return false;
+
+	opcode = FIELD_GET(CMDQ_0_OP, cmd->data[0]);
+	return opcode >= CMDQ_OP_CFGI_STE && opcode < CMDQ_OP_ATC_INV;
+}
+
+int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
+				struct arm_smmu_cmdq *cmdq,
+				struct arm_smmu_cmd *cmds, int n,
+				bool sync)
+{
+	int ret = __arm_smmu_cmdq_issue_cmdlist(smmu, cmdq, cmds, n, sync);
+
+	/*
+	 * A bare CMD_SYNC can be issued with n == 0 (e.g. an empty
+	 * batch_submit()), in which case there is no cmds[0] to inspect
+	 * and nothing to repeat.
+	 */
+	if (!n || ret || !sync)
+		return ret;
+
+	if (arm_smmu_erratum_cmd_needs_repeating(&cmds[0]))
+		ret = __arm_smmu_cmdq_issue_cmdlist(smmu, cmdq, cmds, n, sync);
+
+	return ret;
+}
+
 static int arm_smmu_cmdq_issue_cmd_p(struct arm_smmu_device *smmu,
 				     struct arm_smmu_cmd *cmd, bool sync)
 {
@@ -860,6 +901,14 @@ static bool arm_smmu_cmdq_batch_force_sync(struct arm_smmu_device *smmu,
 	    (smmu->options & ARM_SMMU_OPT_CMDQ_FORCE_SYNC))
 		return true;
 
+	/*
+	 * See the description at arm_smmu_erratum_repeat_tlbi_cfgi_key. Batches
+	 * never mix CFGI/TLBI with others, so checking cmds[0] alone is enough.
+	 */
+	if (cmds->num == CMDQ_BATCH_ENTRIES &&
+	    arm_smmu_erratum_cmd_needs_repeating(&cmds->cmds[0]))
+		return true;
+
 	return false;
 }
 
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index c909c9a88538..9508bf50d14c 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -1207,10 +1207,15 @@ void arm_smmu_attach_commit(struct arm_smmu_attach_state *state);
 void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master,
 				  const struct arm_smmu_ste *target);
 
+int __arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
+				  struct arm_smmu_cmdq *cmdq,
+				  struct arm_smmu_cmd *cmds, int n,
+				  bool sync);
 int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 				struct arm_smmu_cmdq *cmdq,
 				struct arm_smmu_cmd *cmds, int n,
 				bool sync);
+bool arm_smmu_erratum_cmd_needs_repeating(struct arm_smmu_cmd *cmd);
 
 #ifdef CONFIG_ARM_SMMU_V3_SVA
 bool arm_smmu_sva_supported(struct arm_smmu_device *smmu);
-- 
2.50.1

[PATCH v8 3/3] iommu/arm-smmu-v3: Enable CFGI/TLBI-repeat workaround on Tegra264

[Ashish Mhetre]

Nvidia Tegra264 SMMU is affected by an erratum where a TLB entry can
survive an invalidation that races with concurrent traffic targeting
the same entry. The hardware-recommended software workaround is to
issue every CFGI/TLBI command (each followed by CMD_SYNC) twice, and
that infrastructure is already in place behind
arm_smmu_erratum_repeat_tlbi_cfgi_key.

Neither IDR nor IIDR flags this Tegra264-specific bug, so hardware
detection is not possible. Tegra264 is device-tree-only (no ACPI/IORT
support) and already has a dedicated "nvidia,tegra264-smmu" compatible,
so DT-probe is the only viable detection path.

Enable the workaround on instances matching the existing
"nvidia,tegra264-smmu" compatible by calling static_branch_enable() on
arm_smmu_erratum_repeat_tlbi_cfgi_key. Document the erratum in
Documentation/arch/arm64/silicon-errata.rst.

The host repeats only the invalidations it issues itself. On a nesting
setup the guest submits its own invalidations, either directly through
VCMDQ or via the iommufd user-invalidation path, so the guest is the one
that has to apply the workaround there. Report the erratum to user space
through the new IOMMU_HW_INFO_ARM_SMMUV3_ERRATA_REPEAT_TLBI_CFGI hw_info
flag so a VMM can confirm the guest runs a matching "nvidia,tegra264-smmu"
(or otherwise double the commands itself). The host therefore does not
repeat the guest's user-issued invalidations, which would otherwise be
doubled a second time.

Signed-off-by: Ashish Mhetre <amhetre@nvidia.com>
---
 Documentation/arch/arm64/silicon-errata.rst         |  2 ++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c |  3 +++
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c         | 11 +++++++++--
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h         |  1 +
 include/uapi/linux/iommufd.h                        | 12 +++++++++++-
 5 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/Documentation/arch/arm64/silicon-errata.rst b/Documentation/arch/arm64/silicon-errata.rst
index b0c68b64f5ac..744591598dfc 100644
--- a/Documentation/arch/arm64/silicon-errata.rst
+++ b/Documentation/arch/arm64/silicon-errata.rst
@@ -312,6 +312,8 @@ stable kernels.
 |                |                 | T241-MPAM-4,    |                             |
 |                |                 | T241-MPAM-6     |                             |
 +----------------+-----------------+-----------------+-----------------------------+
+| NVIDIA         | T264 SMMU       | T264-SMMU-3     | N/A                         |
++----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | Freescale/NXP  | LS2080A/LS1043A | A-008585        | FSL_ERRATUM_A008585         |
 +----------------+-----------------+-----------------+-----------------------------+
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
index 032d0c272f3f..25982bdbcbd9 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-iommufd.c
@@ -33,6 +33,9 @@ void *arm_smmu_hw_info(struct device *dev, u32 *length,
 	info->iidr = readl_relaxed(master->smmu->base + ARM_SMMU_IIDR);
 	info->aidr = readl_relaxed(master->smmu->base + ARM_SMMU_AIDR);
 
+	if (arm_smmu_erratum_repeat_tlbi_cfgi())
+		info->flags |= IOMMU_HW_INFO_ARM_SMMUV3_ERRATA_REPEAT_TLBI_CFGI;
+
 	*length = sizeof(*info);
 	*type = IOMMU_HW_INFO_TYPE_ARM_SMMUV3;
 
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index f0f8b2daf8d5..67e19c240ea5 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -829,11 +829,16 @@ int __arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 	return ret;
 }
 
+bool arm_smmu_erratum_repeat_tlbi_cfgi(void)
+{
+	return static_branch_unlikely(&arm_smmu_erratum_repeat_tlbi_cfgi_key);
+}
+
 bool arm_smmu_erratum_cmd_needs_repeating(struct arm_smmu_cmd *cmd)
 {
 	u8 opcode;
 
-	if (!static_branch_unlikely(&arm_smmu_erratum_repeat_tlbi_cfgi_key))
+	if (!arm_smmu_erratum_repeat_tlbi_cfgi())
 		return false;
 
 	opcode = FIELD_GET(CMDQ_0_OP, cmd->data[0]);
@@ -5357,8 +5362,10 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
 	if (of_dma_is_coherent(dev->of_node))
 		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
 
-	if (of_device_is_compatible(dev->of_node, "nvidia,tegra264-smmu"))
+	if (of_device_is_compatible(dev->of_node, "nvidia,tegra264-smmu")) {
 		tegra_cmdqv_dt_probe(dev->of_node, smmu);
+		static_branch_enable(&arm_smmu_erratum_repeat_tlbi_cfgi_key);
+	}
 
 	return ret;
 }
diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
index 9508bf50d14c..6465c0357fb5 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h
@@ -1216,6 +1216,7 @@ int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
 				struct arm_smmu_cmd *cmds, int n,
 				bool sync);
 bool arm_smmu_erratum_cmd_needs_repeating(struct arm_smmu_cmd *cmd);
+bool arm_smmu_erratum_repeat_tlbi_cfgi(void);
 
 #ifdef CONFIG_ARM_SMMU_V3_SVA
 bool arm_smmu_sva_supported(struct arm_smmu_device *smmu);
diff --git a/include/uapi/linux/iommufd.h b/include/uapi/linux/iommufd.h
index fddaf14cfdd2..be23922b053a 100644
--- a/include/uapi/linux/iommufd.h
+++ b/include/uapi/linux/iommufd.h
@@ -602,11 +602,21 @@ struct iommu_hw_info_vtd {
 	__aligned_u64 ecap_reg;
 };
 
+/**
+ * enum iommu_hw_info_arm_smmuv3_flags - Flags for ARM SMMUv3 hw_info
+ * @IOMMU_HW_INFO_ARM_SMMUV3_ERRATA_REPEAT_TLBI_CFGI: If set, user space must
+ *	issue TLBI/CFGI+SYNC commands twice due to a hardware erratum T264-SMMU-3.
+ *	See the description at arm_smmu_erratum_repeat_tlbi_cfgi_key.
+ */
+enum iommu_hw_info_arm_smmuv3_flags {
+	IOMMU_HW_INFO_ARM_SMMUV3_ERRATA_REPEAT_TLBI_CFGI = 1 << 0,
+};
+
 /**
  * struct iommu_hw_info_arm_smmuv3 - ARM SMMUv3 hardware information
  *                                   (IOMMU_HW_INFO_TYPE_ARM_SMMUV3)
  *
- * @flags: Must be set to 0
+ * @flags: Combination of enum iommu_hw_info_arm_smmuv3_flags
  * @__reserved: Must be 0
  * @idr: Implemented features for ARM SMMU Non-secure programming interface
  * @iidr: Information about the implementation and implementer of ARM SMMU,
-- 
2.50.1