[PW_SID:1091690] Support non-leaf and range invalidation features in RISC-V by linux-riscv-bot · Pull Request #1888 · linux-riscv/linux

linux-riscv-bot · 2026-05-08T15:14:35Z

PR for series 1091690 applied to workflow__riscv__fixes

Name: Support non-leaf and range invalidation features in RISC-V
URL: https://patchwork.kernel.org/project/linux-riscv/list/?series=1091690
Version: 2

Use in-line member documentation and add some small clarifications to the members. This is preparation to add more members. - Note that pgsize is only used by arm-smmuv3 - Note that freelist is only used by iommupt - Reword queued to emphasize the flush-all behavior Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

Add a struct to keep track of all the things that are pending to be merged into the gather. The way gather merging works, the pending range is checked against the current gather, and the current gather can be flushed before the pending things are added. Thus, if new things have to be recorded in the gather they need to be kept in the pending struct until after the gather is optionally flushed. The next patch adds new items to the gather and the pending struct. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

Generating the ARM SMMUv3 and RISC-V invalidation commands optimally requires some additional details from iommupt: - leaf_levels_bitmap is used to compute the ARM Range Invalidation Table Top Level hint - leaf_levels_bitmap is also used to compute the stride when generating single invalidations to invalidate once per leaf - table_levels_bitmap also computes the ARM TTL for future cases when there are no leaves Put these under a feature since only two drivers need to calculate them. This is also useful for the coming kunit iotlb invalidation test to know more about what invalidation is happening. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

…inval RISC-V can use the information from PT_FEAT_DETAILED_GATHER to compute the best stride to generate the single TLB invalidations. Pass the gather down to the lower functions and create a full-range gather for the flush-all callback. Reviewed-by: Tomasz Jeznach <tjeznach@rivosinc.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

Replace the per-page IOTLB invalidation loop with stride-based invalidation that uses the level bitmaps from iommu_iotlb_gather. Pre-calculate the invalidation information before running over the bonds loop as it is the same for every entry. The lowest set bit in the PT_FEAT_DETAILED_GATHER bitmaps indicates the stride. This design ignores the SVNAPOT contiguous pages on the assumption that they still have to be individually invalidated like ARM requires, though it is not clear from the spec. Replace the 2M cutoff for global invalidation with a 512 command limit. This is the same for a 4k stride and now scales with the stride size. Reviewed-by: Tomasz Jeznach <tjeznach@rivosinc.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

Non-leaf invalidation allows the single invalidate command to also clear the walk cache. If NL is available, set the NL bit if the gather indicates tables have been changed. The stride is already calculated properly. Reviewed-by: Tomasz Jeznach <tjeznach@rivosinc.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

The command queue entry format is 128 bits. Follow the pattern of the other drivers and encode the 64 bit dword number in the macro itself. RISC-V further has similarly named macros that are not field layout macros, but field content macros which won't get a new number. Overall this is clearer to understand the code and check for errors like using the wrong macro in the wrong spot. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

Use the RISC-V IOMMU Address Range Invalidation extension (capabilities.S, spec section 9.3) to invalidate an IOVA range with a single IOTINVAL.VMA command using NAPOT-encoded addressing. One iommu_iotlb_gather maps to one NAPOT invalidation command. The smallest power-of-two aligned range covering the gather is used since over-invalidation is always safe. S and NL seem to be orthogonal in the spec, so if NL is not supported then global invalidation is probably always going to happen as wiping a large range without a table change is not common. Reviewed-by: Tomasz Jeznach <tjeznach@rivosinc.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Linux RISC-V bot <linux.riscv.bot@gmail.com>

linux-riscv-bot · 2026-05-08T15:20:09Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
build-rv32-defconfig
Desc: Builds riscv32 defconfig
Duration: 116.19 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T15:39:33Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
build-rv64-clang-allmodconfig
Desc: Builds riscv64 allmodconfig with Clang, and checks for errors and added warnings
Duration: 1162.04 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:07:42Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
build-rv64-gcc-allmodconfig
Desc: Builds riscv64 allmodconfig with GCC, and checks for errors and added warnings
Duration: 1687.30 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:08:03Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
build-rv64-nommu-k210-defconfig
Desc: Builds riscv64 defconfig with NOMMU for K210
Duration: 19.15 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:08:25Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
build-rv64-nommu-k210-virt
Desc: Builds riscv64 defconfig with NOMMU for the virt platform
Duration: 20.84 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:08:28Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
checkpatch
Desc: Runs checkpatch.pl on the patch
Duration: 0.61 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:46Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
dtb-warn-rv64
Desc: Checks for Device Tree warnings/errors
Duration: 76.72 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:48Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
header-inline
Desc: Detects static functions without inline keyword in header files
Duration: 0.24 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:51Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
kdoc
Desc: Detects for kdoc errors
Duration: 0.76 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:53Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
module-param
Desc: Detect module_param changes
Duration: 0.25 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:55Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
verify-fixes
Desc: Verifies that the Fixes: tags exist
Duration: 0.23 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:09:56Z

Patch 1: "[v2,1/8] iommu: Split the kdoc comment for struct iommu_iotlb_gather"
verify-signedoff
Desc: Verifies that Signed-off-by: tags are correct
Duration: 0.27 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:12:06Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
build-rv32-defconfig
Desc: Builds riscv32 defconfig
Duration: 114.25 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:30:49Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
build-rv64-clang-allmodconfig
Desc: Builds riscv64 allmodconfig with Clang, and checks for errors and added warnings
Duration: 1120.88 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:58:06Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
build-rv64-gcc-allmodconfig
Desc: Builds riscv64 allmodconfig with GCC, and checks for errors and added warnings
Duration: 1635.54 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:58:27Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
build-rv64-nommu-k210-defconfig
Desc: Builds riscv64 defconfig with NOMMU for K210
Duration: 19.51 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:58:49Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
build-rv64-nommu-k210-virt
Desc: Builds riscv64 defconfig with NOMMU for the virt platform
Duration: 20.82 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T16:58:53Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
checkpatch
Desc: Runs checkpatch.pl on the patch
Duration: 1.69 seconds
Result: WARNING
Output:

CHECK: Lines should not end with a '('
#97: FILE: drivers/iommu/generic_pt/iommu_pt.h:513:
+		.pending.free_list = IOMMU_PAGES_LIST_INIT(

CHECK: Lines should not end with a '('
#123: FILE: drivers/iommu/generic_pt/iommu_pt.h:536:
+			gather_range_pending(

CHECK: Lines should not end with a '('
#157: FILE: drivers/iommu/generic_pt/iommu_pt.h:1086:
+		.pending.free_list = IOMMU_PAGES_LIST_INIT(

CHECK: Lines should not end with a '('
#178: FILE: drivers/iommu/generic_pt/iommu_pt.h:1132:
+		.pending.free_list = IOMMU_PAGES_LIST_INIT(

total: 0 errors, 0 warnings, 4 checks, 160 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.

Commit 9f4b2d530d5f ("iommupt: Add struct iommupt_pending_gather") has style problems, please review.

NOTE: Ignored message types: ALLOC_SIZEOF_STRUCT CAMELCASE COMMIT_LOG_LONG_LINE GIT_COMMIT_ID MACRO_ARG_REUSE NO_AUTHOR_SIGN_OFF

NOTE: If any of the errors are false positives, please report
      them to the maintainer, see CHECKPATCH in MAINTAINERS.
total: 0 errors, 0 warnings, 4 checks, 160 lines checked
CHECK: Lines should not end with a '('

linux-riscv-bot · 2026-05-08T17:00:13Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
dtb-warn-rv64
Desc: Checks for Device Tree warnings/errors
Duration: 78.44 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T17:00:15Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
header-inline
Desc: Detects static functions without inline keyword in header files
Duration: 1.20 seconds
Result: ERROR
Output:

Detected static functions without inline keyword in header files:
+
+static void gather_add_table(struct iommupt_pending_gather *pending,
+			     const struct pt_state *pts,
--
+
+static void gather_range_pending(struct iommupt_pending_gather *pending,
+				 struct pt_iommu *iommu_table, pt_vaddr_t iova,
Detected static functions without inline keyword in header files: 1

linux-riscv-bot · 2026-05-08T17:00:17Z

Patch 2: "[v2,2/8] iommupt: Add struct iommupt_pending_gather"
kdoc
Desc: Detects for kdoc errors
Duration: 0.72 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T20:23:58Z

Patch 6: "[v2,6/8] iommu/riscv: Add RISCV_IOMMU_CAPABILITIES_NL"
verify-fixes
Desc: Verifies that the Fixes: tags exist
Duration: 0.23 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T20:24:00Z

Patch 6: "[v2,6/8] iommu/riscv: Add RISCV_IOMMU_CAPABILITIES_NL"
verify-signedoff
Desc: Verifies that Signed-off-by: tags are correct
Duration: 0.27 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T20:26:10Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
build-rv32-defconfig
Desc: Builds riscv32 defconfig
Duration: 115.15 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T20:44:49Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
build-rv64-clang-allmodconfig
Desc: Builds riscv64 allmodconfig with Clang, and checks for errors and added warnings
Duration: 1116.82 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:12:00Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
build-rv64-gcc-allmodconfig
Desc: Builds riscv64 allmodconfig with GCC, and checks for errors and added warnings
Duration: 1629.46 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:12:21Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
build-rv64-nommu-k210-defconfig
Desc: Builds riscv64 defconfig with NOMMU for K210
Duration: 19.37 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:12:44Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
build-rv64-nommu-k210-virt
Desc: Builds riscv64 defconfig with NOMMU for the virt platform
Duration: 20.88 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:12:46Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
checkpatch
Desc: Runs checkpatch.pl on the patch
Duration: 0.63 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:04Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
dtb-warn-rv64
Desc: Checks for Device Tree warnings/errors
Duration: 75.72 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:06Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
header-inline
Desc: Detects static functions without inline keyword in header files
Duration: 0.23 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:08Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
kdoc
Desc: Detects for kdoc errors
Duration: 0.73 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:10Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
module-param
Desc: Detect module_param changes
Duration: 0.24 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:12Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
verify-fixes
Desc: Verifies that the Fixes: tags exist
Duration: 0.25 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:14:14Z

Patch 7: "[v2,7/8] iommu/riscv: Include the dword number in RISCV_IOMMU_CMD macros"
verify-signedoff
Desc: Verifies that Signed-off-by: tags are correct
Duration: 0.26 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:16:23Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
build-rv32-defconfig
Desc: Builds riscv32 defconfig
Duration: 114.26 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T21:35:04Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
build-rv64-clang-allmodconfig
Desc: Builds riscv64 allmodconfig with Clang, and checks for errors and added warnings
Duration: 1120.24 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:02:28Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
build-rv64-gcc-allmodconfig
Desc: Builds riscv64 allmodconfig with GCC, and checks for errors and added warnings
Duration: 1641.84 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:02:48Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
build-rv64-nommu-k210-defconfig
Desc: Builds riscv64 defconfig with NOMMU for K210
Duration: 19.16 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:03:11Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
build-rv64-nommu-k210-virt
Desc: Builds riscv64 defconfig with NOMMU for the virt platform
Duration: 20.65 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:03:14Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
checkpatch
Desc: Runs checkpatch.pl on the patch
Duration: 1.84 seconds
Result: WARNING
Output:

CHECK: Lines should not end with a '('
#55: FILE: drivers/iommu/riscv/iommu-bits.h:741:
+static inline void riscv_iommu_cmd_inval_set_napot(

total: 0 errors, 0 warnings, 1 checks, 111 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.

Commit f6f9b27f7d97 ("iommu/riscv: Add NAPOT range invalidation support") has style problems, please review.

NOTE: Ignored message types: ALLOC_SIZEOF_STRUCT CAMELCASE COMMIT_LOG_LONG_LINE GIT_COMMIT_ID MACRO_ARG_REUSE NO_AUTHOR_SIGN_OFF

NOTE: If any of the errors are false positives, please report
      them to the maintainer, see CHECKPATCH in MAINTAINERS.
total: 0 errors, 0 warnings, 1 checks, 111 lines checked
CHECK: Lines should not end with a '('

linux-riscv-bot · 2026-05-08T22:04:32Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
dtb-warn-rv64
Desc: Checks for Device Tree warnings/errors
Duration: 76.89 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:04:34Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
header-inline
Desc: Detects static functions without inline keyword in header files
Duration: 0.24 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:04:37Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
kdoc
Desc: Detects for kdoc errors
Duration: 0.73 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:04:39Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
module-param
Desc: Detect module_param changes
Duration: 0.25 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:04:40Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
verify-fixes
Desc: Verifies that the Fixes: tags exist
Duration: 0.24 seconds
Result: PASS

linux-riscv-bot · 2026-05-08T22:04:42Z

Patch 8: "[v2,8/8] iommu/riscv: Add NAPOT range invalidation support"
verify-signedoff
Desc: Verifies that Signed-off-by: tags are correct
Duration: 0.27 seconds
Result: PASS

Linux RISC-V bot and others added 9 commits May 2, 2026 08:13

Adding CI files

2c3b264

linux-riscv-bot force-pushed the workflow__riscv__fixes branch 2 times, most recently from 2d4fcdd to cd9d421 Compare May 14, 2026 08:49

linux-riscv-bot closed this May 16, 2026

linux-riscv-bot deleted the pw1091690 branch May 16, 2026 02:12

Conversation

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026

Uh oh!

linux-riscv-bot commented May 8, 2026