You are viewing a plain text version of this content. The canonical link for it is here.
- Re: [PR] [TIRx] tcgen05 dispatch paths, buffer dim-surgery views, FlashMLA lowering, and typed-buffer migration fixes [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/01 00:54:19 UTC, 4 replies.
- [GH] (tvm/main): Workflow run "tvm-bot" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/01 00:54:40 UTC, 8 replies.
- [GH] (tvm-ffi/main-dev/2026-07-29/rust_visitor): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/01 01:28:19 UTC, 1 replies.
- [PR] [Python] Keep object tying dormant on legacy allocation [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/01 01:47:07 UTC, 1 replies.
- Re: [PR] [CORE] Staging Object Tying Feature [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/01 01:53:27 UTC, 1 replies.
- [GH] (tvm/main): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/01 02:09:41 UTC, 7 replies.
- [GH] (tvm-ffi/main): Workflow run "mainline-only" failed! - posted by GitBox <gi...@apache.org> on 2026/08/01 02:31:13 UTC, 4 replies.
- [GH] (tvm-ffi/main): Workflow run "mainline-only" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/01 02:45:33 UTC, 3 replies.
- [GH] (tvm-ffi/kathrync/map): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/01 04:18:45 UTC, 13 replies.
- [GH] (tvm-ffi/kathrync/map): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/01 07:04:04 UTC, 4 replies.
- [PR] [TIRx] Flatten cuda/trn backend operator folder [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/01 19:32:41 UTC, 2 replies.
- [I] Segfault in TVMFFIEnvRegisterCAPI on macOS arm64 with apache-tvm-ffi==0.1.13, resolved by pinning 0.1.12 [tvm-ffi] - posted by "rap7239 (via GitHub)" <gi...@apache.org> on 2026/08/01 22:55:59 UTC, 1 replies.
- Re: [PR] [FEAT] Add structural map, maybe_inplace_mutate, and var_remap APIs [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 07:23:48 UTC, 21 replies.
- [PR] tirx: represent buffer parameters with BufferType [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 09:30:56 UTC, 10 replies.
- [GH] (tvm/tvm-phase-out-buffer-map): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/02 09:31:31 UTC, 0 replies.
- [GH] (tvm/tvm-phase-out-buffer-map): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/02 09:41:21 UTC, 0 replies.
- [PR] fix(version): Keep source patch ahead of post releases [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 12:36:50 UTC, 2 replies.
- [PR] [CORE] Mark type-info lookup as pure [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 13:50:05 UTC, 2 replies.
- [GH] (tvm-ffi/task-309-type-info-pure): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/02 13:51:01 UTC, 0 replies.
- [GH] (tvm-ffi/main): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/02 15:19:25 UTC, 0 replies.
- [GH] (tvm-ffi/main-dev/2026-07-29/rust_visitor): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/02 20:51:29 UTC, 0 replies.
- Re: [PR] [FEAT][RUST] Add rust structural_visit and structural_walk. [tvm-ffi] - posted by "gemini-code-assist[bot] (via GitHub)" <gi...@apache.org> on 2026/08/02 22:28:45 UTC, 2 replies.
- Re: [I] Cannot build 0.1.13rc2 with 3.15 from source [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 22:30:58 UTC, 0 replies.
- [PR] [FFI] Bump tvm-ffi to latest Aug 3 [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/02 23:20:04 UTC, 3 replies.
- [PR] [BUILD] Migrate the Z3 dependency to mlc-z3-static [tvm] - posted by "Ubospica (via GitHub)" <gi...@apache.org> on 2026/08/03 01:46:38 UTC, 3 replies.
- [GH] (tvm/2026-08-02-mlc-z3-static): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/03 01:55:20 UTC, 0 replies.
- Re: [PR] [Fix][Codegen] Preserve NaNs in floating-point min and max [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/03 06:35:19 UTC, 0 replies.
- Re: [PR] [Runtime][Disco][Relax] Add CPUCCL support for distributed execution [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/03 06:44:53 UTC, 1 replies.
- [PR] feat(lower-tirx): align NVIDIA IKET profiling with the official ABI [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/03 21:02:03 UTC, 9 replies.
- [PR] Bump cryptography from 41.0.6 to 50.0.0 in /docker/python [tvm] - posted by "dependabot[bot] (via GitHub)" <gi...@apache.org> on 2026/08/04 03:35:41 UTC, 0 replies.
- [GH] (tvm/main): Workflow run "tvm-bot" failed! - posted by GitBox <gi...@apache.org> on 2026/08/04 08:32:05 UTC, 8 replies.
- [PR] Make ModuleObj ABI compatible [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/04 10:09:52 UTC, 1 replies.
- Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] - posted by "actinks (via GitHub)" <gi...@apache.org> on 2026/08/04 11:56:53 UTC, 0 replies.
- [GH] (tvm-ffi/tvm-ffi-abi-compact-check): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/04 12:01:29 UTC, 0 replies.
- Re: [PR] [CORE] Make ModuleObj Stable [tvm-ffi] - posted by "cyx-6 (via GitHub)" <gi...@apache.org> on 2026/08/04 12:31:07 UTC, 0 replies.
- [GH] (tvm-ffi/tvm-ffi-abi-compact-check): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/04 13:46:28 UTC, 0 replies.
- [GH] (tvm-ffi/main): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/04 16:02:41 UTC, 1 replies.
- Re: [PR] [Relax][Test] Cover default GPU pipeline scheduling for R.power/elementwise kernels [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/04 19:41:35 UTC, 1 replies.
- Re: [PR] [Relax][ONNX] Fix Resize coordinate error with non-integer scales [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/04 19:57:07 UTC, 0 replies.
- [PR] [FFI] Keep object optionals pointer-sized [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/04 21:57:52 UTC, 1 replies.
- [GH] (tvm-ffi/task313-optional-objectref-abi): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/04 22:08:19 UTC, 0 replies.
- [GH] (tvm-ffi/task313-optional-objectref-abi): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/04 22:28:39 UTC, 0 replies.
- [PR] [TEST] Relax shared DAG hash scaling threshold [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/04 23:05:15 UTC, 1 replies.
- Re: [PR] [FIX][TIRx] Use physical order for Buffer.local views [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/04 23:11:41 UTC, 0 replies.
- [PR] [FFI] Bump tvm-ffi to 0.1.13.post2 [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/04 23:14:06 UTC, 1 replies.
- [PR] [FIX][TIRx] Traverse pointer expressions in tile calls [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/04 23:58:42 UTC, 1 replies.
- [PR] [FIX][TIRx] Remap typed buffer expressions during specialization [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/04 23:58:44 UTC, 3 replies.
- [PR] [Fix][Relax][ONNX] Handle Split initializer with keep_params_in_input [tvm] - posted by "hahalfx (via GitHub)" <gi...@apache.org> on 2026/08/05 07:21:11 UTC, 1 replies.
- [PR] [FIX][Relax][ONNX] Keep the static shape of a rank-0 Shape input [tvm] - posted by "adityasingh2400 (via GitHub)" <gi...@apache.org> on 2026/08/05 14:55:17 UTC, 4 replies.
- [PR] [Web] Bump tvmjs version to 0.26.0 on v0.26.0 [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/05 15:00:39 UTC, 0 replies.
- [PR] [Python] Bump apache-tvm-ffi floor to >=0.1.13.post2 [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/05 15:24:13 UTC, 2 replies.
- [PR] [Web] Bump tvmjs version to 0.27.0-dev0 on main [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/05 15:24:20 UTC, 1 replies.
- [GH] (tvm/tvm-dev/2026-08-05-main-ffi-floor): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/05 16:54:30 UTC, 1 replies.
- [GH] (tvm/tvm-dev/2026-08-05-main-web-bump): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/05 16:55:51 UTC, 0 replies.
- Re: [PR] [Web] Bump tvmjs version to 0.26.0 and apache-tvm-ffi floor to >=0.1.13.post2 [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/05 17:55:19 UTC, 0 replies.
- [GH] (tvm/main): Workflow run "Publish TVM wheels" failed! - posted by GitBox <gi...@apache.org> on 2026/08/05 19:12:32 UTC, 0 replies.
- [PR] [Fix][TIRx] Fix MSVC build of IndexDataTypeNormalizer [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/05 19:20:42 UTC, 3 replies.
- [PR] [Fix][Arith] Isolate Z3 contexts and make memoization deterministic [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/05 19:23:05 UTC, 1 replies.
- Re: [I] [Bug][Relax][ONNX] Split._impl_v13 rejects graph-initializer split operand under keep_params_in_input=True [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/05 19:31:57 UTC, 0 replies.
- [GH] (tvm/main): Workflow run "Publish TVM wheels" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/05 22:13:31 UTC, 0 replies.
- [I] [VOTE] Release Apache TVM v0.26.0.rc0 [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/05 22:20:02 UTC, 9 replies.
- Re: [I] [Bug] [FRONTEND][ONNX] Error converting operator Slice: TVMError: Check failed: (IsBaseOf(relax::TensorStructInfo(DataType::Void(), kUnknownNDim), GetStructInfo(data))) is false [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/05 23:53:36 UTC, 0 replies.
- [PR] [FIX][CUDA] Use virtual architecture targets for NVRTC [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/06 00:37:16 UTC, 0 replies.
- [PR] [BugFix][Metal] Preserve pointer address spaces for byte offsets [tvm] - posted by "GY-Bai (via GitHub)" <gi...@apache.org> on 2026/08/06 05:13:13 UTC, 3 replies.
- [PR] [FIX][TIRx] Use typed buffer parameter in pointer config test [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/06 07:54:35 UTC, 1 replies.
- Re: [PR] [ONNX] Fix Split operator for uneven splits with num_outputs [tvm] - posted by "Mr-Neutr0n (via GitHub)" <gi...@apache.org> on 2026/08/06 14:01:38 UTC, 2 replies.
- Re: [PR] [FIX][CUDA] Select NVRTC architecture for output format [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/06 21:51:16 UTC, 0 replies.
- [PR] fix: create wheel for GIL build of Python 3.14 [tvm-ffi] - posted by "tmct (via GitHub)" <gi...@apache.org> on 2026/08/06 22:08:12 UTC, 3 replies.
- Re: [I] [Bug] build error: tvm-ffi expects cxx_std_17 while tvm now upgraded to cxx_std_20 [tvm] - posted by "petersktang (via GitHub)" <gi...@apache.org> on 2026/08/07 00:31:22 UTC, 0 replies.
- [PR] [TIRx][CUDA] Add a table-driven PTX dialect and retire tirx.ptx.* [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/07 04:47:07 UTC, 1 replies.
- [PR] [PageKVCache] Add support of sinks params and fix sliding windnow [tvm] - posted by "krishnaraj36 (via GitHub)" <gi...@apache.org> on 2026/08/07 05:10:59 UTC, 0 replies.
- [GH] (tvm/gpt_oss_kvCache): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/07 05:37:28 UTC, 0 replies.
- [PR] [Feat][Rust] Add native structural_map and structural_mutate [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/07 06:27:44 UTC, 9 replies.
- [GH] (tvm-ffi/feat/rust-structural-map): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/07 06:39:30 UTC, 2 replies.
- [GH] (tvm-ffi/feat/rust-structural-map): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/07 07:14:55 UTC, 1 replies.
- [GH] (tvm/apache-pr-ptx-dialect): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/07 07:52:59 UTC, 0 replies.
- [I] [Bug] [Relax] adaptive_avg_pool2d returns incorrect LLVM CPU results [tvm] - posted by "tangyixuan01 (via GitHub)" <gi...@apache.org> on 2026/08/08 03:01:27 UTC, 3 replies.
- [I] [Bug] TVM CUDA `cumprod` launch grid overflow (CUDA_ERROR_INVALID_VALUE) [tvm] - posted by "XYZboom (via GitHub)" <gi...@apache.org> on 2026/08/08 03:02:43 UTC, 1 replies.
- [PR] [SCRIPT] Support PEP 695 symbolic variables in Relax and TIR [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/08 10:25:28 UTC, 1 replies.
- [PR] [Fix][TOPI] Fuse GPU scan blocks to avoid CUDA gridDim.y overflow [tvm] - posted by "chenmiaoming (via GitHub)" <gi...@apache.org> on 2026/08/08 13:29:30 UTC, 6 replies.
- [GH] (tvm/tvm-type-var-parser-syntax): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/08 13:33:55 UTC, 0 replies.
- [PR] [Relax][Frontend][ONNX] Support dynamic Range bounds [tvm] - posted by "napronald (via GitHub)" <gi...@apache.org> on 2026/08/09 01:21:56 UTC, 4 replies.
- [PR] [TIRx][CUDA] uint32 index dtypes, directive fixes, and PTX ISA coverage [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/09 06:03:40 UTC, 1 replies.
- Re: [I] [Bug] [Relax][BYOC] JSONSerializer fails to handle tir_vars in BYOC for models with symbolic arithmetic [tvm] - posted by "zupengwang (via GitHub)" <gi...@apache.org> on 2026/08/10 05:08:56 UTC, 1 replies.
- [PR] fix: preserve LLVM vector value dominance across control flow [tvm] - posted by "mvanhorn (via GitHub)" <gi...@apache.org> on 2026/08/10 12:34:35 UTC, 1 replies.
- [GH] (tvm/fix/20015-llvm-vector-dominance): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/10 12:35:14 UTC, 0 replies.
- [I] [RESULT][VOTE] Release Apache TVM v0.26.0.rc0 [tvm] - posted by "MasterJH5574 (via GitHub)" <gi...@apache.org> on 2026/08/11 00:37:47 UTC, 0 replies.
- [PR] [Fix][WebGPU] Preserve read-only buffer access modes [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/11 05:53:29 UTC, 1 replies.
- [PR] [Relax][Frontend][TFLite] Support StableHLO shape ops [tvm] - posted by "Aharrypotter (via GitHub)" <gi...@apache.org> on 2026/08/11 07:09:26 UTC, 9 replies.
- Re: [PR] [Relax][ONNX] Add CastLike support and dynamic-k Trilu to expand backend coverage [tvm] - posted by "Aharrypotter (via GitHub)" <gi...@apache.org> on 2026/08/11 09:44:16 UTC, 1 replies.
- [PR] [Relax][ONNX] Support lower-rank PRelu slopes [tvm] - posted by "Aharrypotter (via GitHub)" <gi...@apache.org> on 2026/08/11 10:27:06 UTC, 2 replies.
- [PR] [BugFix][TE] Initialize nested reductions at the outermost reduction scope [tvm] - posted by "Gunse11er (via GitHub)" <gi...@apache.org> on 2026/08/11 10:40:08 UTC, 1 replies.
- [I] [Bug] `Mul` + `ReduceSum` crashes in onnx frontend [tvm] - posted by "XYZboom (via GitHub)" <gi...@apache.org> on 2026/08/11 12:01:40 UTC, 1 replies.
- [I] [Bug] dlight GPU GEMV scheduling crashes on Relax dataflow block due to imperfect loop nest after cache_read insertion [tvm] - posted by "XYZboom (via GitHub)" <gi...@apache.org> on 2026/08/11 12:43:43 UTC, 1 replies.
- [PR] [Fix][Relax][ONNX] Fold Min/Max/Sum/Mean constants elementwise [tvm] - posted by "aryanputta (via GitHub)" <gi...@apache.org> on 2026/08/11 17:37:12 UTC, 1 replies.
- [I] Cross-compiling against free-threaded Python 3.14 (3.14t) fails to find Development.Module [tvm-ffi] - posted by "iamthebot (via GitHub)" <gi...@apache.org> on 2026/08/11 18:46:52 UTC, 0 replies.
- [PR] [FIX] Support free-threaded Python cross-builds [tvm-ffi] - posted by "iamthebot (via GitHub)" <gi...@apache.org> on 2026/08/11 19:14:30 UTC, 0 replies.
- [PR] [FEAT][RUST] Resolve reflected type methods and constructors from Rust [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/11 19:20:33 UTC, 1 replies.
- [GH] (tvm-ffi/feat/rust-reflected-method-lookup): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/11 19:25:38 UTC, 0 replies.
- [PR] [TIRX] Split backend.cuda.intrinsics, fold tcgen05 descriptors, and fix two PTX dialect gaps [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/11 21:15:19 UTC, 1 replies.
- [GH] (tvm/main): Workflow run "npm_and_yarn in /web for js-yaml - Update #1520990934" failed! - posted by GitBox <gi...@apache.org> on 2026/08/11 23:07:36 UTC, 0 replies.
- [GH] (tvm-ffi/feat/rust-reflected-method-lookup): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/12 00:09:41 UTC, 0 replies.
- [PR] [Feature][Relax] Support shared-KV attention with configurable sliding windows [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/12 02:03:14 UTC, 1 replies.
- [PR] [Fix][DLight] Reject unsupported GEMV access patterns before scheduling [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/12 05:34:32 UTC, 0 replies.
- [GH] (tvm/bugfix-20106): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/12 06:54:14 UTC, 0 replies.
- [GH] (tvm/upstream/dlight-gemv-access-legality): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/12 07:04:49 UTC, 0 replies.
- [PR] [Fix][Relax][ONNX] Fold binary ops on symbolic PrimExpr scalars without numpy dispatch [tvm] - posted by "Irfanuet191 (via GitHub)" <gi...@apache.org> on 2026/08/12 09:34:27 UTC, 1 replies.
- [I] [RFC] Expanding ONNX operator coverage in the Relax frontend [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/12 09:40:58 UTC, 0 replies.
- [I] [Bug] LLVM backend truncates large AllocBuffer size from int64 to int32 [tvm] - posted by "littlehamsterxu-bot (via GitHub)" <gi...@apache.org> on 2026/08/12 09:46:43 UTC, 0 replies.
- [PR] [Relax][ONNX] Support scalar QDQ inputs [tvm] - posted by "Aharrypotter (via GitHub)" <gi...@apache.org> on 2026/08/12 16:55:00 UTC, 3 replies.
- [PR] [Web] Avoid redundant memory byte copies [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/13 02:47:51 UTC, 6 replies.
- [PR] [Fix][Relax][Torch] Preserve derived exported input dimensions [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/13 04:07:08 UTC, 5 replies.
- [PR] [Fix][TIRx] Preserve index semantics when narrowing to int32 [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/13 04:37:56 UTC, 4 replies.
- [GH] (tvm/issue-19570): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/13 14:22:15 UTC, 0 replies.
- [GH] (tvm/issue-19570): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/13 14:35:04 UTC, 1 replies.
- [PR] [Web] Retain evicted shape tuples until teardown [tvm] - posted by "atirna (via GitHub)" <gi...@apache.org> on 2026/08/15 14:46:51 UTC, 2 replies.
- [PR] [BugFix] Align default (C) tirx.round lowering to ties-to-even [tvm] - posted by "LngelKyo (via GitHub)" <gi...@apache.org> on 2026/08/15 15:58:35 UTC, 4 replies.
- [PR] [Fix][WebGPU] Validate and bound symbolic stack allocations [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/15 23:09:49 UTC, 3 replies.
- [PR] [FIX][RELAX] Lower non-contiguous WebGPU cumsum [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 03:10:27 UTC, 0 replies.
- [PR] [Fix][Relax] Track lowered reshape storage aliases [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 03:18:04 UTC, 3 replies.
- [PR] [Fix][Relax] Preserve Torch integral power semantics [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 03:47:17 UTC, 4 replies.
- [PR] [Feat][Web] Support per-parameter tensor cache encoding [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 04:44:15 UTC, 8 replies.
- [PR] [Fix][Relax][Torch] Align retained expand dimensions by trailing rank [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 04:52:28 UTC, 4 replies.
- [PR] [Fix][Relax][Torch] Materialize runtime scalar shape values [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 04:55:06 UTC, 7 replies.
- [PR] [Fix][Relax] Preserve tensor-derived symbols during fusion [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/16 05:05:34 UTC, 5 replies.
- Re: [PR] [Relax][Frontend][ONNX] Fix Resize coordinate error with non-integer scales [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/16 05:24:09 UTC, 2 replies.
- [PR] [TIRx][CUDA] Replace source helpers with typed PTX forms [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/16 06:27:59 UTC, 6 replies.
- [PR] [Fix][LLVM] Preserve 64-bit AllocBuffer extents [tvm] - posted by "fallenmi (via GitHub)" <gi...@apache.org> on 2026/08/16 06:58:43 UTC, 2 replies.
- Re: [PR] [Fix][Relax] Lower non-contiguous WebGPU cumsum [tvm] - posted by "yongwww (via GitHub)" <gi...@apache.org> on 2026/08/17 02:19:12 UTC, 3 replies.
- [I] [BYOC][oneDNN] Update and maintain oneDNN BYOC integration [tvm] - posted by "jagadeeshvx (via GitHub)" <gi...@apache.org> on 2026/08/17 04:37:12 UTC, 2 replies.
- [I] [Bug][Relax][Frontend][ONNX] ConvTranspose with 3D (5D) input fails in from_onnx: NotImplementedError [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 09:16:29 UTC, 2 replies.
- Re: [PR] [Runtime] Add PagedAttentionKVCache checkpoint primitives [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/17 10:42:57 UTC, 3 replies.
- Re: [PR] [Fix][DLight] Reject GEMV accesses unsupported by scheduling [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/17 10:43:24 UTC, 2 replies.
- [I] [Bug][Relax][Frontend][ONNX] Flatten with out-of-range `axis` silently accepted by `from_onnx`; onnxruntime rejects the same model [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 18:02:32 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Validate Flatten axis range in `from_onnx` [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 18:39:57 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] `Mean` with constant (initializer) inputs: single input returns a 0-d scalar, multi-input raises TypeError [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 19:00:12 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Fix Mean/Sum/Min/Max with all-constant inputs [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 19:42:35 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] PRelu with lower-rank broadcastable `slope` rejected by `from_onnx`: ValueError "Unsupported PRelu slope shape" [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 19:59:27 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Support broadcastable multi-axis PRelu slopes [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 20:46:34 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] Pad `mode="wrap"` (opset 18+) and `axes` input rejected by `from_onnx`: `OpAttributeInvalid` [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 20:49:04 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] Reshape mishandles 0 in `shape`: constant-fold path breaks 0-copy semantics (default allowzero=0); allowzero=1 ignored [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 20:57:05 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Support Pad mode="wrap" and axes input for op… [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/17 22:11:13 UTC, 2 replies.
- [PR] [TIRx][CUDA] Add PTX address expressions with immediate byte offsets [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/17 23:21:59 UTC, 1 replies.
- [PR] [TIRx][CUDA] Version-gate CUDA 12.8 tensor-map enums and fix registry-test lock leak [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/18 00:07:28 UTC, 1 replies.
- [PR] [CUDA][BugFix] Guard tensor map enums by toolkit version [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/18 02:22:51 UTC, 2 replies.
- Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/18 04:58:10 UTC, 1 replies.
- [PR] [Web] Avoid tensor-cache record copies [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/18 10:33:16 UTC, 1 replies.
- [I] [Bug][Metal] Metal codegen produces silently-wrong output on a HiFi-GAN-style Relax graph (cosine 0.756 vs LLVM 1.0000) [tvm] - posted by "zacharywhitley (via GitHub)" <gi...@apache.org> on 2026/08/18 10:33:50 UTC, 0 replies.
- [PR] [Fix][DLight] Handle rank-one GEMV cache loads [tvm] - posted by "SamJSui (via GitHub)" <gi...@apache.org> on 2026/08/18 14:02:09 UTC, 1 replies.
- [GH] (tvm/issue-19517): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/18 14:41:59 UTC, 1 replies.
- [GH] (tvm/issue-19517): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/18 15:47:41 UTC, 0 replies.
- [PR] [TIRx][CUDA] Add register and cluster launch controls [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/19 01:46:06 UTC, 1 replies.
- [PR] [Fix][DLight] Localize private scalar reduction buffers [tvm] - posted by "SamJSui (via GitHub)" <gi...@apache.org> on 2026/08/19 01:56:58 UTC, 1 replies.
- [PR] [Fix][Relax] Honor ONNX Reshape zero semantics [tvm] - posted by "tandede (via GitHub)" <gi...@apache.org> on 2026/08/19 18:46:44 UTC, 3 replies.
- [GH] (tvm/main): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/20 00:16:34 UTC, 8 replies.
- [PR] fix: isolate tvm-ffi from TVM's C++20 build standard [tvm] - posted by "mvanhorn (via GitHub)" <gi...@apache.org> on 2026/08/20 06:55:41 UTC, 1 replies.
- Re: [I] [Bug] build CXX standard conflicts: tvm-ffi expects cxx_std_17 while tvm now upgraded to 20 [tvm] - posted by "mvanhorn (via GitHub)" <gi...@apache.org> on 2026/08/20 06:55:43 UTC, 3 replies.
- Re: [PR] [DLight] Fix GPU reduction ScheduleError on 4D non-last-axis reductions [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/20 13:30:00 UTC, 3 replies.
- [PR] [Fix][Relax] Run destructors for non-trivially-destructible types in Arena [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/20 21:40:03 UTC, 4 replies.
- Re: [I] [Bug][Relax] FuseOpsByPattern leaks native memory on every call (arena-allocated Group::attrs never destructed) [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/20 21:53:50 UTC, 1 replies.
- [PR] [FEAT] Add core opaque pointer handler map [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/21 00:53:41 UTC, 1 replies.
- [PR] [FIX] Make ffi.Error picklable and clarify unpicklable-type errors [tvm-ffi] - posted by "cyx-6 (via GitHub)" <gi...@apache.org> on 2026/08/21 10:09:15 UTC, 2 replies.
- Re: [PR] [CI] Enable Windows ARM64 wheels [tvm-ffi] - posted by "contentis (via GitHub)" <gi...@apache.org> on 2026/08/21 10:11:16 UTC, 1 replies.
- Re: [I] [Tracking Issue][TFlite] Expand Missing / Partial Operator Coverage [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/21 15:28:46 UTC, 0 replies.
- [PR] [Fix][Rust] Serialize structural visit test registration [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/21 18:00:05 UTC, 1 replies.
- [PR] [FEAT][RUST] Complete structural walk/map/visit/mutate dispatch APIs [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/21 18:43:18 UTC, 1 replies.
- [PR] [TIRx][CUDA] Allow newer CUTLASS packages for IKET [tvm] - posted by "jinhongyii (via GitHub)" <gi...@apache.org> on 2026/08/22 02:02:24 UTC, 1 replies.
- [PR] [Web] Avoid copies when uploading WASM memory to WebGPU [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/22 05:45:59 UTC, 1 replies.
- [PR] [Web] Upload pass-through tensor-cache records directly to WebGPU [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/22 07:19:22 UTC, 1 replies.
- [PR] [Web] Decode packed BF16 tensor records in place [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/22 07:27:26 UTC, 10 replies.
- Re: [PR] [TOPI][CUDA] Fix topk/sort gridDim overflow by remapping grid axes in sort_ir [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/23 07:57:57 UTC, 1 replies.
- [PR] [IR][TIRX] Add first-class tuple expressions [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/23 16:12:36 UTC, 1 replies.
- [PR] [BugFix][Arith] Reject padded IterMapSimplify fallback [tvm] - posted by "zupengwang (via GitHub)" <gi...@apache.org> on 2026/08/23 20:20:57 UTC, 0 replies.
- [PR] [Fix][Relax][ONNX] Correct fmod mapping in Mod constant folding [tvm] - posted by "shoemoney (via GitHub)" <gi...@apache.org> on 2026/08/24 08:07:50 UTC, 1 replies.
- [PR] feat(lower-tirx): support PTX movmatrix [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/24 09:05:25 UTC, 1 replies.
- [PR] [TIRx][CUDA] Resolve clusterCtaIdx for single-CTA launches [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/24 13:34:04 UTC, 0 replies.
- [PR] [TIRx][CUDA] Accept packed sub-byte tensor-map dtypes in force_cu_dtype [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/24 13:34:11 UTC, 2 replies.
- Re: [PR] [TIRx][CUDA] Fix single-CTA clusterCtaIdx resolution and accept packed sub-byte tensor-map dtypes [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/24 15:20:54 UTC, 0 replies.
- Re: [PR] [FIX] Make ffi.Error picklable [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/24 18:28:53 UTC, 0 replies.
- [I] [Bug] [Relax][ONNX] Runtime-shape Reshape import passes TensorType to relax.op.reshape [tvm] - posted by "Anemone220 (via GitHub)" <gi...@apache.org> on 2026/08/25 02:28:50 UTC, 0 replies.
- [I] [Bug] [Relax][ONNX] Min import fails on symbolic broadcast [tvm] - posted by "Anemone220 (via GitHub)" <gi...@apache.org> on 2026/08/25 02:29:53 UTC, 0 replies.
- [I] [Bug] [Relax][ONNX] Gather import fails when indices come from Shape [tvm] - posted by "Anemone220 (via GitHub)" <gi...@apache.org> on 2026/08/25 02:31:08 UTC, 1 replies.
- [I] [Bug] [Relax][VMShapeLower] Static-shape Size/Slice/Reshape hits an uncomputed shape slot [tvm] - posted by "Anemone220 (via GitHub)" <gi...@apache.org> on 2026/08/25 02:32:47 UTC, 0 replies.
- [I] [Bug] [Relax][FuseTIR] Symbolic Flatten/Slice/Cast triggers a duplicate remap assertion [tvm] - posted by "Anemone220 (via GitHub)" <gi...@apache.org> on 2026/08/25 02:33:50 UTC, 0 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Support Shape outputs as Gather indices [tvm] - posted by "Gunse11er (via GitHub)" <gi...@apache.org> on 2026/08/25 08:47:24 UTC, 1 replies.
- [PR] [TIRx][CUDA] Preserve explicit single-CTA cluster launches [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/25 18:53:40 UTC, 1 replies.
- [PR] [Fix][Arith] Preserve nested floormod semantics [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/25 20:11:07 UTC, 2 replies.
- [I] [Bug][Relax][Frontend][ONNX] ONNX Scatter (opset 9/10) with broadcastable `indices`/`updates` silently returns wrong numeric output [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:07:29 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] Softmax ignores opset<=12 coerce-to-2D semantics and defaults `axis` to -1 instead of 1, silently producing wrong output for valid models [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:34:35 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][ONNX] Softplus: frontend injects a `threshold` that does not exist in the ONNX spec — observable accuracy loss in float64 for `20 < x ≲ 30` [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:37:30 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] Squeeze silently ignores axes that select non-unit dimensions instead of raising an error (violates ONNX spec) [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:38:41 UTC, 1 replies.
- [I] [Bug][Relax][Frontend][ONNX] `Where` mishandles shape-tensor (`ShapeExpr`) inputs: rejects valid size-1 broadcasting and crashes on mixed tensor/`ShapeExpr` inputs [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:42:14 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Fix Scatter with indices smaller than data [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/25 21:56:48 UTC, 3 replies.
- [GH] (tvm/main): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/25 22:48:36 UTC, 10 replies.
- [PR] [Fix][Relax] Raise error on non-unit dim ONNX Squeeze axis [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/25 23:52:26 UTC, 8 replies.
- [GH] (tvm/fix-onnx-non-unit-axis): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/25 23:53:04 UTC, 1 replies.
- [GH] (tvm/fix-onnx-non-unit-axis): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/26 00:04:31 UTC, 0 replies.
- [GH] (tvm/main): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/26 03:54:14 UTC, 1 replies.
- [GH] (tvm/memory/04-inplace-bf16-decode): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/26 04:27:39 UTC, 1 replies.
- [GH] (tvm/memory/04-inplace-bf16-decode): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/26 05:25:46 UTC, 1 replies.
- [PR] [Codegen][LLVM] Add LLVM 23 compatibility [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/26 05:50:31 UTC, 1 replies.
- [GH] (tvm/fix/llvm-23-compat): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/26 05:51:06 UTC, 0 replies.
- [GH] (tvm/fix/llvm-23-compat): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/26 05:54:49 UTC, 0 replies.
- Re: [PR] [Relax][VM] Improve diagnostics for unlowered Relax operators [tvm] - posted by "yinli-systems (via GitHub)" <gi...@apache.org> on 2026/08/26 11:20:14 UTC, 0 replies.
- [I] [Bug] [Relax] default pipeline reuses storage across alias boundary and produces silent wrong-code [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:06:02 UTC, 1 replies.
- [I] [Bug] [Bug][Relax] StaticPlanBlockMemory can emit non-dominating storage across if branches for reshape chains [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:15:46 UTC, 0 replies.
- [I] [Bug][Relax] Default LLVM pipeline can leave high-level Relax ops after FoldConstant resolves dynamic reshape shape [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:17:51 UTC, 0 replies.
- [I] [Bug][Relax] Default LLVM pipeline crashes in FoldConstant when reshape target is a runtime Shape [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:20:04 UTC, 0 replies.
- [I] [Bug][Relax] FuseTIR crashes when FuseOps creates a private function named fused for nested tuple getitem[Bug] [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:25:59 UTC, 0 replies.
- [I] [Bug][Relax] FuseOps can reorder call_tir before symbolic shape binding used in out_ty [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:28:34 UTC, 0 replies.
- [I] [Bug][Relax][LLVM] scatter_elements with symbolic H/W extents times out during build[Bug] [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:30:10 UTC, 0 replies.
- [I] [Bug][Relax] CombineParallelMatmul crashes on 1-D matmul boundary cases [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:32:01 UTC, 0 replies.
- [I] [Bug][Relax] Matmul rewrite passes drop explicit out_dtype and change float32 outputs to float16 [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/26 12:33:00 UTC, 0 replies.
- Re: [PR] [Fix][S-TIR][DLight] Guard non-affine reduction write-back [tvm] - posted by "Junius-Wynn (via GitHub)" <gi...@apache.org> on 2026/08/26 15:38:15 UTC, 4 replies.
- [PR] [TIRx][CUDA] Honor required CUDA thread extents in PTX [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/26 15:45:27 UTC, 2 replies.
- [GH] (tvm/tirx-reqntid): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/26 16:01:31 UTC, 0 replies.
- [GH] (tvm/tirx-reqntid): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/26 16:58:18 UTC, 0 replies.
- [PR] [FEAT][RUST] Add AnyCompatible::to_any. [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/26 18:33:21 UTC, 2 replies.
- [PR] [FEAT][RUST] Support Any-valued maps and typed default mutation [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/27 02:25:58 UTC, 1 replies.
- [I] [Relax][VM] Re-running KillAfterLastUse on a default-lowered module crashes compiled VM codegen[Bug] [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 03:45:01 UTC, 0 replies.
- [GH] (tvm/fix/relax-onnx-flatten-axis-validation): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/27 04:44:03 UTC, 0 replies.
- [I] [Relax][Transform] ReorderTakeAfterMatmul drops relax.take mode and changes out-of-bounds semantics [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 05:47:19 UTC, 1 replies.
- [I] [Bug][Relax] AdjustMatmulOrder drops the outer matmul out_dtype [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 06:35:52 UTC, 0 replies.
- [I] [Bug][Relax] CombineParallelMatmul changes branch dtype when out_dtype differs [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 07:53:10 UTC, 1 replies.
- [I] [Bug][Relax] ReorderPermuteDimsAfterConcat asserts on valid unknown-rank input [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 07:55:47 UTC, 2 replies.
- [I] [Bug][Relax] CombineParallelMatmul fails valid branches with broadcast-compatible biases [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 07:57:06 UTC, 0 replies.
- [GH] (tvm/fix/relax-onnx-scatter-broadcast): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/27 08:32:44 UTC, 3 replies.
- [GH] (tvm/fix/relax-onnx-scatter-broadcast): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/27 09:22:31 UTC, 0 replies.
- [PR] [BugFix][Relax] Preserve take mode in ReorderTakeAfterMatmul [tvm] - posted by "katrinagui (via GitHub)" <gi...@apache.org> on 2026/08/27 09:47:28 UTC, 1 replies.
- [GH] (tvm/fix/relax-onnx-scatter-broadcast): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/27 10:42:08 UTC, 0 replies.
- [I] [Relax] [Bug] Default build silently computes wrong values for match_cast followed by split (second half of split output is misplaced) [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/27 12:17:39 UTC, 1 replies.
- [PR] [BugFix][Relax] Skip parallel matmul fusion for mixed output dtypes [tvm] - posted by "katrinagui (via GitHub)" <gi...@apache.org> on 2026/08/27 12:34:00 UTC, 1 replies.
- [GH] (tvm-ffi/feat/rust-ir-binding-support): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/27 20:12:48 UTC, 1 replies.
- [PR] [DOCS][TIRx] Expand API reference coverage [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/27 20:35:01 UTC, 0 replies.
- [PR] [Fix][Relax][ONNX] Where: broadcast size-1 shape expressions, materialize ShapeExpr inputs [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/27 20:37:56 UTC, 1 replies.
- [GH] (tvm/fix/relax-onnx-where-shape-broadcast): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/27 20:38:35 UTC, 0 replies.
- [GH] (tvm/fix/relax-onnx-scatter-broadcast): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/27 20:57:01 UTC, 0 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Squeeze: raise an error when axes select a non-unit dimension [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/27 21:08:57 UTC, 3 replies.
- [GH] (tvm/fix-onnx-non-unit-axis): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/27 21:37:20 UTC, 0 replies.
- [PR] [Fix][Relax][Frontend][ONNX] Fix Softplus accuracy loss from hardcoded threshold [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/27 21:47:05 UTC, 1 replies.
- [GH] (tvm-ffi/feat/rust-ir-binding-support): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/28 00:58:50 UTC, 0 replies.
- [PR] fix(tirx): stabilize multi-GPU correctness tests [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/28 02:03:23 UTC, 1 replies.
- [GH] (tvm/fix/issue-20048-dlight-reduction): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 05:57:58 UTC, 0 replies.
- [GH] (tvm/fix/issue-20048-dlight-reduction): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 05:57:58 UTC, 0 replies.
- Re: [I] [Bug] `relax.build` crashes with a `ScheduleError` during the dlight GPU scheduling pass [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 05:57:58 UTC, 0 replies.
- [PR] [FIX][TIRx] Restore IterVar span reflection [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 06:10:42 UTC, 0 replies.
- [PR] [REFACTOR][RUST] Clarify structural mutator dispatch and recursion APIs [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 06:53:43 UTC, 1 replies.
- [GH] (tvm/fix-onnx-mod-constant-fold): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 06:55:51 UTC, 0 replies.
- [GH] (tvm/fix-onnx-mod-constant-fold): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 06:55:52 UTC, 0 replies.
- [PR] [FEAT][RUST] Support Any elements in FFI containers [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 07:04:47 UTC, 2 replies.
- [GH] (tvm/fix/reorder-take-mode): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:05:32 UTC, 0 replies.
- [GH] (tvm/fix/reorder-take-mode): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:05:32 UTC, 0 replies.
- [PR] [FEAT][RUST] Add object binding runtime support [tvm-ffi] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 07:15:24 UTC, 2 replies.
- [PR] [Perf][Arith] Materialize Z3 solvers lazily on first query [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 07:30:33 UTC, 1 replies.
- [GH] (tvm/perf-lazy-z3-solver): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:35:34 UTC, 0 replies.
- [GH] (tvm/fix/combine-parallel-matmul-dtype): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:44:17 UTC, 0 replies.
- [GH] (tvm/fix/combine-parallel-matmul-dtype): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:44:17 UTC, 0 replies.
- [GH] (tvm/fix/reflect-itervar-span): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:45:21 UTC, 0 replies.
- Re: [I] [Bug] CUDA Check failed: scope != "global" when using `argmin` [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 07:55:28 UTC, 0 replies.
- [GH] (tvm/perf-lazy-z3-solver): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/28 07:59:00 UTC, 0 replies.
- [PR] [Fix][Relax] Skip ReorderPermuteDimsAfterConcat for unknown-rank inputs [tvm] - posted by "yanght27 (via GitHub)" <gi...@apache.org> on 2026/08/28 09:10:36 UTC, 1 replies.
- [I] [Bug][Relax][ONNX] GatherElements/GatherND/ScatterND/OneHot mishandle negative indices [tvm] - posted by "lackjava-creator (via GitHub)" <gi...@apache.org> on 2026/08/28 10:41:12 UTC, 1 replies.
- [PR] [Relax][Frontend][ONNX] Add test cases to validate symbolic shapes in Min/Max broadcast [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/28 11:42:12 UTC, 0 replies.
- [GH] (tvm/issue-20175): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 11:42:53 UTC, 2 replies.
- [GH] (tvm/issue-20175): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/28 11:47:07 UTC, 1 replies.
- [PR] [Fix][Relax] Normalize negative indices in Gather/Scatter/OneHot Ops [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/28 13:43:43 UTC, 1 replies.
- [PR] [Fix][Relax] Preserve match-cast storage liveness [tvm] - posted by "zupengwang (via GitHub)" <gi...@apache.org> on 2026/08/28 15:38:24 UTC, 1 replies.
- [GH] (tvm/fix/relax-onnx-pad-wrap-axes): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/28 16:49:59 UTC, 1 replies.
- [GH] (tvm/fix/relax-onnx-flatten-axis-validation): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/28 16:50:15 UTC, 0 replies.
- [PR] [Fix][Arith] Give each materialized Z3 solver a private context [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/28 19:56:52 UTC, 1 replies.
- [PR] [Fix][Support] Use sbsa-linux CUDA include dir on ARM64 Linux [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/28 21:25:32 UTC, 1 replies.
- [PR] [TIRx][CUDA] Support exact required block dimensions [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/29 00:28:44 UTC, 1 replies.
- [PR] [TIRx][CUDA] Align the PTX dialect with PTX ISA 9.2 [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/29 01:49:18 UTC, 1 replies.
- [GH] (tvm/fix-20204-unknown-rank-permute): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 04:23:30 UTC, 0 replies.
- [GH] (tvm/fix-20204-unknown-rank-permute): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 04:23:30 UTC, 0 replies.
- [PR] [REFACTOR][TE] Represent tensor loads with opaque callees [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/29 04:38:57 UTC, 1 replies.
- Re: [PR] [Fix][TIRx] Restore IterVar span reflection [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/29 04:56:40 UTC, 0 replies.
- [GH] (tvm/disco_cpu_ccl): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 05:11:59 UTC, 1 replies.
- [GH] (tvm/disco_cpu_ccl): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 05:11:59 UTC, 1 replies.
- [GH] (tvm/upstream/torch-runtime-scalar-shapes): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/29 05:54:52 UTC, 0 replies.
- [PR] [TIRx][CUDA] Allow launch bounds with required block size [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/29 07:25:09 UTC, 1 replies.
- [GH] (tvm/fix/tirx-required-block-launch-bounds): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/29 07:25:44 UTC, 0 replies.
- [GH] (tvm/upstream/torch-runtime-scalar-shapes): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/29 07:29:05 UTC, 0 replies.
- [GH] (tvm/yanxu/cudnn-sdpa-fixes): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 09:47:03 UTC, 0 replies.
- [GH] (tvm/yanxu/cudnn-sdpa-fixes): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/29 09:47:03 UTC, 0 replies.
- [GH] (tvm/issue-20175): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/29 11:28:30 UTC, 0 replies.
- Re: [PR] [Relax][Frontend][ONNX] Support symbolic shapes in Min/Max broadcast [tvm] - posted by "cchung100m (via GitHub)" <gi...@apache.org> on 2026/08/29 15:26:26 UTC, 2 replies.
- [I] [Bug][Relax][Frontend][Torch] `from_fx` crashes with obscure internal `TypeError: reduce() of empty iterable` for `flatten` with `start_dim > end_dim`, instead of a clear "invalid flatten args" error [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 15:43:35 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][Torch] `torch.einsum` with repeated subscripts (diagonal/trace, e.g. `'ii->i'`, `'ii->'`, `'...ii->...i'`) fails to convert: `aten.diagonal` unsupported [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 15:49:21 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][Torch] `Tensor.expand` mishandles `-1` when `sizes` has more dims than the input (right-alignment): valid models crash at import with `IndexError` [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 15:51:34 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][PyTorch] `torch.Tensor.mean(..., dtype=...)` silently ignores the `dtype` argument and returns the input dtype [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 15:54:40 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][Torch] `torch.round` rounds half values away from zero (violates ties-to-even), and `torch.round(x, decimals=...)` fails to convert via `from_exported_program` [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 15:58:51 UTC, 0 replies.
- [I] [Bug][Relax][Frontend][PyTorch] `x.split(split_size)` with a non-divisible `split_size > D/2` produces wrong chunk shapes: frontend `_split` converts the per-chunk size to a section count, but `relax.op.split`'s int argument means "equal sections" [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/29 16:00:04 UTC, 0 replies.
- [PR] [CORE] Inherit behavioral type attributes [tvm-ffi] - posted by "junrushao (via GitHub)" <gi...@apache.org> on 2026/08/29 16:27:55 UTC, 4 replies.
- [PR] [FIX] Stabilize Python annotation metadata [tvm-ffi] - posted by "junrushao (via GitHub)" <gi...@apache.org> on 2026/08/29 16:28:04 UTC, 1 replies.
- [PR] [CUDA][TIRx] Preserve device state during cleanup and skip invalid Top-K references [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/29 19:13:40 UTC, 1 replies.
- [PR] [TIRx] Restore BufferRegion as typed expression [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/29 19:45:20 UTC, 1 replies.
- [GH] (tvm/fix/relax-match-cast-storage-liveness): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/30 03:03:11 UTC, 0 replies.
- [GH] (tvm/fix/relax-match-cast-storage-liveness): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/30 03:03:11 UTC, 0 replies.
- [GH] (tvm/fix/relax-onnx-pad-wrap-axes): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/30 03:19:20 UTC, 0 replies.
- Re: [I] [Bug][S-TIR][DLight] Reduction schedule rule crashes with ScheduleError on 4D tensor non-last-axis reduction [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/30 03:46:44 UTC, 1 replies.
- Re: [PR] [Fix][Relax][ONNX] Fix Range dtype for PrimExpr bounds [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/30 03:47:39 UTC, 0 replies.
- Re: [PR] [TIRx][Schedule] Support rfactor for arg reducers selecting last index [tvm] - posted by "tlopex (via GitHub)" <gi...@apache.org> on 2026/08/30 03:51:43 UTC, 0 replies.
- [PR] [Fix][Relax][Metal] Constrain wide-head prefill tiling [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/30 05:12:54 UTC, 0 replies.
- [PR] [Fix][Metal] Bound symbolic stack allocations [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/30 05:13:42 UTC, 0 replies.
- Re: [PR] [Fix][Relax][Frontend][ONNX] Support Pad-18 axes input, keep wrap for Pad-19 [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/30 10:30:38 UTC, 0 replies.
- [PR] [Fix][Relax][Frontend][Torch] Support `aten.diagonal` from decomposed repeated-subscript einsum [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/30 10:52:15 UTC, 2 replies.
- [GH] (tvm/fix/relax-torch-einsum-diagonal): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/30 10:52:56 UTC, 1 replies.
- [GH] (tvm/fix/relax-torch-einsum-diagonal): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/30 11:04:46 UTC, 0 replies.
- [I] [Bug][Relax] CPU default pipeline: fused PrimFunc keeps symbolic height/width unbound (MakePackedAPI ICHECK) [tvm] - posted by "lrcyyds1 (via GitHub)" <gi...@apache.org> on 2026/08/30 15:36:26 UTC, 1 replies.
- [PR] [Relax][Frontend][Torch] Fix `torch.round(x, decimals)` via `from_exported_program` and negative-decimals rounding [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/30 15:36:53 UTC, 0 replies.
- [PR] [Relax][Frontend][PyTorch] Fix x.split(int) with a non-divisible split_size [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/30 15:37:09 UTC, 0 replies.
- [GH] (tvm/fix/relax-torch-round): Workflow run "Lint" failed! - posted by GitBox <gi...@apache.org> on 2026/08/30 15:37:23 UTC, 0 replies.
- [GH] (tvm/fix/relax-torch-split-int-split-size): Workflow run "Lint" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/30 15:37:36 UTC, 0 replies.
- [PR] [Fix][Relax][Frontend][Torch] Honor the `dtype` argument of `aten.mean` (`torch.Tensor.mean` / `torch.mean`) [tvm] - posted by "siyiweigeHEW (via GitHub)" <gi...@apache.org> on 2026/08/30 15:47:05 UTC, 0 replies.
- [PR] [Fix][S-TIR] Preserve general reduction predicates [tvm] - posted by "akaashrp (via GitHub)" <gi...@apache.org> on 2026/08/30 17:02:59 UTC, 0 replies.
- [PR] [CORE] Propagate static object type bounds in Any conversions [tvm-ffi] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/30 17:56:08 UTC, 2 replies.
- [PR] [Fix][Relax] Validate non-concat axes before fusing parallel matmul biases [tvm] - posted by "OmarAzizi (via GitHub)" <gi...@apache.org> on 2026/08/30 20:58:53 UTC, 0 replies.
- [GH] (tvm/fix/relax-torch-split-int-split-size): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/30 21:22:25 UTC, 0 replies.
- [PR] [REFACTOR][TIR] Split masked buffer access into special calls [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/31 00:21:40 UTC, 1 replies.
- [PR] [Fix][Relax][Frontend][Torch] Validate flatten dims in `from_fx` [tvm] - posted by "hiyufan (via GitHub)" <gi...@apache.org> on 2026/08/31 14:47:34 UTC, 0 replies.
- [PR] [STUBGEN] Carry native layout facts through the language-neutral data. [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/31 15:11:31 UTC, 0 replies.
- [GH] (tvm-ffi/main-dev/2026-08-31/stubgen_layout_facts): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/31 15:43:27 UTC, 0 replies.
- [PR] [REFACTOR][IR] Unify expression subscription realization [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/31 16:16:02 UTC, 1 replies.
- [PR] [CORE] Publish native layout traits from ObjectDef [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/31 17:26:46 UTC, 0 replies.
- [GH] (tvm-ffi/main-dev/2026-08-31/refl_type_final_polymorphic): Workflow run "CI" failed! - posted by GitBox <gi...@apache.org> on 2026/08/31 17:41:24 UTC, 0 replies.
- [PR] [REFACTOR][IR] Introduce TensorLoad in core IR [tvm] - posted by "tqchen (via GitHub)" <gi...@apache.org> on 2026/08/31 18:27:16 UTC, 1 replies.
- [GH] (tvm-ffi/main-dev/2026-08-31/refl_type_final_polymorphic): Workflow run "CI" is working again! - posted by GitBox <gi...@apache.org> on 2026/08/31 18:35:17 UTC, 0 replies.
- Re: [PR] [Docs][TIRx] Reorganize and align documentation with current APIs [tvm] - posted by "spectrometerHBH (via GitHub)" <gi...@apache.org> on 2026/08/31 19:23:14 UTC, 0 replies.
- [PR] [FEAT][RUST] Accept any Into return value in structural map/mutate callbacks [tvm-ffi] - posted by "Seven-Streams (via GitHub)" <gi...@apache.org> on 2026/08/31 21:05:36 UTC, 1 replies.