commits  

Messages by Thread

• (tvm) branch nightly updated (45a2a4082e -> 0701aaba4b) github-bot
• svn commit: r80752 - dev/tvm/tvm-ffi-v0.1.3-rc0 syfeng
• (tvm-ffi) tag v0.1.3-rc0 created (now 70d6bf0) syfeng
• (tvm-ffi) branch main updated: Add Siyuan Feng's GPG key to KEYS file (#269) tqchen
• (tvm-ffi) branch main updated: feat(stubgen): Generate `__all__` for proper exporting (#268) junrushao
• (tvm) branch main updated: [Relax][PyTorch]: Fix the sqrt operation requires float dtype but receives int64 in attention scaling (#18454) tlopex
• (tvm) branch main updated: [CI] Fix crash when grep finds no matches (#18457) tqchen
• (tvm) branch main updated: [Relax][PyTorch] Fix MultiheadAttention complie (#18459) tlopex
• (tvm) branch main updated: [Relax][PyTorch] Add decomposed operator support for normalization (#18460) tlopex
• (tvm) branch nightly updated (b6ac0721a0 -> 45a2a4082e) github-bot
• (tvm-ffi) branch main updated: feat(stubgen): Refactor into staged pipeline, Introduce directive `import` (#259) tqchen
• (tvm) branch main updated: [Relax][PyTorch] Add decomposed operator support for Binary (#18458) tlopex
• (tvm) branch nightly updated (506a0bbc3f -> b6ac0721a0) github-bot
• (tvm) branch main updated: [DataType] Update to use explicit Bool Type Aligning with DLPack (#18453) tlopex
• (tvm-ffi) branch main updated: [DTYPE] Include dtype literals (#263) tqchen
• (tvm-ffi) branch main updated: [DTYPE] Align bool parsing to align with DLPack (#262) tqchen
• (tvm) branch tvm-ffi-bool updated (b19a4c7d24 -> 0067e1a3d6) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm) branch tvm-ffi-bool updated (3458c475f1 -> b19a4c7d24) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@f8471f820a121e9d20ab56f5f26139b461f9afea) tqchen
• (tvm) branch main updated (6c7ed243e5 -> f8471f820a) tlopex
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@11347788478724395fe1b2c0cac268411e3c5c37) tqchen
• (tvm) branch main updated: [DOCS] Update the merge setting (#18451) tlopex
• [PR] [DOCS] Update the merge setting [tvm] via GitHub
  • Re: [PR] [DOCS] Update the merge setting [tvm] via GitHub
  • Re: [PR] [DOCS] Update the merge setting [tvm] via GitHub
  • Re: [PR] [DOCS] Update the merge setting [tvm] via GitHub
• (tvm) branch main updated: [DOCS] Remove prebuilt package references and disable Colab button at tutorials (#18436) tqchen
• (tvm) branch tvm-ffi-bool updated (a7821b8a8c -> 3458c475f1) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm) branch tvm-ffi-bool updated (a975381e85 -> a7821b8a8c) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm) branch main updated: [CI] Update pre-commit configuration (#18448) tqchen
• [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
  • Re: [PR] fix(web): update progress reporting when loading from cache [tvm] via GitHub
• (tvm) branch main updated: [Relax][PyTorch] Add lower bound support for range constraints (#18447) mshr
• [PR] [Relax][PyTorch] Add decomposed operator support for Pad [tvm] via GitHub
• [PR] [CI] Update pre-commit configuration [tvm] via GitHub
  • Re: [PR] [CI] Update pre-commit configuration [tvm] via GitHub
  • Re: [PR] [CI] Update pre-commit configuration [tvm] via GitHub
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@0754ad82d6669af048effcf019cb549ed342605c) tqchen
• [PR] [Relax][PyTorch] Add lower bound support for range constraints [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add lower bound support for range constraints [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add lower bound support for range constraints [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add lower bound support for range constraints [tvm] via GitHub
• Re: [I] [Bug][FRONTEND][ONNX] Error converting operator Expand: TVMError: broadcast_to expects the input tensor shape is broadcastable to the target shape. [tvm] via GitHub
• Re: [I] [FRONTEND][ONNX] Error converting operator Transpose: TVMError: PermuteDims expects the number of input axes to equal the ndim of the input tensor. [tvm] via GitHub
  • Re: [I] [FRONTEND][ONNX] Error converting operator Transpose: TVMError: PermuteDims expects the number of input axes to equal the ndim of the input tensor. [tvm] via GitHub
• (tvm) branch main updated: [FRONTEND][ONNX] Fix operator Transpose: TVMError: PermuteDims expects the number of input axes to equal the ndim of the input tensor (#18435) tlopex
• (tvm) branch main updated: [Relax][PyTorch] Add decomposed operator support for MaxPool (#18446) tlopex
• (tvm) branch nightly updated (4b555a964f -> 506a0bbc3f) github-bot
• [PR] [Relax][PyTorch] Add decomposed operator support for MaxPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for MaxPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for MaxPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for MaxPool [tvm] via GitHub
• [I] relax export json [tvm] via GitHub
  • Re: [I] relax export json [tvm] via GitHub
  • Re: [I] relax export json [tvm] via GitHub
• [PR] Add CPU Sample device API implementation [tvm] via GitHub
  • Re: [PR] Add CPU Sample device API implementation [tvm] via GitHub
  • Re: [PR] Add CPU Sample device API implementation [tvm] via GitHub
  • Re: [PR] Add CPU Sample device API implementation [tvm] via GitHub
• (tvm) branch tvm-ffi-bool updated (11f324d9ea -> a975381e85) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@506a0bbc3f37bbee4bca5ce45972eefb6dc0288c) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@506a0bbc3f37bbee4bca5ce45972eefb6dc0288c) tqchen
• (tvm) branch main updated: [Relax][PyTorch] Add decomposed operator support for AdaptiveAvgPool (#18437) tlopex
• [I] [Bug] InternalError: sqrt operation requires float dtype but receives int64 in attention scaling [tvm] via GitHub
• [I] [Bug] InternalError: Squeeze dimension check too strict compared to PyTorch behavior [tvm] via GitHub
• [I] [Bug] InternalError: PermuteDims dimension mismatch when converting scaled_dot_product_attention with 2D inputs [tvm] via GitHub
• [I] [Bug] InternalError: cannot make uint from negative value -inf when compiling model with MultiheadAttention [tvm] via GitHub
• [I] [Bug] KeyError: 'dtype' when converting PyTorch model with gradient checkpointing using torch.export [tvm] via GitHub
• [I] [Bug] Segmentation fault when converting PyTorch index assignment operation with torch.export [tvm] via GitHub
• [PR] [Relax][PyTorch] Add decomposed operator support for AdaptiveAvgPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for AdaptiveAvgPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for AdaptiveAvgPool [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add decomposed operator support for AdaptiveAvgPool [tvm] via GitHub
• (tvm) branch main updated (4b555a964f -> 6785c8f131) tlopex
• [PR] [DOC] Remove prebuilt package references and disable Colab button at tutorials [tvm] via GitHub
  • Re: [PR] [DOC] Remove prebuilt package references and disable Colab button at tutorials [tvm] via GitHub
  • Re: [PR] [DOC] Remove prebuilt package references and disable Colab button at tutorials [tvm] via GitHub
  • Re: [PR] [DOCS] Remove prebuilt package references and disable Colab button at tutorials [tvm] via GitHub
  • Re: [PR] [DOCS] Remove prebuilt package references and disable Colab button at tutorials [tvm] via GitHub
• (tvm) branch nightly updated (f574031657 -> 4b555a964f) github-bot
• (tvm) branch tvm-ffi-bool updated: Update tqchen
• (tvm-ffi) branch main updated: [DLPack] Leverage exchange api when possible (#260) yaxingcai
• (tvm) branch main updated: Adjusted Longrope embedding function to match Huggingface Implementation (#18422) ruihangl
• (tvm) branch tvm-ffi-bool updated (ee4270e209 -> 4b1670421f) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@d013dad06d38cf0e011ac065815db2609d2c3efe) tqchen
• (tvm) branch tvm-ffi-bool created (now ee4270e209) tqchen
  • (tvm) 01/01: [DataType] Update to use explicit Bool Type Aligning with DLPack tqchen
• (tvm) branch main updated: [TOPI] Support integer type input for log and log2 (#18426) ruihangl
• (tvm) branch main updated: [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests (#18433) tlopex
• [PR] [FRONTEND][ONNX] Fix operator Transpose: TVMError: PermuteDims expects the number of input axes to equal the ndim of the input tensor [tvm] via GitHub
  • Re: [PR] [FRONTEND][ONNX] Fix operator Transpose: TVMError: PermuteDims expects the number of input axes to equal the ndim of the input tensor [tvm] via GitHub
• (tvm-ffi) branch main updated: [ADDON] Add torch c dlpack for macos (#256) tqchen
• [I] [Docs] Remove Google Analytics from the TVM Website [tvm] via GitHub
  • Re: [I] [Docs] Remove Google Analytics from the TVM Website [tvm] via GitHub
  • Re: [I] [Docs] Remove Google Analytics from the TVM Website [tvm] via GitHub
  • Re: [I] [Docs] Remove Google Analytics from the TVM Website [tvm] via GitHub
  • Re: [I] [Docs] Remove Google Analytics from the TVM Website [tvm] via GitHub
• (tvm) branch main updated: [Relax][Pytorch] Support basic range constraints (#18429) mshr
• Re: [I] [Bug] 3 tests fail [tvm] via GitHub
  • Re: [I] [Bug] 3 tests fail [tvm] via GitHub
  • Re: [I] [Bug] 3 tests fail [tvm] via GitHub
  • Re: [I] [Bug] 3 tests fail [tvm] via GitHub
• (tvm) branch nightly updated (26db8bfd7e -> f574031657) github-bot
• [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests [tvm] via GitHub
• (tvm-ffi) branch main updated: [Fix] Enhanced check for CUDA availability (#258) ruihangl
• svn commit: r80636 - release/tvm/tvm-ffi-v0.1.2 tqchen
• (tvm-ffi) tag v0.1.2 created (now c1df05f) tqchen
• (tvm) branch main updated: [TEST][CODEGEN] Fix the test scripts tries to tell numpy a dtype name that it cannot recognise (#18430) tlopex
• [PR] Enable username checks in PR title and body [tvm] via GitHub
  • Re: [PR] [CI] Enable username checks in PR title and body [tvm] via GitHub
  • Re: [PR] [CI] Enable username checks in PR title and body [tvm] via GitHub
  • Re: [PR] [CI] Enable username checks in PR title and body [tvm] via GitHub
• (tvm) branch nightly updated (33fa9262fa -> 26db8bfd7e) github-bot
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@26db8bfd7e527198f43f3cc379f404c7513a82ef) tqchen
• (tvm-ffi) branch main updated: [ADDON] Add torch c dlpack ext wheels for windows (#252) tqchen
• (tvm-ffi) branch main updated: [PyTorch] Allow tensor conversion on rocm backend (#253) tqchen
• (tvm) branch dependabot/pip/docker/python/pip-25.3 created (now 872198ecc5) github-bot
• [PR] Bump pip from 22.1.1 to 25.3 in /docker/python [tvm] via GitHub
• (tvm) branch dependabot/pip/docker/python/pip-25.2 deleted (was 80b1bc229f) github-bot
• (tvm) branch main updated: [Web] Fix arrayDecodeStorage scope issue for q0f32 models (#18415) tqchen
• [PR] [TEST][CODEGEN] Fix the test scripts tries to tell numpy a dtype name that it cannot recognise [tvm] via GitHub
  • Re: [PR] [TEST][CODEGEN] Fix the test scripts tries to tell numpy a dtype name that it cannot recognise [tvm] via GitHub
• [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
  • Re: [PR] [Relax][Pytorch] Support basic range constraints [tvm] via GitHub
• (tvm) branch main updated: [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) (#18428) mshr
• [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(8) [tvm] via GitHub
• (tvm-ffi) branch main updated: [misc] support various arg types (#229) tqchen
• (tvm-ffi) branch main updated: feat: Introduce `tvm.registry.get_registered_type_keys()` (#249) tqchen
• (tvm-ffi) branch main updated: chore(cython): Specify `--module-name` when building Cython (#248) tqchen
• (tvm) branch nightly updated (3d136588d9 -> 33fa9262fa) github-bot
• (tvm) branch main updated: [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) (#18427) mshr
• (tvm-ffi) branch main updated: fix(cython): Make sure `TypeInfo` is properly registered by all classes (#246) junrushao
• (tvm-ffi) branch main updated: feat(cython): Expose TypeAttr via `_lookup_type_attr` (#247) junrushao
• svn commit: r80574 - dev/tvm/tvm-ffi-v0.1.2-rc0 tqchen
• (tvm-ffi) tag v0.1.2.rc0 created (now c1df05f) tqchen
• (tvm-ffi) branch main updated: [PYTHON] Further streamline number handling (#242) tqchen
• (tvm-ffi) branch main updated (5a87749 -> 8ee0e49) tqchen
• (tvm-ffi) branch main updated: [CYTHON] Improve fallback and dtype convert behavior (#241) tqchen
• (tvm-ffi) branch main updated: [FIX] Fix missing static registration for DLTensor* (#239) junrushao
• (tvm-ffi) branch main updated: [ERROR] Make Error conform more to std (#240) junrushao
• [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Add support for decomposed operators and fix IR of ops tests(7) [tvm] via GitHub
• (tvm-ffi) branch main updated: [STREAM] Enable compact with cuda-python driver stream (#236) junrushao
• (tvm-ffi) branch main updated: Allow handling of load-bearing compiler flags for dlpack (#231) junrushao
• (tvm-ffi) branch main updated: [CYTHON] Fix ctypes.c_void_p for nullptr (#235) tqchen
• Re: [PR] feat(relax/frontend/torch): Add basic range constraint support [tvm] via GitHub
  • Re: [PR] feat(relax/frontend/torch): Add basic range constraint support [tvm] via GitHub
  • Re: [PR] feat(relax/frontend/torch): Add basic range constraint support [tvm] via GitHub
  • Re: [PR] feat(relax/frontend/torch): Add basic range constraint support [tvm] via GitHub
• [PR] Support integer type input for log and log2 [tvm] via GitHub
  • Re: [PR] Support integer type input for log and log2 [tvm] via GitHub

• Earlier messages
• Later messages