commits  

Messages by Thread

• (tvm) branch refl-cleanup updated (a7f593113e -> 32e7738c47) tqchen
  • (tvm) 01/01: [FFI][REFACTOR] Phase out TVM_FFI_REGISTER_GLOBAL in favor of GlobalDef tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@ea4369c221a26875c63d29aab435fc453d25b953) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@ea4369c221a26875c63d29aab435fc453d25b953) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@ea4369c221a26875c63d29aab435fc453d25b953) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@ea4369c221a26875c63d29aab435fc453d25b953) tqchen
• (tvm) branch refl-cleanup created (now a7f593113e) tqchen
  • (tvm) 01/01: [FFI][REFACTOR] Phase out TVM_FFI_REGISTER_GLOBAL in favor of GlobalDef tqchen
• [PR] [FFI][REFACTOR] Modularize refelection [tvm] via GitHub
  • Re: [PR] [FFI][REFACTOR] Modularize refelection [tvm] via GitHub
• (tvm) branch main updated: [TIR] Add `T.thread_return()` for early thread exit in CUDA kernels (#18134) tqchen
• (tvm) branch main updated: [BugFix][NNAPI] Fix type mismatch and test_mean annotation (#18140) tqchen
• Re: [I] [Bug] [CUDA] "hsinh" is undefined when the target is CUDA [tvm] via GitHub
• (tvm) branch main updated: [CodeGen][CUDA] Add sinhf CUDA Math API for CodeGen (#18144) tqchen
• (tvm) branch main updated: [Test] Use roi_list variable instead of hardcoded values in ROI tensor creation (#18145) mshr
• [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
  • Re: [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
  • Re: [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
  • Re: [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
  • Re: [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
  • Re: [I] [Bug] In the latest version of TVM, PackedFunc has been removed. [tvm] via GitHub
• (tvm) branch main updated: [FRONTEND][ONNX] Extend axes for layer_norm when gamma/beta are multi-dimensional (#18143) mshr
• (tvm) branch nightly updated (27a314eb73 -> a61e3ca0ab) github-bot
• [PR] [Test] Use roi_list variable instead of hardcoded values in ROI tensor creation [tvm] via GitHub
  • Re: [PR] [Test] Use roi_list variable instead of hardcoded values in ROI tensor creation [tvm] via GitHub
  • Re: [PR] [Test] Use roi_list variable instead of hardcoded values in ROI tensor creation [tvm] via GitHub
• (tvm) branch main updated: [Fix][ONNX] Fixed constant ROI handling in resize2d when loading onnx models (#18138) mshr
• [PR] [CodeGen][CUDA] Add sinhf CUDA Math API for CodeGen [tvm] via GitHub
  • Re: [PR] [CodeGen][CUDA] Add sinhf CUDA Math API for CodeGen [tvm] via GitHub
  • Re: [PR] [CodeGen][CUDA] Add sinhf CUDA Math API for CodeGen [tvm] via GitHub
• (tvm) branch main updated (6ecf8bd8d2 -> a51951795c) yongwww
• (tvm) branch main updated (99f5e662fd -> 6ecf8bd8d2) yongwww
• (tvm) branch main updated (27a314eb73 -> 99f5e662fd) yongwww
• [PR] [FRONTEND][ONNX] Extend axes for layer_norm when gamma/beta are multi-dimensional [tvm] via GitHub
  • Re: [PR] [FRONTEND][ONNX] Extend axes for layer_norm when gamma/beta are multi-dimensional [tvm] via GitHub
  • Re: [PR] [FRONTEND][ONNX] Extend axes for layer_norm when gamma/beta are multi-dimensional [tvm] via GitHub
• [PR] [FFI][PYTHON] Improve the traceback generation in python [tvm] via GitHub
  • Re: [PR] [FFI][PYTHON] Improve the traceback generation in python [tvm] via GitHub
• [PR] [BugFix][NNAPI] Fix type mismatch and test_mean annotation [tvm] via GitHub
  • Re: [PR] [BugFix][NNAPI] Fix type mismatch and test_mean annotation [tvm] via GitHub
• [PR] [REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style [tvm] via GitHub
  • Re: [PR] [REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style [tvm] via GitHub
  • [PR] [REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style [tvm] via GitHub
  • Re: [PR] [REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style [tvm] via GitHub
  • Re: [PR] [REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style [tvm] via GitHub
• [PR] [Fix][ONNX] Fixed constant ROI handling in resize2d when loading onnx models [tvm] via GitHub
  • Re: [PR] [Fix][ONNX] Fixed constant ROI handling in resize2d when loading onnx models [tvm] via GitHub
  • Re: [PR] [Fix][ONNX] Fixed constant ROI handling in resize2d when loading onnx models [tvm] via GitHub
• [PR] [Fix][ONNX] Fix CumSum conversion when loading ONNX model [tvm] via GitHub
  • Re: [PR] [Fix][ONNX] Fix CumSum conversion when loading ONNX model [tvm] via GitHub
  • Re: [PR] [Fix][ONNX] Fix CumSum conversion when loading ONNX model [tvm] via GitHub
• (tvm) branch nightly updated (c3d3596151 -> 27a314eb73) github-bot
• [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator LayerNormalization: InternalError: Op(relax.nn.layer_norm) requires the input gamma to have as many dimensions as the length of input axes. [tvm] via GitHub
• [I] [Bug] Error converting operator CumSum when 'reverse=1': (ptr) is false: The struct_info is not populated, check if you have normalized the expr [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator CumSum when 'reverse=1': (ptr) is false: The struct_info is not populated, check if you have normalized the expr [tvm] via GitHub
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@27a314eb73d2805d8ffe0c4d6948eeace4f06fd7) tqchen
• (tvm) branch main updated: [MISC] Fix compilation warnings of unnecessary `std::move()` calls (#18130) tqchen
• [PR] [TIR] Add `T.thread_return()` for early thread exit in CUDA kernels [tvm] via GitHub
  • Re: [PR] [TIR] Add `T.thread_return()` for early thread exit in CUDA kernels [tvm] via GitHub
  • Re: [PR] [TIR] Add `T.thread_return()` for early thread exit in CUDA kernels [tvm] via GitHub
• [I] [Bug] Error converting operator Resize: Mismatched type on argument [tvm] via GitHub
  • Re: [I] [Bug] Error converting operator Resize: Mismatched type on argument [tvm] via GitHub
• [I] [Bug] InternalError: Check failed: (offset + needed_size <= this->buffer.size) is false: storage allocation failure [tvm] via GitHub
  • Re: [I] [Bug] InternalError: Check failed: (offset + needed_size <= this->buffer.size) is false: storage allocation failure [tvm] via GitHub
  • Re: [I] [Bug] InternalError: Check failed: (offset + needed_size <= this->buffer.size) is false: storage allocation failure [tvm] via GitHub
  • Re: [I] [Bug] InternalError: Check failed: (offset + needed_size <= this->buffer.size) is false: storage allocation failure [tvm] via GitHub
• [PR] Create test [tvm] via GitHub
  • Re: [PR] Create test [tvm] via GitHub
• [PR] [MISC] Fix compilation warnings of unnecessary `std::move()` calls [tvm] via GitHub
  • Re: [PR] [MISC] Fix compilation warnings of unnecessary `std::move()` calls [tvm] via GitHub
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@c3d3596151e5a4e6efd499232c3839181b22b5c1) tqchen
• (tvm) branch nightly updated (f7b57ec3e5 -> c3d3596151) github-bot
• (tvm) branch main updated: Delete redundant imports (#18129) syfeng
• (tvm) branch main updated: [Target] Support CUDA device function calls (#18055) syfeng
• Re: [I] [Bug] Install on H100 problem about cutlass_fpA_intB_gemm [tvm] via GitHub
  • Re: [I] [Bug] Install on H100 problem about cutlass_fpA_intB_gemm [tvm] via GitHub
• (tvm) branch main updated (f7b57ec3e5 -> 458b0ab5e7) tqchen
• [PR] Delete redundant imports [tvm] via GitHub
  • Re: [PR] Delete redundant imports [tvm] via GitHub
  • Re: [PR] Delete redundant imports [tvm] via GitHub
• (tvm) branch nightly updated (c6969d7238 -> f7b57ec3e5) github-bot
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@f7b57ec3e5a066515b68b54a605972b254282a60) tqchen
• (tvm) branch main updated: Revert "[Refactor] Build cython with isolate environment" (#18127) tqchen
• [PR] [Relax][Doc] Visualize Relax [tvm] via GitHub
  • Re: [PR] [Relax][Doc] Visualize Relax [tvm] via GitHub
  • Re: [PR] [Relax][Doc] Visualize Relax [tvm] via GitHub
  • Re: [PR] [Doc] Visualize the architecture using a UML sequence diagram [tvm] via GitHub
  • Re: [PR] [Doc] Visualize the architecture using a UML sequence diagram [tvm] via GitHub
  • Re: [PR] [Doc] Visualize the architecture using a UML sequence diagram [tvm] via GitHub
  • Re: [PR] [Doc] Visualize the architecture using a UML sequence diagram [tvm] via GitHub
• (tvm) branch revert-18124-fix_0708_cython created (now ff55bb2a0f) tqchen
• [PR] Revert "[Refactor] Build cython with isolate environment" [tvm] via GitHub
  • Re: [PR] Revert "[Refactor] Build cython with isolate environment" [tvm] via GitHub
• (tvm) branch main updated (c6969d7238 -> 6fce4b767a) tqchen
• [I] [Tracking Issue] when TVM can support nvidia blackwell arch? [tvm] via GitHub
  • Re: [I] [Tracking Issue] when TVM can support nvidia blackwell arch? [tvm] via GitHub
  • Re: [I] [Tracking Issue] when TVM can support nvidia blackwell arch? [tvm] via GitHub
  • Re: [I] [Tracking Issue] when TVM can support nvidia blackwell arch? [tvm] via GitHub
  • Re: [I] [Tracking Issue] when TVM can support nvidia blackwell arch? [tvm] via GitHub
• (tvm) branch nightly updated (d9f0838dad -> c6969d7238) github-bot
• [PR] Phase out StackVM runtime support [tvm] via GitHub
  • Re: [PR] Phase out StackVM runtime support [tvm] via GitHub
• Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
  • Re: [PR] [Target] Support CUDA device function calls [tvm] via GitHub
• (tvm) branch main updated: [Refactor] Build cython with isolate environment (#18124) tqchen
• Re: [I] [Bug] TVM FFI encountered a Segfault when calling the invoke_stateful function [tvm] via GitHub
• (tvm) branch main updated: [Relax][ONNX][Transform] Add mode choice, new mode, and warning for take() (#18061) syfeng
• (tvm) branch nightly updated (49b139be8b -> d9f0838dad) github-bot
• [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
  • Re: [PR] [Refactor] Build cython with isolate environment [tvm] via GitHub
• Re: [PR] [MetaSchedule][M3c] Add Per-Store-Feature [tvm] via GitHub
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@d9f0838dad8cdea4f6d6cba361ac35436f4b2f97) tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@351dd8061bf0055ec369c2f94d91b86eb3196349) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@351dd8061bf0055ec369c2f94d91b86eb3196349) tqchen
• (tvm) branch main updated: [KVCache] Fix kernel dispatch based on attention kinds (#18122) yongwww
• (tvm) branch main updated (045eb5bc9c -> 351dd8061b) yongwww
• [PR] [Codegen] Update LLVM version requirement for `insertDeclare` [tvm] via GitHub
  • Re: [PR] [Codegen] Update LLVM version requirement for `insertDeclare` [tvm] via GitHub
• [PR] [KVCache] Fix kernel dispatch based on attention kinds [tvm] via GitHub
  • Re: [PR] [KVCache] Fix kernel dispatch based on attention kinds [tvm] via GitHub
• Re: [I] [Bug] Direct function calls fail when using VM over RPC [tvm] via GitHub
  • Re: [I] [Bug] Direct function calls fail when using VM over RPC [tvm] via GitHub
• (tvm) branch main updated (11f8606ef4 -> 045eb5bc9c) syfeng
  • (tvm) 02/02: [release] Update version to 0.22.dev0 on main branch syfeng
  • (tvm) 01/02: [release] Update version to 0.21.0 on main branch syfeng
• (tvm) branch main updated: Revert "[FFI] Replace `Arg2Str` with a more powerful `for_each`" (#18121) syfeng
• Re: [PR] [release][Dont Squash] Update version to 0.21.0 and 0.22.0.dev on main branch [tvm] via GitHub
  • Re: [PR] [release][Dont Squash] Update version to 0.21.0 and 0.22.0.dev on main branch [tvm] via GitHub
  • Re: [PR] [release][Dont Squash] Update version to 0.21.0 and 0.22.0.dev on main branch [tvm] via GitHub
  • Re: [PR] [release][Dont Squash] Update version to 0.21.0 and 0.22.0.dev on main branch [tvm] via GitHub
• [PR] Revert "[FFI] Replace `Arg2Str` with a more powerful `for_each`" [tvm] via GitHub
  • Re: [PR] Revert "[FFI] Replace `Arg2Str` with a more powerful `for_each`" [tvm] via GitHub
• (tvm) branch revert-18117-ffi-for-each-2 created (now 94969340f5) tqchen
  • (tvm) 01/01: Revert "[FFI] Replace `Arg2Str` with a more powerful `for_each` (#18117)" tqchen
• Re: [PR] [Relax][ONNX][Transform] Add mode choice, new mode, and warning for take() [tvm] via GitHub
  • Re: [PR] [Relax][ONNX][Transform] Add mode choice, new mode, and warning for take() [tvm] via GitHub
  • Re: [PR] [Relax][ONNX][Transform] Add mode choice, new mode, and warning for take() [tvm] via GitHub
  • Re: [PR] [Relax][ONNX][Transform] Add mode choice, new mode, and warning for take() [tvm] via GitHub
• (tvm) branch nightly updated (8e478f57b9 -> 49b139be8b) github-bot
• Re: [I] [Bug] Check failed: (reporter->AssertEQ(data->shape[data->shape.size() - 1], weight->shape[1])) is false: DenseRel: input dimension doesn’t match, data shape=[1, 512], weight shape=[512, 1000] [tvm] via GitHub
• [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Test] Fix tuple pattern in unittest `tests/python/relax/test_dataflow_rewriter.py` [tvm] via GitHub
  • Re: [PR] [Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode` [tvm] via GitHub
  • Re: [PR] [Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode` [tvm] via GitHub
  • Re: [PR] [Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode` [tvm] via GitHub
  • Re: [PR] [Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode` [tvm] via GitHub
  • Re: [PR] [Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode` [tvm] via GitHub
• [PR] [Relax] Fix a bug that occurred due to shape inference not handling static dim vs symbolic dim [tvm] via GitHub
  • Re: [PR] [Relax] Fix a bug that occurred due to shape inference not handling static dim vs symbolic dim [tvm] via GitHub
  • Re: [PR] [Relax] Fix a bug that occurred due to shape inference not handling static dim vs symbolic dim [tvm] via GitHub
  • Re: [PR] [Relax] Fix a bug that occurred due to shape inference not handling static dim vs symbolic dim [tvm] via GitHub
• (tvm) branch main updated: [FFI] Replace `Arg2Str` with a more powerful `for_each` (#18117) syfeng
• (tvm) branch main updated: bump cutlass_fpA_intB_gemm (#18118) syfeng
• (tvm) branch nightly updated (a4dc883b19 -> 8e478f57b9) github-bot
• [PR] bump cutlass_fpA_intB_gemm [tvm] via GitHub
  • Re: [PR] bump cutlass_fpA_intB_gemm [tvm] via GitHub
• [PR] [FFI] Replace Arg2Str with a more powerful `for_each` [tvm] via GitHub
  • Re: [PR] [FFI] Replace `Arg2Str` with a more powerful `for_each` [tvm] via GitHub
  • Re: [PR] [FFI] Replace `Arg2Str` with a more powerful `for_each` [tvm] via GitHub
  • Re: [PR] [FFI] Replace `Arg2Str` with a more powerful `for_each` [tvm] via GitHub
• (tvm) branch main updated: [FFI] Use fold expression to simplify for_each (#18116) syfeng
• (tvm) branch nightly updated (ed4c92c5ce -> a4dc883b19) github-bot
• [PR] [FFI] Use fold expression to simplify for_each [tvm] via GitHub
  • Re: [PR] [FFI] Use fold expression to simplify for_each [tvm] via GitHub
• (tvm) branch main updated: [FFI] Replace `__attribute__` with C++ standard attributes (#18114) tqchen
• (tvm) branch main updated: [Fix][Serialization] Add support for NaN value serialization (#18115) tqchen
• Re: [PR] [Fix][Serialization] Add support for NaN value serialization [tvm] via GitHub
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@2b6b9e442d3baa8e80f267c5a17d9d5be06b7648) tqchen
• (tvm) branch main updated: [CMake] Refine C++/CUDA standard settings in CMakeLists.txt (#18113) syfeng
• Re: [PR] [CMake] Refine C++/CUDA standard settings in CMakeLists.txt [tvm] via GitHub
  • Re: [PR] [CMake] Refine C++/CUDA standard settings in CMakeLists.txt [tvm] via GitHub
• Re: [PR] [FFI] Replace `__attribute__` with C++ standard attributes [tvm] via GitHub
  • Re: [PR] [FFI] Replace `__attribute__` with C++ standard attributes [tvm] via GitHub
• (tvm) branch main updated: [FFI] Cleanup visit_attrs attribute after refactor (#18112) syfeng

• Later messages