discuss-archive  

Messages by Thread

• • Re: [PR] [REFACTOR][IR] Inline ApplyPassToFunction into relax decompose_ops, delete the util [tvm] via GitHub
• [PR] [REFACTOR] Move src/ir/script_printer.cc to src/script/printer/ [tvm] via GitHub
  • Re: [PR] [REFACTOR] Move src/ir/script_printer.cc to src/script/printer/ [tvm] via GitHub
• [GH] (tvm/main): Workflow run "Ping Reviewers" is working again! GitBox
• [I] Support in Hugging Face kernels [tvm-ffi] via GitHub
  • Re: [I] Support in Hugging Face kernels [tvm-ffi] via GitHub
  • Re: [I] Support in Hugging Face kernels [tvm-ffi] via GitHub
• [PR] [REFACTOR][S-TIR] Merge shared memory per thread_extent scope [tvm] via GitHub
  • Re: [PR] [REFACTOR][S-TIR] Merge shared memory per thread_extent scope [tvm] via GitHub
  • Re: [PR] [REFACTOR][S-TIR] Merge shared memory per thread_extent scope [tvm] via GitHub
  • Re: [PR] [REFACTOR][S-TIR] Merge shared memory per thread_extent scope [tvm] via GitHub
• [GH] (tvm/main): Workflow run "Ping Reviewers" failed! GitBox
  • [GH] (tvm/main): Workflow run "Ping Reviewers" failed! GitBox
• [I] [Bug] TensorRT 10 compatibility issues in Relax TensorRT BYOC (TVM 0.24, CUDA 12.4+) [tvm] via GitHub
  • Re: [I] [Bug] TensorRT 10 compatibility issues in Relax TensorRT BYOC (TVM 0.24, CUDA 12.4+) [tvm] via GitHub
  • Re: [I] [Bug] TensorRT 10 compatibility issues in Relax TensorRT BYOC (TVM 0.24, CUDA 12.4+) [tvm] via GitHub
• Re: [I] [Tracking Issue][TFLite] Support quantized operator import in Relax frontend [tvm] via GitHub
• [GH] (tvm/tvm-attrs-h-cleanup): Workflow run "Lint" failed! GitBox
• [PR] [BugFix][Vulkan][CodeGen] Change OpControlBarrier to AcquireRelease. [tvm] via GitHub
  • Re: [PR] [BugFix][Vulkan][CodeGen] Change OpControlBarrier to AcquireRelease [tvm] via GitHub
  • Re: [PR] [BugFix][Vulkan][CodeGen] Change OpControlBarrier to AcquireRelease [tvm] via GitHub
  • Re: [PR] [BugFix][Vulkan][CodeGen] Change OpControlBarrier to AcquireRelease [tvm] via GitHub
• [GH] (tvm/fix226): Workflow run "CI" is working again! GitBox
• [PR] [REFACTOR][IR] Cleanup attrs.h: drop NullValue, AttrsNodeReflAdapter, legacy BaseAttrsNode methods [tvm] via GitHub
  • Re: [PR] [REFACTOR][IR] Cleanup attrs.h: drop NullValue, AttrsNodeReflAdapter, legacy BaseAttrsNode methods [tvm] via GitHub
  • Re: [PR] [REFACTOR][IR] Cleanup attrs.h: drop NullValue, AttrsNodeReflAdapter, legacy BaseAttrsNode methods [tvm] via GitHub
• [PR] [Docs] Reorganize development guide content [tvm] via GitHub
  • Re: [PR] [Docs] Reorganize development guide content [tvm] via GitHub
  • Re: [PR] [Docs] Reorganize development guide content [tvm] via GitHub
  • Re: [PR] [Docs] Reorganize development guide content [tvm] via GitHub
  • Re: [PR] [Docs] Reorganize development guide content [tvm] via GitHub
• [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR] Tie AnnotateDeviceRegions/SplitHostDevice/LowerDeviceKernelLaunch together [tvm] via GitHub
• [PR] [REFACTOR][TIR][ARITH] Phase out ControlFlowGraph, NarrowPredicateExpression, and rename Simplify to StmtSimplify [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR][ARITH] Phase out ControlFlowGraph, NarrowPredicateExpression, and rename Simplify to StmtSimplify [tvm] via GitHub
  • Re: [PR] [REFACTOR][TIR][ARITH] Phase out ControlFlowGraph, NarrowPredicateExpression, and rename Simplify to StmtSimplify [tvm] via GitHub
• [GH] (tvm/cos-refinement): Workflow run "CI" failed! GitBox
• [PR] [REFACTOR][RELAX] Fold CalleeCollector into relax DeadCodeElimination [tvm] via GitHub
  • Re: [PR] [REFACTOR][RELAX] Fold CalleeCollector into relax DeadCodeElimination [tvm] via GitHub
  • Re: [PR] [REFACTOR][RELAX] Fold CalleeCollector into relax DeadCodeElimination [tvm] via GitHub
• [GH] (tvm/fix226): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/cos-refinement): Workflow run "Lint" failed! GitBox
• [PR] Fix PytestUnknownMarkWarning: Unknown pytest.mark.adreno_clml [tvm] via GitHub
  • Re: [PR] Fix PytestUnknownMarkWarning: Unknown pytest.mark.adreno_clml [tvm] via GitHub
  • Re: [PR] Fix PytestUnknownMarkWarning: Unknown pytest.mark.adreno_clml [tvm] via GitHub
  • Re: [PR] Fix PytestUnknownMarkWarning: Unknown pytest.mark.adreno_clml [tvm] via GitHub
• [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add UNIDIRECTIONAL_SEQUENCE_RNN converter [tvm] via GitHub
• [PR] [CI] Add TIRX tests to CI [tvm] via GitHub
  • Re: [PR] [CI] Add TIRX tests to CI [tvm] via GitHub
  • Re: [PR] [CI] Add TIRX tests to CI [tvm] via GitHub
  • Re: [PR] [CI] Add TIRX tests to CI [tvm] via GitHub
  • Re: [PR] [CI] Add TIRX tests to CI [tvm] via GitHub
• [PR] [TIRx][Script] Canonicalize buffer-like tile op arguments [tvm] via GitHub
  • Re: [PR] [TIRx][Script] Canonicalize buffer-like tile op arguments [tvm] via GitHub
  • Re: [PR] [TIRx][Script] Canonicalize buffer-like tile op arguments [tvm] via GitHub
  • Re: [PR] [TIRx][Script] Canonicalize buffer-like tile op arguments [tvm] via GitHub
  • Re: [PR] [TIRx][Script] Canonicalize buffer-like tile op arguments [tvm] via GitHub
• [PR] [TIRX] Expose analyzer-aware Python stmt/expr functors [tvm] via GitHub
  • Re: [PR] [TIRX] Expose analyzer-aware Python stmt/expr functors [tvm] via GitHub
  • Re: [PR] [TIRX] Expose analyzer-aware Python stmt/expr functors [tvm] via GitHub
  • Re: [PR] [TIRX] Expose analyzer-aware Python stmt/expr functors [tvm] via GitHub
• [GH] (tvm/fix222): Workflow run "CI" is working again! GitBox
• [GH] (tvm/fix222): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/fix222): Workflow run "Lint" failed! GitBox
• [PR] [IR] Add annotations to Call nodes [tvm] via GitHub
  • Re: [PR] [IR] Add annotations to Call nodes [tvm] via GitHub
  • Re: [PR] [IR] Add annotations to Call nodes [tvm] via GitHub
• [GH] (tvm/pass1): Workflow run "CI" is working again! GitBox
• Re: [PR] [Relax][Frontend][TFLite] Add bucketize/unique/sign + fixes [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add bucketize/unique/sign + fixes [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Add bucketize/unique/sign + fixes [tvm] via GitHub
• [GH] (tvm/pass1): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/pass1): Workflow run "Lint" failed! GitBox
• [PR] [CodeGen][CUDA] Move fast math intrinsic lowering option to PassContext [tvm] via GitHub
  • Re: [PR] [CodeGen][CUDA] Move fast math intrinsic lowering option to PassContext [tvm] via GitHub
• [GH] (tvm/tvm-for-cooperative-tensor): Workflow run "CI" failed! GitBox
• [PR] [Metal] Enable Metal 4 shader compilation [tvm] via GitHub
  • Re: [PR] [Metal] Enable Metal 4 shader compilation [tvm] via GitHub
  • Re: [PR] [Metal] Enable Metal 4 shader compilation [tvm] via GitHub
  • Re: [PR] [Metal] Enable Metal 4 shader compilation [tvm] via GitHub
  • Re: [PR] [Metal] Enable Metal 4 shader compilation [tvm] via GitHub
• [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm/main): Workflow run "CI" is working again! GitBox
• [GH] (tvm/tflite_quantization_params): Workflow run "CI" failed! GitBox
  • [GH] (tvm/tflite_quantization_params): Workflow run "CI" failed! GitBox
• [GH] (tvm/tflite_quantization_params): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/runtime-modularize): Workflow run "CI" is working again! GitBox
• [GH] (tvm/runtime-modularize): Workflow run "Lint" is working again! GitBox
• [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
  • [GH] (tvm-ffi/main): Workflow run "CI" is working again! GitBox
• [GH] (tvm/runtime-modularize): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "Lint" failed! GitBox
• [GH] (tvm/runtime-modularize): Workflow run "PR" is working again! GitBox
• [GH] (tvm/runtime-modularize): Workflow run "CI" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "CI" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "CI" failed! GitBox
  • [GH] (tvm/runtime-modularize): Workflow run "CI" failed! GitBox
• [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
  • Re: [PR] [BUILD] Modularize device runtime into per-backend DSOs [tvm] via GitHub
• Re: [I] [Bug] Relax ONNX Gather mishandles negative indices [tvm] via GitHub
• [I] [Bug] TVM ONNX LayerNormalization lacks numerical stability for large float32 inputs [tvm] via GitHub
  • Re: [I] [Bug] TVM ONNX LayerNormalization lacks numerical stability for large float32 inputs [tvm] via GitHub
• [GH] (tvm/issue-19585): Workflow run "CI" is working again! GitBox
• [GH] (tvm/issue-19585): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/issue-19585): Workflow run "CI" failed! GitBox
• [GH] (tvm/issue-19585): Workflow run "Lint" failed! GitBox
• [PR] [RPC][Tracker] Reject invalid tracker message sizes and consume frame header [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Reject invalid tracker message sizes and consume frame header [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Reject invalid tracker message sizes and consume frame header [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Reject invalid tracker message sizes and consume frame header [tvm] via GitHub
• Re: [PR] perf: improve field access overhead [tvm-ffi] via GitHub
• Re: [PR] doc: Add version switcher [tvm-ffi] via GitHub
• Re: [PR] chore(skill): Add skills to better develop with TVM-FFI [tvm-ffi] via GitHub
• Re: [PR] feat(python): add Function.__init__ constructor for direct callable wrapping [tvm-ffi] via GitHub
• [PR] [Relay/ONNX] Add RMSNormalization converter for ONNX opset 23 [tvm] via GitHub
  • Re: [PR] [Relay/ONNX] Add RMSNormalization converter for ONNX opset 23 [tvm] via GitHub
  • Re: [PR] [Relay/ONNX] Add RMSNormalization converter for ONNX opset 23 [tvm] via GitHub
  • Re: [PR] [Relay/ONNX] Add RMSNormalization converter for ONNX opset 23 [tvm] via GitHub
• [GH] (tvm-ffi/kathryncat/visitor-mutator): Workflow run "CI" failed! GitBox
• [PR] wip: visitor/mutator [tvm-ffi] via GitHub
  • Re: [PR] wip: visitor/mutator [tvm-ffi] via GitHub
  • Re: [PR] wip: visitor/mutator [tvm-ffi] via GitHub
• [GH] (tvm/issue-19575): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/issue-19575): Workflow run "Lint" failed! GitBox
• [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
  • Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
  • Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
  • Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
  • Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
  • Re: [PR] [TIR] Fixup max/min with input containing NAN [tvm] via GitHub
• [PR] [Relax] Normalize negative concat axis in `ReorderPermuteDimsAfterConcat` [tvm] via GitHub
  • Re: [PR] [Relax] Normalize negative concat axis in `ReorderPermuteDimsAfterConcat` [tvm] via GitHub
  • Re: [PR] [Relax] Normalize negative concat axis in ReorderPermuteDimsAfterConcat [tvm] via GitHub
  • Re: [PR] [Relax] Normalize negative concat axis in ReorderPermuteDimsAfterConcat [tvm] via GitHub
• [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
  • [GH] (tvm-ffi/pyobject): Workflow run "CI" failed! GitBox
• [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub
  • Re: [PR] [FEAT][Python] Tie Python wrapper lifetime to underlying C++ FFI object [tvm-ffi] via GitHub

• Earlier messages
• Later messages