discuss-archive  

Messages by Thread

• • Re: [I] Metal support: MTLBuffer access and current_work_stream for kDLMetal [tvm-ffi] via GitHub
• [PR] [FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [POC][FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [POC][FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [POC][FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
  • Re: [PR] [POC][FIX] Zero cause_chain and extra_context in SafeCallContext to prevent SIGBUS on macOS arm64 [tvm-ffi] via GitHub
• [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
  • [GH] (tvm/main): Workflow run "Lint" is working again! GitBox
• Re: [I] [Bug] Relax ONNX Atan returns 0 instead of ±pi/2 for huge finite float32 inputs [tvm] via GitHub
• Re: [I] [Bug] Relax ONNX Asinh returns inf instead of finite values for huge float32 inputs [tvm] via GitHub
• Re: [I] [Bug] Relax ONNX Sinh/Cosh overflow to inf while ONNX Runtime returns finite values [tvm] via GitHub
• Re: [I] [Bug] Relax ONNX Erf returns zero for tiny non-zero float32 inputs [tvm] via GitHub
• [GH] (tvm/stablehlo_tflite_ops_pr2): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/stablehlo_tflite_ops_pr2): Workflow run "Lint" failed! GitBox
• [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][TFLite] Support StableHLO region-based ops and multi-subgraph models [tvm] via GitHub
• Re: [I] [CUDA][TIR] Standard math intrinsics should not lower to fast-math `__*f` by default [tvm] via GitHub
  • Re: [I] [CUDA][TIR] Standard math intrinsics should not lower to fast-math `__*f` by default [tvm] via GitHub
• Re: [PR] [RFC][CodeGen][CUDA]: Gate fast math intrinsic lowering behind target option [tvm] via GitHub
• Re: [I] [Bug] Segmentation fault when importing `torchmetrics` after creating a TVM CUDA target (LLVM initialization / COFF OptTable) [tvm] via GitHub
• Re: [I] [Bug] Relax ONNX Asin/Acos lose precision for mid-range float32 inputs [tvm] via GitHub
• Re: [PR] [BugFix][Target][LLVM] Use libm for asin/acos instead of buggy inline Taylor [tvm] via GitHub
• Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
  • Re: [I] [Tracking Issue][TFLite] Remaining builtin operator coverage beyond #19412 [tvm] via GitHub
• [GH] (tvm/upstream/arith-knowns-by-key): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/upstream/arith-knowns-by-key): Workflow run "Lint" failed! GitBox
• [PR] [RPC][Tracker] Bound msg_size to MAX_TRACKER_MSG_BYTES to prevent unbounded buffer growth [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Bound msg_size to MAX_TRACKER_MSG_BYTES to prevent unbounded buffer growth [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Bound msg_size to MAX_TRACKER_MSG_BYTES to prevent unbounded buffer growth [tvm] via GitHub
  • Re: [PR] [RPC][Tracker] Bound msg_size to MAX_TRACKER_MSG_BYTES to prevent unbounded buffer growth [tvm] via GitHub
• [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
  • Re: [I] [Bug][RPC] RPC tracker buffers unbounded data on a single TCP connection [tvm] via GitHub
• [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
  • Re: [PR] [ARITH] Index TransitiveComparisonAnalyzer knowns by Key [tvm] via GitHub
• [PR] [TVMScript] Handle undefined functions when dumping IRModule [tvm] via GitHub
  • Re: [PR] [TVMScript] Handle undefined functions when dumping IRModule [tvm] via GitHub
  • Re: [PR] [TVMScript] Handle undefined functions when dumping IRModule [tvm] via GitHub
  • Re: [PR] [TVMScript] Handle undefined functions when dumping IRModule [tvm] via GitHub
• Re: [PR] [WebGPU] Fix UI jank caused by compositor starvation in batched dispatch [tvm] via GitHub
  • Re: [PR] [WebGPU] Fix UI jank caused by compositor starvation in batched dispatch [tvm] via GitHub
• [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" is working again! GitBox
• Re: [PR] [BugFix][Target][LLVM] Route sinh/cosh/atan/asinh/erf through libm extern [tvm] via GitHub
  • Re: [PR] [BugFix][Target][LLVM] Route sinh/cosh/atan/asinh/erf through libm extern [tvm] via GitHub
  • Re: [PR] [BugFix][Target][LLVM] Route sinh/cosh/atan/asinh/erf through libm extern [tvm] via GitHub
  • Re: [PR] [BugFix][Target][LLVM] Route sinh/cosh/atan/asinh/erf through libm extern [tvm] via GitHub
• [I] [Bug] layer_norm` Relax/TIR module crashes the process on Windows [tvm] via GitHub
• [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
  • [GH] (tvm/main): Workflow run "tvm-bot" failed! GitBox
• [GH] (tvm/tirx): Workflow run "CI" is working again! GitBox
• [GH] (tvm/tirx): Workflow run "CI" failed! GitBox
  • [GH] (tvm/tirx): Workflow run "CI" failed! GitBox
• [GH] (tvm/tirx): Workflow run "Lint" is working again! GitBox
• [PR] Add TIRX support [tvm] via GitHub
  • Re: [PR] Add TIRx support [tvm] via GitHub
  • Re: [PR] [TIRx] Bringup TIRx Infrastructure [tvm] via GitHub
  • Re: [PR] [TIRx] Bringup TIRx Infrastructure [tvm] via GitHub
  • Re: [PR] [TIRx] Bringup TIRx Infrastructure [tvm] via GitHub
• [GH] (tvm/xnnpack-byoc): Workflow run "Lint" failed! GitBox
  • [GH] (tvm/xnnpack-byoc): Workflow run "Lint" failed! GitBox
• [PR] [WIP][DO NOT MERGE] XNNPACK BYOC backend for Relax CPU inference [tvm] via GitHub
  • Re: [PR] [WIP][DO NOT MERGE] XNNPACK BYOC backend for Relax CPU inference [tvm] via GitHub
  • Re: [PR] [WIP][DO NOT MERGE] XNNPACK BYOC backend for Relax CPU inference [tvm] via GitHub
• [I] [Bug] maximum/minimum/relu/clip do not propagate NaN (IEEE 754 violation) [tvm] via GitHub
  • Re: [I] [Bug] maximum/minimum/relu/clip do not propagate NaN (IEEE 754 violation) [tvm] via GitHub
• [I] [Bug] Matmul and reductions with zero-size dimension return uninitialized memory [tvm] via GitHub
• [I] [Bug] DataflowUseInplaceCalls corrupts concat input via in-place buffer reuse [tvm] via GitHub
  • Re: [I] [Bug] DataflowUseInplaceCalls corrupts concat input via in-place buffer reuse [tvm] via GitHub
• [I] [Bug] AdjustMatmulOrder crashes on batched matmul chains (3D @ 2D) [tvm] via GitHub
• [I] [Bug] ReorderPermuteDimsAfterConcat crashes with negative concat axis [tvm] via GitHub
  • Re: [I] [Bug] ReorderPermuteDimsAfterConcat crashes with negative concat axis [tvm] via GitHub
• [I] [Bug] LegalizeOps batch_norm uses training-mode computation, ignoring provided running stats [tvm] via GitHub
  • Re: [I] [Bug] LegalizeOps batch_norm uses training-mode computation, ignoring provided running stats [tvm] via GitHub
  • Re: [I] [Bug] LegalizeOps batch_norm uses training-mode computation, ignoring provided running stats [tvm] via GitHub
• [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Fix TopK scalar K extraction in from_onnx [tvm] via GitHub
• Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
  • Re: [PR] [Web][COS] Persist URL→hash mapping across page loads [tvm] via GitHub
• [GH] (tvm/issue-19541): Workflow run "CI" is working again! GitBox
• [GH] (tvm/issue-19541): Workflow run "Lint" is working again! GitBox
• [GH] (tvm/issue-19541): Workflow run "Lint" failed! GitBox
• Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
  • Re: [PR] [Relax][ONNX] Prevent `Div` divide-by-zero crashes [tvm] via GitHub
• Re: [I] [Bug] NPE since 0.1.8 [tvm-ffi] via GitHub
  • Re: [I] [Bug] NPE since 0.1.8 [tvm-ffi] via GitHub
  • Re: [I] [Bug] NPE since 0.1.8 [tvm-ffi] via GitHub
  • Re: [I] [Bug] NPE since 0.1.8 [tvm-ffi] via GitHub
  • Re: [I] [Bug] NPE since 0.1.8 [tvm-ffi] via GitHub

• Later messages