You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2024/05/23 23:27:13 UTC
(tvm) branch ci-docker-staging updated (63697b59be -> 41a13e7f5f)
This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch ci-docker-staging
in repository https://gitbox.apache.org/repos/asf/tvm.git
discard 63697b59be [CI] Set NVIDIA_DISABLE_REQUIRE=true
add 70d80cc126 [Unity] Relax VM (#13878)
add 25e6174770 [Unity] Relax expressions and types (#13901)
add 51cf110c0c [Unity][IR] First-class StructInfo (#13907)
add 40c43108c2 [Unity][CI] Unity specific jenkins setup (do not upstream to main) (#13910)
add e68ef58c84 [Unity] Basic StructInfo Analysis and Expr construction (#13916)
add aa59d86ac7 [Unity] Relax BlockBuilder and ExprMutator (#13926)
add 540ba28f5c [Unity] Relax TVMScript Parser. (#13932)
add 334fd8be51 [Unity] Relax TVMScript Printer (#13944)
add 574084d340 [Unity] Relax VM codegen (#13954)
add 518a7bdd96 [Unity] Relax VM shape lowering pass (#13956)
add b3d913a737 [Unity] e2e Relax minimum build flow (#13961)
add caad7cdf48 [Unity][TVMScript] Use explicit `R.shape` in TVMScript (#13979)
add 903628f106 [Unity] Relax op: index (#13987)
add 0319939e47 [Unity] Relax op: datatype (#13986)
add 11e5e1d6d4 [Unity] Relax op: set (#13990)
add 128042204c [Unity] Relax op: image (#13994)
add 64f281749a [Unity] Relax op: arithmetic, comparison (#13983)
add be4373a1d2 [Unity] Relax op: statistical (#13991)
add fa3cfbd60e [Unity] Relax op: neural networks (#13993)
add 45e6d0bade [Unity] Relax op: creation (#13984)
add 206762f0e6 [Unity] Relax op: linear algebra (#13988)
add 088afc5d54 [Unity] Relax op: search (#13992)
add 4430d26394 [Unity] Relax op: manipulation (#13989)
add 471ff84280 [Unity] NestedMsg Support utility (#13995)
add 604754a6c8 [Unity][Pass] Operator Fusion Passes (#14001)
add 50f616c95c [Unity][Pass] LambdaLift pass (#14012)
add 0a0166e81f [Unity][VM] Supporting "compiled" exec mode. (#14015)
add 660a1f361c [Unity][Pass] BindParams pass, FoldConstant pass (#14016)
add 30a5669b8a [Unity][Pass][TuningAPI] Introduce TuningAPI and MetaSchedule pass (#14014)
add b792751aaa [Unity] Relay -> Relax translator (#14026)
add 554e609aca [Unity][Pass] Normalize Pass (#14031)
add 477d011093 [Unity][BlockBuilder] CallTE convert PrimValue args (#14028)
add f61c330ec6 [Unity][Pass] Wellformed Analysis (#14032)
add 4df587aba6 [Unity][TVMScript] Move tir/relax import in script out of __init__.py (#14033)
add d2b8ea330d [Unity][Pass] Operator legalization (#14029)
add e0c6ca0383 [Unity][Op] Add ShapeExpr Tests for Reshape Op (#14035)
add e15f59bcf0 [Unity] Initial PyTorch Frontend (#14037)
add e8ebd68e92 [Unity][Pass] Block-level static memory planning (#14038)
add 2e1d85149b [Unity] Disallow inline prim_func in relax IR (#14040)
add 60d3796e20 [Unity] Update tests to adapt to latest TVMScript syntax (#14039)
add 7d70af3723 [Unity] Relax dataflow pattern language (matching) (#14041)
add 44b05d68bb [Unity] Statement rewriter for DataflowBlock (#14043)
add 7a71e53b69 [Unity][Pass] FuseOps FuseTIR fixes (#14044)
add e7cc2876ec [Unity][TVMScript] Overload `__neg__` for relax expr (#14045)
add 5239022d97 [Unity][VM] Add per-op profiling support (#14053)
add 38178a3298 [Unity][BYOC] Add pattern-based partitioning pass (#14054)
add a7e936fca1 [Unity] Relax op: collapse sum (#14059)
add 7ce0a519e7 [Unity][Fix][Pass] Fix FuseOps for lack graph edges (#14058)
add 34d5944135 [Unity][Pass] Remove Unused Function (#14061)
add 2c5ccfa7a6 [Unity][BYOC] Add pass to merge composite functions to offload large subgraphs (#14062)
add 7f5f816084 [Unity][Frontend] Annotate number of non-static input of FX function (#14067)
add 73fe88d04a [Unity][Transform] Add LiftTransformParams pass (#14069)
add b137d22ed4 [Unity][BYOC][Pass] RunCodegen and TensorRT (#14078)
add 38f920d480 [Unity][Pass] Canonicalize Bindings (#14079)
add 2a51dcae46 [Unity] Add testcases for `expr_args_converter` (#14080)
add 026217d998 [Unity][BYOC] Add CUTLASS backend (#14081)
add 4365d1e29d [Unity][BYOC] Add DNNL backend (#14082)
add 2409be1934 [Unity][Op] `log_softmax` and `cross_entropy_with_logits` (#14083)
add 9caab4d9d8 [Unity][Analysis] TIR pattern kind analysis for multi-buffer write block (#14075)
add 872c9f23c6 [Unity][Fix][Pass] FoldConstant with DCE in dataflow block (#14087)
add 6b87e35617 [Unity] Refactor Relax Build JIT UX (#14088)
add 3959707f19 [Unity][Relax] Set Shape Function to Be Host Function (#14090)
add 79fe0a22aa [Unity] Fix typo in the comment (#14096)
add 779c54d9ad [Unity] Lower `shape_of` to a builtin (#14093)
add 3173f2855d [Unity] Relax Recursive function (#14092)
add 62a4c735fb [Unity][Layout] Add layout transformation analysis for PrimFunc (#14066)
add 1bf8b63648 [Unity] Remove attributes of relax.print, assert and unique (#14101)
add 30d204bb4a [Unity][BYOC]Add relax backend pattern registry (#14106)
add a1f8d72178 [Unity] Update tests again to adapt to latest TVMScript syntax (#14115)
add 6a9caf40e7 [Unity][Fix] Fix bug in MergeCompositeFunctions (#14117)
add 3515f95ba0 [Unity][BlockBuilder] Add `name_hint` argument for `emit` and `emit_output` (#14126)
add c1f304497e [Unity][WEB] Relax vm on web runtime (#14131)
add aaa457d304 [Unity] Add Global info (#14132)
add 837dd6c228 [Unity][BYOC] Add transposed matmul support to Relax CUTLASS BYOC (#14128)
add a4b1ec097c [Unity][TVMScript] emit_te sugar (#14123)
add be170bb3d2 [Unity][BYOC] Assign group to unused bindings and ignroe PrimFunc (#14139)
add 77f4c719f3 [Unity] Add callback to FuseOpsByPattern to check match result is accepted (#14109)
add 1b6d104f68 [Unity][Legalize] Fix Scalar Constant Legalization (#14127)
add 788c12f84f [Unity][Pass] Enhance constant folding to fold relax ops by evaluating them. (#14146)
add f17a7285bb [Unity][Debugging] AST printer (#14152)
add c78e68bf4d [Unity][Pass] Support Symbolic Shape Deduction during BindParam (#14154)
add ab5aaa0d1b [Unity][Analysis] Checking function return struct info in well-formed check (#14155)
add 89cff055d3 [Unity][BYOC] Use Relax legalize + CPU build for reference in tests (#14162)
add 2ca340a30b [Unity] Add bind_constants option to FuseOpsByPattern (#14151)
add 4fe3a6d1f4 [Unity][Analysis] Analysis for detecting recursion in Relax (#14149)
add 2de852f203 [Unity][BYOC] Add batch matmul support to Relax CUTLASS BYOC (#14166)
add 8de0ff78f5 [Unity][Op] Full support of Relax op `power` (#14171)
add 9797924d54 [Unity][Analysis] Restore Python bindings for var analyses (#14180)
add 30dc2365e7 [Unity][OP] Add an operator for fused multi head attention (#14150)
add 279317d4cd [Unity][WEBGPU] Codegen improvements and WebRuntime (#14187)
add 3f66edc8ed [Unity][Transform] LiftTransformParams handling multiple functions (#14192)
add 283a3db675 [Unity][Op] Group normalization (#14194)
add c25659d88f [Unity][Op] Argmax and argmin (#14195)
add 2f114d28b3 [Unity][Op] Legalize `round`, `floor`, `ceil`, `sign` (#14198)
add 012923572a [Unity][Frontend] FX translator supporting more ops (#14196)
add a115d03e76 [Unity][Frontend] FX translator returning weights with `keep_params_as_input` (#14197)
add 27b6fa7cb4 [Unity][Fix] FX translating dtype (#14201)
add 6ab0abc08d [Unity][TIR][Pass] ForceNarrowIndexToInt32 (#14203)
add 3f4417c6db [Unity][Frontend] FX translator support torch.baddbmm (#14202)
add 2d562f8926 [Unity] Introduce Default GPU Schedule Pass (#14182)
add 97dc05ee18 [Unity][Frontend] Attach imported model weights, deprecate ImporterOutput (#14211)
add 033fc0e34b [Unity][Frontend] FX translator supports unwrapping unit return tuple (#14212)
add 9f9cf2967c [Unity][WEB] Improve webgpu codegen options to skip readonly (#14213)
add 8f2153cfb4 [Unity][Frontend] from_fx keeps parameters in order (#14214)
add 30848742cb [Unity][BYOC] Add dynamic shape support to CUTLASS matmul (#14216)
add 73b03dadb9 [Unity][Transform] Memory plan across the IRModule (#14220)
add 780a579c72 [Unity][WEB] Simplify WebGPU Codegen per spec (#14225)
add 3345aeda4a [Unity] ensure memory.alloc_tensor/storage roundtrippable (#14226)
add 30a9403729 [Unity] Improve implementation of FuseOps (#14229)
add 9bc1ecc672 [Unity][Transform] SimplifyNormInference (#14221)
add a6b37b80ce [Unity][Bugfix] Do not include `PrimFunc`s in the dependency graph when checking for recursion (#14228)
add 3e03ca5abe [Unity][TVMScript] Enable Context-Aware Parsing (#14234)
add cea0161bcb [Unity][TVMScript] Fix prim_func lost issue in relax.emit_te (#14189)
add 14a7de57b3 [Unity][WEB] Update text prompts for syntactical correctness (#14237)
add 663c8211a4 [Unity][WEB] Improve ndarray cache (#14236)
add c7c4dee36b [Unity][Op][Tweak] Improve `StructInfo` inference for `shape_of` (#14243)
add 633c9002e3 [Unity][Op] Add repeat, tile, conv2d_transpose, avg_pool2d (#14238)
add 07f7fc0a9c [Unity] Allow user defined func attrs in emit_te (#14255)
add 04d6d15984 [Unity][BYOC] Add conv2d and residual block patterns for Relax cutlass BYOC (#14252)
add df9d084553 [Unity][Frontend] Import `tanh` and fix `layer_norm` (#14247)
add 71c23b7a11 [Unity] Consider target context for Relay to Relax conversion (#14269)
add 70386ecc91 [Unity] Introduce call_dps_packed (#14183)
add c676d1392f [Unity][ci] Use CPU-SMALL instances (#14256)
add 121e1e7a03 [TVMScript][Unity] Improve PyLint Compatibility (#14276)
add 6be5c19c9c [Unity] Mark tests that need python3.8 compact.
add 68827fa1d6 [Unity] Fix StructInfo Infer for `vm.alloc_tensor` (#14283)
add e3f975543f [Unity][Op] Cumsum (#14297)
add 67e183c33f [Unity][DEBUG] Add Instrument (#14302)
add f96e09a9ce [Unity][Pass] Add a pass to alter the TIR implementation of an operator (#14215)
add a46fd90478 [Unity][Op] Enable special dimension value 0 in reshape (#14311)
add e0c4456f50 [Unity][Web] WebGPU explicit max buffer size (#14321)
add 4bfbd3a703 [Unity] Support pattern-based rewriting (#14312)
add 2f1642f128 [Unity][BYOC] Support matmul + residual block fusion in CUTLASS BYOC (#14317)
add 778b343ed9 [Unity][BYOC] Improve expressiveness of the pattern check function in FuseOpsByPattern (#14310)
add 7c5a570dc1 [Unity][Transform] Enhance RewriteDataflowReshape transform (#14265)
add b10e2a2a31 [Unity][Contrib] Introduce several features of cutlass profiler (#14275)
add f67657fe09 [Unity][TOPI] fp16 LayerNorm & GroupNorm (#14264)
add 5cd30b00ef [Unity][Transform] Automatic Layout Conversion (#14257)
add 59746ea9b0 [Unity][Transform] Simple Dead Code Elimination (#14262)
add 371c3a2f2f [Unity][Transform] Automatic Mixed Precision (#14242)
add 691983096c [Unity][Lint] Fix cpplint casting (#14333)
add 5229827888 [Unity][Transform] DefaultSchedule pass (#14266)
add 375a2921ea [Unity][Op] Fix Strided Slice Shape Inference (#14324)
add 920f98d8f9 [Unity][Pass] Reuse prior infra to implement more complete DCE (#14334)
add f4a451d273 [Unity] Remove Python interface of RemoveUnusedFunction (#14336)
add 050562bfe9 [Unity][BYOC] Update testcases to follow recent changes (#14339)
add 7534531a6e [Unity][Frontend] FX exp and strided_slice fix (#14338)
add 021ce24d0c [Unity] Support model kwargs in dynamo_capture_subgraph (#14349)
add fe34a55bee [Unity][BYOC] Check leaked intermediate variables in cutlass patterns (#14350)
add 9ae5895aae [Unity][Transform] AMP out_dtype=float16 testcases (#14358)
add cd48ba9e33 [Unity][Fix] Fix block memory plan to handle bool (#14357)
add 57b90700d2 [Unity][Transform] Introduce data-dependent operation of reshape and its constant folding (#14282)
add 27f0ad3230 [Unity][Transform] Fix AMP tests (#14360)
add 20cc2ff300 [Unity] Add support to append relay op attrs in translator (#14356)
add 79492b81d9 [Unity][WEB] Support async pipeline creation (#14362)
add 7974e0f816 [Unity][Pass] Fix FuseOps error if there is no output of a given group (#14354)
add b5cdcec5d6 [Unity][Fix] Infer Layout must support negative axes (#14365)
add eb7b020c22 [Unity] Add More Ops For FX Translator (#14348)
add 14f7b5af61 [Unity][TVMScript] Update GlobalVar `checked_type_` when `emit_te` (#14367)
add a0b28c486e [Unity][Fix] Allow scalar layout initialization (#14370)
add 9253f0fef1 [Unity] Also include output dtype in simt MathInstruction (#14372)
add 8d8ced72a1 [Unity][VM] Add CUDA graph vm builtins (#14371)
add 983b569325 [Unity] Add missing #include <array> (#14383)
add d6efe9f042 [Unity][Transform] SplitCallTIRByPattern and CUTLASS backend (#14274)
add f7f45dcc8e [Unity] Support simple dynamic-shape-aware fusion (#14396)
add 34b32cdf68 [Unity][Op] Add stop_lift_params (#14368)
add 6eb12ff85a [Unity][TVMScript] Fix Shape Var occurrence in Tensor annotation (#14404)
add ae3cf5f5f1 [Unity][Transform] Common Subexpression Elimination (#14361)
add 6818799e74 [Unity][QNN][Hexagon]Support Relax Constants in the QNN TOPI operations (#14386)
add a8205ed0c1 [Unity][Op] Conv1d (#14388)
add 906f69056d [Unity] Fix getting shapes for cutlass BYOC kernels (#14411)
add 5426085988 [Unity][Op] Expose scale in `R.nn.attention` and add its legalize op (#14412)
add 7f44914317 [Unity][Hexagon] Enable Relax VM for Hexagon (#14415)
add 94b9bbc6bc [Unity][Fix] Copy over module attrs in FuseTIR (#14418)
add 4d29a7cff8 [Unity] Handle extern func calls in static memory planning (#14419)
add e424eea3dd [Unity] Include constant shapes in the profiler result (#14428)
add b1346b97a9 [Unity][Fix] Annotate TIR op pattern could have no stores. (#14420)
add beb25a701c [Unity] Minor updates to DataFlowBlockRewrite (#14431)
add 22c363a72a [Unity] Remove non-deterministic behavior from graph pattern matching (#14417)
add 61f7edc1af [Unity][Graph matching] Automatically add `used-by` constraints for `is_op` pattern (#14439)
add ad4bfbc5f9 [Unity][Op][Docs] Update comment for `call_tir_dyn` (#14441)
add 0a4ce4dfc3 [Unity][Graph matching] Clean up undo stack for parent and child nodes properly (#14440)
add 52f6e639de [Unity] Pattern-based rewriting for dataflow block (#14446)
add 5c8b7af59c [Unity] Fix lint during for upstream black
add 1527bfba04 [TVMScript][Bugfix] Tuple on the RHS of AssignDoc (#14452)
add 97ab25c33f [Unity][BYOC] Faster cutlass codegen (#14465)
add db01567b09 [Unity][Op] introduce `shape_to_tensor` op (#14447)
add ea56542f96 [Unity][CI] Update gpu and lint image (#14473)
add 751da24bce [Unity][BYOC] Fix `RunCodegen` pass on symbolic shape (#14472)
add e54e04d520 [Unity][BYOC] Add cuBLAS backend (#14291)
add 5c2058dc3c [Unity][OP] Add `arange` op (#14463)
add 3b8aa040e4 [Unity][op] Relax rsqrt and sinh (#14479)
add aec773d690 [Unity][VM] Add Attention KV cache builtin (#14478)
add b7d3762f39 [Unity][Hexagon] Don't pass raw scalars in hexagon/qnn/nn.py (#14474)
add 6bcd375369 [Unity] Torch-like NN module enhancement (#14499)
add 1eea30d54a [Unity] LiftTransformParams with symbolic shape robustness (#14500)
add e93ee9fc56 [Unity][Op] vm.alloc_tensor infer struct info (#14503)
add 22878a52fa [Unity][Refactor] Use NameSupply for unique symbol generation and remove NameTable (#14507)
add d9f3aa32ef [Unity][Analysis] Get symbolic TIR vars from struct info (#14509)
add 8e4f94a689 [Unity][Op] Extend `relax.op.take` op to match behavior with `topi.take`. (#14481)
add bd003e847b [Unity] Memory planning with TIR var upper bound (#14511)
add 3bd516701c [Unity] handle bf16 in dump_ndarray_cache and load_ndarray_cache (#14514)
add a92258ec55 [Unity] Support Simple Dynamic-Shape-Aware in FuseTIR (#14515)
add 8474255a04 [Fix] symbolic thread extent program compilation (#14516)
add 603f8bd721 [Unity][Op] Negative Log Likelihood Loss (#14517)
add bb479e66dd [Unity][Transform] Some Improvements on pass DecomposeOps (#14512)
add cea447cf37 [Unity][Graph matching] Improved matching algorithm and implementation (#14501)
add 34695afa5c [Unity][Op] introduce `ScatterElement` op (#14493)
add 8ea976276c [Unity] Properly handle tuple-outputting function in `FuseOpsByPattern` (#14525)
add d5182388a5 [Unity][Op] Gradient functions for high-level Relax operators (#14527)
add 0c447d6f9c [Unity][Transform] High-level reverse-mode automatic differentiation pass (#14542)
add 97f4db0f91 [Unity] DefaultGPUSchedule working for targets other than CUDA (#14540)
add b816aa26db [Unity][Pass] Enhance Dynamic-aware FuseOps (#14543)
add add45b5c1c [Unity] Make FuseOps work on a call_tir which has a ShapeExpr arg (#14553)
add ac645b2348 [Unity] Fix emit_te with symbolic input (#14552)
add d6c8f7933b [Unity] Preserve symbolic var args when applying call_tir (#14555)
add a6831ba9c4 [Unity] Enable pod args in WebGPU (#14560)
add 876ed385ce [Unity][Op] Symbolic shape support of take grad (#14559)
add fadf144f27 [Unity][Op] add einsum and flip in Relax (#14545)
add 8b4dcfd1f1 [Unity][TOPI] Symbolic shape support for `collapse_sum` (#14535)
add c51bd8b76d [Unity] Enhance Dynamic-aware FuseTIR (#14577)
add ddbd3d7ec0 [Unity][TVMScript] Update struct_info for GlobalVar (#14579)
add eeae66b301 [Unity][MetaSchedule] Skip Scheduled PrimFuncs in Task Generation (#14402)
add 88f5b8f590 [Unity][PyTorch] Disable gradient during dynamo subgraph capture to save RAM (#14602)
add 61fbf4262d [Unity] Fix FuseTIR when the same buffer is read multiple times with different access pattern (#14603)
add e416226bd7 [Unity][MetaSchedule] BlockCollector focusing on current func (#14595)
add 7db0b984de [Unity][Op] Dynamic Strided Slice (#14548)
add a59c09de77 [Unity] Add pass for combining parallel matmul (#14583)
add f762b4e833 [MERGE] Bring changes from main into unity 2023-04-12
add 181e559008 [MERGE-FIX] Fix regresions after merge
add 14150f60d6 [Unity][BYOC] Add fused patterns for stacked attention (#14608)
add 4ba82881f5 [Unity] Fix ForceNarrowI32 with pod arguments (#14605)
add 9ddd39be56 [Unity][UX] Symbolic Variables Used in Multiple Functions (#14606)
add 3fd22484cf [Unity][Bugfix] Resolve failure on `test_e2e_op_dynamic.py` (#14616)
add 5135461b53 [Unity][VM] Converting tuple arg to Python tuple (#14620)
add bcad5a9abe [Unity][VM] LibComparator using dtype from input (#14623)
add 89d539d746 [Unity] Update specific builtins for LM (#14617)
add 93945ce640 [Unity][CODEGEN] Fix metal codegen when with only single working dim (#14627)
add d5725a8443 [Unity][CUTLASS] Support batched matmul + residual fusion (#14613)
add 7b3fe3aa08 [Unity][CI] Update images to include jax deps (#14610)
add 6892b01c0b [Unity] hotfix webgpu codegen for vec load (#14630)
add 9daf9acf05 [Unity][Frontend] Some changes on the PyTorch FX Frontend (#14625)
add b3e9adc1cb [Unity] update ci cpu/gpu images (#14631)
add 8569b9fe63 [Unity][AMP] Fix merging concrete type and "unknown" type (#14612)
add 0f05116453 [Unity][MetaSchedule] Add the module_equality param for tune_relax flow (#14537)
add d9d9172cf9 [Unity][TARGET] Updates vulkan codegen for DeclBuffer (#14641)
add ff2f3c861a [Unity] `enable_warning` option for LegalizeOps and MSApplyDatabase (#14634)
add b663c58c6b [Unity] BlockBuilder assigning unique tensor names in call_te (#14632)
add 3138fc6abc [Unity] Improve error message in webgpu request (#14640)
add 81d6778947 [Unity][Frontend] Add `no_bind_return_tuple` for PyTorch FX Translator (#14639)
add 9dd85dd703 Adding powerPreference argument to navigator.gpu.requestAdapter (#14650)
add ccca0f5ecf [Unity][BYOC] Fuse attention pattern with `strided_slice` (#14649)
add ee53619661 [Unity] Improve and reduces possible memory leak RPC debug (#14662)
add f19e6835fe [Unity][BYOC] Add check for stacked attention patterns (#14664)
add bb1e10c81f [Unity] Add rewriting for CUDA graph capturing (#14513)
add f7835a6f80 [Unity][CUTLASS] Require the residual input to have the same shape as input (#14657)
add ec89242fbd [Unity] Update docs for operators (#14659)
add 814afe6b78 [Unity] Improve WebGPU codegen for large grid (#14674)
add 6c662eb631 [Unity] Use custom hash in `BlockBuilder` to avoid hashing large constants (#14675)
add 981e0e8762 [Unity][Training] Optimizer library (#14670)
add b4fbac785b [Unity] Fix `DataflowReshapeRewrite` when input has multiple buffers from tuple (#14669)
add c91fde574c [WebGPU] This PR fixes the webgpu runtime when there is no pod params (#14685)
add a8f3a22cdd [Unity][TuningAPI] Temporary patch for large models (#14691)
add 4b59ee6e9d [Unity] FuseOps skipping PrimValues (#14687)
add bc1597ce2b [Unity][CUTLASS] Fix CUTLASS codegen for occasional variable name conflict (#14692)
add efa8282aa6 [Unity] Use split rather than slice in `CombineParallelMatmul` (#14688)
add 51fa5763aa [Unity][WebGPU] Move NDArrayCache Support to relax runtime (#14689)
add 167bc874e3 [Unity][Training] Loss functions and AppendLoss pass (#14668)
add ee6e26f2cb [Unity][Op] Avoid indices in TIR matmul being 0 in legalization (#14701)
add 19e82ecb3a [Unity] MetaScheduleApplyDatabase using workload from records (#14702)
add d6af4af23e [MERGE] Merge main into unity 2023-04-23
add 129fd8f84f [Unity] Reduce cast to fp32 for constant input in AMP (#14679)
add e47bf761d4 [Unity] Smart parameter fetching (#14708)
add d468419ea0 [Unity][Training] Trainer and SetupTrainer (#14706)
add 24480ea573 [Unity][CUTLASS] Fixed memory leak in attention kernel offload (#14723)
add 84e286be71 [Unity] Cache ndarray-cache.json (#14722)
add 9bff70ab06 [Unity] Allow KVCache Access without Shape (#14726)
add 94e3d512dd [Unity][CUTLASS] Fixed stacked attention offload when QKV reshape uses the same shape expression (#14728)
add 9c36056a12 [Unity] Allow modifying function signature by AMP to accept fp16 inputs (#14719)
add 2232207afd [Unity] Fix ForceNarrowIndexToI32 so it ignores i16 (#14733)
add b40391961a [Unity] Add system lib build option to relax (#14734)
add 9914a77b0f [Unity] Process all Relax functions in CompositeFunctionAnnotator (#14736)
add 126f8d082b [Unity] Add options to MS tuning pass to enable more fine-grained tuning (#14730)
add dd02791e5a [Uniy][Op] Expand support of attention bias layout (#14737)
add 65d99bd751 [Unity][CUTLASS] Add layer norm support (#14731)
add 2bd506fa92 [Unity][BYOC] Support implicit attention patterns (#14744)
add 107ef2b65e [Unity][Relax] Memory planning for call_tir_dyn (#14750)
add 7a13d1e807 Fix super().visit_var_binding_ in PyExprVisitor and PyExprMutator (#14754)
add f9b07ae9ed [Unity][VM] `kill_tensor` and `kill_storage` releasing NDArray in VM at runtime (#14753)
add a0e0f275ab [Unity] Limit number of characters in logger names (#14752)
add bda94c3544 [Unity] Fix FX translator no output issue (#14761)
add e10ddf110d [Unity] Enhance CopyWithNewVars utility (#14764)
add 56516c2bc5 [Unity][CI] Use the upgraded images (#14768)
add 40fab65be3 [Unity] Update LM Sample builtins (#14793)
add 98a25dfec9 [Unity][Pass] Lazy transform params (#14769)
add bf11516632 [Unity][Training] More Relax operators gradient supported (#14777)
add f7eed214d1 [MERGE] Merge main into unity 2023-05-07
add 4795988944 Use latest lint image
add e1c2ed4863 [LINT] Fix clang-format script for newest clang-format
add d837fd0738 [MERGE] Fix lint after lint image upgrade
add 96aca9d85f [Unity] Fix Unary Op Legalization (#14789)
add ff333acaa9 [Unity] Introduce FewShotTuning Pass (#14624)
add 7c56517545 [Unity][BYOC] Fix incorrect bias stride in matmul cutlass offload (#14807)
add 5ad8941252 [Unity] Add pass to allocate big workspace and pass it to all functions that need temp storage (#14802)
add 0c9aa58907 [Unity][BYOC] Add shape validation for bias arg in cuBLAS (#14809)
add aab076e2bf [Unity] NDArray Cache Efficient Load in OpenCL (#14816)
add b5928e775c [Unity] Fix CUDA graph rewrite var used before def (#14800)
add 3b33caf757 [Unity] Cover all Relax functions in implicit attention rewrite (#14818)
add 28567bdc24 [Unity] Improve Error Message loading NDArray Shards (#14823)
add 25a491d52b [Unity][Training] Categorical cross entropy loss (#14757)
add 773620f267 [Unity][CI] Update CPU image to install PyTorch (#14848)
add e2839c0fe8 [Unity][CI] Update CPU image to install PyTorch (fix) (#14852)
add d159f73d2a [MERGE] Merge main into unity 2023-05-14
add e812a219f6 [Unity][FX] Add support for PT2.0 scaled_dot_product_attention (#14841)
add c365a675cd [Unity] Support Mixed-Precision FMA Pattern (#14865)
add 3bcfc293bc [Unity]Lazy transform param now only work on non-dataflow block (#14864)
add cb39cc1a97 [Unity] Support multilib relax build (#14873)
add e4f16cb9e7 [Unity][Relax] gelu-tanh operator (#14814)
add 7aff6bfbb0 [Unity][IR] Purity Tracking (#14394)
add 84a5e72084 [Unity] Fix broken test in cutlass codegen (#14881)
add b3dd058ee3 [Unity][WEB] Enable String object (#14882)
add d3ad96e43c [Unity] Only process relax functions in workspace annotation (#14896)
add 6dd431b430 [TIR][Doc] Fix formatting in pop_trace docstring (#14894)
add 201002e78e [Unity][CUTLASS] Fix for purity tracking (#14891)
add 9a3bbc1d37 [MERGE] merge main into unity 2023-05-20
add 7d2579e62e [Unity][WebGPU] Try F16 support for WebGPU Backend (#14904)
add afa53ded93 [Unity][Transform] Fix scalar case in DefaultGPUSchedule (#14928)
add 6bb531fcb1 [Unity] Allow eliminating only call nodes in CSE pass (#14895)
add 17be2d71ad [Unity] Support causal mask for `R.nn.attention` (#14907)
add 4b072e05f6 [Unity][WebGPU] Fix WebGPU adapter requiring features (#14933)
add eb60d4cac1 [Unity] Cleanup Web runtime wasi (#14929)
add f15f612aa7 [Unity] Improve caching logic in webruntime (#14940)
add 286c6722e0 [Unity] Enhance web cache to add scopes (#14953)
add 66863f53f6 [Unity] Fix MergeCompositeFunctions for non-CallNode dataflow inputs (#14959)
add ec127cb0d6 [Unity][BYOC] Make CUTLASS attention rewriting aware of fp16 <-> f32 casting (#14957)
add dd2f452e4d [Unity][Transform] Fix bug for tir expression in shape in fuse_tir (#14931)
add 7d78b278d3 [Unity] support update KV cache (#14964)
add 240111fe64 [Unity][CUTLASS] Attention hot fix (#14966)
add be9f0781a9 [Unity][Frontend] Translate StableHLO to Relax (#14460)
add cb3bf4014a [Unity][Training] Enhance op gradient (#14932)
add c448c50cd1 [Unity][CUTLASS] Support more residual input shape (#14968)
add 9c0b41bc74 [Unity] Allow filtering out unwanted branches in matmul combining pass (#14971)
add a2bddcf3b2 [Unity] Add popn to kvcache (#14970)
add f796b69a3b [Unity] Fix LazyTransformParams use-def analysis and binding emission (#14974)
add 34e87ae7f8 [Unity] Fix ConvertLayout on binary elemwise ops involving scalar input (#14961)
add 44c849f144 [Unity] Reset match state when backtracking (#14984)
add 0a0b11982e [Unity][Training] Avoid problematic inputs to nll_loss in test_op_gradient_numeric (#14987)
add f360556a0f [Unity][CUTLASS] Fix circular import bug in relax cutlass backend (#15001)
add 153f70aa60 [MERGE] Merge main into unity 2023-06-01
add 2f13f2d3bf [Unity] Fix importing tvm.contrib.cutlass (#15010)
add 6d8fe1c0e3 [Unity] Fix cutlass BYOC after merge (#15012)
add 1a4e08e441 [Unity][FIX] add init file to `relax.backend.contrib` (#15023)
add 3dd9345024 [Unity] Allow name_hint in additional locations (#15027)
add 7dd76cca03 [Unity] Cutlass attention with dynamic sequence length (#15028)
add 91b6cee634 [Unity][BYOC] Cache cuBlasLt handle with thread entry (#15030)
add 0328e98b94 [Unity][NN] Allow nn.Placeholder/Parameter prior to BlockBuilder (#15025)
add 7c15b55aaa [Unity] Add hexp for compute capacity <= 5.2 (#15070)
add e9ddd47a80 [Unity] Add an API to create multiple kv caches with single allocation (#15064)
add 0406b9af22 [Unity][Frontend] Add relax onnx importer and tests (#14999)
add bbe5d6b18c [Unity] Added bounds checking on TupleGetItem index (#15024)
add f61015c20c [Unity] Optimize SampleTopPFromProb (#15072)
add dc1e580e61 [Unity][Relax] Add masked_fill operator (#15077)
add c1aab7d185 [Unity][Relax] Add bitwise and logical ops (AND, NOT, OR, XOR) (#15075)
add 208dd71f18 [Bugfix][CUDA] Fix codegen for hexp for sm >= 52 (#15079)
add 4347b30ffe [Unity] Fix FewShotTuning Failure When Missing Global Symbol (#15097)
add 9eafe7c022 [MERGE] Merge main into unity 2023-06-13
add 4f903d4f30 [MERGE] Fix after merge
add 990c768e9c [Unity][Analysis] Reshape TIR detection with iter-map-simplify (#15099)
add f6050de6f3 [MERGE] recover cutlass in unity
add c68a00a37d [Unity][FuseTIR] Flatten and add tuple fields to parameters / arguments only when they are used (#15113)
add 16158e7cc3 [Unity][Relax][UX] Specify function purity in the @R.function decorator (#15109)
add 121bca6027 [Unity][Relax] Make RewriteDataflowReshape only rewrite volume-preserving ops (#15112)
add 2260bba731 [Unity] Allocate workspace for all functions (#15118)
add 84ce484650 [Unity][BYOC] Integrate fp16 A - int4 B GEMM kernel from FasterTransformer into CUTLASS BYOC (#15111)
add 0a672548e8 [Unity] Minor fix to `RewriteDataflowReshape` condition (#15125)
add cbdee2ff34 [Unity] Hotfix webgpu runtime (#15135)
add 6587faaec0 [Unity][Relax] Generalize CSE to work outside DataflowBlocks (#15047)
add cb451e27d0 [Unity] Fix handling of vm builtins in cuda graph (#15145)
add 6e2889f3c5 [Unity][Bugfix] Fix purity annotation in CSE test (#15143)
add 7d47f7b92d [Unity] Scaffolding DLight (#15141)
add 7c3369c72f [Unity][Pass] FuseOps with partially accessed Tuple param (#15152)
add 01ea438e6e [MERGE] Merge main to unity 2023-06-24
add 730d380ff6 [Fix] Fix merge error
add 7ef7ce7a61 [Unity][Dlight] Add reduction rules (#15156)
add daf9c202ac [Unity][IR][UX] Privacy annotation in Relax (#15140)
add 0dba0e3ed2 [Unity][UX][Tweak] Make it an error to mark a function private and specify a global symbol (#15170)
add ee4e657116 [Unity] Support clear global memory allocators (#15172)
add 7828489c6c Merge remote-tracking branch 'apache-upstream/main' into unity
add 9d27a6aeb8 Fix super() visit function in PyExprVisitor and PyExprMutator (#15189)
add c07f67ceda [RPC] Disable socket SO_REUSEADDR for Windows (#15188)
add f6c138347d [Unity] Legalization for LayoutTransform (#15184)
add e571dc9262 [Unity] Add memory scope and nd allocation support in allocators (#15178)
add 741ca41814 [Unity][Dlight] general reduction rule for gemv-decode (#15169)
add 918fc4ecf7 [Unity][Dlight] Matmul Rules (#15191)
add 40300a3b92 [MERGE] Merge main into unity 2023-07-03
add 1631c3d4ce [MERGE] Fix testcase after merge
add 11db81effd [Unity] Fix dlight reduction rule (#15194)
add 05278ea77b [VM] Add repetition penalty functions to Relax VM (#15219)
add 780d6e6d12 [Unity] Allow specifying struct_info for relax constant (#15220)
add 04f22a9257 [Dlight] Enhance Decode-GEMV Schedule (#15195)
add c45f72b4cc [Unity][TIR][Transform] Support no spatial axes cases for DefaultGPUSchedule (#15232)
add 5a44262502 [Unity] Fix memory statistics issues in estimate_memory_usage (#15224)
add a4fa5e7c59 [Unity][NestedMsg] Add NestedMsgTo helper function (#15223)
add 1637b1436f [Unity][Dlight] Avoid TransformBlockLayout in GEMV Rule (#15248)
add 15a6b475bb [Unity][Dlight] Handle Epilogue Broadcasting (#15252)
add 5828f1e9ee [Unity] Add a Standalone VM Version Number (#15254)
add 2bb0ebbb8b Merge remote-tracking branch 'apache-upstream/main' into unity-staging
add 525f5aee5f [Unity][TIR] Allow symbolic bounds in IndexMap analysis (#15262)
add 399c5eaafe [Unity][Training] Registering te gradient (#15231)
add 6297535440 [MERGE] Hotfix layout transform related change after last merge from main
add b00ed76318 [Unity][Dlight] Minor performance improvement for gemm and gemv (#15278)
add d05c7e3844 [Unity][BYOC] `PrimValue` handling in `FuseOpByPattern` for BYOC (#15217)
add ab8ac58241 [Unity][CUTLASS] Offload RMS norm (#15288)
add 6ab2e960fa [Unity] [DistIR] Introducing DistIR (#15289)
add fae8308813 [Dlight] Enhance fallback schedule with DecomposeReduction (#15302)
add 5dcc25a15c [Unity][Relax][Transform] Do not remove MatchCast for RemoveAllUnused (#15290)
add 1e45ca4ce3 [Unity] [Relax] [ONNX frontend] [op] Add support for Trilu operator (#15299)
add 804ce09195 [Unity] Update CUTLASS Attention to incorprate upstream change (#15309)
add f63fcba404 [Unity] Fix FuseOpsByPattern when a subgraph can be matched by multiple residual patterns (#15308)
add 0924abad15 [Unity][Op] Add leaky relu operator (#15296)
add fd5a41f6f0 [Unity][Dlight] Add schedule rule for decode transpose (#15304)
add b47b2695fb [Unity] CUDA Graph update (#15320)
add 98ef29a852 [Unity][Training] Enhance gradient system (#15230)
add aa28859340 [Unity][Dlight] Fix decode-GeMV rule when spatial-inner without broadcasting (#15330)
add 6294aada46 [Unity][Dlight] Rule matmul avoiding blockIdx.z (#15333)
add 231653cca5 [Unity] fp16 A x int B GEMM update - support int8, more bias shape (#15318)
add 959b7e5e09 [Unity][Dlight] Fix DecodeGeMV rule for spatial-inner with grouping (#15340)
add d8f1ac4e87 Merge remote-tracking branch 'apache-upstream/main' into unity
add c861351bae [Unity][OP] Sync `rms_norm` with main (#15355)
add 057e61575b Add the ability to differentiate between model loads from remote fetch v/s model loads from cache (#15357)
add 7d98f60144 [Unity][Dlight] Fix matmul schedule when out_dtype = fp32 and bias add is fp32 (#15363)
add e38eb9003a [Unity][CUTLASS] Support `out_dtype = "float32"` for FasterTransformer kernel (#15377)
add efa6cdcc6b [Unity] Avoid overloaded-virtual warnings (#15382)
add 0681b3959f [Dlight] Benchmarking Tools for Dynamic Shape PrimFuncs & Relax Function (#15322)
add b0566cdb87 [Unity][DLight] GEMV Rules (#15381)
add b54f9ff194 [Unity] Add support for AXIS_SEPARATOR in AlterOpImpl Pass (#15315)
add 92668c7d05 [Unity][Dlight] Fix reduction rule, aligning last block's iters (#15383)
add 91733636d0 [Unity][Module] Add Core Data Structure (#15398)
add 51d7128326 [Unity][DLight] Fix Reduction Rule (#15412)
add 4c65ceee91 [Unity] Lowering of axis separator in Layout Transform (#15390)
add e8995272d1 [Unity][Dlight] Tensorization Rule in GPU Matmul (#15389)
add c3862b573f [Unity] nn.Module Spec (#15416)
add bac72697ac [Unity] nn.Module Op (#15418)
add 7a4c85e5fb [Unity] nn.Module Torch Integration (#15424)
add 7bb37a54b4 [Unity][Dlight] Improve Dlight Tensorization Rule (#15427)
add 3f1469ea82 [Unity][DLight] Update GEMV rules (#15429)
add 7903ae1451 [Unity][Frontends][Onnx] Improve ConstantOfShape behavior (#15434)
add 0d462a012d [Unity] nn.Module Module and Effect (#15438)
add d3d350b858 [Unity] cuda graph support for cublas (#15435)
add 3727c048c6 [Unity][Op] Implement basic `call_tir_inplace` operator (#15372)
add f64225ee38 [Unity] dynamo with dynamic shape (#15441)
add b89d2fed6a [Unity][Ops] Support for erf in relax (#15445)
add 72bd41d3e8 [Unity][Dlight] Avoid too large vectorization factor in caching (#15443)
add 23edbff40a Merge remote-tracking branch 'upstream/main' into unity-staging
add 53fd712bf7 [MERGE-FIX] Update the code to fix merge issues
add ef00a96394 [Unity]: fix error on enum (#15451)
add 54e304467e [Unity] Eslint and TypeScript configuration fix (#15452)
add 605a61450b [Unity][Op] Support symbolic shape inference for slice op. (#15450)
add 1aecc9b8b6 [Unity][Frontend][Onnx] Simplify gemm (#15458)
add c3f26858ba [Unity][Op] Conv1dTranspose (#15456)
add d9e1659d2a [Unity][Transform] Elide redundant bindings of dataflow vars (#15341)
add f3501d85b7 [Unity][Fix][Op] Add groups to conv1d (#15457)
add 13bd656546 [Bugfix][CUTLASS] CUTLASS path finding (#15476)
add 5e1414275f [Cherry-Pick][BugFix][TIR] ThreadSync with shared.dyn awareness (#15481)
add 9d6e153497 [Unity][BugFix][Op] fix error when calling full_like in masked_fill (#15477)
add a8218b31c2 [Unity][DLight] Use less shared memory for gemv (#15482)
add 6c38001fe1 [MERGE] Merge main into unity 2023-08-04
add 9fb191159f [Unity] Commutative pattern match based on relax.Expr op (#15494)
add 1a1af0c700 [Unity][NN] Allow nn.Module to generate subroutines (#15453)
add e7cb3a848d [Unity][BYOC] Support ND A and 2D B batched matmul for cublasLt (#15499)
add 38e2b886dd [Unity][BYOC] Integrate Flash attention v2 kernel into CUTLASS BYOC (#15467)
add c0b3949765 [Unity][Frontend] Fix torch addmm op param alpha&beta (#15497)
add 5b15824ec5 [Unity][Minor] Fix compilation warning (#15504)
add c97753ea7c [Unity][Dlight] Matmul tensorization for SM70 (#15503)
add 1afbf20129 [Unity][BYOC] Unify the interface between `FuseOpsByPattern(.., annotate_codegen=True)`and `MergeCompositeFunctions()` (#15491)
add bd359685c7 [Unity][DLight] Update gemv rule (#15490)
add cd14d4d7e2 [Unity][Hot fix] Flash attention offload bug due to typo (#15512)
add f500a0dd1a [Unity] Update relax.Function.ret_struct_info when mutated (#15510)
add a4cea71687 [Unity] Add batched A matrix support for FasterTransformer fp16A intB matmul (#15501)
add 4ed34013de [Relax] CuDNN Fallback Support through BYOC. (#15462)
add bcc9a68b3e [Unity] Remove unused local function definitions (#15507)
add c8494dc004 [Unity][ONNX] Support ONNX Symbolic Shape Deduction (#15498)
add 61c693dc42 [Unity] Recursive pattern match in rewrite_call (#15495)
add 7cc7ea8ce6 [Unity] Propagate dynamic begin/end in strided_slice when legalizing (#15515)
add 13398c1ada [Unity][Bugfix] Handle assignment of non-dataflow var to a dataflow var in CanonicalizeBindings (#15523)
add a012fe0df4 [Unity][DLight] Update GEMV Rule for Mali GPUs (#15531)
add 2226a1f558 [Unity] Multi-device support for Relax (#15447)
add 92125d79cf [BugFix][Unity] Fix for TupleStructInfo (#15538)
add 59b5f6d9ae [Unity][BYOC] Make `MergeCompositeFunctions` append codegen name at the end of function name (#15534)
add aa21782828 [Unity][MSC][M0.1] Enable set name and layout for exprs (#15500)
add 36af50439b [Unity] Fix the ambiguity issue in `vDevice` constructor on Mac with Apple clang version 14.0.3 (#15546)
add fa0d35be2c [Unity][Relax][FuseOps] Add MatchCastNode to node_map of IndexedForwardGraph (#15544)
add 0e4c99cdb8 [Unity][ONNX] Improved symbolic handling and reshape functionality (#15550)
add 0092e97840 [Unity] Preserve shape parameters in LazyTransformParams (#15563)
add 74a547b029 [Unity] Infer concatenated shape with multiplication where possible (#15562)
add e4e3d1bc11 [Unity][Minor] Fix Compilation Warning (#15560)
add 54423aa822 [Unity][TVMScript] In-line literals for int8, int16, and float16 (#15565)
add 7b5a89eba1 [Unity] Fix handling of null values in cuda graph rewriting (#15567)
add 2e2126f9e3 [Unity] Implement relax.Function.bind_symbolic_vars (#15509)
add fd08c1c6cd [unity][frontend] add relax frontend torch op (#15557)
add a66e1cd1f5 [WEB] Fix WebGPU runtime build (#15575)
add 688d9dd068 [Unity][CUTLASS][Cherry-pick] Skip profiling all conv2d output alignments when possible (#15583)
add 684218f72d [Unity][Topi] Handle variable begin/end axes in topi.strided_slice (#15520)
add ac3078c7f5 [Unity][FX] Add fx support for functional linear and conv2d (#15566)
add 4eabefbe85 [Minor][BugFix] Remove PyTest Dependency (#15585)
add 497b7481d2 [Unity][Frontend][NN] Add GroupNorm Layer (#15592)
add f922f7cca2 [Unity][Frontend][NN] gelu NN module (#15593)
add b959645454 [Unity] nn.Module external modules (#15487)
add 567848e3a0 [Unity] Avoid trivial `var2 = var1` bindings in pattern matcher (#15578)
add f18e2976f9 [Unity][Frontend][NN] Conv2D NN module (#15586)
add d9d8b19ecd [Unity][Frontend][NN] Conv1D/LayerNorm NN module (#15587)
add b3658480e2 [Unity][MSC][M0.2] MSCGraph core (#15569)
add b99f34d56a [Unity] Support float in vm executable (#15608)
add 619cbeca84 [Unity] Prevent Relax VM from being corrupted after getting incorrect shape (#15612)
add 71cdd460b4 [Unity][Frontend][NN] Op print_ (#15604)
add 4a513825bc [relax][frontend]add relax frontend torch op: tan,asin,acos,atan,sinh,cosh,tanh,asinh,… (#15610)
add 7d1a101e24 [Unity][Frontend][NN] Add Timesteps layer to NN Module API (#15603)
add 0443482f7e [Unity] UpdateVDevice pass and infer vdevice (#15570)
add 5d0ef94ae4 [Unity][Transform] Track callees from external functions in DeadCodeElimination (#15561)
add d28613f3cf Change == to === operator in compact.ts file (#15620)
add d5dcabf2b2 [Unity][Frontend][NN] Add diffusers style Attention layer (#15609)
add 082d98e8ab Fix equality comparison and code consistency in web (#15621)
add c9a73789e7 [Unity] fix masked_fill operation (#15617)
add d2972f3b42 [Unity] Add instruments to relay translator (#15601)
add d3856d34e7 [Unity] Disco: A Framework-Agnostic SPMD Runtime for Distributed Inference/Training (#15622)
add 54d3fec26f [Unity][Dlight] GeMV rule skip cases of "outer dim being grouped" (#15571)
add 7556592dd2 [Disco][Test] MultiHeadAttention Testcase (#15632)
add 0230c77cb7 [Disco] Support ShapeTuple in Disco Protocol (#15634)
add 06ae46e567 [Runtime] Fix ICE from Clang (#15635)
add 71b81127a2 [Disco][Op] broadcast_from_worker0 (#15633)
add 14d9cf804e [Unity][MSC][M0.3] MSCGraph Builder (#15615)
add e0501b527d [Unity][Frontend][NN] Collection of new NN operators (#15642)
add 07ee13f663 [Unity][Dlight] GeMV rule max_num_threads awareness (#15647)
add 6e2ca87cf8 [Runtime] Support Loading Standalone `ndarray-cache.json` (#15654)
add 11d5f677ec [Unity][Frontend][NN] Make effects optional in nn module. (#15650)
add 71c6839b1a [Unity] Implement R.macro for Relax macros (#15455)
add c604012310 [Runtime] ShapeTuple.Product and ShapeTuple Printing (#15652)
add 1d56770213 [Runtime] Refactor NDArrayCache Support (#15659)
add c2eaa1d370 [Disco] Add `Scatter-From-Worker0` (#15653)
add f1246d0078 [Disco] Introduce ShardLoader (#15655)
add 935c91314a [Unity][Disco] Correct document for CopyFromWorker0 and CopyToWorker0 (#15660)
add 04211b3f69 [Unity][DLight] Improve the performance of matmul on Adreno GPU (#15661)
add 4ad3f3389c [Unity][Frontend][NN] Add nn.MultiLinear (#15662)
add a0e0390f0b [Unity] RealizeVDevice pass (#15636)
add 37ad0497fe [Unity] Add support for Google Pixel OpenCL (#15675)
add 90c64c6dce [MERGE] Merge main into unity 2023-09-06
add 3b6a788d77 Fix after merge
add 89c40611b3 [Unity][MSC][M0.4 && M0.5] Codegen && Test (#15645)
add c5b7afc721 [Unity] Implemented BundleModelParams transform (#15657)
add ec4a8b3011 [Unity] Implement relax.Function.bind_params (#15626)
add f0869fecc4 [Unity][Dlight] Matmul rule on int32 workloads (#15486)
add e1725ac01d [Unity][Dlight] Fallback rule supporting more spatial workloads (#15687)
add 8820ce48d1 [Unity][MSC][special] Change special names (#15691)
add 755af1fec7 [Unity] Added known tir.Expr to relax.PrimValue (#15577)
add 98ba395b6f [Disco][Op] gather_to_worker0 (#15690)
add 60e90a96a8 [Unity] Update remaining calls to `export_library` with keyword arguments (#15695)
add bec2b8f48d [Unity] Preserve ShapeExpr in EliminateCommonSubexpr transform (#15701)
add 7df7b7b1b0 [Unity][Analysis] Check for usage of DataflowVar in all_vars() (#15698)
add 9e7227e084 [Disco][Op] scatter_from_worker0 (#15680)
add e9aef6a479 [Unity] Add `has_function` to RelaxVM (#15674)
add 4956e4f260 [Disco] Add LoadAll method to Disco Shard Loader (#15673)
add 32ed4f00de [Unity] Fix CUTLASS tests following LiftTransformParams signature change (#15707)
add 70d23100f6 [Unity] Clear slots before each shape lowering (#15712)
add c0b8953a1f [Unity] Dynamic-shape param support in LazyTransformParams (#15713)
add 1c90ca7c7c [Unity][Frontends][Onnx] fixup resize2d dynamic input support (#15692)
add 071e94f653 Merge remote-tracking branch 'main' into 'unity'
add ea54a01d6b Fix after merging
add b82c10f497 [Disco] Make Session.CallWithPacked public (#15720)
add 0e800cf3f8 [UNITY][Pass] Optimize redundant layout transform ops (#15678)
add cdf40be5ca [Unity][Frontend][NN] Enable tuple/list input (#15670)
add 93bb64708f [DLight] Update GEMV rule to support Adreno outer reduction (#15730)
add fb883b77d8 [Unity][disco] Change collective communication to destination-passing style (#15735)
add 5e02385c0c [Unity] Implemented SameShapeConstraint for dataflow pattern matches (#15694)
add 7bf1203795 [Unity][MSC][M1.1] Add translate && codegen for torch (#15704)
add e0e6c1d9c3 [Relax][VM] Add additional flags for picojson (#15737)
add 40b9a926b9 [Disco] Pipe-based Multi-processing Session (#15727)
add bd51b5b592 [Unity] [Bugfix] Fix KeyError: 'padding' in _avg_pool2d implementation (#15734)
add 25d6c45ff4 [Unity] Fix BYOC codegen for dynamic shapes (#15750)
add afb2e421f0 [Unity][Disco] separate computation and communication into 2 stream (#15742)
add 21cdf6c1c4 [Unity] Unwrap tuples during CanonicalizeBindings (#15696)
add dd3bfb3424 [Unity][DLight] Fix outer_reduction for dynamic shape workloads (#15743)
add dddce91b3f [NN][Op]ConvTranspose1D (#15741)
add 6d085ae6f0 [Runtime][CUDA] NVTX Integration (#15760)
add b9d7bba19f [Disco] Integrate NCCL with CUDAGraph (#15766)
add 61e0e98a49 [Hotfix] NVTX for CPU-only Build (#15774)
add 8a98ffecc9 Merge branch 'main' into unity
add 25c6218e64 [Unity] Skip shape checking on transformed params (#15736)
add 1965315108 [Unity][BugFix] FuseTIR not keeping ExternFunc in IRModule (#15782)
add 9fd45741e8 [nn.Module] Add Missing Import nn.MultiLinear (#15784)
add 14f06adca2 [Unity] Keep num_input attribute after ModelBundleParams (#15789)
add 0fe3ed4424 [Unity] Mark result of LazyTransformParams as impure function (#15697)
add 2fdedf1ea8 [Disco] Integrate RCCL (#15776)
add cb3763b84d [Disco] Improve RCCL Integration Logic (#15796)
add c8ebc6c520 [Unity] Fix dlight bench when func_name is global var (#15793)
add 82ee2164d1 [Unity] Update test cases of LazyTransformParams as impure (#15803)
add 676def7c9d [Unity][Frontend][NN] Switch to always using fp32 for timestep embedding calculation (#15811)
add 8ff15c1626 Merge remote-tracking branch 'apache-upstream/main' into unity-staging
add b7471f8315 [Bugfix] Remove duplicate code in nccl.cc (#15800)
add e809d64658 [Unity][Dlight] Fix inline consumer in matmul tensorize rule (#15781)
add 6ea16b98e7 [Unity] Delegate DataflowVar visitor to Var by default (#15688)
add 732ae53653 [Unity][TVMScript] Produce var = R.ExternFunc("") statements (#15703)
add dfc77eb129 [Unity] Extend RemoveAllUnused to support relax::Expr (#15807)
add beef1f7c75 [TE] Support using tir::Var as CreatePrimFunc args (#15817)
add d06a658b10 [Unity][Disco] Use default stream in disco (#15827)
add 2fd23843fd Remove create_relax_prim_func (#15828)
add 230f8b2491 [Disco] Add AllGather (#15764)
add ea329cba02 [Unity][Analysis] Change warning to info for non-affine transform (#15820)
add 82586fb9ea [Unity] Extend EliminateCommonSubexpr to operate on relax::Expr (#15815)
add 1230b4009b [Unity] Dtype check in legalization of R.matmul (#15825)
add f78d9ed0ed [UNITY][Pass] Remove redundant reshape (#15806)
add 7494bc4580 [Unity][BYOC] Support offloading multi-query attention by Flash Attention (#15831)
add aa4587feb5 [Unity] Implement relax.transform.KillAfterLastUse (#15810)
add 225d067fb8 [Unity] Support Padding Reversal in Alter-Op pass (#15679)
add 4a8a7b9c63 [Unity] Implement LowerAllocTensor to remove R.builtin.alloc_tensor (#15809)
add d85ea256fa [Unity][TIR][Unittest] Fix failed 0-rank rfactor test (#15846)
add 374b03f7e9 [Unity][Doc] Fix Python Docstring for Sphinx Build (#15848)
add 19c378dbe6 [Unity] Avoid Using Wildcard Imports (#15850)
add dd21ed37a8 [Unity][Relax]fix behaviors for importing torch.cat/concat (#15845)
add 233ac7aee0 [Unity] Avoid trivial var-to-var bindings in CanonicalizeBindings (#15840)
add 407e20d74a [Unity][Bugfix] Check for trivial re-bindings in MatchCast nodes (#15854)
add 88a08ae47a [Disco] Loading-time sharding support (#15826)
add d618112932 [Unity][CUTLASS] Fix hardcoded epsilon parameter in RMS norm (#15856)
add 8dda1cb22b [Unity][BYOC] Support attention over batched sequences of variable sequence lengths (#15837)
add 4735a5b8e5 [UNITY] Handle pattern with remove_pad (#15843)
add 063cd7fb14 [Unity][Doc] Fix Relax part fix_docstring (#15860)
add 11c73a2ea6 Merge remote-tracking branch 'apache-upstream/main' into unity-staging
add d2a9f39f34 [Fix] Fix Build with RCCL (#15863)
add 8954139bf6 [Unity] Add R.quantize and R.dequantize ops (#15849)
add 7d53953adc [Unity] Check the PassContext within RewriteCUDAGraph transform (#15861)
add 9f0ac49a04 [CMake] Prefer a static NCCL (#15869)
add 6d2b44dbbf [Bugfix] Fix Disco-CUDAGraph Integration (#15870)
add b486210b1a [Disco] Add -lrt to TVM runtime for NCCL (#15876)
add 2e30dbe0ff [Unity][Fix] Remove duplicated words from comments, NFC (#15875)
add 8a2ca3405b [Unity] Use PrimValue as offset in R.tril and R.triu (#15783)
add 59ec81f125 [Unity][MSC][pre M1.2] Reconstruct codegen (#15813)
add 7c4c2c2f42 [Disco][Fix] Remove Dependency to PyTest (#15886)
add 969e31a978 [Unity][NN] Enhance ReLU and GELU support (#15885)
add b9a02b1e84 Merge remote-tracking branch 'apache-upstream/main' into apache-upstream-unity
add 9f4d0fa5d3 [Unity][Transform] Allow static Relax arguments to dynamic PrimFunc (#15883)
add dd57556660 [Unity][Op] Introduce `call_inplace_packed` as a counterpart to `call_tir_inplace` (#15878)
add ec1184e105 [Unity] [Bugfix] Fix KeyError:'None' in layer_norm and correctly use the normalized_shape (#15894)
add b1380059f0 [Unity] [Bugfix] Fix TypeError in TVM PyTorch frontend for LayerNorm operator (#15902)
add 67d61935df [Unity] [Bugfix] Fix MaxPool TypeError in ONNX frontend (#15908)
add 58e00f3267 [Unity]Add FastMathTransform pass to Relax (#15814)
add f154026db1 [Unity] Improve FuseOps error messages (#15899)
add 465d691e66 [Unity] Ignore R.ExternFunc in EliminateCommonSubexpr (#15900)
add bde8a87381 [Unity][Relax] Support Dynamic Tensor as Index, torch frontend (#15884)
add 95a89d2b13 [Unity] Fix wrong variable name in test_optimize_layout_transform (#15893)
add 1e789f3f08 [Fix][Unity] Fix TVMError when loading ONNX model with CumSum operator (#15804)
add 589d919c59 [Unity] Fix TVMScript Issues in Testcases (#15920)
add c62dcfa89e [Unity] Propagate extra symbolic vars through LiftTransformParams (#15699)
add 559e94a0c8 [Unity][TVMScript] Avoid dangling reference when printing Call attrs (#15923)
add e67678066d [Unity][VM] Improved error message in CodeGenVM::EmitKillObject (#15822)
add 9725144bef [Unity][Transform] Canonicalize and use CSE between pattern matches (#15904)
add 9892675e68 [Unity][Relax]support torch.arange()+ (int) in torch frontend (#15917)
add c606fcf420 [Unity] Paged KV Cache as LM Support (#15910)
add 10d5cae4e9 [Unity][MSC][M1.2] Add translate && codegen for tensorflow (#15905)
add 354c5f1008 [Unity] [Bugfix] Fix bug in interpolate operator's default mode parameter in PyTorch frontend (#15933)
add 5f4412b104 [Unity][Frontend][Onnx] Add support for Elu operator (#15937)
add ecb689e51f [Unity] [Bugfix] Fix TypeError in interpolate caused by scale_factor as tuple (#15935)
add 4a0151c68a [Web] Further increase default EMCC compilation total memory size (#15944)
add 8dc4bd67c3 [Unity][Dlight] Add software pipeline to matmul tensorization rule (#15943)
add 3acb2ca44e [Unity] Fix bug in onnx_frontend when multi inputs have same symbolic shape (#15919)
add 8874f8b6d6 [Unity][MSC][Bugfix] Trilu bugfix && special ops support (#15938)
add ec9e0a0655 [Unity] Allow FLegalize to produce Relax operations (#15842)
add 1ba11f69a7 [Unity][BYOC] Add support for sliding window in attention op (#15951)
add ea70b02961 [Unity] `nn.Mutator` for `nn.Module` level transform (#15958)
add b8a1d63665 [Fix][Unity] Fix TVMError when using relax to load model with Trilu operator (#15924)
add 9ff245078c [Unity][SWA] Overriding windowed cache support (#15963)
add 5c6b6bba84 [Unity][Pass] Include `FoldDataflowBlockOutput` in `CanonicalizeBindings` (#15791)
add ea588c1c37 [Unity] Update relax_dynamo (#15962)
add bcdbc3e244 [UNITY] Fix the symbolic var handling (#15973)
add 4d19c8ab1f [Unity][Transform] Improved canonicalization of non-dataflow Var (#15941)
add 5808cea9af [Unity][BYOC] CoreML Scaffolding (#15556)
add 7ef36ebb5d [Unity] Support symbolic PrimValue arguments (#15980)
add 04c6863a25 [Unity][BYOC] Fix cuBLAS BYOC compatibilty with Disco + `ThreadedSession` (#15977)
add ef39f37f49 [Unity][BYOC] Support variable-length attention by flash attention (#15959)
add 5d62e9bba5 [Unity][MSC][M1.3] Add translate && codegen for tensorrt (#15950)
add ebbe38f328 [Unity] Include LegalizeOps in the default relax.build lowering flow (#15864)
add 8f97a763a0 [Unity][Testing] Show failing module in WellFormedInstrument (#15898)
add 484bd4421b [Unity][BYOC] Do not use cudaMemcpy for max_seqlen in var len attention (#15989)
add 6408cc414f [Disco] Explicitly set the session on DPackedFunc and DModule (#15996)
add 3184a80492 [MERGE] Merge main into unity 2023-10-29
add 393aaa350f Fix after merge
add 53ccf18625 [Unity] Allow Pipeline Registration (#16008)
add 6936829aac [Unity] Remove end-of-life handling from StaticPlanBlockMemory (#15841)
add d93260824b [Unity][UnitTest] Enable BindParams test for R.Prim (#15978)
add 90bb10e2bf [Unity][nn.Module] Support Parameter Packing (#16007)
add 16af021e3d [Unity][nn.Module] Support `nn.SourceModule` (#16006)
add 7486476b6f [Unity] Deterministic Ordering when Iterating IRModule::functions (#16020)
add 8c7aaa6a4f [Unity][UnitTest] Cleanup test_vm_build.py (#15981)
add 49a3a51ff1 [Unity] Ensure one VM register for each relax binding (#15855)
add a9c81a7cc1 [Unity] Replace relax_vm/memory_manager with memory/memory_manager (#15882)
add 3c8603789e [Unity] Avoid Emitting Redandunt Bindings in TensorExpr Op (#16026)
add 0f8186f7dc [Fix] Windows Build (#16028)
add 853732e5ef [Unity] Support getting variable mapping for FunctionCopier (#16012)
add 7833f4ec29 [Unity][BlockBuilder] Allow emitting nested tuple (#15993)
add a801064b5b [Unity] Alias IntTuple <= ShapeTuple (#16035)
add 23371cadb4 [Unity][MSC][M1.4] Add Runner and test with relax (#15997)
add 2329b1a9a9 [Fix] Update mutator name rule (#16046)
add 0ed1e30df5 [Unity][Support] Sample from top-p supports offset (#16069)
add 9202f4b8fc [Bugfix] Compilation Error with Clang (#16071)
add 151aa74370 [Unity][Dlight] Metal Performance (#15985)
add 021f31b737 [Unity] Fix FuseTIR pass for gather/take cases (#16064)
add 7d0e60ad0e Revert "[Unity][Support] Sample from top-p supports offset" (#16077)
add 664beaaf1a [Unity] Loading NDArrayCache by parameter names (#16078)
add cf013c2f6b [Unity] Handle duplicate outputs in LazyTransformParams (#15942)
add 3f1347cbd4 [Unity] Enhance Python Annotations for Relax Expr (#16075)
add 3de77f8def Merge remote-tracking branch 'main' into unity
add e7d12be07a [Unity][Training] Support intermediate vars as require_grads for Gradient pass (#16011)
add e506bffa72 [Unity] Implement FNormalize attribute for operators (#16067)
add eb20534f18 [Unity][Bugfix] Track variable usage from input to impure functions (#16036)
add 8e8799d709 [Unity][Dlight] Enhance matmul tensorizer with Int8 support (#16084)
add 6ce16026f7 [Unity] support symbolic var in RewriteDataflowReshape (#16086)
add 9eeb5bcbcb [Unity] Support Regular expression matching in globalvar dataflow pattern (#16085)
add 9100a8e8c9 [Unity] [LiftTransformParams] Treat symbolic var in weight shape as constant (#16049)
add e37165f40f [Unity][MSC][M1.5-1.7] Add Runner and test with torch, tensorflow && tensorrt (#16072)
add 6b9c277bf6 [Unity][Training] Simplify matmul patterns after gradient (#16082)
add e359e7a210 [Disco] Add loader for presharded params. (#15957)
add 384f9b69ea [Unity] Add `axis` field to scatter_from_worker0 (#16092)
add 276b4cedbd [Unity][Fix] Fix `topi.rms_norm` with float32 upscale (#16099)
add 171ef61797 [Unity] make LazyTransformParam more general (#16088)
add 7dd248b0be [Unity][Dlight] Choose perfect spatial factor in reduction rule (#16101)
add bc10f769d8 [Unity] Add LoadParamOnWorker0 function in shard loader (#16093)
add a3d9108050 [Smallfix][WEB] Change memory manager import for web (#16107)
add 06a48996cc [Unity][Fix] Fix `rms_norm` tests (#16109)
add 7892af0d46 [Unity][WebGPU] Allow lower max storage buffer binding size (#16108)
add 7a0c3f9e05 [Unity][Support] PagedKVCache support growth control (#16112)
add 835bc82665 [Unity][TVMJS] Add md5sum to weight shards (#16122)
add d5daa9806d [Unity] Allow Customized Pipeline in `relax.build` (#16121)
add 26a20ee638 [Unity] Improved error message in relax::Normalizer (#16114)
add e7c7314046 [Unity][Transform] Keep R.ExternFunc in dead-code elimination (#16118)
add 6f650db042 [Unity][DistIR] Legalize redistribute (#16098)
add c9de001490 [Unity] [Transform] Skip constants in CSE pass (#16125)
add 0ddfc657db [Unity] Implement FNormalize for relax.op.call_tir (#16068)
add 684a8ca6a4 [Unity][DLight] Enhance the inline consumer rule (#16124)
add bc2415bcc8 [Unity][BYOC] Support IGEMM in cuBLASLt (#16134)
add 2dbab710e8 Merge branch 'main' into 'unity'
add 44d80e5107 [Unity][Bugfix] Reset window cache current pos when clearing (#16132)
add 8a2ffee38e [Unity] Do not import SciPy by default (#16136)
add 90edf76716 [Unity][LLM] Add NaN checks during sampling for better error reporting (#16141)
add 5e61adc0a3 [Unity][DistIR] Enhance PropagateSharding pass (#16094)
add 165b84bc7f Always use int64 in JSON parser (#16145)
add 4c07f6af43 [Runtime] Introduce Type-Checked `TVMArgs::At<T>(i)` (#16147)
add 4e70c28217 [Runtime] Allowing Packed Arguments in TVM Module VTable (#16148)
add 29450b927e [Unity][MSC] Enable add attributes while fuse ops (#16128)
add bafd49d4b8 [Unity] Flash infer integration (#16146)
add 9a985714f3 [Unity] Migrate Relax Executable/VM to `TVM_MODULE_VTABLE` Convention (#16149)
add 925cb2bbaa [Unity][BYOC] Add cutlass finegrained decode matmul (#16144)
add aae1112a65 [Unity] Support constant args in `nn.ExternModule` (#16130)
add 756ce9917f [Unity][3rdparty] Remove TVM in 3rdparty of FlashInfer (#16155)
add 1de8b347d1 [Unity][DistIR] LowerGlobalViewToLocalView (#16095)
add 2dcb8716e8 [Unity][BlockBuilder] Depracate `BlockBuilder.get()` and change it to `BlockBuilder.finalize()` (#16090)
add 8f24a272a0 [Unity][MSC][M2.1] Add Manager for compile pipeline (#16163)
add af803cf7b4 [Unity][DLight] Fix `general_reduction` for GroupNorm (#16161)
add 64fe5a8a89 [Unity][DistIR] Add DTensor struct info propagation rule for stop_lift_params (#16170)
add c640d0a3c9 [Unity][Web] Fix missing function NVTXScopedRange for web (#16177)
add 8a6184ccfa [Unity, BYOC] Add check for leaking intemediate variables for cublas and cudnn (#16175)
add a6adaae5ef [Unity][DistIR] LowerDistIR (#16169)
add 85389efa2c [Unity][BYOC] Fix Flash var_len attention with sliding window (#16185)
add 68443482c9 [Unity][Bugfix] Handle symbolic matching with non-structural match (#15994)
add d52a9bf388 [Unity][Transform] Implement RemoveUnusedOutputs (#16117)
add d6015c5643 [Unity][BugFix] Fix a bug in relax gelu_tanh computation (#16188)
add fe9d2fe57d [Unity][Transform] Implement ExpandTupleArguments (#16115)
add fc324d0f2c [Unity][Transform] Implement RemoveUnusedParameters (#16116)
add 74667b97f0 [Unity] Enable ccache for `nn.SourceModule` (#16189)
add ed2772f9c8 [Unity][MSC][M2.1] Add pruner for model pruning (#16186)
add 9e4e17ca88 [Unity][WebGPU] Get params from cache by name (#16198)
add a2f55a8812 [WEBGPU] Update to latest compilationHints API (#16197)
add 8f95f6147a [Unity] [Transform] Remove iteration over functions in function pass (#16173)
add 3c7067d6ed [Unity] Minor: Remove debug logging (#16200)
add 34fd234f55 [Unity] Check usage location when canonicalizing trivial bindings (#16193)
add 4e8c975700 [Unity][Bugfix] Fix `tests/python/topi/test_topi_transform.py::test_relax_dynamic_strided_slice` (#16205)
add d0504027bb [Unity] Update FlashInfer (#16208)
add ebbad09cd5 [Unity] Upgrade cutlass_fpA_intB_gemm (#16206)
add 03fc4f6f03 [Dlight] Change max_threads on CUDA (#16203)
add 58e622b74d [Unity][Transform] Implement Relax function inlining (#16194)
add e0518da2a5 [Unity][MSC][M2.3] Add tracker for track layer datas (#16207)
add 35e8404f17 [Disco] Expose `DiscoWorker` and `ndarray_cache_support` in header (#16153)
add f18d186559 [Unity] Speed up NormalizeGlobalVar (#16219)
add b5b980e33a [Unity] Support out dtype for nn.Linear and nn.MultiLinear (#16220)
add 8241385f59 [Unity] De-duplicate calls to TensorStructInfo constructor (#16209)
add 2772fb072a [Unity] Fix upstream tests that fail on unity branch (#16196)
add c6d4926529 [Dlight] Fix NormalizePrimFunc with scalar block (#16156)
add af14fbbbe1 [Relax] Fix to enable emit_te of topi scan/sort kernels (#16226)
add 943508a295 [Unity] Fix typo in dlight fallback rule (#16230)
add cbcb67c047 [Unity][Frontend] Add the `sum` op to frontend ops (#16225)
add fe89ccc360 [Unity][Transform] Pass for automatically extracting DataflowBlocks (#16204)
add f7b0193f9d [Unity] Fix IndexDataTypeNormalizer so that it correctly handles corner case (#16235)
add e100a13737 [Unity] Fix legalizing strided slice (#16232)
add 674167805c Revert "[Unity] Fix IndexDataTypeNormalizer so that it correctly handles corner case" (#16241)
add 6118b770b1 [Unity] Improved error checking for DataflowBlock in nested SeqExpr (#16195)
add e1964eceb5 [Unity] Add runtime debugging method to RelaxVM (#16238)
add cd9445d63b [Unity][lm_support] window kvcache sink (#16240)
add a2e19d21eb [Unity] Fix IndexDataTypeNormalizer so that it correctly handles corner case (#16245)
add 8edfee8574 [Unity][MSC][M2.4] Add quantizer for quantize model (#16228)
add 5b1fa29838 [Unity][VM] Allow `pipeline=None` in `relax.build` (#16246)
add f794db4373 [Unity] Avoid to use `std::regex` (#16249)
add 2d0d4e46a9 [Unity] Enable spot nodes in CI (#16253)
add a0e58987b0 [Unity][nn.Module] Refactor `ExternModule` (#16247)
add 76e239e2e1 [Unity] Fix Cutlass Codegen for Dense (#16252)
add 95f1b5c0e8 [Unity] Hot Fix Unity CI (#16256)
add e98fdea654 [Unity] Bump fpA_intB_gemm (#16244)
add 4e66690a4d [Fix] add TVM_DLL to disco functions (#16258)
add 45eeb8c838 [Unity] Fix ccache env for `nn.SourceModule` (#16257)
add f328e9bde3 [Unity] Add missing library import (#16263)
add 1c35c39264 [Unity] Add Relax multi-device e2e cases (#15823)
add 3de5e865df [Unity][nn.Module] Support Runtime-Calling Any PackedFunc via `op.extern` (#16274)
add 5c8caa6e35 [Unity] Unified KV cache interface and PagedKVCache refactor (#16273)
add 2bf3a0a428 [Unity][MSC][M3.1] Add distiller for distill model (#16264)
add 889d2f6cef [Unity][Frontend] NNModule `tensor_ir_op` support (#16278)
add 8946efa62e Update FlashInfer (#16281)
add 58daeb4905 Update FlashInfer (#16292)
add 303afdbccc [Unity][MSC][M3.2] Add gym for pruning and quantization, enable auto prune/quantize (#16280)
add 2f7e0d578f [Unity] Ensure memory planning cross-function independence (#16318)
add beb832616d [Unity] Update cutlass FpA IntB GeMM submodule (#16320)
add 8867de843d [Unity][MSC][Bugfix] Use random workspace for test (#16322)
add 9030522960 [Unity][Frontend] Introducing Object (#16316)
add b1df4b0856 [Unity][Web][Fix] Fix fetchNDArray for f32-to-bf16 (#16294)
add faa8a0ad46 [Unity][nn.Module] Introduce operator `empty` (#16327)
add ac568eb30a [Unity] Fix PagedKVCache per FlashInfer update (#16317)
add 09c44e6a93 [Unity] Upgrade flashinfer 3rdparty submodule (#16323)
add 163c7ac436 [Unity] Cutlass kernel compatibility with cmake 3.18+ (#16302)
add b3f0e55f24 Change metal dtype of ceil_log2 to fp32 (#16332)
add 4a7e4fec37 [Unity] Fix nn.op.tensor_ir_op signature (#16333)
add 1af82ad666 [Unity] Validate struct info in relax::Call constructor (#16311)
add ec542da8cf [Unity][Transform] Extract partial-tuple-usage from FuseTIR (#16120)
add 6f2fe457c2 [Unity][UnitTest] Increase atol to resolve flaky CI failure (#16340)
add 0cf5f47a1e [Unity] Dispatch cumsum and sort (#16254)
add 7dfc863df8 [Unity] Alter op impl handling empty transform for output (#16331)
add d88cc4267d [Unity][Transform] Implement UpdateParamStructInfo (#16305)
add d509661f89 [Unity][Analysis] Handle PrimStructInfo in EraseToWellDefined (#16304)
add 49fc613a3c [Unity][WEBGPU] Enable wasm exception propagation (#16330)
add c3aa71a53e [Unity][Analysis] Add utility for collecting compile-time bindings (#16312)
add 8d72091b27 [DLight] Skip rule if target is not suitable (#16321)
add 31659b617f [Unity][Dlight] Support dlight gemv rule on nested inner block (#16251)
add fe5f6163e6 [Unity][MSC][Legalize] legalize codes and mute logging (#16325)
add f215a417bf [Unity][NN] Use Linear name for nn.op.permute_dims (#16303)
add ded4be4b39 enhance shared memory merge.
add 6b6419c55f merge from unity upstream
add 047211f1da revert the change for dyanmic test
add 76ceff2c5e fix typo
add e3216a6972 lint fix
add 3fad0109bf [Unity][Contrib] Add vLLM paged attention kernel (#16350)
add b47280b1fa Merge branch 'main' into 'unity'
add 155dd73ac6 Fix after merging 'main' into 'unity'
add 0b13b5c844 [Unity] Enhance Torch-consistency in rehsape (#16360)
add 8e54a9e91d [Unity][DLight] Introduce Specific Rule for RMSNorm (#16338)
add 4c77f0fc24 [TIR] Extend DP4A tensor intrin (#16293)
add ef2a9139c1 [Unity] Improved error message for matmul shape mismatch (#16308)
add b22aa0f7c7 [Unity] Improved error message in ExprMutator::ReEmitBinding (#16307)
add fab6db20e7 [Unity][Transform] Use parameter name in BundleModelParams (#16309)
add 9caa179ce6 [CI] Update docker images tag to 20240105-165030-51bdaec6 (#16344)
add a1a1a7ca03 [AUTOTVM][FIX] Typo fixes and add a warning in the Droplet Search (#16289)
add 1e95b63fcc Merge branch 'main' of github.com:apache/tvm into unity
add 2d53e6ac63 [Unity][Transform] Handle replacement at both var binding and usage (#16367)
add a796023342 [Unity][Fix] Memory planning check value type of 'tir_var_upper_bound' (#16362)
add 4a37cfefa0 [Unity][Analysis] Show objects instead of names in WellFormedChecker (#16310)
add 4e05eb4e1c [CI] Upgrade Unity ci images (#16369)
add 298ad2c3f6 [Unity][Transform] Update LambdaLift to use name of lifted lambda (#16306)
add e1d71b3720 [Unity] Add dlight.gpu.Fallback in DispatchSortScan, add argsort, topk, and cumprod (#16351)
add 9d483d462e [LLVM] Update Host.h path (#16373)
add 3166366af6 [CI] Upgrade sccache version to 0.7.* (#16366)
add 524ec5f1b0 [Runtime] Use cudaGetDeviceCount to check if device exists (#16377)
add 45532d791c Merge branch 'main' of github.com:apache/tvm into unity
add 474c06b8b3 [Unity] Set CMAKE_CUDA_ARCHITECTURES default to native (#16335)
add 0b2358c2e4 [Relay] make "ToScalar" support directly obtaining "int64_t" (#16324)
add e53a8bcfb9 [TOPI][Target] Add fp16 SIMD support for conv2d on `arm_cpu` targets (#16383)
add 8e67e2a3a1 [TVMC] Add tvmc flag to print ir before and print ir after named pass (#16261)
add e2e33ddd54 [Bugfix] Disable SingleEnvThreadVerifier (#16361)
add 5d4c01e0fc [Thrust] Use no sync exec policy and caching allocator (#16386)
add c40d96b59e Merge remote-tracking branch 'upstream/main' into unity
add b8230f6e47 [Unity] Update dispatch test cases following the merge from main (#16388)
add 81a6c51ba4 [Unity] Fix creation of disco ProcessSession (#16375)
add d1b890a4e3 [Unity][Contrib] Fix a bug due to typo in vllm `reconstruct_from_cache` kernel and add test (#16376)
add b69d720593 [Unity][MSC] Avoid depending on trivial bindings in Relax intermediate (#16349)
add 4c7c010513 [Unity][Transform] Implement relax.transform.AdjustMatmulOrder (#16314)
add 7798e93529 [Unity] Support TIR kernel for PagedKVCache (#16374)
add 138cb651e0 [Unity][BlockBuilder] Restore bb.get() (#16378)
add 07d8e02367 [Unity][nnModule] Dynamic shape support in nn Module (#16284)
add 5c87bfe09b [Unity][Relax][Op] Add Conv3D Operator (#16385)
add e9bea9d2e2 [Relax][Frontend][ONNX]fix onnx frontend parse (#16395)
add 98d5153918 [Unity] PagedKVCache supporting on-the-fly RoPE calculation (#16396)
add cf14eddebb [Unity][Transform] Memory planning for dynamic-shape func return (#16111)
add a2a1b53402 [Unity] Split DecomposeOpsForTraining into two steps (#15954)
add cbe9c14879 [Bugfix][Unity] Recover MSVC/NVCC/ROCm/Vulkan (#16414)
add a763b22119 [Unity][Transform] Replace eligible operators with in-place versions in dataflow blocks (#16129)
add c470e1a922 [Unity][Fix] Fix mismatched intrinsic name (#16418)
add 7336379dea [Unity][Frontend][NN] Better support for dynamic convolutions (#16427)
add ae8d398b88 [CI] In jenkins.cmd_utils.Sh.tee, check for failing subprocess (#16382)
add f1bf20a950 [RPC] Fix tuning on macOS and Windows (#15771) (#16357)
add 4258c864b9 [RUNTIME][RPC] Enable RPCObjectRef return in RPC (#16387)
add 196b413813 [Relay][Frontend][Torch] fix a typo mistake in nonzero_numpy (#16390)
add 3e52c3dba5 [CI] Remove NVIDIA_DISABLE_REQUIRE (#16384)
add fe9814c73e [OpenCL][CMake] Fix OpenCL tests compilation (#16394)
add a7dd32cc16 [DeviceAPI] Support querying total global memory (#16398)
add 68be158d35 [ROCm] Some fixes of ROCm codegen (#16404)
add 3053f65da7 Add NVIDIA Hopper H100 target tag (#16407)
add 12ad4fbcf4 [Relay][Frontend][Torch] fix pytorch frontend not support logical or (#16400)
add 7ef521fad6 [COMMUNITY] Add new key for release signing (#16419)
add a5e883e846 [RUNTIME][CLML] Fix for Softmax op for 4D tensors (#16328)
add e1c430c7e3 [Relay][Frontend][Torch] fix pytorch frontend linspace op (#16417)
add 827beed0d6 [CMake] Enable cuda lang if USE_CUDA is on (#16426)
add 614a7a9e31 [CI][WASM] Update emsdk and nodejs version (#16420)
add 5a2949bd70 Merge branch 'main' of into branch 'unity'
add a42ca7f189 [Unity][Dlight][Fix] Reduction rule support dyn-shape epilogue (#16429)
add 4e754a7803 [Unity] downgrade cmake version requirement (#16432)
add 35551d4fdf [Unity] Support cumsum with pure int32 (#16439)
add 8f2e820434 [Unity] Improve buffer allocation for handling duplicated buffer names. (#16437)
add b0b8746fbb Revert "[Unity] Split DecomposeOpsForTraining into two steps" (#16442)
add 0a3c736591 [Unity][FIX] fix thread dtype mismatch (#16443)
add 61d952c701 [Unity][WEB] Temp disable wasm exception (#16444)
add 28c68e8073 [RUNTIME][Memory] Implement support for non-zero offset within a storage object in AllocNDArr… (#16415)
add ffa404fb48 [Metal] Dispatch numerically stable tanh for metal (#16438)
add ccca00a078 [BugFix]Ensure that bf16 arrays are created as expected (#16436)
add 81f8690aaa [TIR] Fix of inter thread reduction with shared memory prefetch (#16406)
add ed5c46aebc [ROCm][WebGPU] Intrin Dispatch: `tanh`, `erf`, `log` (#16441)
add bc46e86190 Merge branch 'main' into unity
add 9e94bca5a3 update jenkisfile to enable non-unity tests
add c7814cc2b6 fix typo in storage rewrite
add ed3eb6e72f [skip ci] update branch rule to prepare for unity transition (#16445)
add e4b1d684b4 Merge remote-tracking branch 'upstream/main'
add ee5c994a59 [Disco][3rdparty] Add latency optimized all reduce kernels.
add 20efa23e37 Revert "[Disco][3rdparty] Add latency optimized all reduce kernels."
add 8621517d3d [skip ci] Post unity transition
add bac6564614 Bump pillow from 10.0.1 to 10.2.0 in /apps/microtvm/cmsisnn (#16452)
add 38375bb7a8 Bump pillow from 10.0.1 to 10.2.0 in /apps/microtvm/ethosu (#16451)
add 7789b248fb [Unity][Transform] Handle symbolic variables in LambdaLift (#16411)
add 030ca4a74a [Unity][Transform] Implement relax.transform.ExpandMatmulOfSum (#16313)
add a7478cc2b1 [VM][Hexagon] Implement dma_copy and dma_wait builtin for hexagon (#16448)
add 2c49e01a0a [Unity][Transform] Implement relax.transform.ReorderTakeAfterMatmul (#16315)
add 7eb45ee5db [Schedule] Loop-Partition Scheduling Primitive (#16431)
add a86e41bcd2 [Unity][TVMScript] Update `call_packed` semantics to support empty sinfo_args (#16379)
add 1ee0cff6f9 [Minor] Fix Docstring for sphinx-build (#16461)
add 20b08a5c8d [Runtime][KVCache] AttentionWithFusedQKV and RoPE mode (#16456)
add 0e8e421766 [Cherry-pick][MSC][M4.1] Add plugin && plugin_builder, enable build and test in different frameworks (#16397) (#16460)
add 593a4bd9d7 [Relax] NDArray Cache Update with DLTensor Support (#16464)
add b98bc66584 [Relax] Memory planning for "partially dynamic" shapes (#16466)
add bbbc8956a9 [Contrib] Workspace for cuBLAS backend (#16413)
add 5012208e4b [LINT] Fix pylint issues in test_dma_builtin.py (#16468)
add 47424b374b [Unity] Improve error message in tensor_to_shape struct inference (#16473)
add 8ade9c30ed [Unity] Improved error message in tvm::relax::UpdateStructInfo (#16472)
add a15fd1d6a2 [CI] Update image tag to 20240126-070121-8ade9c30e (#16435)
add b4fc01444c [Doc] Fix Docstring in `extern.py` for Sphinx (#16482)
add 90320b2424 [Frontend][PaddlePaddle] Fixed the bug that prevented the model from being successfully converted to microTVM on MacOS (#16478)
add e7125a7af2 [Unity][Frontend] Add Sigmoid and Square Op (#16483)
add 5092c003de [BugTIR] fix thread_sync occurs in letstmt (#16454)
add ff9ebadf6a [wasm] Enlarge initial memory for emcc (#16485)
add 147ed5e27d [Unity][CodeGen] RunCodegen based on externally-exposed functions (#16422)
add fd1f40f989 [Unity][MSC][Refactor] Reconstruct BYOC and runner (#16467)
add 3700ca4740 Bump pillow from 10.0.1 to 10.2.0 in /apps/microtvm (#16453)
add 6e0bd70391 [Runtime] KV cache providing workspace for attn kernel (#16486)
add 06f7810c4f [SVE] Add vscale builtin (#16484)
add 0628bdba93 [Relax][Pass] Skip data type node for CSE pass (#16493)
add 00a2c7d325 [CI] Support tvm-bot rerun for tvm-unity task (#16502)
add f21c5a4245 [DPL] Support tir_vars field in is_call_tir pattern (#16494)
add 5c45ae8e53 [Relax][Web] Add ApplyPresenceAndRequencyPenalty (#16504)
add 5b2f641b52 [Unity] Support storage reuse for dynamic shapes (#16500)
add 7a303d96c9 [Instrument] Add default instrument to print all passes (#16497)
add 5c68932cae [Relax][Frontent] "tensor_ir_inplace" op (#16498)
add b5f44100e5 [Runtime] ParallelFor skipping thread backend for unit extent (#16508)
add 6ab99df3cb [Misc] Add Relax, Unity Tags in make_notes.py (#16516)
add a3ec544124 [Minor] Fix Clang compilation warning in fuse_tir.cc and codegen_c_host.cc (#16511)
add cdc2303292 [TIR] Require exactly same-dtype matching for Vulkan smem reuse (#16515)
add a6e8fee133 Bump 3rdparty/cutlass_fpA_intB_gemm version (#16506)
add 5ebdd4982f [BugFix] Use x*x*x instead of pow(x,3) (#16518)
add 343435a529 [Unity][MSC][M4.2][Step1] Enable plugin with manager, test plugins in compile pipeline (#16495)
add ff36526a4c [CI] Cut pytest-lazy-fixture (#16512)
add 467528ac87 Revert "[Unity][WEB] Temp disable wasm exception (#16444)" (#16527)
add 8da3de1522 [Redo][Unity] Split DecomposeOpsForTraining into two steps (#16465)
add cb6e4ee147 [Unity] Infer struct info for relax.op.split on dynamic-sized index (#16355)
add 6a2459f185 [Unity][Doc] Document passes that depend on `DataflowBlock`s and encourage using `ConvertToDataflow` (#16514)
add 6a3fadc065 [Unity][Transform] Handle `call_tir_inplace` in `FuseTIR` and `FuseOps` (#16487)
add 2dcf9ec5a6 [Keras] Enable Dense operator for any input dims (#16526)
add 268d15c987 [CI] Fix CI Script and Broken Tests (#16521)
add 1698793226 [CI] Disable testing that downloads from mxnet (#16546)
add 2d9c6c4b33 [Runtime] Fix dtype conversion for bf16 and fp8 (#16545)
add ae93d221d8 [RUNTIME][ROCM] Properly align rocm parameter buffer (#16550)
add 0449a16680 [KVCache] Fix attention kernel for ROCm (#16551)
add 3fd9bac4e0 [TIR] Fix get_block_access_region for let bindings (#16553)
add a6d644b4c3 [Build] Improving debug and build-dir options (#16524)
add 32f83d581c [Runtime] Add "TVM_DLL" to NDArray cache load func (#16541)
add bf4394762a [Unity][TVMScript] Optionally hide StructInfo that can be inferred (#16356)
add b04b1acf40 [Web] Compatibility with PagedKVCache in WebGPU (#16554)
add bb2adbfc4a Improve error message in NDArray::CopyFromTo (#16539)
add 3c5ee30630 [Relax] Support callback as argument (#16542)
add 685355e2c7 [Relax] Minor updates for NN frontend (#16558)
add c5aaa99cc5 [Unity] Check for symbolic vars in PrimValue in when lowering to TIR (#16564)
add 7336debd8d [Bugfix][Relax] Remove call to tvm.build for empty TIR module (#16561)
add 07ecb34006 [microNPU][ETHOSU] Add fixed point for matmul (#16401)
add 059f629ec2 [BYOC] Skip processed functions in FuseOpsByPattern and RunCodegen (#16567)
add 2b813ec04f [Upd] Enable lld search to include /opt/rocm/llvm/bin for rocm (#16540)
add 274c368dba [Bugfix][Transform] Keep private non-primitive functions in FuseTIR (#16565)
add 1a01102f87 [Doc] Remove MxNet related tutorials (#16572)
add 67bd739bed [MISC] Fix compile warnings (#16571)
add daa37e7e95 [Relax][VM] Re-implementation of callback functions (#16573)
add efc2ae9846 [KVCache] Support returning query positions (#16578)
add 6333d86105 [KVCache] Support mode "None" for Rotary Embebdding (#16580)
add c7e3510f59 [MISC] Update the 3rdparty/libflash_attn submodule (#16576)
add 5645c52c6d [Marvell BYOC]: Marvell AI Accelerator Integration - Phase 1 (#16570)
add a692453612 [Relax] Ignore non-relax functions in relax.transform.RunCodegen (#16586)
add 9ed9f7a93f [Debug] Improve error message for codegen pattern mismatches (#16587)
add 94e83f27cc [Unity][VM] Recursively visit match bindings in VMShapeLowerMutator (#16583)
add f0c28a0791 [RUNTIME][METAL] Fix multithreading access of metal runtime (#16605)
add 36ebcd0a87 [Relax][Transform] Canonicalize `let var = R.const` bindings (#16601)
add 1e6482c808 [Relax] Expose name_hint field for BlockBuilder.match_cast (#16600)
add dd709412eb [Unity][TVMScript] Parse R.Object return type from call_pure_packed (#16593)
add 40ec46970f [WEB] Move ArtifactCache to Interface, Support Cache delete and Batch Delete, Remove typo (#16525)
add 270c793574 Extend gpu memory bandwidth test to work through RPC (#16608)
add 460000202e [KVCache] Support passing in attn_score_scaling_factor into KV cache (#16606)
add 2066ce9612 [Unity][MSC][M4.2][Step2] Enable plugin with manager, test plugins in compile pipeline (#16581)
add a6157a6369 [SVE] Change the dtype of Ramp and Broadcast lanes to PrimExpr (#16523)
add b21855758e [Relax] Implement operators to read runtime DLTensor* information (#16563)
add e5bfb028b9 [Doc] Fixed Docstring usage example in `tvm.ir.make_node` (#16610)
add d91fe450c8 [Transform][Bugfix] Handle non-composite lambda functions in FuseOps (#16598)
add 3ef478b486 [Relax][Runtime] RNNState for Space State Models (#16568)
add bd79374a01 [Bugfix][TVMScript] Handle R.match_cast as last binding in if/else (#16562)
add ff0b99c5ce [Dlight] Scheduling Low batch GEMM using GEMV-like rule (#16579)
add ad3dfb4c1c [Bugfix][Executor] fix debug_executor function debug_get_output (#16492)
add 9fd3461c32 [Frontend][PaddlePaddle] Support conv2d when data_format is NHWC (#16616)
add cf575b8f83 [Relay][ONNX] fix the wrong default value about dtype in Multinomial converter (#16624)
add 8fe01647d1 [Relax][Frontend][Onnx] fix clip unsqueeze opset implement (#16604)
add 5308ef135d [Transform] Improvements to LazyTransformParams (#16602)
add fcfc05bb29 [Transform] Allow explicit name of bundled model parameters (#16597)
add 8f4259710a [Unity][Transform] Check for permute_dims in ExpandMatmulOfSum (#16590)
add 4b7d78d157 [Relax] Handle dynamic arguments in legalization of nn.attention (#16592)
add fc4abee022 [Relax] Fix error message in BlockBuilder (#16629)
add aa55528714 [Relay][ONNX] Fix the Resize operator in ONNX frontend (#16626)
add 72ce7013e4 [Relay][ONNX] Fix the attribute mode parse of operator Upsample (#16622)
add a6ab8cb857 [Web] Fix NDArrayCache loading report callback (#16631)
add bde28ae95a [TIR] Expand debug symbol output for CodeGenLLVM (#16544)
add fac9520921 [Unity][Transform] Raise error in FuseOpsByPattern for SSA violation (#16421)
add 33a6f75e52 [Unity][Analysis] Include impure call in VerifyWellFormed errors (#16585)
add faa6628215 [Relax] Additional unit tests for RemoveUnusedParameters (#16574)
add 84b3f69edb [Unity][SLM] GPU sampling (#16575)
add b5815753dc [Transform] Implement relax.transform.ReorderPermuteDimsAfterConcat (#16596)
add e715814985 [Relax][Transform] Preserve param names in LiftTransformParams (#16594)
add 864fd5c706 [Transform] De-duplicate MatchCast nodes in EliminateCommonSubexpr (#16599)
add 89cc09c621 [Unity][Transform] Handle dynamic shapes in CombineParallelMatmul (#16591)
add 2ca8f3131e [Bugfix][Cutlass] Check if function attributes is None (#16619)
add 8194b484e7 [Runtime] Add TVM_DLL to threading backend funcs (#16630)
add 7e269dcfc8 [RUNTIME][RPC] Enable RPCObjectRef over multi-hop RPC (#16635)
add 99e22328bf [Disco] Implement `Session.import_python_module` method (#16617)
add 3ec0ca5b0b [Disco] Expose functions to query the per-worker device/rank (#16639)
add b3fa6cb873 [AOT][Testing] Print output values on test failure (#16611)
add 563ef9587c [SVE] Add support for scalable data type strings (#16612)
add ff3716b83a [TVMScript] Represent tir::builtin::ret() using python "return" (#16640)
add c2c579bb0a [BugFix][FFI] Add a missing default for datatype lanes (#16649)
add e261a27036 [Transform] Check for zero-param operators in LiftTransformParams (#16595)
add e56c5e1ef2 [Bugfix][Transform] Preserve symbolic variables in FuseOps (#16637)
add e5420436a0 [Dlight] Skip GeMV when normalization fails (#16665)
add 3b25588926 [TOPI] improve inclusive_scan for thrust (#16652)
add 5718ff35ef [Relax][Runtime] Support Unpack API for NDArrayCache (#16648)
add 73b01eec4a Fixing workload comment (#16662)
add ad1da4ee57 [Runtime][Builtin] Using float32 accumulation in attention kernel (#16667)
add ae2ab58ad6 [KVCache] Fix the reference counter in sequence fork (#16666)
add 31bb4b58fe [TVMScript] Infer T.reads() for DeclBuffer nodes (#16663)
add 880af308ee Simplify Windows CMake Command (#16656)
add 46aaf61119 [BugFix] add the default value for DFT in ONNX frontend (#16659)
add fe5a350f47 [Relax] add sample_indices in sampling (#16675)
add 22dd8d895a Minor update docs instructions (#16609)
add a0f57a0570 [Relax] Eager free original weights in transform_params (#16674)
add ad3722f7eb [skip ci] Fix wasm exception flag (#16683)
add d284cf4212 [Relax][Frontend][NN] Add support for Conv3D (#16654)
add 6ca2341460 [Relax] Remove the legalization of cumsum/cumprob (#16676)
add e005f8574c [Frontend][PaddlePaddle] PaddlePaddle model with NCHW data format that supports quantization (#16651)
add 657880cdce [Bugfix][TIR] Fix duplicate AllocateConst in CacheReadWrite schedule primitive (#16660)
add 7b7677fc75 [TIR] Enhance and fix tensorize schedule for some case (#16560)
add 898f87ffd6 [Bugfix][TIR] Handle AttrStmt of upcoming tir.Var in ConvertSSA (#16682)
add ab5602607f Use target name instead of node name as function name (#16690)
add 48992a4093 [DeviceAPI] Support "GetCurrentStream" (#16689)
add 5bbe1aba6d [Dlight] LowBatchGemv rule only apply to function with spatial symbolic var (#16678)
add 682a62c54d [TIR] Support Vector Reinterpret Calls (#16673)
add 40dd376375 [Unity][TIR] Clear struct info when specializing PrimFunc (#16584)
add af82d97043 [Relax][Frontend][Onnx] support MaxPool1/2/3D and AveragePool1/2/3D (#16681)
add 9b3621bf39 [CI] add merge_with_main in unity (#16661)
add 7ab970d38e [Lint] Add check to prevent usage of #include <regex> (#16412)
add c43fad1d60 [Relax] Implement StructInfoPattern for dataflow pattern matching (#16685)
add 86c5df8923 [Runtime][RPC] Fix FreeObject in minrpc server (#16647)
add d894334800 [COMMUNITY] Add new key for release signing (#16695)
add f988cb48d1 [Bugfix][SLM] Produce well-formed Relax for nn.modules.KVCache (#16684)
add dc7d6873ba [Runtime] PagedKVCache execute data copy on a separate stream (#16692)
add 639a6e4f3c [Contrib] Support NDArray cache taking generator (#16693)
add 7ac03ca960 [WEB] Initial support for asyncify (#16694)
add 596db033bb [MSC][M5.1] Build wrapper to support compression (#16668)
add 254e90a82a [microNPU][ETHOSU] Fix LUT size for int16 activations (#16680)
add 95f97e881a [Relax] CUDA graph rewrite treating StringImm as static (#16691)
add cae1af62f9 [LLVM][RUNTIME] Add optional LLVM ORCJIT runtime executor (#15964)
add ca12cb6d47 [Relax][Frontend][Onnx] add sum and globalavgpool 1d/3d op (#16669)
add 1278c3544d [Disco] Propagate structlog configuration to disco workers (#16618)
add ef8c42813c [Fix][Relax] Fix top-p/top-k sampling kernel (#16703)
add 436d8f9691 [TVMScript] Allow use of relax.Expr with void type as a statement (#16641)
add f34a58a8ca [Fix][Arith] Fix canonical simplification of LE (#16704)
add fe340c9245 [Dlight] Add fallback for low batch gemv with outer reduction (#16701)
add 05e218c543 [Runtime] Add TVM_DLL to memory manager functions (#16705)
add 831d769f47 [Transform] Remove R.Object parameters after LazyTransformParams (#16699)
add 3b25697e9d [MSC][M5.2] Enable quantize && prune with gym by wrapper (#16702)
add 6a877df173 [CMake] Add "USE_FLASHINFER" to libinfo (#16710)
add dffdc3e592 [Relax][Frontend] Add op `tanh`, `exp`, `negative`, and `permute` (#16711)
add 8023a981a4 [Relax] Normalize use of void-type variable to inline R.tuple() (#16658)
add 981009d457 [Fix] PagedKVCache fetching compute stream when copy stream is needed (#16714)
add c00cc031de [Target] Automatically detect system triple when not specified by the user (#16513)
add 695f958bc9 [TIR] Improve well-formed check's handling of match buffer (#16655)
add af0c038f2e [SVE] Add codegen support for scalable buffer accesses (#16696)
add 071fb8a429 [RUNTIME] Ensure NDArray.CopyTo(Device) always sync (#16716)
add 0978ab656c [RUNTIME][METAL] Provide richer runtime when error happens (#16713)
add 939b8b9ce7 [Web] Seperate parallel shard download and iterative shard loading (#16650)
add 9ec72494cf [TIR] Implement max/min_value for fp8 data types (#16723)
add 94866f769a [VM] [Hexagon] Add buffers to `dma_wait` builtin (#16706)
add 45df1247c6 [Web] Implement linear congruential generator, make runtime seedable (#16722)
add feb104393c [TIR][CUDA] Add native FP8 support to codegen (#16548)
add e57ab7a9dc [Fix] Introduce TVM_DEBUG_WITH_ABI_CHANGE to warn ABI changes in debug mode (#16728)
add 174f46ea1d Fix cpp_rtvm cmake build on Windows (#16724)
add b8f64c21c5 [Builtin] Sliding window and sink support for PagedKVCache (#16729)
add 1c73491669 [Dlight] Fix GeMV shared memory estimation (#16731)
add fbfa926585 [Relax] Implement relax.transform.TopologicalSort (#16697)
add 6df42d4b98 [Bugfix][TIR] Avoid overwrite of unmanaged buffer allocations (#16726)
add 5cbcaf45a4 [MetaSchedule] Make the `opt_level` of `tune_relay()` adjustable (#16725)
add 95ec38be98 [Arith] Provide tighter ConstIntBounds for special cases (#16588)
add 8b5bd555d1 [Target][CUDA] Allow non-numeric arch as needed for latest gpu (#16736)
add 97d92b5483 [Refactor][Runtime] Always specify device in allocator interface (#16738)
add 7641c6e20e [CLML] Fix build TVM with CLML on MacOS (#16672)
add ff6ce9c2b3 Enable Shared Function in LiftTransformParam Pass (#16717)
add 48cedc7d2e [Arith][Fixup] Require feature flag for tighter inequality bounds (#16735)
add a9436b8154 [Fix][Builtin] Fix "GetQueryPosition" of PagedKVCache (#16746)
add 89e9028849 [Cutlass] Add group gemm kernels (#16751)
add 7683bc23b1 [Fix] Lazy import of "psutil" in disco process pool (#16752)
add 0f38ef2d6e [Bugfix][TIR] Fix cache_read update buffer region (#16742)
add e257fb8a41 [Runtime] CUDA IPC Memory support and custom allreduce kernels (#16750)
add 62beb0251e [microNPU][ETHOSU] Add fixed point for tanh (#16266)
add 858486fe8e [Relax][Pass] Lowering passes for GPU IPC memory and allreduce (#16759)
add f9b38ab711 [SME][Docker] Add Fixed Virtual Platform (FVP) and toolchain install (#16755)
add 6c701fe5b8 [Unity][Parser] Check well-formedness in the parser (#16569)
add 89cd74c07d [CONTRIB] Add nm symbol dump (#16763)
add a2de07c772 [SME][AOT] Add Fixed Virtual Platform (FVP) functional testing infrastructure (#16749)
add 31803e6ec7 [LLVM] Lack of DWARF type is not an error (#16748)
add 1cccc3b5d6 [SLM] Allow modules to define pre-processing of weights (#16757)
add 122398995d [VM][Hexagon] Cache operations when bypass mode is enabled (#16762)
add 134f8fd208 [Fix] Remove redundant "remove_all_unused" in IPC memory lowering (#16771)
add 21e1380063 [Hotfix] Revert driver API pass ordering that breaks MLC, mark failing test (#16770)
add 77a7b01081 [RUNTIME][OPENCL] Bugfix for ciImage create with host ptr (#16768)
add 80bcf4cdb7 [Fix][Dlight] (Low-batched-)GeMV on small spatial loops (#16775)
add 6d97b95eed [Fix] Fix the purity flag of "vm.call_tir_dyn" and "kill" ops (#16773)
add d79b1dd955 [Contrib] Remove thrust "built but not used" warning (#16776)
add cb08f0d57b [TIR][Driver] Use `BindTarget` to specify target for FP8 legalization (#16767)
add 9899f9cd28 [AOT][Testing] Improve output mismatch information on test failure (#16765)
add b3981d2f77 [Fix] Fix numpy dtype map (#16780)
add 5a8d928b84 [Relax] Improve malform error msg (#16779)
add ef46f4e8d3 Revert "[SLM] Allow modules to define pre-processing of weights" (#16777)
add b2204ae698 [IR] Default to empty attributes, instead of NULL (#16745)
add 69c091400a [Fix] Fix build errors with VS2022 (#16790)
add ae7b8d9aed [Codegen, Cuda] Add overload for fp8x4 e5m2 <-> half4 conversion (#16787)
add 72f0326a88 [Analysis] Allow calls to GlobalVar in @R.function (#16778)
add bf2d43e314 [IR][Relax] Improve highlighting in assert_structural_equal (#16756)
add bcfbcabff8 [Bugfix][Cutlass] Remove a typo in cutlass build (#16789)
add 016b512ad4 [Relax] Refactor PatternRewriter into separate Block/Expr mutators (#16730)
add 8274d142a3 [Relax] Implement operators to inspec DLTensor::strides and offset (#16721)
add 571fdaf1eb [Web] Add `kv_state` and `rnn_state` to wasm_runtime (#16791)
add 4f3a863c1f [Cutlass] Add check for group gemm param shapes (#16788)
add ac2f47867f [SME] Add support for inserting processor state annotations (#16761)
add a768ee4900 [Fix] fix for numpy 2.0 compatibility (#16793)
add d43e1ab71d [Doc] Fix set_axis_separator example (#16792)
add 2f889774ec [3rdparty] AUTO mode for custom all-reduce strategy (#16797)
add ceb8e224b9 [Relax] Improve CanonicalizeBindings in DataflowVar edge case (#16783)
add 1891b4db49 [Disco] Propagate structlog/logging config to workers (#16715)
add 726a141649 [Target] Use LLVM target parser for determining Arm(R) A-Profile Architecture features (#16425)
add 86b5a1301c [Relax] Allow composition of DFPattern replacements (#16732)
add 83e7e9b2eb [Debug] Improve error messages in LiftTransformParams (#16802)
add 4c45b828be [Relax] Unit-test for structural equal of recursive function (#16796)
add cb31cb3e4f [Debug] Improve error message in VMShapeLower (#16806)
add eb5458e0e9 [Relax] Allow R.Prim('bool') in relax::If and assert_op (#16642)
add 8ee8d0d0b8 [Runtime] Add "TVM_DLL" to NVTX header (#16809)
add 3ce87cba21 Fix includes of custom allreduce kernel (#16814)
add 64db9f78a0 [Runtime] Introduce MSCCLPP with NCCL equivalent interface (#16804)
add d109573cb4 [Runtime][LLVM] Fix errors during loading of target tags (#16808)
add f8b9a5faa4 [SLM] Add unit tests for SLM to Relax exporter (#16784)
add d2c7167913 [Cutlass] Fix usage of cuda stream for group gemm (#16818)
add cd60f6d4fe [Cmake] Allow using custom CCCL path for thrust (#16816)
add 109804cc6a [Codegen] Add check to disable invalid reinterpret (#16786)
add 5daa303ce7 [Fix] PAPI docs (#16820)
add c3be89a407 [KVCache] Support forking sequence at specific posotion (#16813)
add 5053a4f29f [LLVM] Fix compilation failure due to minor change (#16812)
add eb4175bd3d [VM] Recycle VMFrame (#16822)
add ef32a611e3 [Relax] Enable capturing symbolic shapes in cuda graph (#16815)
add 78ba385fcb [BugTIR]fix error merging shared memory for ptx_cp_async (#16800)
add a39067bf7a [Fix] Add TVM_DLL to Disco session (#16821)
add b4b97f8278 [Web] Allow custom bc files in emcc making (#16825)
add ffa9cfd0dd [BugFix][Ansor] Fixing Ansor Gradient Bug (#16739)
add 384b7f7a74 [KVCache] Introducing auxiliary data manager (#16824)
add b5fda2d93a [TIR] Ramp and Broadcast lanes fixed to int32 dtype (#16795)
add 3f615dcc3e [Bugfix][Relax] BlockBuilder may not assume unique input functions (#16805)
add 00395ae43d [Relax][Bugfix] Provide the full Expr to pattern-match rewriter (#16828)
add fc78b22fbc [Relax][VM] Refactor CUDA graph builtins as VM extension (#16823)
add f83a32906f [Relax] Share storage allocs among functions after cuda graph rewriting (#16830)
add ca99a98dfe [Disco] Reduce Process/ThreadSession message queue reads and writes (#16817)
add c20cdafcbc [SME] Target parser support for SME (#16794)
add 3a423615ee [Disco] Support setting workers' CPU affinity (#16807)
add ef80af65dd [Web] Support building tvm/web on Windows (#16810)
add 9862c84b9f [KVCache] Reducing CacheAuxDataManager copy size (#16831)
add 54e31f374c [Relax] Capture symbolic vars in struct info of weights (#16834)
add 35c614303a [TVMScript] Do not throw error for duplicate definitions (#16811)
add 545e0977e3 [Relax] Allow DeadCodeElimination within ApplyPassToFunction (#16801)
add 61249b41ce [Relax][Transform] Provide callback versions of LazyTransformParams (#16798)
add 6f74762743 [Relax] Provide well-formed output in `transform.LazyGetInput` (#16841)
add c84f6bb4fd Bump pillow from 10.2.0 to 10.3.0 in /apps/microtvm/ethosu (#16838)
add dd384906e3 Bump pillow from 10.2.0 to 10.3.0 in /apps/microtvm/cmsisnn (#16839)
add 53f05d8ded [Debug][Disco] Check if a PackedFunc exists before calling it (#16845)
add cd08356e66 [TIR] Fix segfaults from ordering of Let/Assert in MakePackedAPI (#16543)
add c93f0bae9b [Meta-Schedule][OpenCL] Enable MS tuning for Android OpenCL (#16846)
add ab94ca3b91 [CI] Disable flaky unit test (#16837)
add b01de08715 [DLight] Fix a corner case for reduction rule (#16848)
add b91d4e55b3 [TVMScript] Produce empty DictAttrs when R.func_attrs is absent (#16844)
add ee3f7bc855 [MSC][M5.3] Support torch.dynamo for dynamic models (#16772)
add 9b5a7a457f [IR] Provide well-formed intermediate in ApplyPassToFunction (#16843)
add 3e802d12f1 [Relax,Topi] Allow passing workspace to thrust to avoid allocations (#16851)
add a156181ee3 [Relax] Fix EliminiateCommonSubexpr removing alloc tensor (#16852)
add a7be540457 [KVCache] Initialize one extra page than specified (#16849)
add 97d7a3512b Fixing probability comment (#16850)
add 81a850693d [TIR] Use constructor for new PrimFunc in TransformLayout (#16832)
add d1e24ca721 [Web] Support web indexDB cache for larger model storage (#16733)
add 0594994c7d [ONNX] Fix interpreting auto_pad parameters in ConvTranspose operator (#16001)
add a309b6b857 [Thrust] Use pointer to tls pool to prevent creating new pool (#16856)
add 4d4f0508a2 [SVE] Support scalable vectors in LoopVectorizer (#16782)
add 95cb0de27a [VULKAN] Fix CLZ support for Vulkan (#16858)
add a482b4c191 [Picojson] Let the key of objects in json be ordered by default (#16863)
add 2829b59e1c [TVMScript] Add parser and printer support for e4m3/e5m2 fp8 (#16864)
add 6748215b42 [Codegen, CUDA] Add handling of fp8 broadcast / const (#16865)
add 4617efac7b [Relax] Dispatch sort/scan for non-cuda gpu backends (#16867)
add f9e36fcbf8 [3rdparty] Bump FlashInfer (#16866)
add c67a055386 [BugFix][Target] Added null check to fix segfault at ->defined() in cpu.cc DetectSystemTriple() (#16766)
add 4b906554af [OpenCL] Add OpenCL device for automatic target detection (#16854)
add 0aae97d8e4 [PageKV] allow PopN to pop all the tokens in last block (#16871)
add 88a1c6560c [3rdparty] Bump flashinfer (#16868)
add 3f09e7f5ce [Thrust] Fix thrust workspace allocation (#16873)
add 0a3fe22208 [Relax] Enhance symbolic expr estimation in memory planning (#16872)
add 5c80691c81 [Dlight] Enhance vectorization loading weight for gemv (#16878)
add 64969035fd [release] Update version to 0.16.0 on main branch
add d0cbb02e1d [release] Update version to 0.17.dev0 on main branch
add 64911ab5da [Runtime] Implemented Datatype.itemsize() (#16880)
add a64d1f1cc3 [TIR] Make T.reinterpret nop when dtype is the same (#16879)
add f267691fa4 [Relax] Stabilize relax pass mutation order (#16883)
add d4056ca795 [SVE] Support splitting by vscale in `tir::split` and `te::split` (#16862)
add cdfdd0e4ec [Contrib] Enable fp16 for thrust sort (#16887)
add e738f1d4f1 [Relax][Frontend] Fix sort, argsort and topk in nn module (#16886)
add 95d6778908 [dlight] Add check for matmul dtype and fix reduction rule (#16884)
add d1ac73ca2d [CUBLAS][FP8] Support e4m3 gemm in cuBLAS BYOC (#16888)
add 3680a0d5a2 [RUNTIME][VULKAN] Support total_global_memory (#16890)
add 08965f08ff [CUBLAS] Set fp32 compute and scale dtypes in fp16 matmul (#16892)
add 4cb4605ba3 [TVMScript][Bug] Add test case for missing symbolic bounds (#16877)
add 94a44d7d62 [QoL][Relax] Return well-formed IR from relax::Function::CreateEmpty (#16861)
add 460f6f1d3e [QoL][Relax] Infer StructInfo for relax::Tuple on construction (#16860)
add d030ce27a1 [TVMScript] Optionally use `ruff format` instead of `black` (#16876)
add 857fe614ab [Target] Don't register AArch64 target tags without LLVM compiler support (#16897)
add b3ffd97569 [BYOC] Add layout check and update shape check for cublas FP8 BYOC (#16895)
add da56c89f32 [Dlight] Enhance vectorization for gpu matmul (#16894)
add de91c5ca94 [Bugfix] rocm shared memory issue on MI250 (#16901)
add 7dc0472aef [Bugfix] CudaDeviceAPI::GetAttr may check kExist when GPUs absent (#16903)
add 59376eeca3 [Relax] Allow specifying entry_funcs for BYOC (#16902)
add fe5270956d [CMAKE] Misc improvment of Util (#16900)
add 622bd150dd [Relax] Handle binary operations between Tensor and PrimValue (#16827)
add 36efa36f53 [Upd] Fixed lld search in rocm (#16907)
add 6afbc12e27 [Bugfix][Relax] Raise exception for OOM allocation (#16905)
add 2978427c2a [Relax] Prevent to generate duplicate func in dispatch_sort_scan (#16904)
add a2511cc516 [QoL][Relax] Use SeqExpr in IR types when SeqExpr is required (#16859)
add 6b77cbabe8 [Misc] Enhance Release Note Script and Remove Useless File (#16913)
add 57316dae14 [Web] Support string[] in setPackedFunc() and exceptionally long arrays (#16910)
add 29534b70fa [SVE] Check for SVE target in VectorizeLoop (#16893)
add b0143d106f [CMAKE] Make LOG_BEFORE_THROW explicit (#16914)
add 342f4721c4 [Disco] Improve error message for CallPacked (#16919)
add 11f2253b9c Restore "pytest.mark.gpu" for RELAX tests (#16741)
add 2f395f1756 [SVE][TOPI] Add conv2d NHWC hybrid SVE schedule for `arm_cpu` (#16899)
add 5cf4ca6d15 [Marvell BYOC]: Marvell AI Accelerator Integration - Phase 2 (#16915)
add 4f8c03fad3 [TVMScript] Support `T.launch_thread` with i64 dtype (#16916)
add 39f2482580 [Fix] Fix SSA conversion for SizeVar retention (#16924)
add 51cfb70f86 [Fix][Dlight] Fix GeneralReduction for log-sum-exp (#16923)
add 5bd10472e9 [SCRIPT][ADRENO] Fix in build config for adreno (#16927)
add 278a6af085 [Relax][TIR] Introduce new `cumsum` op for gpu (#16934)
add 97ff7cc4f1 [VM][OPENCL] Take advantage of OpenCL host ptr for improved copy (#16929)
add 1453893be0 [CLML] Fix in clml pattern check condition (#16933)
add 3ff3daa26d [CI] Upgrade CUDA to 12.4 (#16939)
add 63e0a0ff82 [Thrust] Increase static workspace size (#16937)
add 0b09ed0185 [3rdparty] Bump FlashInfer for sampling functions (#16935)
add b54f57aa72 [TFLite] Add support for GELU conversion (#16936)
add 081c23becf [Relax] Allow PrimValue as index in relax.op.take (#16940)
add b00fc55654 [CI] Enable Conda setup v3 (#16942)
add e10cdc5d4c [tir][Compute-at] Make compute-ated block simple when the predicate could be merged (#16945)
add 2d7663ceeb [CI] Use LLVM17 for tests on `ci_cpu` (#16931)
add dd09c85f87 [CI] Update image tag to 20240428-060115-0b09ed018 (#16948)
add c0385c7523 [Runtime] Allow offset to be specified in NDArray::CreateView (#16938)
add b4a69de47b Enable gemv schedule for adreno (#16932)
add 114ad70a22 [TOPI] Revert unification of conv2d NHWC hybrid scheduling for `arm_cpu` targets (#16951)
add c8deb7fa36 Overriding the StructuralEqual() for easy usage (#16908)
add bc8742b4c9 [Misc] Add script for testing release package (#16956)
add 6252fa5802 [TIR] Enhance CLZ intrinsic support (#16952)
add a320b63198 [Unity][Cutlass] Fix C source generation of dense operation (#16476)
add 20d769617f [Relax] Express dynamic arguments of strided_slice as arguments (#16826)
add effa5d7993 [CUBLAS] Enable offloading of R.matmul + R.dequantize (#16896)
add 944d180fba [SVE] Add get_active_lane_mask builtin (#16965)
add 59ef0ee9d8 [Bugfix][ONNX] Improve broadcast and batch_matmul conversion (#16961)
add 9cfebca136 [TVMScript] Fix error reporting inside Macro func (#16967)
add 876f52805d [LLVM] Stringref API deprecation fixes (#16968)
add 28d32b52cb [TIR] Support narrow dtype for let binding (#16947)
add 819b0023e4 [Relax] Support nested ModuleList in nn.Module (#16971)
add 02c4c55eaa [SVE] Add codegen support for `vscale_range()` function attribute (#16962)
add c0a47ed139 [CUBLAS][FP8] Enable R.matmul + R.multiply offloading (#16974)
add 4c1ebcf81a [Relax] Implement relax.op.view (#16955)
add fffd168d00 [Unity][BYOC] Use arith.Analyzer to check batch equality of matmul in cublas (#16982)
add 2565aa38ef [BugFix][Relax] change FuseOpsByPattern strategy to pattern-match maximal subgraph (#16922)
add 825dc1ffb5 [TOPI] Remove `blockIdx.z` in topi sort (#16977)
add 4403379e39 [JVM] Automatic Compatibility of JVM AttachCurrentThread (#16987)
add d1ac1c0202 [KVCache] Fix the aux data syncing order of paged KV cache (#16988)
add 1d4b9ea5c3 [UnitTest] Use pytest's scope='session' for tvm.testing.parameter (#16930)
add fd820ade5f [Disco] Expose disco.Session.shutdown through the python API (#16979)
add 29337449db [Cuda] Skip FreeDataSpace when CUDA driver is in inconsistent state (#16980)
add eb242ec77b [DLight] Check for target in function attributes (#16958)
add 0dfc5f955e [Unity] Check for transpose and dynamic shape in AdjustMatmulOrder (#16589)
add 5b5f8d0f77 [QoL][IR] Provide std::hash and std::equal_to for IR Variable types (#16909)
add c2d14ae872 [Relax][Transform] Handle identical PrimFunc with distinct VDevice (#16959)
add c6a8a80009 [Disco] Allow allocation that only exists on worker0 (#16993)
add 93233a988e [Disco] Treat hangup of disco worker process as kShutdown (#16989)
add 54c68d6af4 [Disco] Implement `num_workers` property for `disco.Session` (#16978)
add d9dbbc9154 [Bugfix][Disco] Handle NDArray larger than OS buffer for pipe (#16992)
add b7467aa27b [Relay] fixed to make TupleGetItem inherits the previous span (#16996)
add cfe1711934 chore: remove repetitive words (#16957)
add b49468ddf1 [SME] Introduce scalable fp32 dense schedule (#16921)
add f044eefd0e [Runtime][Disco] Restore checks for hangup of disco pipe (#16997)
add afb6416234 [WebGPU] Handle device OOM in createBuffer (#17005)
add 3cd6673890 [Runtime] Allow query of available device memory through DeviceAPI (#16994)
add 18a2a250f8 [KVCache] Support KVCache decode from forked sequence and pop more tokens (#16995)
add 209971a62e [DLIGHT][GPU] Improved gemv outer fallback schedule (#16973)
add 3b976585c7 [DLIGHT][GPU] Enhance opencl thread limit for schedules (#16972)
add 2e56421dda [DLight] Update Adreno GEMV Rules (#17016)
add a5862a5c69 [SVE] Use only powers of two as possible vscale values (#17001)
add ac9a943c4d [TOPI][Testing] Enable conv2d NHWC fp16 topi testing for `arm_cpu` (#17007)
add e978a449f9 [COMMUNITY] New committer: Balint Cristian (#17018)
add b1951a7811 [USMP] add missing const specifier for global_const_workspace (#16999)
add 41a13e7f5f Update
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (63697b59be)
\
N -- N -- N refs/heads/ci-docker-staging (41a13e7f5f)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
No new revisions were added by this update.
Summary of changes:
.asf.yaml | 2 +-
.github/actions/setup/action.yml | 12 +-
.github/workflows/main.yml | 18 +-
.gitmodules | 3 +
3rdparty/cutlass | 2 +-
3rdparty/cutlass_fpA_intB_gemm | 2 +-
3rdparty/flashinfer | 1 +
3rdparty/libflash_attn | 2 +-
3rdparty/mscclpp/include/common.h | 107 +
3rdparty/mscclpp/include/msccl.cuh | 323 +
3rdparty/mscclpp/include/msccl.h | 494 ++
3rdparty/picojson/picojson.h | 107 +-
3rdparty/picojson/test_picojson.cpp | 65 +
3rdparty/tensorrt_llm/custom_allreduce_kernels.cu | 401 +
3rdparty/tensorrt_llm/custom_allreduce_kernels.h | 81 +
CMakeLists.txt | 115 +-
CONTRIBUTORS.md | 1 +
KEYS | 118 +
LICENSE | 1 +
apps/benchmark/adreno/adreno_gpu_bench_clml.py | 11 -
apps/benchmark/adreno/adreno_gpu_bench_texture.py | 11 -
apps/benchmark/util.py | 11 -
apps/cpp_rtvm/CMakeLists.txt | 4 +-
apps/cpp_rtvm/main.cc | 2 +-
apps/ios_rpc/CMakeLists.txt | 3 +-
apps/microtvm/cmsisnn/requirements.txt | 125 +-
apps/microtvm/ethosu/requirements.txt | 125 +-
apps/microtvm/poetry.lock | 134 +-
apps/microtvm/pyproject.toml | 2 +-
apps/relax_examples/e2e_auto_tir.py | 253 +
apps/relax_examples/mlp.py | 57 +
apps/relax_examples/nn_module.py | 69 +
apps/relax_examples/resnet.py | 53 +
ci/jenkins/docker-images.ini | 20 +-
ci/jenkins/generated/arm_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/cortexm_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/cpu_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/docker_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/gpu_jenkinsfile.groovy | 17 +-
ci/jenkins/generated/hexagon_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/i386_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/lint_jenkinsfile.groovy | 9 +-
.../generated/minimal_cross_isa_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/minimal_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/riscv_jenkinsfile.groovy | 9 +-
ci/jenkins/generated/wasm_jenkinsfile.groovy | 9 +-
ci/jenkins/templates/gpu_jenkinsfile.groovy.j2 | 2 +-
ci/jenkins/templates/utils/Build.groovy.j2 | 5 +-
ci/jenkins/templates/utils/base.groovy.j2 | 2 +-
ci/jenkins/unity_jenkinsfile.groovy | 377 +
ci/scripts/github/github_tvmbot.py | 1 +
ci/scripts/jenkins/cmd_utils.py | 44 +-
cmake/config.cmake | 29 +
cmake/modules/CUDA.cmake | 40 +-
cmake/modules/Hexagon.cmake | 3 +
cmake/modules/LLVM.cmake | 1 +
cmake/modules/LibInfo.cmake | 7 +
cmake/modules/OpenCL.cmake | 2 +-
cmake/modules/contrib/CLML.cmake | 10 +
cmake/modules/contrib/CUTLASS.cmake | 54 +-
cmake/modules/contrib/DNNL.cmake | 8 +-
cmake/modules/contrib/MSC.cmake | 30 +
cmake/modules/contrib/MSCCLPP.cmake | 50 +
cmake/modules/contrib/Mrvl.cmake | 31 +
cmake/modules/contrib/TensorRT.cmake | 2 +-
cmake/modules/contrib/vllm.cmake | 25 +
cmake/utils/FindCUDA.cmake | 9 +
cmake/utils/FindLLVM.cmake | 18 +
cmake/utils/FindNCCL.cmake | 5 +-
cmake/utils/Utils.cmake | 7 +-
conda/build-environment.yaml | 2 +-
conda/recipe/conda_build_config.yaml | 4 +-
conda/recipe/meta.yaml | 7 +-
docker/Dockerfile.ci_cpu | 5 +
docker/Dockerfile.ci_gpu | 2 +-
docker/Dockerfile.demo_mrvl | 47 +
docker/bash.sh | 11 +
.../install/ubuntu2004_install_python_package.sh | 1 -
docker/install/ubuntu_install_aprofile_aem.sh | 54 +
docker/install/ubuntu_install_boost.sh | 2 +-
docker/install/ubuntu_install_emscripten.sh | 4 +-
docker/install/ubuntu_install_nodejs.sh | 2 +-
docker/install/ubuntu_install_papi.sh | 2 +-
docker/install/ubuntu_install_python_package.sh | 1 -
docker/install/ubuntu_install_sccache.sh | 2 +-
docs/conf.py | 3 +-
docs/how_to/deploy/index.rst | 1 +
docs/how_to/deploy/mrvl.rst | 252 +
docs/how_to/profile/papi.rst | 4 +-
docs/install/from_source.rst | 2 +-
gallery/how_to/compile_models/from_mxnet.py | 147 -
.../how_to/deploy_models/deploy_model_on_nano.py | 24 +-
.../how_to/deploy_models/deploy_model_on_rasp.py | 26 +-
.../how_to/deploy_models/deploy_prequantized.py | 2 +-
gallery/how_to/deploy_models/deploy_quantized.py | 166 -
.../how_to/extend_tvm/bring_your_own_datatypes.py | 409 -
.../tune_with_autoscheduler/tune_network_arm.py | 14 +-
.../tune_with_autoscheduler/tune_network_cuda.py | 16 +-
.../tune_with_autoscheduler/tune_network_mali.py | 17 +-
.../tune_with_autoscheduler/tune_network_x86.py | 15 +-
gallery/how_to/tune_with_autotvm/tune_relay_arm.py | 11 -
.../how_to/tune_with_autotvm/tune_relay_cuda.py | 11 -
.../tune_with_autotvm/tune_relay_mobile_gpu.py | 11 -
gallery/how_to/tune_with_autotvm/tune_relay_x86.py | 11 -
gallery/how_to/work_with_msc/_resnet.py | 350 +
gallery/how_to/work_with_msc/using_tools.py | 140 +
gallery/how_to/work_with_msc/utils.py | 112 +
include/tvm/arith/analyzer.h | 29 +
include/tvm/ir/attrs.h | 7 +-
include/tvm/ir/expr.h | 38 +
include/tvm/ir/function.h | 133 +-
include/tvm/ir/global_info.h | 128 +
include/tvm/ir/module.h | 26 +-
include/tvm/ir/name_supply.h | 46 +-
include/tvm/ir/transform.h | 79 +-
include/tvm/ir/type.h | 3 +-
include/tvm/node/script_printer.h | 42 +
include/tvm/node/structural_equal.h | 4 +-
include/tvm/relax/analysis.h | 604 ++
include/tvm/relax/attrs/ccl.h | 62 +
include/tvm/relax/attrs/create.h | 54 +
include/tvm/relax/attrs/datatype.h | 53 +
include/tvm/relax/attrs/distributed.h | 48 +
include/tvm/relax/attrs/image.h | 81 +
include/tvm/relax/attrs/index.h | 57 +
include/tvm/relax/attrs/linear_algebra.h | 53 +
include/tvm/relax/attrs/manipulate.h | 162 +
include/tvm/relax/attrs/nn.h | 583 ++
include/tvm/relax/attrs/op.h | 95 +
include/tvm/relax/attrs/qdq.h | 50 +
include/tvm/relax/attrs/search.h | 48 +
include/tvm/relax/attrs/sorting.h | 99 +
include/tvm/relax/attrs/statistical.h | 67 +
include/tvm/relax/backend.h | 51 +
include/tvm/relax/binding_rewrite.h | 115 +
include/tvm/relax/block_builder.h | 296 +
include/tvm/relax/dataflow_matcher.h | 104 +
include/tvm/relax/dataflow_pattern.h | 968 +++
include/tvm/relax/dataflow_pattern_functor.h | 187 +
include/tvm/relax/distributed/axis_group_graph.h | 478 ++
include/tvm/relax/distributed/global_info.h | 92 +
include/tvm/relax/distributed/struct_info.h | 192 +
include/tvm/relax/distributed/transform.h | 76 +
include/tvm/relax/exec_builder.h | 181 +
include/tvm/relax/expr.h | 1144 +++
include/tvm/relax/expr_functor.h | 551 ++
include/tvm/relax/nested_msg.h | 602 ++
include/tvm/relax/op_attr_types.h | 96 +
include/tvm/relax/struct_info.h | 473 ++
include/tvm/relax/struct_info_functor.h | 157 +
include/tvm/relax/tir_pattern.h | 75 +
include/tvm/relax/transform.h | 650 ++
include/tvm/relax/tuning_api.h | 396 +
include/tvm/relax/type.h | 166 +
include/tvm/relax/utils.h | 150 +
include/tvm/relay/function.h | 2 +-
include/tvm/relay/transform.h | 14 +-
include/tvm/runtime/c_runtime_api.h | 2 +-
include/tvm/runtime/container/map.h | 20 +-
include/tvm/runtime/container/shape_tuple.h | 29 +
include/tvm/runtime/data_type.h | 84 +-
include/tvm/runtime/device_api.h | 10 +
include/tvm/runtime/disco/builtin.h | 124 +
include/tvm/runtime/disco/cuda_ipc_memory.h | 102 +
include/tvm/runtime/disco/disco_worker.h | 95 +
include/tvm/runtime/disco/session.h | 349 +
include/tvm/runtime/logging.h | 18 +-
include/tvm/runtime/memory/memory_manager.h | 40 +-
include/tvm/runtime/module.h | 4 +
include/tvm/runtime/ndarray.h | 32 +-
include/tvm/runtime/nvtx.h | 54 +
include/tvm/runtime/object.h | 18 +-
include/tvm/runtime/packed_func.h | 47 +-
include/tvm/runtime/relax_vm/builtin.h | 89 +
include/tvm/runtime/relax_vm/bytecode.h | 245 +
include/tvm/runtime/relax_vm/executable.h | 225 +
.../tvm/runtime/relax_vm/ndarray_cache_support.h | 95 +
include/tvm/runtime/relax_vm/vm.h | 236 +
include/tvm/runtime/threading_backend.h | 27 +-
include/tvm/script/ir_builder/ir/frame.h | 4 +
include/tvm/script/ir_builder/relax/frame.h | 297 +
include/tvm/script/ir_builder/relax/ir.h | 146 +
include/tvm/script/ir_builder/tir/frame.h | 2 +-
include/tvm/script/ir_builder/tir/ir.h | 15 +-
include/tvm/script/printer/ir_docsifier.h | 7 +
include/tvm/te/operation.h | 2 +-
include/tvm/te/schedule.h | 20 +-
include/tvm/tir/analysis.h | 15 +-
include/tvm/tir/buffer.h | 14 +-
include/tvm/tir/builtin.h | 58 +
include/tvm/tir/data_type_rewriter.h | 14 +-
include/tvm/tir/expr.h | 8 +-
include/tvm/tir/function.h | 9 +-
include/tvm/tir/op.h | 11 +-
include/tvm/tir/schedule/schedule.h | 22 +-
include/tvm/tir/transform.h | 28 +-
include/tvm/tir/var.h | 30 +
include/tvm/topi/nn/group_norm.h | 31 +-
include/tvm/topi/nn/layer_norm.h | 28 +-
include/tvm/topi/nn/rms_norm.h | 24 +-
include/tvm/topi/transform.h | 166 +-
.../src/main/native/org_apache_tvm_native_c_api.cc | 29 +-
jvm/pom.xml | 4 +-
..._fpA_intB_gemm.txt => LICENSE.tensorrt_llm.txt} | 0
licenses/LICENSE.vllm.txt | 201 +
python/tvm/__init__.py | 6 +
python/tvm/_ffi/libinfo.py | 14 +-
python/tvm/_ffi/runtime_ctypes.py | 58 +-
python/tvm/arith/__init__.py | 2 +-
python/tvm/arith/analyzer.py | 38 +-
python/tvm/auto_scheduler/task_scheduler.py | 6 +
python/tvm/auto_scheduler/workload_registry.py | 2 +-
python/tvm/autotvm/tuner/__init__.py | 8 +-
python/tvm/autotvm/tuner/droplet_tuner.py | 142 +
python/tvm/autotvm/tuner/droplet_turner.py | 127 -
python/tvm/contrib/cc.py | 60 +-
python/tvm/contrib/coreml_runtime.py | 1 +
python/tvm/contrib/cudnn.py | 21 +-
python/tvm/contrib/cutlass/attention_operation.py | 333 +
python/tvm/contrib/cutlass/build.py | 506 +-
python/tvm/contrib/cutlass/conv2d_operation.py | 61 +-
python/tvm/contrib/cutlass/gemm_operation.py | 224 +-
python/tvm/contrib/cutlass/gemm_profiler.py | 4 +-
python/tvm/contrib/cutlass/gen_gemm.py | 70 +-
python/tvm/contrib/cutlass/gen_tensor_op.py | 424 +-
python/tvm/contrib/cutlass/layer_norm_operation.py | 48 +
python/tvm/contrib/cutlass/library.py | 8 +
python/tvm/contrib/cutlass/rms_norm_operation.py | 47 +
python/tvm/contrib/debugger/debug_executor.py | 10 +-
python/tvm/contrib/emcc.py | 41 +-
python/tvm/contrib/hexagon/session.py | 37 +-
python/tvm/contrib/mrvl.py | 457 ++
python/tvm/contrib/msc/__init__.py | 17 +
python/tvm/contrib/msc/core/__init__.py | 17 +
python/tvm/contrib/msc/core/_ffi_api.py | 21 +
python/tvm/contrib/msc/core/codegen/__init__.py | 20 +
python/tvm/contrib/msc/core/codegen/codegen.py | 256 +
python/tvm/contrib/msc/core/codegen/sources.py | 219 +
python/tvm/contrib/msc/core/frontend/__init__.py | 19 +
python/tvm/contrib/msc/core/frontend/translate.py | 343 +
python/tvm/contrib/msc/core/gym/__init__.py | 21 +
python/tvm/contrib/msc/core/gym/agent/__init__.py | 20 +
.../tvm/contrib/msc/core/gym/agent/base_agent.py | 325 +
python/tvm/contrib/msc/core/gym/agent/method.py | 83 +
.../tvm/contrib/msc/core/gym/agent/search_agent.py | 179 +
.../tvm/contrib/msc/core/gym/control/__init__.py | 20 +
.../tvm/contrib/msc/core/gym/control/configer.py | 95 +
.../tvm/contrib/msc/core/gym/control/controller.py | 105 +
.../tvm/contrib/msc/core/gym/control/namespace.py | 40 +
python/tvm/contrib/msc/core/gym/control/service.py | 826 ++
python/tvm/contrib/msc/core/gym/control/worker.py | 220 +
.../contrib/msc/core/gym/environment/__init__.py | 21 +
.../contrib/msc/core/gym/environment/base_env.py | 429 +
.../tvm/contrib/msc/core/gym/environment/method.py | 206 +
.../contrib/msc/core/gym/environment/prune_env.py | 94 +
.../msc/core/gym/environment/quantize_env.py | 77 +
python/tvm/contrib/msc/core/gym/namespace.py | 40 +
python/tvm/contrib/msc/core/ir/__init__.py | 19 +
python/tvm/contrib/msc/core/ir/graph.py | 1027 +++
python/tvm/contrib/msc/core/ir/translate.py | 342 +
python/tvm/contrib/msc/core/runtime/__init__.py | 20 +
python/tvm/contrib/msc/core/runtime/hook.py | 194 +
python/tvm/contrib/msc/core/runtime/jit.py | 365 +
python/tvm/contrib/msc/core/runtime/runner.py | 1577 ++++
python/tvm/contrib/msc/core/tools/__init__.py | 24 +
python/tvm/contrib/msc/core/tools/configer.py | 108 +
.../tvm/contrib/msc/core/tools/distill/__init__.py | 21 +
.../tvm/contrib/msc/core/tools/distill/configer.py | 57 +
.../contrib/msc/core/tools/distill/distiller.py | 268 +
.../tvm/contrib/msc/core/tools/distill/method.py | 74 +
python/tvm/contrib/msc/core/tools/execute.py | 403 +
.../tvm/contrib/msc/core/tools/prune/__init__.py | 21 +
.../tvm/contrib/msc/core/tools/prune/configer.py | 93 +
python/tvm/contrib/msc/core/tools/prune/method.py | 120 +
python/tvm/contrib/msc/core/tools/prune/pruner.py | 534 ++
.../contrib/msc/core/tools/quantize/__init__.py | 21 +
.../contrib/msc/core/tools/quantize/configer.py | 125 +
.../tvm/contrib/msc/core/tools/quantize/method.py | 474 ++
.../contrib/msc/core/tools/quantize/quantizer.py | 260 +
python/tvm/contrib/msc/core/tools/tool.py | 1629 ++++
.../tvm/contrib/msc/core/tools/track/__init__.py | 21 +
.../tvm/contrib/msc/core/tools/track/configer.py | 66 +
python/tvm/contrib/msc/core/tools/track/method.py | 98 +
python/tvm/contrib/msc/core/tools/track/tracker.py | 196 +
python/tvm/contrib/msc/core/transform/__init__.py | 20 +
python/tvm/contrib/msc/core/transform/pattern.py | 874 ++
python/tvm/contrib/msc/core/transform/transform.py | 170 +
python/tvm/contrib/msc/core/utils/__init__.py | 27 +
python/tvm/contrib/msc/core/utils/arguments.py | 261 +
python/tvm/contrib/msc/core/utils/dataset.py | 563 ++
python/tvm/contrib/msc/core/utils/expr.py | 222 +
python/tvm/contrib/msc/core/utils/file.py | 536 ++
python/tvm/contrib/msc/core/utils/info.py | 433 +
python/tvm/contrib/msc/core/utils/log.py | 186 +
python/tvm/contrib/msc/core/utils/message.py | 172 +
python/tvm/contrib/msc/core/utils/namespace.py | 80 +
python/tvm/contrib/msc/core/utils/register.py | 402 +
python/tvm/contrib/msc/framework/__init__.py | 17 +
.../contrib/msc/framework/tensorflow/__init__.py | 24 +
.../contrib/msc/framework/tensorflow/_ffi_api.py | 21 +
.../msc/framework/tensorflow/codegen/__init__.py | 19 +
.../msc/framework/tensorflow/codegen/codegen.py | 72 +
.../msc/framework/tensorflow/frontend/__init__.py | 19 +
.../msc/framework/tensorflow/frontend/translate.py | 82 +
.../msc/framework/tensorflow/runtime/__init__.py | 19 +
.../msc/framework/tensorflow/runtime/runner.py | 305 +
.../msc/framework/tensorflow/tools/__init__.py | 22 +
.../framework/tensorflow/tools/distill/__init__.py | 19 +
.../tensorflow/tools/distill/distiller.py | 56 +
.../framework/tensorflow/tools/prune/__init__.py | 19 +
.../msc/framework/tensorflow/tools/prune/pruner.py | 56 +
.../tensorflow/tools/quantize/__init__.py | 19 +
.../tensorflow/tools/quantize/quantizer.py | 56 +
.../framework/tensorflow/tools/track/__init__.py | 19 +
.../framework/tensorflow/tools/track/tracker.py | 56 +
.../tvm/contrib/msc/framework/tensorrt/__init__.py | 17 +
.../tvm/contrib/msc/framework/tensorrt/_ffi_api.py | 21 +
.../msc/framework/tensorrt/codegen/__init__.py | 19 +
.../msc/framework/tensorrt/codegen/codegen.py | 210 +
.../msc/framework/tensorrt/codegen/sources.py | 488 ++
.../msc/framework/tensorrt/codegen/utils.py | 98 +
.../msc/framework/tensorrt/frontend/__init__.py | 19 +
.../msc/framework/tensorrt/frontend/translate.py | 86 +
.../msc/framework/tensorrt/runtime/__init__.py | 19 +
.../msc/framework/tensorrt/runtime/runner.py | 186 +
.../msc/framework/tensorrt/tools/__init__.py | 22 +
.../framework/tensorrt/tools/distill/__init__.py | 19 +
.../framework/tensorrt/tools/distill/distiller.py | 56 +
.../msc/framework/tensorrt/tools/prune/__init__.py | 19 +
.../msc/framework/tensorrt/tools/prune/pruner.py | 56 +
.../framework/tensorrt/tools/quantize/__init__.py | 20 +
.../framework/tensorrt/tools/quantize/method.py | 147 +
.../framework/tensorrt/tools/quantize/quantizer.py | 363 +
.../msc/framework/tensorrt/tools/track/__init__.py | 19 +
.../msc/framework/tensorrt/tools/track/tracker.py | 160 +
.../msc/framework/tensorrt/transform/__init__.py | 20 +
.../msc/framework/tensorrt/transform/pattern.py | 465 ++
.../msc/framework/tensorrt/transform/transform.py | 42 +
python/tvm/contrib/msc/framework/torch/__init__.py | 17 +
python/tvm/contrib/msc/framework/torch/_ffi_api.py | 21 +
.../msc/framework/torch/codegen/__init__.py | 19 +
.../contrib/msc/framework/torch/codegen/codegen.py | 80 +
.../msc/framework/torch/frontend/__init__.py | 19 +
.../msc/framework/torch/frontend/translate.py | 129 +
.../msc/framework/torch/runtime/__init__.py | 20 +
.../tvm/contrib/msc/framework/torch/runtime/jit.py | 213 +
.../contrib/msc/framework/torch/runtime/runner.py | 346 +
.../contrib/msc/framework/torch/tools/__init__.py | 22 +
.../msc/framework/torch/tools/distill/__init__.py | 20 +
.../msc/framework/torch/tools/distill/distiller.py | 145 +
.../msc/framework/torch/tools/distill/method.py | 114 +
.../msc/framework/torch/tools/prune/__init__.py | 19 +
.../msc/framework/torch/tools/prune/pruner.py | 56 +
.../msc/framework/torch/tools/quantize/__init__.py | 20 +
.../msc/framework/torch/tools/quantize/method.py | 267 +
.../framework/torch/tools/quantize/quantizer.py | 56 +
.../msc/framework/torch/tools/track/__init__.py | 19 +
.../msc/framework/torch/tools/track/tracker.py | 56 +
python/tvm/contrib/msc/framework/tvm/__init__.py | 17 +
python/tvm/contrib/msc/framework/tvm/_ffi_api.py | 21 +
.../contrib/msc/framework/tvm/codegen/__init__.py | 19 +
.../contrib/msc/framework/tvm/codegen/codegen.py | 58 +
.../contrib/msc/framework/tvm/runtime/__init__.py | 19 +
.../contrib/msc/framework/tvm/runtime/runner.py | 329 +
.../contrib/msc/framework/tvm/tools/__init__.py | 22 +
.../msc/framework/tvm/tools/distill/__init__.py | 19 +
.../msc/framework/tvm/tools/distill/distiller.py | 56 +
.../msc/framework/tvm/tools/prune/__init__.py | 19 +
.../msc/framework/tvm/tools/prune/pruner.py | 56 +
.../msc/framework/tvm/tools/quantize/__init__.py | 20 +
.../msc/framework/tvm/tools/quantize/method.py | 203 +
.../msc/framework/tvm/tools/quantize/quantizer.py | 168 +
.../msc/framework/tvm/tools/track/__init__.py | 19 +
.../msc/framework/tvm/tools/track/tracker.py | 159 +
python/tvm/contrib/msc/pipeline/__init__.py | 20 +
python/tvm/contrib/msc/pipeline/config.py | 172 +
python/tvm/contrib/msc/pipeline/dynamic.py | 492 ++
python/tvm/contrib/msc/pipeline/manager.py | 287 +
python/tvm/contrib/msc/pipeline/pipeline.py | 845 ++
python/tvm/contrib/msc/pipeline/utils.py | 220 +
python/tvm/contrib/msc/pipeline/worker.py | 786 ++
python/tvm/contrib/msc/pipeline/wrapper.py | 281 +
python/tvm/contrib/msc/plugin/__init__.py | 19 +
python/tvm/contrib/msc/plugin/_ffi_api.py | 21 +
python/tvm/contrib/msc/plugin/build.py | 286 +
python/tvm/contrib/msc/plugin/codegen/__init__.py | 19 +
python/tvm/contrib/msc/plugin/codegen/codegen.py | 319 +
python/tvm/contrib/msc/plugin/codegen/sources.py | 1157 +++
python/tvm/contrib/msc/plugin/op/__init__.py | 17 +
python/tvm/contrib/msc/plugin/op/_ffi_api.py | 21 +
python/tvm/contrib/msc/plugin/register.py | 85 +
python/tvm/contrib/msc/plugin/utils.py | 109 +
python/tvm/contrib/ndk.py | 43 +-
python/tvm/contrib/nvcc.py | 13 +-
python/tvm/contrib/relay_viz/interface.py | 23 +-
python/tvm/contrib/rocm.py | 1 +
python/tvm/contrib/thrust.py | 2 -
python/tvm/contrib/tvmjs.py | 355 +
python/tvm/dlight/__init__.py | 27 +
python/tvm/dlight/base/__init__.py | 29 +
python/tvm/dlight/base/analysis.py | 265 +
python/tvm/dlight/base/common_schedules.py | 98 +
python/tvm/dlight/base/schedule_rule.py | 120 +
python/tvm/dlight/base/transform.py | 93 +
python/tvm/dlight/benchmark/__init__.py | 24 +
python/tvm/dlight/benchmark/bench.py | 313 +
python/tvm/dlight/benchmark/extract.py | 358 +
python/tvm/dlight/benchmark/utils.py | 172 +
python/tvm/dlight/gpu/__init__.py | 28 +
python/tvm/dlight/gpu/base.py | 40 +
python/tvm/dlight/gpu/fallback.py | 90 +
python/tvm/dlight/gpu/gemv.py | 802 ++
python/tvm/dlight/gpu/general_reduction.py | 126 +
python/tvm/dlight/gpu/low_batch_gemv.py | 623 ++
python/tvm/dlight/gpu/matmul.py | 932 +++
python/tvm/dlight/gpu/reduction.py | 298 +
python/tvm/dlight/gpu/rmsnorm.py | 142 +
python/tvm/dlight/gpu/transpose.py | 133 +
python/tvm/dlight/gpu/utils.py | 93 +
python/tvm/driver/build_module.py | 30 +-
python/tvm/driver/tvmc/compiler.py | 31 +-
python/tvm/driver/tvmc/composite_target.py | 5 +
python/tvm/error.py | 1 +
python/tvm/exec/disco_worker.py | 124 +
python/tvm/exec/gpu_memory_bandwidth.py | 98 +-
python/tvm/exec/microtvm_debug_shell.py | 8 +-
python/tvm/exec/rpc_proxy.py | 38 +-
python/tvm/ir/__init__.py | 1 +
python/tvm/ir/attrs.py | 6 +-
python/tvm/ir/base.py | 7 +-
python/tvm/ir/expr.py | 62 +-
python/tvm/ir/function.py | 27 +-
python/tvm/ir/global_info.py | 54 +
python/tvm/ir/instrument.py | 36 +
python/tvm/ir/json_compact.py | 72 +-
python/tvm/ir/module.py | 75 +-
python/tvm/ir/supply.py | 7 +-
python/tvm/ir/transform.py | 138 +-
python/tvm/meta_schedule/__init__.py | 1 +
python/tvm/meta_schedule/builder/local_builder.py | 12 +
python/tvm/meta_schedule/logging.py | 3 +-
python/tvm/meta_schedule/relax_integration.py | 420 +
python/tvm/meta_schedule/relay_integration.py | 4 +
python/tvm/meta_schedule/runner/local_runner.py | 12 +
python/tvm/meta_schedule/testing/tune_utils.py | 5 -
python/tvm/meta_schedule/tir_integration.py | 87 +-
python/tvm/meta_schedule/tune_context.py | 3 +-
python/tvm/meta_schedule/utils.py | 50 +-
python/tvm/micro/testing/aot_test_utils.py | 10 +
python/tvm/relax/__init__.py | 118 +
python/tvm/relax/_ffi_api.py | 20 +
python/tvm/relax/analysis/__init__.py | 47 +
python/tvm/relax/analysis/_ffi_api.py | 19 +
python/tvm/relax/analysis/analysis.py | 582 ++
python/tvm/relax/analysis/estimate_memory_usage.py | 182 +
python/tvm/relax/backend/__init__.py | 21 +
python/tvm/relax/backend/_ffi_api.py | 21 +
python/tvm/relax/backend/contrib/__init__.py | 17 +
python/tvm/relax/backend/contrib/coreml.py | 490 ++
python/tvm/relax/backend/contrib/cublas.py | 243 +
python/tvm/relax/backend/contrib/cudnn.py | 108 +
python/tvm/relax/backend/contrib/cutlass.py | 612 ++
python/tvm/relax/backend/dispatch_sort_scan.py | 268 +
python/tvm/relax/backend/pattern_registry.py | 119 +
python/tvm/relax/backend/patterns.py | 617 ++
python/tvm/relax/backend/utils.py | 43 +
python/tvm/relax/backend_tir/__init__.py | 21 +
python/tvm/relax/backend_tir/contrib/__init__.py | 20 +
python/tvm/relax/backend_tir/contrib/cutlass.py | 720 ++
python/tvm/relax/backend_tir/cumsum.py | 193 +
python/tvm/relax/backend_tir/pattern.py | 576 ++
python/tvm/relax/binding_rewrite.py | 155 +
python/tvm/relax/block_builder.py | 799 ++
python/tvm/relax/distributed/__init__.py | 23 +
python/tvm/relax/distributed/_ffi_api.py | 20 +
python/tvm/relax/distributed/global_info.py | 70 +
python/tvm/relax/distributed/struct_info.py | 141 +
python/tvm/relax/distributed/transform/__init__.py | 24 +
python/tvm/relax/distributed/transform/_ffi_api.py | 19 +
.../tvm/relax/distributed/transform/transform.py | 67 +
python/tvm/relax/dpl/__init__.py | 22 +
python/tvm/relax/dpl/_ffi.py | 20 +
python/tvm/relax/dpl/context.py | 80 +
python/tvm/relax/dpl/pattern.py | 1187 +++
python/tvm/relax/dpl/rewrite.py | 118 +
python/tvm/relax/exec_builder.py | 147 +
python/tvm/relax/expr.py | 1224 +++
python/tvm/relax/expr_functor.py | 1538 ++++
python/tvm/relax/frontend/__init__.py | 19 +
python/tvm/relax/frontend/common.py | 55 +
python/tvm/relax/frontend/nn/__init__.py | 39 +
python/tvm/relax/frontend/nn/_tensor_op.py | 104 +
python/tvm/relax/frontend/nn/core.py | 658 ++
python/tvm/relax/frontend/nn/exporter.py | 333 +
python/tvm/relax/frontend/nn/extern.py | 399 +
python/tvm/relax/frontend/nn/modules.py | 949 +++
python/tvm/relax/frontend/nn/op.py | 2694 ++++++
python/tvm/relax/frontend/nn/spec.py | 256 +
python/tvm/relax/frontend/nn/subroutine.py | 179 +
python/tvm/relax/frontend/nn/torch.py | 135 +
python/tvm/relax/frontend/nn/visitor.py | 145 +
python/tvm/relax/frontend/onnx/__init__.py | 20 +
python/tvm/relax/frontend/onnx/onnx_frontend.py | 2414 ++++++
python/tvm/relax/frontend/stablehlo/__init__.py | 20 +
.../frontend/stablehlo/stablehlo_translator.py | 445 +
python/tvm/relax/frontend/torch/__init__.py | 21 +
python/tvm/relax/frontend/torch/dynamo.py | 207 +
python/tvm/relax/frontend/torch/fx_translator.py | 1695 ++++
python/tvm/relax/ir/instrument.py | 58 +
python/tvm/relax/op/__init__.py | 148 +
python/tvm/relax/op/_ffi_api.py | 19 +
python/tvm/relax/op/_op_gradient.py | 1318 +++
python/tvm/relax/op/base.py | 824 ++
python/tvm/relax/op/binary.py | 388 +
python/tvm/relax/op/builtin/__init__.py | 19 +
python/tvm/relax/op/builtin/_ffi_api.py | 19 +
python/tvm/relax/op/builtin/builtin.py | 84 +
python/tvm/relax/op/ccl/__init__.py | 18 +
python/tvm/relax/op/ccl/_ffi_api.py | 20 +
python/tvm/relax/op/ccl/ccl.py | 106 +
python/tvm/relax/op/create.py | 267 +
python/tvm/relax/op/datatype.py | 59 +
python/tvm/relax/op/distributed/__init__.py | 24 +
python/tvm/relax/op/distributed/_ffi_api.py | 20 +
python/tvm/relax/op/distributed/distributed.py | 138 +
python/tvm/relax/op/grad/__init__.py | 19 +
python/tvm/relax/op/grad/_ffi_api.py | 20 +
python/tvm/relax/op/grad/grad.py | 217 +
python/tvm/relax/op/image/__init__.py | 18 +
python/tvm/relax/op/image/_ffi_api.py | 20 +
python/tvm/relax/op/image/image.py | 128 +
python/tvm/relax/op/index.py | 136 +
python/tvm/relax/op/linear_algebra.py | 112 +
python/tvm/relax/op/manipulate.py | 507 ++
python/tvm/relax/op/mask.py | 38 +
python/tvm/relax/op/memory/__init__.py | 20 +
python/tvm/relax/op/memory/_ffi_api.py | 19 +
python/tvm/relax/op/memory/memory.py | 126 +
python/tvm/relax/op/memory/view.py | 94 +
python/tvm/relax/op/nn/__init__.py | 50 +
python/tvm/relax/op/nn/_ffi_api.py | 20 +
python/tvm/relax/op/nn/nn.py | 1952 +++++
python/tvm/relax/op/op_attrs.py | 184 +
python/tvm/relax/op/qdq.py | 88 +
python/tvm/relax/op/search.py | 104 +
python/tvm/relax/op/set.py | 110 +
python/tvm/relax/op/sorting.py | 116 +
python/tvm/relax/op/statistical.py | 337 +
python/tvm/relax/op/ternary.py | 43 +
python/tvm/relax/op/unary.py | 601 ++
python/tvm/relax/op/vm/__init__.py | 19 +
python/tvm/relax/op/vm/_ffi_api.py | 19 +
python/tvm/relax/op/vm/vm.py | 133 +
python/tvm/relax/pipeline.py | 151 +
python/tvm/relax/struct_info.py | 272 +
python/tvm/relax/testing/__init__.py | 23 +
python/tvm/relax/testing/ast_printer.py | 374 +
python/tvm/relax/testing/lib_comparator.py | 132 +
python/tvm/relax/testing/matmul.py | 62 +
python/tvm/relax/testing/nn.py | 358 +
python/tvm/relax/testing/relay_translator.py | 273 +
python/tvm/relax/testing/runtime_builtin.py | 36 +
python/tvm/relax/testing/transform.py | 226 +
python/tvm/relax/testing/vm.py | 90 +
python/tvm/relax/training/__init__.py | 26 +
python/tvm/relax/training/_ffi_api.py | 20 +
python/tvm/relax/training/loss.py | 384 +
python/tvm/relax/training/optimizer.py | 713 ++
python/tvm/relax/training/setup_trainer.py | 213 +
python/tvm/relax/training/trainer.py | 393 +
python/tvm/relax/training/utils.py | 205 +
python/tvm/relax/transform/__init__.py | 96 +
python/tvm/relax/transform/_ffi_api.py | 19 +
.../tvm/relax/transform/attach_external_modules.py | 52 +
python/tvm/relax/transform/fast_math.py | 64 +
.../tvm/relax/transform/ipc_allreduce_rewrite.py | 146 +
.../tvm/relax/transform/lazy_transform_params.py | 399 +
.../tvm/relax/transform/legalize_ops/__init__.py | 33 +
python/tvm/relax/transform/legalize_ops/binary.py | 69 +
python/tvm/relax/transform/legalize_ops/ccl.py | 125 +
python/tvm/relax/transform/legalize_ops/common.py | 120 +
python/tvm/relax/transform/legalize_ops/create.py | 86 +
.../tvm/relax/transform/legalize_ops/datatype.py | 31 +
.../relax/transform/legalize_ops/distributed.py | 43 +
python/tvm/relax/transform/legalize_ops/grad.py | 230 +
python/tvm/relax/transform/legalize_ops/image.py | 39 +
python/tvm/relax/transform/legalize_ops/index.py | 135 +
.../tvm/relax/transform/legalize_ops/inspect_op.py | 128 +
.../relax/transform/legalize_ops/linear_algebra.py | 117 +
.../tvm/relax/transform/legalize_ops/manipulate.py | 215 +
python/tvm/relax/transform/legalize_ops/nn.py | 719 ++
python/tvm/relax/transform/legalize_ops/qdq.py | 119 +
python/tvm/relax/transform/legalize_ops/search.py | 41 +
.../relax/transform/legalize_ops/statistical.py | 87 +
python/tvm/relax/transform/legalize_ops/unary.py | 66 +
.../relax/transform/lower_gpu_ipc_alloc_storage.py | 83 +
.../relax/transform/optimize_layout_transform.py | 86 +
.../relax/transform/remove_redundant_reshape.py | 81 +
python/tvm/relax/transform/transform.py | 1856 +++++
python/tvm/relax/transform/tuning_api/__init__.py | 22 +
python/tvm/relax/transform/tuning_api/_ffi_api.py | 19 +
python/tvm/relax/transform/tuning_api/database.py | 273 +
.../transform/tuning_api/default_functions.py | 306 +
.../tvm/relax/transform/tuning_api/primitives.py | 419 +
python/tvm/relax/ty.py | 75 +
python/tvm/relax/type_converter.py | 179 +
python/tvm/relax/utils.py | 384 +
python/tvm/relax/vm_build.py | 357 +
python/tvm/relay/analysis/sparse_conv2d.py | 11 +-
python/tvm/relay/analysis/sparse_dense.py | 11 +-
.../tvm/relay/backend/contrib/ethosu/legalize.py | 181 +-
.../contrib/ethosu/te/binary_elementwise.py | 4 +-
.../tvm/relay/backend/contrib/ethosu/te/common.py | 26 +
.../relay/backend/contrib/ethosu/te/convolution.py | 4 +-
.../relay/backend/contrib/ethosu/te/depthwise.py | 4 +-
.../relay/backend/contrib/ethosu/te/identity.py | 3 +-
.../tvm/relay/backend/contrib/ethosu/te/pooling.py | 4 +-
python/tvm/relay/backend/contrib/ethosu/util.py | 2 +-
python/tvm/relay/expr_functor.py | 2 +-
python/tvm/relay/frontend/keras.py | 14 +-
python/tvm/relay/frontend/onnx.py | 47 +-
python/tvm/relay/frontend/paddlepaddle.py | 134 +-
python/tvm/relay/frontend/pytorch.py | 15 +-
python/tvm/relay/frontend/tensorflow_ops.py | 6 +-
python/tvm/relay/frontend/tflite.py | 21 +
python/tvm/relay/function.py | 3 +
python/tvm/relay/op/contrib/__init__.py | 1 +
python/tvm/relay/op/contrib/clml.py | 119 +-
python/tvm/relay/op/contrib/ethosu.py | 137 +-
python/tvm/relay/op/contrib/mrvl.py | 908 +++
python/tvm/relay/op/op_attrs.py | 10 -
python/tvm/relay/op/strategy/arm_cpu.py | 99 +-
python/tvm/relay/qnn/op/legalizations.py | 19 +-
.../tvm/relay/quantize/_partition_conversions.py | 4 +-
python/tvm/relay/testing/py_converter.py | 6 +-
python/tvm/rpc/proxy.py | 22 +-
python/tvm/rpc/server.py | 19 +-
python/tvm/rpc/tracker.py | 8 +-
python/tvm/runtime/__init__.py | 3 +
python/tvm/runtime/disco/__init__.py | 25 +
python/tvm/runtime/disco/_ffi_api.py | 20 +
python/tvm/runtime/disco/process_pool.py | 187 +
python/tvm/runtime/disco/session.py | 486 ++
python/tvm/runtime/module.py | 10 +-
python/tvm/runtime/ndarray.py | 58 +-
python/tvm/runtime/relax_vm.py | 513 ++
python/tvm/runtime/script_printer.py | 98 +-
python/tvm/runtime/support.py | 69 +
python/tvm/script/__init__.py | 1 -
python/tvm/script/highlight.py | 95 +-
python/tvm/script/ir_builder/ir/__init__.py | 4 +
python/tvm/script/ir_builder/ir/ir.py | 71 +-
python/tvm/script/ir_builder/relax/__init__.py | 19 +
python/tvm/script/ir_builder/relax/_ffi_api.py | 20 +
.../ir_builder/relax/distributed/__init__.py | 19 +
.../ir_builder/relax/distributed/_ffi_api.py | 22 +
.../tvm/script/ir_builder/relax/distributed/ir.py | 162 +
python/tvm/script/ir_builder/relax/frame.py | 55 +
python/tvm/script/ir_builder/relax/ir.py | 808 ++
python/tvm/script/ir_builder/tir/ir.py | 60 +-
python/tvm/script/parser/__init__.py | 5 +-
python/tvm/script/parser/core/entry.py | 53 +-
python/tvm/script/parser/core/parser.py | 103 +-
python/tvm/script/parser/ir/__init__.py | 12 +-
python/tvm/script/parser/ir/entry.py | 30 +-
python/tvm/script/parser/ir/parser.py | 20 +-
python/tvm/script/parser/relax/__init__.py | 48 +
python/tvm/script/parser/relax/dist.py | 106 +
python/tvm/script/parser/relax/entry.py | 533 ++
python/tvm/script/parser/relax/parser.py | 401 +
python/tvm/script/parser/tir/__init__.py | 4 +-
python/tvm/script/parser/tir/entry.py | 6 +-
python/tvm/script/parser/tir/parser.py | 38 +-
python/tvm/script/relax.py | 18 +
python/tvm/script/tir.py | 18 +
python/tvm/support.py | 5 +-
python/tvm/target/codegen.py | 3 +-
python/tvm/target/detect_target.py | 42 +-
python/tvm/target/tag.py | 6 +-
python/tvm/te/operation.py | 22 +-
python/tvm/te/schedule.py | 14 +-
python/tvm/testing/__init__.py | 1 -
python/tvm/testing/aot.py | 173 +-
python/tvm/testing/utils.py | 73 +-
python/tvm/tir/__init__.py | 1 +
python/tvm/tir/analysis/analysis.py | 10 +
python/tvm/tir/expr.py | 14 +-
python/tvm/tir/function.py | 3 +
python/tvm/tir/op.py | 104 +-
python/tvm/tir/schedule/schedule.py | 123 +-
python/tvm/tir/tensor_intrin/__init__.py | 1 -
python/tvm/tir/tensor_intrin/arm_cpu.py | 371 +-
python/tvm/tir/tensor_intrin/dot_product_common.py | 82 +-
python/tvm/tir/tensor_intrin/rocm.py | 3 +-
python/tvm/tir/transform/function_pass.py | 3 +-
python/tvm/tir/transform/transform.py | 39 +-
python/tvm/topi/arm_cpu/__init__.py | 5 +-
python/tvm/topi/arm_cpu/arm_utils.py | 132 +-
python/tvm/topi/arm_cpu/conv2d.py | 43 +-
python/tvm/topi/arm_cpu/conv2d_gemm.py | 155 +-
python/tvm/topi/arm_cpu/dense.py | 10 +-
python/tvm/topi/arm_cpu/dense_alter_op.py | 75 +
python/tvm/topi/arm_cpu/injective.py | 4 +-
python/tvm/topi/arm_cpu/matmul.py | 124 +
python/tvm/topi/arm_cpu/pstate_attributes.py | 84 +
python/tvm/topi/arm_cpu/qnn_legalize.py | 8 +-
python/tvm/topi/cuda/scan.py | 159 +-
python/tvm/topi/cuda/scatter_elements.py | 2 +-
python/tvm/topi/cuda/sort.py | 127 +-
python/tvm/topi/cuda/sparse.py | 7 +-
python/tvm/topi/hexagon/qnn/nn.py | 31 +-
python/tvm/topi/math.py | 11 +-
python/tvm/topi/nn/conv1d.py | 5 +-
python/tvm/topi/nn/conv2d.py | 9 +-
python/tvm/topi/nn/conv3d_transpose.py | 6 +-
python/tvm/topi/nn/group_norm.py | 2 +
python/tvm/topi/nn/layer_norm.py | 2 +
python/tvm/topi/nn/pooling.py | 14 +-
python/tvm/topi/scan.py | 2 +-
python/tvm/topi/scatter_elements.py | 2 +-
python/tvm/topi/testing/group_norm_python.py | 5 +-
python/tvm/topi/testing/layer_norm_python.py | 3 +
python/tvm/topi/transform.py | 35 +
python/tvm/topi/x86/dense_alter_op.py | 2 +-
rust/tvm/examples/resnet/build.rs | 6 +
rust/tvm/examples/resnet/src/main.rs | 5 +
rust/tvm/src/ir/module.rs | 14 +-
rust/tvm/src/ir/relay/mod.rs | 2 +
src/arith/analyzer.cc | 26 +
src/arith/canonical_simplify.cc | 1 +
src/arith/const_fold.h | 3 +-
src/arith/const_int_bound.cc | 39 +-
src/arith/int_set.cc | 33 +-
src/arith/ir_visitor_with_analyzer.cc | 2 +-
src/arith/iter_affine_map.cc | 19 +-
src/arith/modular_set.cc | 2 +-
src/arith/pattern_match.h | 15 +-
src/arith/rewrite_simplify.cc | 327 +-
src/arith/rewrite_simplify.h | 3 +-
src/arith/scalable_expression.cc | 106 +
src/arith/scalable_expression.h | 89 +
src/auto_scheduler/search_policy/utils.h | 2 +-
src/contrib/msc/core/codegen/base_codegen.h | 231 +
src/contrib/msc/core/codegen/code_stack.cc | 471 ++
src/contrib/msc/core/codegen/code_stack.h | 642 ++
src/contrib/msc/core/codegen/codegen_json.cc | 66 +
src/contrib/msc/core/codegen/codegen_json.h | 108 +
src/contrib/msc/core/codegen/codegen_utils.cc | 70 +
src/contrib/msc/core/codegen/codegen_utils.h | 191 +
src/contrib/msc/core/codegen/cpp_codegen.h | 197 +
src/contrib/msc/core/codegen/py_codegen.h | 214 +
src/contrib/msc/core/ir/graph.cc | 1520 ++++
src/contrib/msc/core/ir/graph.h | 892 ++
src/contrib/msc/core/ir/graph_builder.cc | 1035 +++
src/contrib/msc/core/ir/graph_builder.h | 434 +
src/contrib/msc/core/ir/plugin.cc | 327 +
src/contrib/msc/core/ir/plugin.h | 686 ++
src/contrib/msc/core/printer/cpp_printer.cc | 360 +
src/contrib/msc/core/printer/cpp_printer.h | 157 +
src/contrib/msc/core/printer/msc_base_printer.cc | 170 +
src/contrib/msc/core/printer/msc_base_printer.h | 259 +
src/contrib/msc/core/printer/msc_doc.cc | 91 +
src/contrib/msc/core/printer/msc_doc.h | 344 +
src/contrib/msc/core/printer/print_utils.cc | 110 +
src/contrib/msc/core/printer/print_utils.h | 233 +
src/contrib/msc/core/printer/prototxt_printer.cc | 112 +
src/contrib/msc/core/printer/prototxt_printer.h | 79 +
src/contrib/msc/core/printer/python_printer.cc | 272 +
src/contrib/msc/core/printer/python_printer.h | 105 +
.../msc/core/transform/bind_named_params.cc | 164 +
src/contrib/msc/core/transform/bind_shape.cc | 140 +
src/contrib/msc/core/transform/fuse_tuple.cc | 218 +
src/contrib/msc/core/transform/inline_params.cc | 192 +
src/contrib/msc/core/transform/layout_utils.cc | 225 +
src/contrib/msc/core/transform/layout_utils.h | 131 +
src/contrib/msc/core/transform/set_byoc_attrs.cc | 107 +
src/contrib/msc/core/transform/set_expr_layout.cc | 1355 +++
src/contrib/msc/core/transform/set_expr_name.cc | 556 ++
src/contrib/msc/core/utils.cc | 479 ++
src/contrib/msc/core/utils.h | 373 +
src/contrib/msc/framework/tensorflow/codegen.cc | 162 +
src/contrib/msc/framework/tensorflow/codegen.h | 72 +
.../msc/framework/tensorflow/codegen_utils.h | 58 +
src/contrib/msc/framework/tensorflow/config.h | 57 +
.../msc/framework/tensorflow/tf_v1_opcode.cc | 620 ++
.../msc/framework/tensorflow/tf_v1_opcode.h | 85 +
src/contrib/msc/framework/tensorrt/codegen.cc | 624 ++
src/contrib/msc/framework/tensorrt/codegen.h | 97 +
src/contrib/msc/framework/tensorrt/codegen_utils.h | 131 +
.../msc/framework/tensorrt/tensorrt_opcode.cc | 836 ++
.../msc/framework/tensorrt/tensorrt_opcode.h | 125 +
.../msc/framework/tensorrt/transform_tensorrt.cc | 746 ++
src/contrib/msc/framework/torch/codegen.cc | 163 +
src/contrib/msc/framework/torch/codegen.h | 72 +
src/contrib/msc/framework/torch/codegen_utils.h | 69 +
src/contrib/msc/framework/torch/config.h | 57 +
src/contrib/msc/framework/torch/torch_opcode.cc | 777 ++
src/contrib/msc/framework/torch/torch_opcode.h | 125 +
src/contrib/msc/framework/tvm/codegen.cc | 212 +
src/contrib/msc/framework/tvm/codegen.h | 69 +
src/contrib/msc/framework/tvm/codegen_utils.h | 66 +
src/contrib/msc/framework/tvm/config.h | 60 +
src/contrib/msc/framework/tvm/relax_opcode.cc | 834 ++
src/contrib/msc/framework/tvm/relax_opcode.h | 83 +
src/contrib/msc/plugin/base_codegen.h | 674 ++
src/contrib/msc/plugin/codegen_utils.h | 75 +
src/contrib/msc/plugin/tensorrt_codegen.cc | 901 ++
src/contrib/msc/plugin/tensorrt_codegen.h | 134 +
src/contrib/msc/plugin/torch_codegen.cc | 510 ++
src/contrib/msc/plugin/torch_codegen.h | 137 +
src/contrib/msc/plugin/tvm_codegen.cc | 411 +
src/contrib/msc/plugin/tvm_codegen.h | 124 +
src/driver/driver_api.cc | 18 +-
src/ir/apply_pass_to_function.cc | 136 +
src/ir/expr.cc | 4 +-
src/ir/function.cc | 25 +-
src/ir/global_info.cc | 45 +
src/ir/module.cc | 77 +-
src/ir/name_supply.cc | 8 +-
src/ir/transform.cc | 86 +-
src/ir/type.cc | 3 +-
.../feature_extractor/per_store_feature.cc | 2 +-
src/meta_schedule/module_equality.cc | 13 -
src/meta_schedule/postproc/verify_gpu_code.cc | 2 +-
.../multi_level_tiling_tensor_core.cc | 6 +-
.../space_generator/post_order_apply.cc | 62 -
src/meta_schedule/utils.h | 75 +-
src/node/container_printing.cc | 10 +-
src/node/ndarray_hash_equal.h | 7 +
src/node/script_printer.cc | 41 +-
src/node/structural_equal.cc | 50 +-
src/node/structural_hash.cc | 10 +
src/relax/analysis/analysis.cc | 211 +
src/relax/analysis/computable_at_compile_time.cc | 99 +
src/relax/analysis/detect_recursion.cc | 398 +
src/relax/analysis/layout_transformation.cc | 622 ++
src/relax/analysis/shape_analysis.cc | 55 +
src/relax/analysis/struct_info_analysis.cc | 1429 ++++
src/relax/analysis/tir_op_pattern_kind.cc | 538 ++
src/relax/analysis/udchain.cc | 110 +
src/relax/analysis/var2value.cc | 91 +
src/relax/analysis/well_formed.cc | 586 ++
.../backend/contrib/codegen_json/codegen_json.h | 419 +
src/relax/backend/contrib/cublas/codegen.cc | 133 +
src/relax/backend/contrib/cudnn/codegen.cc | 110 +
src/relax/backend/contrib/cutlass/codegen.cc | 299 +
src/relax/backend/contrib/dnnl/codegen.cc | 105 +
src/relax/backend/contrib/tensorrt/codegen.cc | 267 +
src/relax/backend/contrib/utils.cc | 71 +
src/relax/backend/contrib/utils.h | 156 +
src/relax/backend/pattern_registry.cc | 77 +
src/relax/backend/pattern_registry.h | 73 +
src/relax/backend/task_extraction.cc | 149 +
src/relax/backend/vm/codegen_vm.cc | 482 ++
src/relax/backend/vm/codegen_vm_tir.cc | 543 ++
src/relax/backend/vm/exec_builder.cc | 399 +
src/relax/backend/vm/vm_builtin_lower.cc | 227 +
src/relax/backend/vm/vm_shape_lower.cc | 821 ++
src/relax/distributed/axis_group_graph.cc | 395 +
src/relax/distributed/global_info.cc | 70 +
src/relax/distributed/struct_info.cc | 141 +
.../distributed/transform/legalize_redistribute.cc | 123 +
src/relax/distributed/transform/lower_distir.cc | 271 +
.../transform/lower_global_view_to_local_view.cc | 442 +
.../distributed/transform/propagate_sharding.cc | 623 ++
src/relax/distributed/transform/utils.cc | 81 +
src/relax/distributed/transform/utils.h | 67 +
src/relax/ir/binding_rewrite.cc | 357 +
src/relax/ir/block_builder.cc | 1090 +++
src/relax/ir/dataflow_matcher.cc | 1272 +++
src/relax/ir/dataflow_matcher_impl.h | 102 +
src/relax/ir/dataflow_pattern.cc | 670 ++
src/relax/ir/dataflow_pattern_functor.cc | 115 +
src/relax/ir/emit_te.cc | 78 +
src/relax/ir/emit_te.h | 68 +
src/relax/ir/expr.cc | 706 ++
src/relax/ir/expr_functor.cc | 822 ++
src/relax/ir/py_expr_functor.cc | 685 ++
src/relax/ir/struct_info.cc | 231 +
src/relax/ir/struct_info_functor.cc | 155 +
src/relax/ir/tir_pattern.cc | 37 +
src/relax/ir/transform.cc | 413 +
src/relax/ir/type.cc | 88 +
src/relax/op/ccl/ccl.cc | 152 +
src/relax/op/ccl/ccl.h | 50 +
src/relax/op/distributed/binary.cc | 74 +
src/relax/op/distributed/binary.h | 89 +
src/relax/op/distributed/ccl.cc | 43 +
src/relax/op/distributed/distributed.cc | 223 +
src/relax/op/distributed/distributed.h | 69 +
src/relax/op/distributed/linear_algebra.cc | 103 +
src/relax/op/distributed/linear_algebra.h | 39 +
src/relax/op/distributed/manipulate.cc | 139 +
src/relax/op/distributed/manipulate.h | 41 +
src/relax/op/distributed/nn.cc | 63 +
src/relax/op/distributed/nn.h | 40 +
src/relax/op/distributed/op.cc | 52 +
src/relax/op/distributed/statistical.cc | 87 +
src/relax/op/distributed/statistical.h | 44 +
src/relax/op/distributed/unary.cc | 65 +
src/relax/op/distributed/unary.h | 78 +
src/relax/op/distributed/utils.cc | 104 +
src/relax/op/distributed/utils.h | 68 +
src/relax/op/image/resize.cc | 141 +
src/relax/op/image/resize.h | 43 +
src/relax/op/memory/view.cc | 359 +
src/relax/op/memory/view.h | 38 +
src/relax/op/nn/attention.cc | 183 +
src/relax/op/nn/attention.h | 42 +
src/relax/op/nn/convolution.cc | 807 ++
src/relax/op/nn/convolution.h | 95 +
src/relax/op/nn/nn.cc | 868 ++
src/relax/op/nn/nn.h | 104 +
src/relax/op/nn/pooling.cc | 686 ++
src/relax/op/nn/pooling.h | 52 +
src/relax/op/op.cc | 1126 +++
src/relax/op/op_common.cc | 189 +
src/relax/op/op_common.h | 573 ++
src/relax/op/tensor/binary.cc | 212 +
src/relax/op/tensor/binary.h | 135 +
src/relax/op/tensor/create.cc | 330 +
src/relax/op/tensor/create.h | 105 +
src/relax/op/tensor/datatype.cc | 91 +
src/relax/op/tensor/datatype.h | 53 +
src/relax/op/tensor/grad.cc | 221 +
src/relax/op/tensor/grad.h | 68 +
src/relax/op/tensor/index.cc | 566 ++
src/relax/op/tensor/index.h | 63 +
src/relax/op/tensor/inspect.cc | 471 ++
src/relax/op/tensor/inspect.h | 131 +
src/relax/op/tensor/linear_algebra.cc | 255 +
src/relax/op/tensor/linear_algebra.h | 57 +
src/relax/op/tensor/manipulate.cc | 1533 ++++
src/relax/op/tensor/manipulate.h | 177 +
src/relax/op/tensor/qdq.cc | 207 +
src/relax/op/tensor/qdq.h | 61 +
src/relax/op/tensor/search.cc | 209 +
src/relax/op/tensor/search.h | 49 +
src/relax/op/tensor/set.cc | 141 +
src/relax/op/tensor/set.h | 37 +
src/relax/op/tensor/sorting.cc | 155 +
src/relax/op/tensor/sorting.h | 71 +
src/relax/op/tensor/statistical.cc | 219 +
src/relax/op/tensor/statistical.h | 125 +
src/relax/op/tensor/ternary.cc | 149 +
src/relax/op/tensor/ternary.h | 45 +
src/relax/op/tensor/unary.cc | 96 +
src/relax/op/tensor/unary.h | 156 +
src/relax/testing/transform.cc | 43 +
src/relax/training/utils.cc | 224 +
src/relax/training/utils.h | 60 +
src/relax/transform/adjust_matmul_order.cc | 220 +
src/relax/transform/allocate_workspace.cc | 205 +
src/relax/transform/alter_op_impl.cc | 414 +
src/relax/transform/annotate_tir_op_pattern.cc | 55 +
src/relax/transform/attach_global_symbol.cc | 64 +
src/relax/transform/bind_params.cc | 215 +
src/relax/transform/bind_symbolic_vars.cc | 177 +
src/relax/transform/bundle_model_params.cc | 123 +
src/relax/transform/call_tir_rewrite.cc | 193 +
src/relax/transform/canonicalize_bindings.cc | 397 +
src/relax/transform/combine_parallel_matmul.cc | 398 +
src/relax/transform/compute_prim_value.cc | 94 +
src/relax/transform/convert_dataflow.cc | 151 +
src/relax/transform/convert_layout.cc | 323 +
src/relax/transform/dataflow_inplace.cc | 1039 +++
src/relax/transform/dead_code_elimination.cc | 194 +
src/relax/transform/decompose_ops.cc | 261 +
src/relax/transform/eliminate_common_subexpr.cc | 231 +
src/relax/transform/expand_matmul_of_sum.cc | 111 +
src/relax/transform/expand_tuple_arguments.cc | 187 +
src/relax/transform/few_shot_tuning.cc | 186 +
src/relax/transform/fold_constant.cc | 335 +
src/relax/transform/fuse_ops.cc | 1437 ++++
src/relax/transform/fuse_tir.cc | 1163 +++
src/relax/transform/gradient.cc | 795 ++
src/relax/transform/gradient_simplifier.cc | 200 +
src/relax/transform/gradient_simplifier.h | 46 +
src/relax/transform/infer_amp_utils.cc | 65 +
src/relax/transform/infer_amp_utils.h | 85 +
src/relax/transform/infer_layout_utils.cc | 126 +
src/relax/transform/infer_layout_utils.h | 246 +
src/relax/transform/inline_functions.cc | 228 +
src/relax/transform/kill_after_last_use.cc | 273 +
src/relax/transform/lambda_lift.cc | 510 ++
src/relax/transform/lazy_transform_params.cc | 280 +
src/relax/transform/legalize_ops.cc | 404 +
src/relax/transform/lift_transform_params.cc | 854 ++
src/relax/transform/lower_alloc_tensor.cc | 106 +
src/relax/transform/merge_composite_functions.cc | 413 +
src/relax/transform/meta_schedule.cc | 216 +
src/relax/transform/normalize.cc | 292 +
src/relax/transform/realize_vdevice.cc | 296 +
src/relax/transform/remove_purity_checking.cc | 97 +
src/relax/transform/remove_unused_outputs.cc | 326 +
src/relax/transform/remove_unused_parameters.cc | 260 +
.../transform/reorder_permute_dims_after_concat.cc | 187 +
src/relax/transform/reorder_take_after_matmul.cc | 164 +
src/relax/transform/rewrite_cuda_graph.cc | 905 ++
src/relax/transform/rewrite_dataflow_reshape.cc | 172 +
src/relax/transform/run_codegen.cc | 223 +
src/relax/transform/split_call_tir_by_pattern.cc | 782 ++
src/relax/transform/static_plan_block_memory.cc | 975 +++
src/relax/transform/to_mixed_precision.cc | 610 ++
src/relax/transform/to_non_dataflow.cc | 68 +
src/relax/transform/topological_sort.cc | 377 +
src/relax/transform/tuning_api/database.cc | 350 +
src/relax/transform/tuning_api/primitives.cc | 273 +
src/relax/transform/update_param_struct_info.cc | 111 +
src/relax/transform/update_vdevice.cc | 114 +
src/relax/transform/utils.cc | 45 +
src/relax/transform/utils.h | 443 +
src/relax/utils.cc | 240 +
src/relay/analysis/call_graph.h | 6 +-
src/relay/analysis/type_solver.cc | 2 +-
src/relay/backend/contrib/codegen_c/codegen_c.h | 35 +-
src/relay/backend/contrib/cutlass/codegen.h | 21 +
src/relay/backend/contrib/dnnl/codegen.cc | 1 -
src/relay/backend/contrib/dnnl/query_layout.cc | 18 +-
src/relay/backend/contrib/mrvl/codegen.cc | 1403 ++++
src/relay/backend/contrib/mrvl/compiler_attr.cc | 68 +
src/relay/backend/te_compiler_cache.cc | 4 +-
src/relay/backend/utils.cc | 11 +-
src/relay/backend/vm/lambda_lift.cc | 2 +-
src/relay/ir/dataflow_matcher.cc | 2 +-
src/relay/ir/function.cc | 4 +-
src/relay/ir/transform.cc | 4 +-
src/relay/op/contrib/ethosu/identity.cc | 3 +-
src/relay/op/nn/nn.cc | 1 +
src/relay/printer/tir_text_printer.cc | 4 +-
src/relay/printer/tvmscript_printer.cc | 4 +-
src/relay/transforms/dynamic_to_static.cc | 3 +-
src/relay/transforms/pattern_utils.h | 43 +-
src/relay/transforms/simplify_expr.cc | 2 +-
src/relay/transforms/to_cps.cc | 4 +-
src/relay/transforms/type_infer.cc | 2 +-
src/runtime/c_runtime_api.cc | 3 +-
src/runtime/contrib/cblas/gemm_common.h | 16 +-
src/runtime/contrib/clml/clml_runtime.cc | 62 +-
src/runtime/contrib/coreml/coreml_runtime.mm | 2 +-
src/runtime/contrib/cublas/cublas.cc | 178 +-
src/runtime/contrib/cublas/cublas_json_runtime.cc | 166 +
src/runtime/contrib/cublas/cublas_utils.cc | 25 +
src/runtime/contrib/cublas/cublas_utils.h | 24 +
src/runtime/contrib/cudnn/conv_backward.cc | 40 +-
src/runtime/contrib/cudnn/conv_forward.cc | 36 +-
src/runtime/contrib/cudnn/cudnn_json_runtime.cc | 196 +
src/runtime/contrib/cudnn/cudnn_utils.h | 15 +
src/runtime/contrib/cutlass/fp16_group_gemm.cu | 70 +
src/runtime/contrib/cutlass/fp8_group_gemm.cu | 85 +
src/runtime/contrib/cutlass/group_gemm_runner.cuh | 206 +
src/runtime/contrib/cutlass/weight_preprocess.cc | 61 +
src/runtime/contrib/dnnl/dnnl_json_runtime.cc | 63 +-
src/runtime/contrib/json/json_node.h | 2 +-
src/runtime/contrib/mrvl/mrvl_base64.h | 78 +
src/runtime/contrib/mrvl/mrvl_runtime.cc | 158 +
src/runtime/contrib/mrvl/mrvl_sw_runtime_lib.cc | 175 +
src/runtime/contrib/mrvl/mrvl_sw_runtime_lib.h | 45 +
src/runtime/contrib/msc/tensorrt_runtime.cc | 358 +
src/runtime/contrib/mscclpp/allreduce.cu | 184 +
src/runtime/contrib/papi/papi.cc | 2 +-
src/runtime/contrib/thrust/thrust.cu | 410 +-
src/runtime/contrib/vllm/attention_kernels.cu | 774 ++
src/runtime/contrib/vllm/attention_utils.cuh | 55 +
src/runtime/contrib/vllm/cache_alloc.cc | 55 +
src/runtime/contrib/vllm/cache_kernels.cu | 234 +
src/runtime/contrib/vllm/dtype_float16.h | 697 ++
src/runtime/crt/common/crt_runtime_api.c | 2 +
src/runtime/cuda/cuda_common.h | 3 +
src/runtime/cuda/cuda_device_api.cc | 66 +-
src/runtime/disco/bcast_session.cc | 142 +
src/runtime/disco/bcast_session.h | 90 +
src/runtime/disco/builtin.cc | 152 +
src/runtime/disco/cuda_ipc/cuda_ipc_memory.cc | 235 +
src/runtime/disco/cuda_ipc/custom_allreduce.cc | 118 +
src/runtime/disco/disco_worker.cc | 210 +
src/runtime/disco/disco_worker_thread.h | 83 +
src/runtime/disco/loader.cc | 449 +
src/runtime/disco/nccl/nccl.cc | 257 +
src/runtime/disco/nccl/nccl_context.h | 147 +
src/runtime/disco/process_session.cc | 279 +
src/runtime/disco/protocol.h | 254 +
src/runtime/disco/session.cc | 61 +
src/runtime/disco/threaded_session.cc | 193 +
src/runtime/disco/utils.h | 49 +
.../graph_executor/debug/graph_executor_debug.cc | 23 +-
.../graph_executor/debug/graph_executor_debug.h | 12 +
src/runtime/library_module.cc | 13 +-
src/runtime/memory/memory_manager.cc | 29 +-
src/runtime/memory/naive_allocator.h | 24 +-
src/runtime/memory/pooled_allocator.h | 39 +-
src/runtime/metadata.cc | 3 +-
src/runtime/metal/metal_common.h | 53 +-
src/runtime/metal/metal_device_api.mm | 53 +-
src/runtime/metal/metal_module.mm | 26 +-
src/runtime/minrpc/minrpc_server.h | 17 +-
src/runtime/minrpc/rpc_reference.h | 9 +
src/runtime/module.cc | 4 +
src/runtime/ndarray.cc | 87 +-
src/runtime/nvtx.cc | 42 +
src/runtime/opencl/opencl_device_api.cc | 18 +-
.../opencl/opencl_wrapper/opencl_wrapper.cc | 28 +-
src/runtime/pack_args.h | 27 +-
src/runtime/regex.cc | 41 +
src/runtime/regex.h | 64 +
src/runtime/relax_vm/builtin.cc | 619 ++
src/runtime/relax_vm/bytecode.cc | 68 +
src/runtime/relax_vm/cuda/cuda_graph_builtin.cc | 223 +
src/runtime/relax_vm/executable.cc | 580 ++
src/runtime/relax_vm/hexagon/builtin.cc | 72 +
src/runtime/relax_vm/kv_state.cc | 77 +
src/runtime/relax_vm/kv_state.h | 263 +
src/runtime/relax_vm/lm_support.cc | 622 ++
src/runtime/relax_vm/ndarray_cache_support.cc | 377 +
src/runtime/relax_vm/paged_kv_cache.cc | 1911 +++++
src/runtime/relax_vm/rnn_state.cc | 487 ++
src/runtime/relax_vm/vm.cc | 1091 +++
src/runtime/rocm/rocm_device_api.cc | 23 +-
src/runtime/rocm/rocm_module.cc | 2 +-
src/runtime/rpc/rpc_device_api.cc | 7 +-
src/runtime/rpc/rpc_endpoint.cc | 77 +-
src/runtime/rpc/rpc_local_session.cc | 20 +-
src/runtime/rpc/rpc_module.cc | 7 +
src/runtime/rpc/rpc_session.h | 54 +-
src/runtime/threading_backend.cc | 75 +-
src/runtime/vm/vm.cc | 10 +-
src/runtime/vulkan/vulkan_device.cc | 13 +-
src/runtime/vulkan/vulkan_device.h | 2 +
src/runtime/vulkan/vulkan_device_api.cc | 12 +
src/runtime/vulkan/vulkan_device_api.h | 1 +
src/script/ir_builder/ir/frame.cc | 4 +-
src/script/ir_builder/ir/ir.cc | 74 +-
src/script/ir_builder/ir/utils.h | 11 +
src/script/ir_builder/relax/distributed.cc | 61 +
src/script/ir_builder/relax/frame.cc | 280 +
src/script/ir_builder/relax/ir.cc | 278 +
src/script/ir_builder/relax/utils.h | 140 +
src/script/ir_builder/tir/frame.cc | 16 +-
src/script/ir_builder/tir/ir.cc | 80 +-
.../printer/doc_printer/python_doc_printer.cc | 6 +-
src/script/printer/ir/distributed.cc | 42 +
src/script/printer/ir/ir.cc | 48 +-
src/script/printer/ir/utils.h | 7 +-
src/script/printer/ir_docsifier.cc | 26 +-
src/script/printer/relax/binding.cc | 95 +
src/script/printer/relax/call.cc | 361 +
src/script/printer/relax/distributed.cc | 132 +
src/script/printer/relax/expr.cc | 177 +
src/script/printer/relax/function.cc | 139 +
src/script/printer/relax/region.cc | 100 +
src/script/printer/relax/struct_info.cc | 196 +
src/script/printer/relax/tir.cc | 124 +
src/script/printer/relax/type.cc | 89 +
src/script/printer/relax/utils.h | 158 +
src/script/printer/tir/expr.cc | 4 +-
src/script/printer/tir/function.cc | 2 +-
src/script/printer/tir/ir.cc | 5 +-
src/script/printer/tir/stmt.cc | 22 +
src/script/printer/utils.h | 10 +-
src/support/array.h | 27 +-
src/support/ffi_testing.cc | 54 +
src/support/libinfo.cc | 27 +
src/support/ordered_set.h | 100 +
src/support/pipe.h | 47 +-
src/support/process_id.h | 67 +
src/target/codegen.cc | 1 -
src/target/intrin_rule.cc | 20 +-
src/target/intrin_rule.h | 21 +-
src/target/llvm/codegen_aarch64.cc | 117 +
src/target/llvm/codegen_arm.cc | 2 +-
src/target/llvm/codegen_cpu.cc | 159 +-
src/target/llvm/codegen_cpu.h | 16 +-
src/target/llvm/codegen_hexagon.cc | 13 +-
src/target/llvm/codegen_llvm.cc | 251 +-
src/target/llvm/codegen_llvm.h | 24 +-
src/target/llvm/codegen_nvptx.cc | 2 +-
src/target/llvm/intrin_rule_llvm.cc | 25 +-
src/target/llvm/intrin_rule_rocm.cc | 93 +-
src/target/llvm/llvm_instance.cc | 126 +-
src/target/llvm/llvm_instance.h | 19 +-
src/target/llvm/llvm_module.cc | 225 +-
src/target/parsers/aprofile.cc | 90 +-
src/target/parsers/cpu.cc | 20 +
src/target/source/codegen_c.cc | 14 +-
src/target/source/codegen_c.h | 4 +-
src/target/source/codegen_c_host.cc | 7 +-
src/target/source/codegen_c_host.h | 3 +-
src/target/source/codegen_cuda.cc | 162 +-
src/target/source/codegen_metal.cc | 3 +-
src/target/source/codegen_opencl.cc | 8 +-
src/target/source/codegen_webgpu.cc | 352 +-
src/target/source/codegen_webgpu.h | 23 +-
src/target/source/intrin_rule_cuda.cc | 12 +
src/target/source/intrin_rule_metal.cc | 5 +-
src/target/source/intrin_rule_opencl.cc | 3 +
src/target/source/intrin_rule_webgpu.cc | 2 +-
src/target/source/literal/cuda_half_t.h | 78 +-
src/target/source/literal/cuda_int8_t.h | 64 +
src/target/source/source_module.cc | 2 +-
src/target/spirv/codegen_spirv.cc | 12 +-
src/target/spirv/codegen_spirv.h | 2 +-
src/target/spirv/intrin_rule_spirv.cc | 8 +
src/target/tag.cc | 8 +
src/target/target_kind.cc | 39 +-
src/te/operation/compute_op.cc | 6 +-
src/te/operation/create_primfunc.cc | 77 +-
src/te/operation/create_primfunc.h | 15 +
src/te/operation/op_utils.cc | 10 +-
src/te/schedule/message_passing.cc | 3 +-
src/te/schedule/schedule_lang.cc | 30 +-
src/tir/analysis/block_access_region_detector.cc | 17 +-
src/tir/analysis/identify_memcpy.cc | 2 +-
src/tir/analysis/is_pure_function.cc | 97 +
src/tir/analysis/verify_ssa.cc | 2 +-
src/tir/analysis/verify_well_formed.cc | 9 +-
src/tir/contrib/ethosu/passes.cc | 2 +-
src/tir/ir/data_type_rewriter.cc | 130 +-
src/tir/ir/expr.cc | 96 +-
src/tir/ir/expr_functor.cc | 10 +-
src/tir/ir/function.cc | 47 +
src/tir/ir/script/script_complete.cc | 11 +
src/tir/ir/specialize.cc | 10 +-
src/tir/ir/stmt.cc | 40 +-
src/tir/ir/tir_visitor_with_path.cc | 74 +-
src/tir/ir/tir_visitor_with_path.h | 43 +
src/tir/ir/transform.cc | 4 +-
src/tir/op/builtin.cc | 22 +
src/tir/op/op.cc | 111 +-
src/tir/op/runtime.cc | 41 +
src/tir/schedule/analysis/analysis.cc | 31 +-
src/tir/schedule/analysis/reducer.cc | 32 +-
src/tir/schedule/concrete_schedule.cc | 183 +-
src/tir/schedule/concrete_schedule.h | 4 +-
src/tir/schedule/ir_comparator.cc | 30 +-
src/tir/schedule/ir_comparator.h | 1 +
src/tir/schedule/primitive.h | 21 +-
src/tir/schedule/primitive/blockize_tensorize.cc | 5 +-
src/tir/schedule/primitive/cache_read_write.cc | 23 +-
src/tir/schedule/primitive/compute_at.cc | 4 +
src/tir/schedule/primitive/compute_inline.cc | 21 +-
src/tir/schedule/primitive/decompose_padding.cc | 9 -
.../schedule/primitive/layout_transformation.cc | 28 +-
src/tir/schedule/primitive/loop_transformation.cc | 303 +-
src/tir/schedule/primitive/reduction.cc | 4 +-
src/tir/schedule/schedule.cc | 2 +
src/tir/schedule/traced_schedule.cc | 27 +-
src/tir/schedule/traced_schedule.h | 4 +-
src/tir/schedule/transform.cc | 87 +
src/tir/schedule/transform.h | 9 +
src/tir/transforms/arg_binder.cc | 46 +-
src/tir/transforms/arg_binder.h | 38 +-
src/tir/transforms/bound_checker.cc | 15 +-
src/tir/transforms/compact_buffer_region.cc | 22 +-
src/tir/transforms/default_gpu_schedule.cc | 173 +
src/tir/transforms/force_narrow_index_to_i32.cc | 94 +
src/tir/transforms/hoist_expression.cc | 9 +-
src/tir/transforms/inject_permuted_layout.cc | 2 +-
src/tir/transforms/inject_software_pipeline.cc | 2 +-
src/tir/transforms/ir_utils.cc | 92 +-
src/tir/transforms/ir_utils.h | 3 +-
src/tir/transforms/lower_cross_thread_reduction.cc | 37 +-
src/tir/transforms/lower_custom_datatypes.cc | 2 +-
src/tir/transforms/lower_opaque_block.cc | 4 +-
src/tir/transforms/lower_thread_allreduce.cc | 4 +-
src/tir/transforms/lower_tvm_builtin.cc | 214 +-
src/tir/transforms/make_packed_api.cc | 58 +-
src/tir/transforms/memhammer_lower_auto_copy.cc | 4 +-
.../merge_dynamic_shared_memory_allocations.cc | 642 --
.../transforms/merge_shared_memory_allocations.cc | 703 ++
.../plan_update_buffer_allocation_location.cc | 36 +-
src/tir/transforms/renew_defs.cc | 6 +-
src/tir/transforms/renormalize_split_pattern.cc | 2 +-
src/tir/transforms/simplify.cc | 8 +
src/tir/transforms/simplify.h | 9 +-
src/tir/transforms/storage_access.cc | 14 +
src/tir/transforms/storage_access.h | 1 +
src/tir/transforms/storage_flatten.cc | 2 +-
src/tir/transforms/storage_rewrite.cc | 79 +-
src/tir/transforms/texture_flatten.cc | 2 +-
src/tir/transforms/thread_storage_sync.cc | 2 +-
src/tir/transforms/transform_mma_buffer_layout.cc | 2 +-
src/tir/transforms/unify_thread_binding.cc | 3 +-
src/tir/transforms/unroll_loop.cc | 7 +-
src/tir/transforms/unsupported_dtype_legalize.cc | 43 +-
src/tir/transforms/vectorize_loop.cc | 223 +-
src/tir/usmp/analysis/extract_buffer_info.cc | 2 +-
src/tir/usmp/transform/create_io_allocates.cc | 8 +-
src/topi/einsum.cc | 5 +-
src/topi/nn.cc | 5 +
src/topi/transform.cc | 21 +-
tests/cpp-runtime/opencl/opencl_nativeptr.cc | 40 +-
tests/cpp/container_test.cc | 4 +-
tests/cpp/nested_msg_test.cc | 319 +
tests/cpp/pattern_match_test.cc | 12 +-
tests/cpp/runtime/memory/memory_manager_tests.cc | 20 +-
tests/cpp/target/parsers/aprofile_test.cc | 301 +-
tests/cpp/target_test.cc | 22 +-
tests/cpp/tir_scalable_datatype.cc | 202 +
tests/lint/check_file_type.py | 4 +
tests/lint/cppdocs.sh | 2 +-
tests/lint/cpplint.sh | 10 +
tests/lint/pylint.sh | 3 +
tests/lint/pylintrc | 4 +-
tests/micro/zephyr/test_zephyr.py | 2 +-
.../test_runtime_ndarray.py | 70 +-
.../python/arith/test_arith_canonical_simplify.py | 5 +
tests/python/arith/test_arith_const_int_bound.py | 491 +-
tests/python/arith/test_arith_detect_cse.py | 6 +-
tests/python/arith/test_arith_domain_touched.py | 5 +-
tests/python/arith/test_arith_intset.py | 8 +
tests/python/arith/test_arith_rewrite_simplify.py | 133 +-
tests/python/arith/test_arith_simplify.py | 47 +
.../test_auto_scheduler_search_task.py | 19 +-
.../autotvm/test_autotvm_graph_tuner_core.py | 7 +
tests/python/codegen/test_inject_ptx_ldg32.py | 13 +-
.../python/codegen/test_target_codegen_aarch64.py | 314 +-
tests/python/codegen/test_target_codegen_cuda.py | 12 +
.../python/codegen/test_target_codegen_cuda_fp8.py | 818 ++
.../codegen/test_target_codegen_gpu_common.py | 55 +
tests/python/codegen/test_target_codegen_rocm.py | 53 +
tests/python/contrib/test_android/__init__.py | 18 +
.../python/contrib/test_android/infrastructure.py | 57 +
.../contrib/test_android/test_meta_schedule.py | 71 +
tests/python/contrib/test_clml/test_ops.py | 116 +-
tests/python/contrib/test_coreml_codegen.py | 2 +-
tests/python/contrib/test_cutlass.py | 98 +
tests/python/contrib/test_ethosu/test_codegen.py | 96 +
.../test_ethosu/test_copy_compute_reordering.py | 45 +-
.../contrib/test_ethosu/test_create_tiles.py | 8 -
.../contrib/test_ethosu/test_encode_constants.py | 41 +-
tests/python/contrib/test_ethosu/test_legalize.py | 45 +
.../contrib/test_ethosu/test_merge_constants.py | 48 +-
.../test_ethosu/test_remove_concatenates.py | 6 +-
.../contrib/test_ethosu/test_replace_conv2d.py | 54 +-
.../contrib/test_ethosu/test_replace_copy.py | 12 +-
tests/python/contrib/test_ethosu/test_scheduler.py | 23 +-
.../test_ethosu/test_tir_to_cs_translator.py | 40 +-
tests/python/contrib/test_ethosu/test_vela_api.py | 9 +-
.../metaschedule_e2e/test_resnet50_fp16.py | 2 +-
.../metaschedule_e2e/test_resnet50_int8.py | 4 +-
.../contrib/test_hexagon/test_dma_builtin.py | 172 +
.../contrib/test_hexagon/test_meta_schedule.py | 10 +-
.../test_relax_2d_buffer_allocation.py | 93 +
.../contrib/test_hexagon/test_relax_integration.py | 236 +
.../test_hexagon/topi/slice_op/test_cast_slice.py | 4 +-
tests/python/contrib/test_mrvl/__init__.py | 17 +
tests/python/contrib/test_mrvl/infrastructure.py | 153 +
tests/python/contrib/test_mrvl/test_mrvl.py | 187 +
tests/python/contrib/test_msc/test_graph_build.py | 2035 +++++
tests/python/contrib/test_msc/test_pipeline.py | 287 +
tests/python/contrib/test_msc/test_plugin.py | 367 +
tests/python/contrib/test_msc/test_runner.py | 169 +
tests/python/contrib/test_msc/test_tools.py | 310 +
tests/python/contrib/test_msc/test_transform.py | 150 +
.../contrib/test_msc/test_translate_relax.py | 1143 +++
.../contrib/test_msc/test_translate_relay.py | 1061 +++
.../contrib/test_msc/test_translate_tensorflow.py | 1696 ++++
.../contrib/test_msc/test_translate_tensorrt.py | 852 ++
.../contrib/test_msc/test_translate_torch.py | 1122 +++
tests/python/disco/test_callback.py | 130 +
tests/python/disco/test_ccl.py | 415 +
tests/python/disco/test_custom_allreduce.py | 82 +
tests/python/disco/test_loader.py | 458 ++
tests/python/disco/test_session.py | 238 +
tests/python/dlight/test_benchmark.py | 318 +
tests/python/dlight/test_gpu_fallback.py | 261 +
tests/python/dlight/test_gpu_gemv.py | 1345 +++
tests/python/dlight/test_gpu_general_reduction.py | 606 ++
tests/python/dlight/test_gpu_low_batch_gemv.py | 385 +
tests/python/dlight/test_gpu_matmul.py | 677 ++
tests/python/dlight/test_gpu_matmul_tensorize.py | 702 ++
tests/python/dlight/test_gpu_reduction.py | 1156 +++
tests/python/dlight/test_gpu_rmsnorm.py | 287 +
tests/python/dlight/test_gpu_transpose.py | 189 +
tests/python/dlight/test_primitives.py | 60 +
tests/python/driver/tvmc/test_command_line.py | 56 +-
tests/python/driver/tvmc/test_compiler.py | 18 +
tests/python/driver/tvmc/test_composite_target.py | 1 +
tests/python/driver/tvmc/test_target_options.py | 19 +
tests/python/frontend/keras/test_forward.py | 10 +
tests/python/frontend/mxnet/test_forward.py | 5 +-
tests/python/frontend/onnx/test_forward.py | 135 +-
tests/python/frontend/pytorch/test_forward.py | 25 +-
tests/python/frontend/tensorflow/test_forward.py | 2 +-
tests/python/frontend/tflite/test_forward.py | 78 +-
tests/python/integration/test_arm_aprofile.py | 6 +
tests/python/integration/test_auto_tensorize.py | 4 +-
tests/python/integration/test_legacy_tuning.py | 2 +-
tests/python/integration/test_lower.py | 3 +-
tests/python/ir/test_dtype.py | 40 +
tests/python/ir/test_pass_instrument.py | 63 +
.../test_meta_schedule_cpu_dot_product.py | 2 +-
...meta_schedule_mma_m16n8k8_auto_tensorization.py | 158 +-
.../test_meta_schedule_post_order_apply.py | 14 +-
...est_meta_schedule_postproc_rewrite_tensorize.py | 4 +-
...meta_schedule_schedule_cuda_layout_transform.py | 1 +
.../test_meta_schedule_schedule_rule_mlt_intrin.py | 6 +-
.../test_meta_schedule_schedule_rule_mlt_tc.py | 126 +-
.../test_meta_schedule_trace_apply.py | 190 +-
.../python/micro/test_aot_legalize_packed_call.py | 5 +-
tests/python/micro/test_micro_ms_tuning.py | 1 +
.../quantization/test_quantization_accuracy.py | 19 +-
tests/python/relax/conftest.py | 89 +
.../distributed/test_distributed_dtensor_sinfo.py | 93 +
..._distributed_transform_legalize_redistribute.py | 69 +
.../test_distributed_transform_lower_distir.py | 396 +
...ributed_transform_lower_global_to_local_view.py | 1579 ++++
...est_distributed_transform_propagate_sharding.py | 2029 +++++
.../test_distributed_tvmscript_parser.py | 193 +
.../test_distributed_tvmscript_printer.py | 159 +
tests/python/relax/frontend_nn_extern_module.cc | 69 +
tests/python/relax/test_analysis.py | 791 ++
.../test_analysis_computable_at_compile_time.py | 243 +
.../relax/test_analysis_contains_impure_call.py | 105 +
.../python/relax/test_analysis_detect_recursion.py | 453 +
.../relax/test_analysis_estimate_memory_usage.py | 127 +
.../relax/test_analysis_struct_info_analysis.py | 765 ++
.../test_analysis_suggest_layout_transforms.py | 832 ++
tests/python/relax/test_analysis_well_formed.py | 706 ++
tests/python/relax/test_ast_printer.py | 703 ++
.../relax/test_backend_dispatch_sort_scan.py | 429 +
.../relax/test_backend_transform_shape_lower.py | 897 ++
tests/python/relax/test_bind_params.py | 161 +
tests/python/relax/test_bind_symbolic_vars.py | 290 +
tests/python/relax/test_binding_rewrite.py | 336 +
tests/python/relax/test_blockbuilder_core.py | 994 +++
tests/python/relax/test_blockbuilder_emit_te.py | 125 +
tests/python/relax/test_codegen_coreml.py | 294 +
tests/python/relax/test_codegen_cublas.py | 516 ++
tests/python/relax/test_codegen_cudnn.py | 248 +
tests/python/relax/test_codegen_cutlass.py | 2501 ++++++
tests/python/relax/test_codegen_dnnl.py | 96 +
tests/python/relax/test_codegen_tensorrt.py | 115 +
tests/python/relax/test_codegen_tir_cutlass.py | 708 ++
tests/python/relax/test_contrib_vllm.py | 780 ++
tests/python/relax/test_dataflow_inplace.py | 644 ++
tests/python/relax/test_dataflow_pattern.py | 2003 +++++
tests/python/relax/test_e2e_op_dynamic.py | 100 +
tests/python/relax/test_expr.py | 314 +
tests/python/relax/test_expr_args_converter.py | 146 +
tests/python/relax/test_expr_functor.py | 867 ++
tests/python/relax/test_fast_math_transform.py | 59 +
tests/python/relax/test_frontend_common.py | 41 +
tests/python/relax/test_frontend_dynamo.py | 555 ++
tests/python/relax/test_frontend_from_fx.py | 3524 ++++++++
tests/python/relax/test_frontend_nn_debug.py | 83 +
tests/python/relax/test_frontend_nn_exporter.py | 636 ++
.../python/relax/test_frontend_nn_extern_module.py | 251 +
tests/python/relax/test_frontend_nn_jit.py | 208 +
tests/python/relax/test_frontend_nn_modules.py | 719 ++
tests/python/relax/test_frontend_nn_mutator.py | 225 +
tests/python/relax/test_frontend_nn_op.py | 1221 +++
tests/python/relax/test_frontend_nn_packing.py | 86 +
tests/python/relax/test_frontend_nn_subroutines.py | 101 +
tests/python/relax/test_frontend_nn_tensor.py | 164 +
tests/python/relax/test_frontend_onnx.py | 1889 +++++
tests/python/relax/test_frontend_stablehlo.py | 353 +
tests/python/relax/test_inline_functions.py | 404 +
tests/python/relax/test_json_compact.py | 50 +
tests/python/relax/test_kill_after_last_use.py | 107 +
tests/python/relax/test_lower_alloc_tensor.py | 47 +
.../relax/test_meta_schedule_relax_integration.py | 210 +
tests/python/relax/test_op_binary.py | 321 +
tests/python/relax/test_op_ccl.py | 276 +
tests/python/relax/test_op_create.py | 730 ++
tests/python/relax/test_op_datatype.py | 122 +
tests/python/relax/test_op_distributed.py | 58 +
tests/python/relax/test_op_grad.py | 141 +
tests/python/relax/test_op_gradient_numeric.py | 847 ++
tests/python/relax/test_op_image.py | 250 +
tests/python/relax/test_op_index.py | 1014 +++
tests/python/relax/test_op_inspect.py | 252 +
tests/python/relax/test_op_linear_algebra.py | 341 +
tests/python/relax/test_op_manipulate.py | 3356 ++++++++
tests/python/relax/test_op_misc.py | 161 +
tests/python/relax/test_op_nn.py | 1812 ++++
tests/python/relax/test_op_nn_convolution.py | 1756 ++++
tests/python/relax/test_op_nn_pooling.py | 672 ++
tests/python/relax/test_op_qdq.py | 109 +
tests/python/relax/test_op_search.py | 448 +
tests/python/relax/test_op_set.py | 871 ++
tests/python/relax/test_op_sort.py | 294 +
tests/python/relax/test_op_statistical.py | 279 +
tests/python/relax/test_op_take.py | 158 +
tests/python/relax/test_op_ternary.py | 169 +
tests/python/relax/test_op_unary.py | 218 +
tests/python/relax/test_op_view.py | 776 ++
.../python/relax/test_optimize_layout_transform.py | 410 +
tests/python/relax/test_pipeline.py | 114 +
tests/python/relax/test_relax_operators.py | 477 ++
tests/python/relax/test_relay_translator.py | 346 +
.../python/relax/test_remove_redundant_reshape.py | 115 +
tests/python/relax/test_runtime_builtin.py | 262 +
..._builtin_paged_attention_kv_cache_flashinfer.py | 693 ++
...runtime_builtin_paged_attention_kv_cache_tir.py | 1956 +++++
.../python/relax/test_runtime_builtin_rnn_state.py | 262 +
tests/python/relax/test_struct_info.py | 257 +
tests/python/relax/test_testing_nn.py | 120 +
tests/python/relax/test_training_append_loss.py | 327 +
tests/python/relax/test_training_loss.py | 298 +
tests/python/relax/test_training_optimizer.py | 602 ++
.../relax/test_training_optimizer_numeric.py | 174 +
tests/python/relax/test_training_setup_trainer.py | 232 +
.../python/relax/test_training_trainer_numeric.py | 170 +
tests/python/relax/test_transform.py | 492 ++
.../relax/test_transform_adjust_matmul_order.py | 515 ++
.../relax/test_transform_allocate_workspace.py | 164 +
tests/python/relax/test_transform_alter_op_impl.py | 576 ++
.../test_transform_annotate_tir_op_pattern.py | 440 +
.../relax/test_transform_attach_global_symbol.py | 121 +
tests/python/relax/test_transform_bind_params.py | 179 +
.../relax/test_transform_bind_symbolic_vars.py | 270 +
.../relax/test_transform_bundle_model_params.py | 237 +
.../relax/test_transform_canonicalize_bindings.py | 1015 +++
tests/python/relax/test_transform_codegen_pass.py | 389 +
.../test_transform_combine_parallel_matmul.py | 697 ++
.../relax/test_transform_compute_prim_value.py | 104 +
.../relax/test_transform_convert_dataflow.py | 493 ++
.../python/relax/test_transform_convert_layout.py | 1505 ++++
tests/python/relax/test_transform_cse.py | 663 ++
.../relax/test_transform_dead_code_elimination.py | 662 ++
tests/python/relax/test_transform_decompose_ops.py | 386 +
.../relax/test_transform_expand_matmul_of_sum.py | 157 +
.../relax/test_transform_expand_tuple_args.py | 79 +
.../python/relax/test_transform_few_shot_tuning.py | 391 +
tests/python/relax/test_transform_fold_constant.py | 444 +
tests/python/relax/test_transform_fuse_ops.py | 1646 ++++
.../relax/test_transform_fuse_ops_by_pattern.py | 1247 +++
tests/python/relax/test_transform_fuse_tir.py | 2318 ++++++
tests/python/relax/test_transform_gradient.py | 1326 +++
.../relax/test_transform_gradient_checkpoint.py | 755 ++
.../relax/test_transform_gradient_numeric.py | 227 +
.../relax/test_transform_gradient_te_register.py | 384 +
.../test_transform_inline_private_functions.py | 105 +
.../relax/test_transform_ipc_allreduce_rewrite.py | 151 +
tests/python/relax/test_transform_lambda_lift.py | 512 ++
.../relax/test_transform_lazy_transform_params.py | 1190 +++
tests/python/relax/test_transform_legalize_ops.py | 441 +
.../relax/test_transform_legalize_ops_binary.py | 2143 +++++
.../relax/test_transform_legalize_ops_ccl.py | 146 +
.../test_transform_legalize_ops_create_datatype.py | 860 ++
.../test_transform_legalize_ops_distributed.py | 64 +
.../relax/test_transform_legalize_ops_grad.py | 374 +
.../relax/test_transform_legalize_ops_image.py | 103 +
..._transform_legalize_ops_index_linear_algebra.py | 1205 +++
.../test_transform_legalize_ops_manipulate.py | 1734 ++++
.../python/relax/test_transform_legalize_ops_nn.py | 3596 ++++++++
.../relax/test_transform_legalize_ops_qdq.py | 572 ++
...st_transform_legalize_ops_search_statistical.py | 1070 +++
.../relax/test_transform_legalize_ops_unary.py | 113 +
.../relax/test_transform_lift_transform_params.py | 1825 +++++
.../test_transform_lower_gpu_ipc_alloc_storage.py | 97 +
.../test_transform_merge_composite_functions.py | 1111 +++
.../test_transform_meta_schedule_apply_database.py | 84 +
.../relax/test_transform_meta_schedule_tuning.py | 234 +
tests/python/relax/test_transform_normalize.py | 585 ++
.../relax/test_transform_normalize_global_var.py | 98 +
...st_transform_operator_specific_normalization.py | 378 +
.../python/relax/test_transform_realize_vdevice.py | 330 +
.../relax/test_transform_remove_unused_outputs.py | 123 +
.../test_transform_remove_unused_parameters.py | 204 +
..._transform_reorder_permute_dims_after_concat.py | 264 +
.../test_transform_reorder_take_after_matmul.py | 186 +
.../relax/test_transform_rewrite_cuda_graph.py | 1180 +++
.../test_transform_rewrite_dataflow_reshape.py | 585 ++
.../test_transform_static_plan_block_memory.py | 1453 ++++
.../relax/test_transform_to_mixed_precision.py | 1040 +++
.../relax/test_transform_topological_sort.py | 457 ++
.../test_transform_update_param_struct_info.py | 71 +
.../python/relax/test_transform_update_vdevice.py | 128 +
tests/python/relax/test_tuning_api.py | 781 ++
tests/python/relax/test_tvmscript_ir_builder.py | 259 +
tests/python/relax/test_tvmscript_parser.py | 2321 ++++++
.../relax/test_tvmscript_parser_op_arith_cmp.py | 183 +
.../relax/test_tvmscript_parser_op_create.py | 176 +
.../relax/test_tvmscript_parser_op_datatype.py | 54 +
.../python/relax/test_tvmscript_parser_op_grad.py | 142 +
.../python/relax/test_tvmscript_parser_op_image.py | 54 +
.../python/relax/test_tvmscript_parser_op_index.py | 82 +
.../test_tvmscript_parser_op_linear_algebra.py | 96 +
.../relax/test_tvmscript_parser_op_manipulate.py | 443 +
tests/python/relax/test_tvmscript_parser_op_nn.py | 368 +
.../relax/test_tvmscript_parser_op_search.py | 105 +
tests/python/relax/test_tvmscript_parser_op_set.py | 68 +
.../python/relax/test_tvmscript_parser_op_sort.py | 60 +
.../relax/test_tvmscript_parser_op_statistical.py | 191 +
tests/python/relax/test_tvmscript_printer_relax.py | 883 ++
tests/python/relax/test_utils.py | 275 +
.../relax/test_vm_alloc_storage_with_scope.py | 74 +
tests/python/relax/test_vm_build.py | 1250 +++
tests/python/relax/test_vm_builtin.py | 84 +
tests/python/relax/test_vm_builtin_lower.py | 86 +
tests/python/relax/test_vm_callback_function.py | 124 +
tests/python/relax/test_vm_codegen_only.py | 492 ++
tests/python/relax/test_vm_codegen_tir.py | 224 +
tests/python/relax/test_vm_cuda_graph.py | 108 +
tests/python/relax/test_vm_execbuilder.py | 296 +
tests/python/relax/test_vm_instrument.py | 102 +
tests/python/relax/test_vm_multi_device.py | 186 +
tests/python/relax/test_vm_profiler.py | 130 +
tests/python/relay/aot/aprofile_aem.mk | 98 +
.../relay/aot/aprofile_extra_support_routines.c | 25 +
.../relay/aot/test_aot_create_function_metadata.py | 38 +-
tests/python/relay/aot/test_aot_test_harness.py | 111 +
tests/python/relay/aot/test_crt_aot.py | 1 +
.../relay/strategy/arm_cpu/scalable_utils.py | 53 +
tests/python/relay/strategy/arm_cpu/test_dense.py | 173 +
.../relay/strategy/arm_cpu/test_dense_dsp.py | 88 -
tests/python/relay/strategy/arm_cpu/test_matmul.py | 118 +
.../relay/strategy/test_select_implementation.py | 77 +-
tests/python/relay/test_any.py | 9 +
tests/python/relay/test_auto_scheduler_tuning.py | 20 +-
tests/python/relay/test_autotvm_task_extraction.py | 1 +
tests/python/relay/test_custom_datatypes.py | 3 +
tests/python/relay/test_json_compact.py | 91 +-
tests/python/relay/test_op_level2.py | 10 +
tests/python/relay/test_op_level3.py | 4 +-
tests/python/relay/test_op_qnn_conv2d.py | 7 +
tests/python/relay/test_op_qnn_leaky_relu.py | 2 +-
tests/python/relay/test_pass_alter_op_layout.py | 83 +-
tests/python/relay/test_py_converter.py | 16 +-
tests/python/relay/test_roofline.py | 4 +-
tests/python/relay/test_vm.py | 24 +-
tests/python/runtime/test_runtime_device_api.py | 52 +
.../runtime/test_runtime_module_based_interface.py | 33 +-
tests/python/runtime/test_runtime_module_load.py | 13 +-
tests/python/runtime/test_runtime_nd_array.py | 253 +
tests/python/runtime/test_runtime_rpc.py | 48 +
tests/python/runtime/test_runtime_trace.py | 8 +-
tests/python/target/test_arm_target.py | 125 +
tests/python/target/test_llvm_features_info.py | 26 +-
tests/python/target/test_target_target.py | 19 +
tests/python/te/test_te_create_primfunc.py | 37 +-
tests/python/te/test_te_schedule.py | 42 +-
tests/python/testing/test_tvm_testing_features.py | 11 +
.../test_tir_analysis_get_block_access_region.py | 43 +-
.../test_tir_analysis_identify_memcpy.py | 1 +
.../test_tir_analysis_is_pure_function.py | 104 +
tests/python/tir-analysis/test_tir_analysis_oob.py | 3 +-
.../test_tir_analysis_verify_well_formed.py | 164 +-
tests/python/tir-base/test_debug_info.py | 31 +-
tests/python/tir-base/test_tir_nodes.py | 79 +-
tests/python/tir-base/test_tir_renew_defs.py | 4 +-
.../python/tir-base/test_tir_scalable_datatype.py | 60 +
tests/python/tir-base/test_tir_specialize.py | 41 +-
.../test_tir_schedule_cache_read_write.py | 81 +
.../tir-schedule/test_tir_schedule_compute_at.py | 74 +
.../test_tir_schedule_compute_inline.py | 24 +
.../python/tir-schedule/test_tir_schedule_error.py | 26 +-
.../tir-schedule/test_tir_schedule_partition.py | 460 ++
.../tir-schedule/test_tir_schedule_rfactor.py | 6 +-
.../tir-schedule/test_tir_schedule_split_fuse.py | 142 +-
.../tir-schedule/test_tir_schedule_tensorize.py | 168 +-
.../python/tir-schedule/test_tir_schedule_trace.py | 4 +-
.../test_tir_transform_common_subexpr_elim.py | 17 +-
.../test_tir_transform_convert_blocks_to_opaque.py | 2 +
.../test_tir_transform_convert_ssa.py | 72 +-
...test_tir_transform_force_narrow_index_to_i32.py | 307 +
.../test_tir_transform_fp8_legalize.py | 12 +-
.../tir-transform/test_tir_transform_helpers.py | 20 +-
.../test_tir_transform_inject_ptx_async_copy.py | 12 +-
.../test_tir_transform_inject_rolling_buffer.py | 144 +-
...t_tir_transform_lower_cross_thread_reduction.py | 244 +-
.../test_tir_transform_lower_match_buffer.py | 6 +-
.../test_tir_transform_lower_tvm_builtin.py | 37 +-
.../test_tir_transform_make_packed_api.py | 71 +-
...form_merge_dynamic_shared_memory_allocations.py | 39 +-
...sform_merge_static_shared_memory_allocations.py | 203 +
...sform_plan_update_buffer_allocation_location.py | 33 +-
.../tir-transform/test_tir_transform_simplify.py | 17 +-
.../test_tir_transform_split_host_device.py | 25 +-
.../test_tir_transform_storage_flatten.py | 2 +-
.../test_tir_transform_storage_rewrite.py | 84 +
.../test_tir_transform_thread_sync.py | 77 +
.../tir-transform/test_tir_transform_vectorize.py | 454 +-
.../test_transform_default_gpu_schedule.py | 570 ++
tests/python/tir-usmp/test_tir_usmp_algo.py | 2 -
.../test_tir_usmp_analysis_extract_bufferinfo.py | 5 +-
...ransform_convert_pool_allocations_to_offsets.py | 1 -
tests/python/tir-usmp/test_tir_usmp_utils.py | 1 -
tests/python/topi/test_topi_bitserial_dense.py | 5 +-
tests/python/topi/test_topi_conv2d_nhwc.py | 93 +-
tests/python/topi/test_topi_group_norm.py | 3 +-
tests/python/topi/test_topi_layer_norm.py | 3 +-
tests/python/topi/test_topi_math.py | 4 +-
tests/python/topi/test_topi_matmul.py | 20 +-
tests/python/topi/test_topi_reduce.py | 1 +
tests/python/topi/test_topi_transform.py | 92 +-
tests/python/tvmscript/test_tvmscript_complete.py | 36 +-
.../tvmscript/test_tvmscript_ir_builder_tir.py | 18 +
.../python/tvmscript/test_tvmscript_parser_tir.py | 150 +-
.../python/tvmscript/test_tvmscript_printer_ir.py | 17 +-
.../python/tvmscript/test_tvmscript_printer_tir.py | 77 +-
tests/python/tvmscript/test_tvmscript_roundtrip.py | 147 +-
.../tvmscript/test_tvmscript_syntax_sugar.py | 14 +
tests/scripts/release/PRERELEASE_NOTES.md | 24 -
tests/scripts/release/README.md | 12 +-
tests/scripts/release/make_notes.py | 6 +
tests/scripts/release/test_release_package.sh | 101 +
tests/scripts/setup-adreno-env.sh | 3 +-
tests/scripts/task_build.py | 11 +-
tests/scripts/task_build_adreno_bins.sh | 3 +
tests/scripts/task_config_build_adreno.sh | 3 +-
tests/scripts/task_config_build_cpu.sh | 3 +-
tests/scripts/task_config_build_gpu.sh | 2 +
tests/scripts/task_config_build_mrvl.sh | 33 +
tests/scripts/task_lint.sh | 4 +-
tests/scripts/task_python_integration.sh | 2 +-
tests/scripts/task_python_unittest.sh | 2 +-
tests/scripts/task_show_node_info.sh | 1 +
tests/scripts/unity/README | 2 +
tests/scripts/unity/task_extra_lint.sh | 23 +
tests/scripts/unity/task_python_relax.sh | 41 +
tests/scripts/unity/task_python_relax_gpuonly.sh | 25 +
version.py | 2 +-
vta/scripts/tune_resnet.py | 373 -
vta/tutorials/autotvm/README.txt | 2 -
vta/tutorials/autotvm/tune_alu_vta.py | 340 -
vta/tutorials/autotvm/tune_relay_vta.py | 532 --
vta/tutorials/frontend/deploy_classification.py | 308 -
web/.eslintrc.json | 3 +-
web/.gitignore | 2 +
web/Makefile | 22 +-
web/README.md | 2 +-
web/apps/browser/rpc_plugin.html | 19 +
web/apps/browser/rpc_server.html | 78 +-
web/apps/node/example.js | 5 +-
web/emcc/decorate_as_wasi.py | 21 +-
web/emcc/tvmjs_support.cc | 11 +-
web/emcc/wasm_runtime.cc | 62 +
web/emcc/webgpu_runtime.cc | 44 +-
web/package-lock.json | 8624 +++-----------------
web/package.json | 42 +-
web/rollup.config.js | 36 -
web/rollup.config.mjs | 44 +
web/run_jest.py | 24 +
web/src/artifact_cache.ts | 416 +
web/src/asyncify.ts | 227 +
web/src/compact.ts | 14 +-
web/src/ctypes.ts | 5 +
web/src/index.ts | 16 +-
web/src/memory.ts | 5 +-
web/src/rpc_server.ts | 86 +-
web/src/runtime.ts | 1065 ++-
web/src/support.ts | 88 +
web/src/tvmjs_runtime_wasi.d.ts | 8 +
web/src/webgpu.ts | 786 +-
web/tests/node/test_module_load.js | 3 +-
web/tests/node/test_ndarray.js | 7 +-
web/tests/node/test_object.js | 15 +-
web/tests/node/test_packed_func.js | 52 +-
web/tests/node/test_random_generator.js | 71 +
web/tests/node/test_relax_vm.js | 65 +
web/tests/python/prepare_test_libs.py | 30 +-
web/tests/python/relax_rpc_test.py | 87 +
web/tests/python/webgpu_rpc_test.py | 32 +-
web/tests/python/websock_rpc_test.py | 4 +-
web/tsconfig.json | 5 +-
web/typedoc.json | 5 +-
1759 files changed, 339842 insertions(+), 16041 deletions(-)
create mode 160000 3rdparty/flashinfer
create mode 100644 3rdparty/mscclpp/include/common.h
create mode 100644 3rdparty/mscclpp/include/msccl.cuh
create mode 100644 3rdparty/mscclpp/include/msccl.h
create mode 100644 3rdparty/picojson/test_picojson.cpp
create mode 100644 3rdparty/tensorrt_llm/custom_allreduce_kernels.cu
create mode 100644 3rdparty/tensorrt_llm/custom_allreduce_kernels.h
create mode 100644 apps/relax_examples/e2e_auto_tir.py
create mode 100644 apps/relax_examples/mlp.py
create mode 100644 apps/relax_examples/nn_module.py
create mode 100644 apps/relax_examples/resnet.py
create mode 100644 ci/jenkins/unity_jenkinsfile.groovy
create mode 100644 cmake/modules/contrib/MSC.cmake
create mode 100644 cmake/modules/contrib/MSCCLPP.cmake
create mode 100644 cmake/modules/contrib/Mrvl.cmake
create mode 100644 cmake/modules/contrib/vllm.cmake
create mode 100644 docker/Dockerfile.demo_mrvl
create mode 100755 docker/install/ubuntu_install_aprofile_aem.sh
create mode 100644 docs/how_to/deploy/mrvl.rst
delete mode 100644 gallery/how_to/compile_models/from_mxnet.py
delete mode 100644 gallery/how_to/deploy_models/deploy_quantized.py
delete mode 100644 gallery/how_to/extend_tvm/bring_your_own_datatypes.py
create mode 100644 gallery/how_to/work_with_msc/_resnet.py
create mode 100644 gallery/how_to/work_with_msc/using_tools.py
create mode 100644 gallery/how_to/work_with_msc/utils.py
create mode 100644 include/tvm/ir/global_info.h
create mode 100644 include/tvm/relax/analysis.h
create mode 100644 include/tvm/relax/attrs/ccl.h
create mode 100644 include/tvm/relax/attrs/create.h
create mode 100644 include/tvm/relax/attrs/datatype.h
create mode 100644 include/tvm/relax/attrs/distributed.h
create mode 100644 include/tvm/relax/attrs/image.h
create mode 100644 include/tvm/relax/attrs/index.h
create mode 100644 include/tvm/relax/attrs/linear_algebra.h
create mode 100644 include/tvm/relax/attrs/manipulate.h
create mode 100644 include/tvm/relax/attrs/nn.h
create mode 100644 include/tvm/relax/attrs/op.h
create mode 100644 include/tvm/relax/attrs/qdq.h
create mode 100644 include/tvm/relax/attrs/search.h
create mode 100644 include/tvm/relax/attrs/sorting.h
create mode 100644 include/tvm/relax/attrs/statistical.h
create mode 100644 include/tvm/relax/backend.h
create mode 100644 include/tvm/relax/binding_rewrite.h
create mode 100644 include/tvm/relax/block_builder.h
create mode 100644 include/tvm/relax/dataflow_matcher.h
create mode 100644 include/tvm/relax/dataflow_pattern.h
create mode 100644 include/tvm/relax/dataflow_pattern_functor.h
create mode 100644 include/tvm/relax/distributed/axis_group_graph.h
create mode 100644 include/tvm/relax/distributed/global_info.h
create mode 100644 include/tvm/relax/distributed/struct_info.h
create mode 100644 include/tvm/relax/distributed/transform.h
create mode 100644 include/tvm/relax/exec_builder.h
create mode 100644 include/tvm/relax/expr.h
create mode 100644 include/tvm/relax/expr_functor.h
create mode 100644 include/tvm/relax/nested_msg.h
create mode 100644 include/tvm/relax/op_attr_types.h
create mode 100644 include/tvm/relax/struct_info.h
create mode 100644 include/tvm/relax/struct_info_functor.h
create mode 100644 include/tvm/relax/tir_pattern.h
create mode 100644 include/tvm/relax/transform.h
create mode 100644 include/tvm/relax/tuning_api.h
create mode 100644 include/tvm/relax/type.h
create mode 100644 include/tvm/relax/utils.h
create mode 100644 include/tvm/runtime/disco/builtin.h
create mode 100644 include/tvm/runtime/disco/cuda_ipc_memory.h
create mode 100644 include/tvm/runtime/disco/disco_worker.h
create mode 100644 include/tvm/runtime/disco/session.h
create mode 100644 include/tvm/runtime/nvtx.h
create mode 100644 include/tvm/runtime/relax_vm/builtin.h
create mode 100644 include/tvm/runtime/relax_vm/bytecode.h
create mode 100644 include/tvm/runtime/relax_vm/executable.h
create mode 100644 include/tvm/runtime/relax_vm/ndarray_cache_support.h
create mode 100644 include/tvm/runtime/relax_vm/vm.h
create mode 100644 include/tvm/script/ir_builder/relax/frame.h
create mode 100644 include/tvm/script/ir_builder/relax/ir.h
copy licenses/{LICENSE.cutlass_fpA_intB_gemm.txt => LICENSE.tensorrt_llm.txt} (100%)
create mode 100644 licenses/LICENSE.vllm.txt
create mode 100644 python/tvm/autotvm/tuner/droplet_tuner.py
delete mode 100644 python/tvm/autotvm/tuner/droplet_turner.py
create mode 100644 python/tvm/contrib/cutlass/attention_operation.py
create mode 100644 python/tvm/contrib/cutlass/layer_norm_operation.py
create mode 100644 python/tvm/contrib/cutlass/rms_norm_operation.py
create mode 100644 python/tvm/contrib/mrvl.py
create mode 100644 python/tvm/contrib/msc/__init__.py
create mode 100644 python/tvm/contrib/msc/core/__init__.py
create mode 100644 python/tvm/contrib/msc/core/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/core/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/core/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/core/codegen/sources.py
create mode 100644 python/tvm/contrib/msc/core/frontend/__init__.py
create mode 100644 python/tvm/contrib/msc/core/frontend/translate.py
create mode 100644 python/tvm/contrib/msc/core/gym/__init__.py
create mode 100644 python/tvm/contrib/msc/core/gym/agent/__init__.py
create mode 100644 python/tvm/contrib/msc/core/gym/agent/base_agent.py
create mode 100644 python/tvm/contrib/msc/core/gym/agent/method.py
create mode 100644 python/tvm/contrib/msc/core/gym/agent/search_agent.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/__init__.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/configer.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/controller.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/namespace.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/service.py
create mode 100644 python/tvm/contrib/msc/core/gym/control/worker.py
create mode 100644 python/tvm/contrib/msc/core/gym/environment/__init__.py
create mode 100644 python/tvm/contrib/msc/core/gym/environment/base_env.py
create mode 100644 python/tvm/contrib/msc/core/gym/environment/method.py
create mode 100644 python/tvm/contrib/msc/core/gym/environment/prune_env.py
create mode 100644 python/tvm/contrib/msc/core/gym/environment/quantize_env.py
create mode 100644 python/tvm/contrib/msc/core/gym/namespace.py
create mode 100644 python/tvm/contrib/msc/core/ir/__init__.py
create mode 100644 python/tvm/contrib/msc/core/ir/graph.py
create mode 100644 python/tvm/contrib/msc/core/ir/translate.py
create mode 100644 python/tvm/contrib/msc/core/runtime/__init__.py
create mode 100644 python/tvm/contrib/msc/core/runtime/hook.py
create mode 100644 python/tvm/contrib/msc/core/runtime/jit.py
create mode 100644 python/tvm/contrib/msc/core/runtime/runner.py
create mode 100644 python/tvm/contrib/msc/core/tools/__init__.py
create mode 100644 python/tvm/contrib/msc/core/tools/configer.py
create mode 100644 python/tvm/contrib/msc/core/tools/distill/__init__.py
create mode 100644 python/tvm/contrib/msc/core/tools/distill/configer.py
create mode 100644 python/tvm/contrib/msc/core/tools/distill/distiller.py
create mode 100644 python/tvm/contrib/msc/core/tools/distill/method.py
create mode 100644 python/tvm/contrib/msc/core/tools/execute.py
create mode 100644 python/tvm/contrib/msc/core/tools/prune/__init__.py
create mode 100644 python/tvm/contrib/msc/core/tools/prune/configer.py
create mode 100644 python/tvm/contrib/msc/core/tools/prune/method.py
create mode 100644 python/tvm/contrib/msc/core/tools/prune/pruner.py
create mode 100644 python/tvm/contrib/msc/core/tools/quantize/__init__.py
create mode 100644 python/tvm/contrib/msc/core/tools/quantize/configer.py
create mode 100644 python/tvm/contrib/msc/core/tools/quantize/method.py
create mode 100644 python/tvm/contrib/msc/core/tools/quantize/quantizer.py
create mode 100644 python/tvm/contrib/msc/core/tools/tool.py
create mode 100644 python/tvm/contrib/msc/core/tools/track/__init__.py
create mode 100644 python/tvm/contrib/msc/core/tools/track/configer.py
create mode 100644 python/tvm/contrib/msc/core/tools/track/method.py
create mode 100644 python/tvm/contrib/msc/core/tools/track/tracker.py
create mode 100644 python/tvm/contrib/msc/core/transform/__init__.py
create mode 100644 python/tvm/contrib/msc/core/transform/pattern.py
create mode 100644 python/tvm/contrib/msc/core/transform/transform.py
create mode 100644 python/tvm/contrib/msc/core/utils/__init__.py
create mode 100644 python/tvm/contrib/msc/core/utils/arguments.py
create mode 100644 python/tvm/contrib/msc/core/utils/dataset.py
create mode 100644 python/tvm/contrib/msc/core/utils/expr.py
create mode 100644 python/tvm/contrib/msc/core/utils/file.py
create mode 100644 python/tvm/contrib/msc/core/utils/info.py
create mode 100644 python/tvm/contrib/msc/core/utils/log.py
create mode 100644 python/tvm/contrib/msc/core/utils/message.py
create mode 100644 python/tvm/contrib/msc/core/utils/namespace.py
create mode 100644 python/tvm/contrib/msc/core/utils/register.py
create mode 100644 python/tvm/contrib/msc/framework/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/frontend/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/frontend/translate.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/runtime/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/runtime/runner.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/distill/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/distill/distiller.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/prune/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/prune/pruner.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/quantize/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/quantize/quantizer.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/track/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorflow/tools/track/tracker.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/codegen/sources.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/codegen/utils.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/frontend/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/frontend/translate.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/runtime/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/runtime/runner.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/distill/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/distill/distiller.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/prune/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/prune/pruner.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/quantize/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/quantize/method.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/quantize/quantizer.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/track/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/tools/track/tracker.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/transform/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/transform/pattern.py
create mode 100644 python/tvm/contrib/msc/framework/tensorrt/transform/transform.py
create mode 100644 python/tvm/contrib/msc/framework/torch/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/framework/torch/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/framework/torch/frontend/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/frontend/translate.py
create mode 100644 python/tvm/contrib/msc/framework/torch/runtime/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/runtime/jit.py
create mode 100644 python/tvm/contrib/msc/framework/torch/runtime/runner.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/distill/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/distill/distiller.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/distill/method.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/prune/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/prune/pruner.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/quantize/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/quantize/method.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/quantize/quantizer.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/track/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/torch/tools/track/tracker.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/runtime/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/runtime/runner.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/distill/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/distill/distiller.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/prune/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/prune/pruner.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/quantize/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/quantize/method.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/quantize/quantizer.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/track/__init__.py
create mode 100644 python/tvm/contrib/msc/framework/tvm/tools/track/tracker.py
create mode 100644 python/tvm/contrib/msc/pipeline/__init__.py
create mode 100644 python/tvm/contrib/msc/pipeline/config.py
create mode 100644 python/tvm/contrib/msc/pipeline/dynamic.py
create mode 100644 python/tvm/contrib/msc/pipeline/manager.py
create mode 100644 python/tvm/contrib/msc/pipeline/pipeline.py
create mode 100644 python/tvm/contrib/msc/pipeline/utils.py
create mode 100644 python/tvm/contrib/msc/pipeline/worker.py
create mode 100644 python/tvm/contrib/msc/pipeline/wrapper.py
create mode 100644 python/tvm/contrib/msc/plugin/__init__.py
create mode 100644 python/tvm/contrib/msc/plugin/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/plugin/build.py
create mode 100644 python/tvm/contrib/msc/plugin/codegen/__init__.py
create mode 100644 python/tvm/contrib/msc/plugin/codegen/codegen.py
create mode 100644 python/tvm/contrib/msc/plugin/codegen/sources.py
create mode 100644 python/tvm/contrib/msc/plugin/op/__init__.py
create mode 100644 python/tvm/contrib/msc/plugin/op/_ffi_api.py
create mode 100644 python/tvm/contrib/msc/plugin/register.py
create mode 100644 python/tvm/contrib/msc/plugin/utils.py
create mode 100644 python/tvm/contrib/tvmjs.py
create mode 100644 python/tvm/dlight/__init__.py
create mode 100644 python/tvm/dlight/base/__init__.py
create mode 100644 python/tvm/dlight/base/analysis.py
create mode 100644 python/tvm/dlight/base/common_schedules.py
create mode 100644 python/tvm/dlight/base/schedule_rule.py
create mode 100644 python/tvm/dlight/base/transform.py
create mode 100644 python/tvm/dlight/benchmark/__init__.py
create mode 100644 python/tvm/dlight/benchmark/bench.py
create mode 100644 python/tvm/dlight/benchmark/extract.py
create mode 100644 python/tvm/dlight/benchmark/utils.py
create mode 100644 python/tvm/dlight/gpu/__init__.py
create mode 100644 python/tvm/dlight/gpu/base.py
create mode 100644 python/tvm/dlight/gpu/fallback.py
create mode 100644 python/tvm/dlight/gpu/gemv.py
create mode 100644 python/tvm/dlight/gpu/general_reduction.py
create mode 100644 python/tvm/dlight/gpu/low_batch_gemv.py
create mode 100644 python/tvm/dlight/gpu/matmul.py
create mode 100644 python/tvm/dlight/gpu/reduction.py
create mode 100644 python/tvm/dlight/gpu/rmsnorm.py
create mode 100644 python/tvm/dlight/gpu/transpose.py
create mode 100644 python/tvm/dlight/gpu/utils.py
create mode 100644 python/tvm/exec/disco_worker.py
create mode 100644 python/tvm/ir/global_info.py
create mode 100644 python/tvm/meta_schedule/relax_integration.py
create mode 100644 python/tvm/relax/__init__.py
create mode 100644 python/tvm/relax/_ffi_api.py
create mode 100644 python/tvm/relax/analysis/__init__.py
create mode 100644 python/tvm/relax/analysis/_ffi_api.py
create mode 100644 python/tvm/relax/analysis/analysis.py
create mode 100644 python/tvm/relax/analysis/estimate_memory_usage.py
create mode 100644 python/tvm/relax/backend/__init__.py
create mode 100644 python/tvm/relax/backend/_ffi_api.py
create mode 100644 python/tvm/relax/backend/contrib/__init__.py
create mode 100644 python/tvm/relax/backend/contrib/coreml.py
create mode 100644 python/tvm/relax/backend/contrib/cublas.py
create mode 100644 python/tvm/relax/backend/contrib/cudnn.py
create mode 100644 python/tvm/relax/backend/contrib/cutlass.py
create mode 100644 python/tvm/relax/backend/dispatch_sort_scan.py
create mode 100644 python/tvm/relax/backend/pattern_registry.py
create mode 100644 python/tvm/relax/backend/patterns.py
create mode 100644 python/tvm/relax/backend/utils.py
create mode 100644 python/tvm/relax/backend_tir/__init__.py
create mode 100644 python/tvm/relax/backend_tir/contrib/__init__.py
create mode 100644 python/tvm/relax/backend_tir/contrib/cutlass.py
create mode 100644 python/tvm/relax/backend_tir/cumsum.py
create mode 100644 python/tvm/relax/backend_tir/pattern.py
create mode 100644 python/tvm/relax/binding_rewrite.py
create mode 100644 python/tvm/relax/block_builder.py
create mode 100644 python/tvm/relax/distributed/__init__.py
create mode 100644 python/tvm/relax/distributed/_ffi_api.py
create mode 100644 python/tvm/relax/distributed/global_info.py
create mode 100644 python/tvm/relax/distributed/struct_info.py
create mode 100644 python/tvm/relax/distributed/transform/__init__.py
create mode 100644 python/tvm/relax/distributed/transform/_ffi_api.py
create mode 100644 python/tvm/relax/distributed/transform/transform.py
create mode 100644 python/tvm/relax/dpl/__init__.py
create mode 100644 python/tvm/relax/dpl/_ffi.py
create mode 100644 python/tvm/relax/dpl/context.py
create mode 100644 python/tvm/relax/dpl/pattern.py
create mode 100644 python/tvm/relax/dpl/rewrite.py
create mode 100644 python/tvm/relax/exec_builder.py
create mode 100644 python/tvm/relax/expr.py
create mode 100644 python/tvm/relax/expr_functor.py
create mode 100644 python/tvm/relax/frontend/__init__.py
create mode 100644 python/tvm/relax/frontend/common.py
create mode 100644 python/tvm/relax/frontend/nn/__init__.py
create mode 100644 python/tvm/relax/frontend/nn/_tensor_op.py
create mode 100644 python/tvm/relax/frontend/nn/core.py
create mode 100644 python/tvm/relax/frontend/nn/exporter.py
create mode 100644 python/tvm/relax/frontend/nn/extern.py
create mode 100644 python/tvm/relax/frontend/nn/modules.py
create mode 100644 python/tvm/relax/frontend/nn/op.py
create mode 100644 python/tvm/relax/frontend/nn/spec.py
create mode 100644 python/tvm/relax/frontend/nn/subroutine.py
create mode 100644 python/tvm/relax/frontend/nn/torch.py
create mode 100644 python/tvm/relax/frontend/nn/visitor.py
create mode 100644 python/tvm/relax/frontend/onnx/__init__.py
create mode 100644 python/tvm/relax/frontend/onnx/onnx_frontend.py
create mode 100644 python/tvm/relax/frontend/stablehlo/__init__.py
create mode 100644 python/tvm/relax/frontend/stablehlo/stablehlo_translator.py
create mode 100644 python/tvm/relax/frontend/torch/__init__.py
create mode 100644 python/tvm/relax/frontend/torch/dynamo.py
create mode 100644 python/tvm/relax/frontend/torch/fx_translator.py
create mode 100644 python/tvm/relax/ir/instrument.py
create mode 100644 python/tvm/relax/op/__init__.py
create mode 100644 python/tvm/relax/op/_ffi_api.py
create mode 100644 python/tvm/relax/op/_op_gradient.py
create mode 100644 python/tvm/relax/op/base.py
create mode 100644 python/tvm/relax/op/binary.py
create mode 100644 python/tvm/relax/op/builtin/__init__.py
create mode 100644 python/tvm/relax/op/builtin/_ffi_api.py
create mode 100644 python/tvm/relax/op/builtin/builtin.py
create mode 100644 python/tvm/relax/op/ccl/__init__.py
create mode 100644 python/tvm/relax/op/ccl/_ffi_api.py
create mode 100644 python/tvm/relax/op/ccl/ccl.py
create mode 100644 python/tvm/relax/op/create.py
create mode 100644 python/tvm/relax/op/datatype.py
create mode 100644 python/tvm/relax/op/distributed/__init__.py
create mode 100644 python/tvm/relax/op/distributed/_ffi_api.py
create mode 100644 python/tvm/relax/op/distributed/distributed.py
create mode 100644 python/tvm/relax/op/grad/__init__.py
create mode 100644 python/tvm/relax/op/grad/_ffi_api.py
create mode 100644 python/tvm/relax/op/grad/grad.py
create mode 100644 python/tvm/relax/op/image/__init__.py
create mode 100644 python/tvm/relax/op/image/_ffi_api.py
create mode 100644 python/tvm/relax/op/image/image.py
create mode 100644 python/tvm/relax/op/index.py
create mode 100644 python/tvm/relax/op/linear_algebra.py
create mode 100644 python/tvm/relax/op/manipulate.py
create mode 100644 python/tvm/relax/op/mask.py
create mode 100644 python/tvm/relax/op/memory/__init__.py
create mode 100644 python/tvm/relax/op/memory/_ffi_api.py
create mode 100644 python/tvm/relax/op/memory/memory.py
create mode 100644 python/tvm/relax/op/memory/view.py
create mode 100644 python/tvm/relax/op/nn/__init__.py
create mode 100644 python/tvm/relax/op/nn/_ffi_api.py
create mode 100644 python/tvm/relax/op/nn/nn.py
create mode 100644 python/tvm/relax/op/op_attrs.py
create mode 100644 python/tvm/relax/op/qdq.py
create mode 100644 python/tvm/relax/op/search.py
create mode 100644 python/tvm/relax/op/set.py
create mode 100644 python/tvm/relax/op/sorting.py
create mode 100644 python/tvm/relax/op/statistical.py
create mode 100644 python/tvm/relax/op/ternary.py
create mode 100644 python/tvm/relax/op/unary.py
create mode 100644 python/tvm/relax/op/vm/__init__.py
create mode 100644 python/tvm/relax/op/vm/_ffi_api.py
create mode 100644 python/tvm/relax/op/vm/vm.py
create mode 100644 python/tvm/relax/pipeline.py
create mode 100644 python/tvm/relax/struct_info.py
create mode 100644 python/tvm/relax/testing/__init__.py
create mode 100644 python/tvm/relax/testing/ast_printer.py
create mode 100644 python/tvm/relax/testing/lib_comparator.py
create mode 100644 python/tvm/relax/testing/matmul.py
create mode 100644 python/tvm/relax/testing/nn.py
create mode 100644 python/tvm/relax/testing/relay_translator.py
create mode 100644 python/tvm/relax/testing/runtime_builtin.py
create mode 100644 python/tvm/relax/testing/transform.py
create mode 100644 python/tvm/relax/testing/vm.py
create mode 100644 python/tvm/relax/training/__init__.py
create mode 100644 python/tvm/relax/training/_ffi_api.py
create mode 100644 python/tvm/relax/training/loss.py
create mode 100644 python/tvm/relax/training/optimizer.py
create mode 100644 python/tvm/relax/training/setup_trainer.py
create mode 100644 python/tvm/relax/training/trainer.py
create mode 100644 python/tvm/relax/training/utils.py
create mode 100644 python/tvm/relax/transform/__init__.py
create mode 100644 python/tvm/relax/transform/_ffi_api.py
create mode 100644 python/tvm/relax/transform/attach_external_modules.py
create mode 100644 python/tvm/relax/transform/fast_math.py
create mode 100644 python/tvm/relax/transform/ipc_allreduce_rewrite.py
create mode 100644 python/tvm/relax/transform/lazy_transform_params.py
create mode 100644 python/tvm/relax/transform/legalize_ops/__init__.py
create mode 100644 python/tvm/relax/transform/legalize_ops/binary.py
create mode 100644 python/tvm/relax/transform/legalize_ops/ccl.py
create mode 100644 python/tvm/relax/transform/legalize_ops/common.py
create mode 100644 python/tvm/relax/transform/legalize_ops/create.py
create mode 100644 python/tvm/relax/transform/legalize_ops/datatype.py
create mode 100644 python/tvm/relax/transform/legalize_ops/distributed.py
create mode 100644 python/tvm/relax/transform/legalize_ops/grad.py
create mode 100644 python/tvm/relax/transform/legalize_ops/image.py
create mode 100644 python/tvm/relax/transform/legalize_ops/index.py
create mode 100644 python/tvm/relax/transform/legalize_ops/inspect_op.py
create mode 100644 python/tvm/relax/transform/legalize_ops/linear_algebra.py
create mode 100644 python/tvm/relax/transform/legalize_ops/manipulate.py
create mode 100644 python/tvm/relax/transform/legalize_ops/nn.py
create mode 100644 python/tvm/relax/transform/legalize_ops/qdq.py
create mode 100644 python/tvm/relax/transform/legalize_ops/search.py
create mode 100644 python/tvm/relax/transform/legalize_ops/statistical.py
create mode 100644 python/tvm/relax/transform/legalize_ops/unary.py
create mode 100644 python/tvm/relax/transform/lower_gpu_ipc_alloc_storage.py
create mode 100644 python/tvm/relax/transform/optimize_layout_transform.py
create mode 100644 python/tvm/relax/transform/remove_redundant_reshape.py
create mode 100644 python/tvm/relax/transform/transform.py
create mode 100644 python/tvm/relax/transform/tuning_api/__init__.py
create mode 100644 python/tvm/relax/transform/tuning_api/_ffi_api.py
create mode 100644 python/tvm/relax/transform/tuning_api/database.py
create mode 100644 python/tvm/relax/transform/tuning_api/default_functions.py
create mode 100644 python/tvm/relax/transform/tuning_api/primitives.py
create mode 100644 python/tvm/relax/ty.py
create mode 100644 python/tvm/relax/type_converter.py
create mode 100644 python/tvm/relax/utils.py
create mode 100644 python/tvm/relax/vm_build.py
create mode 100644 python/tvm/relay/op/contrib/mrvl.py
create mode 100644 python/tvm/runtime/disco/__init__.py
create mode 100644 python/tvm/runtime/disco/_ffi_api.py
create mode 100644 python/tvm/runtime/disco/process_pool.py
create mode 100644 python/tvm/runtime/disco/session.py
create mode 100644 python/tvm/runtime/relax_vm.py
create mode 100644 python/tvm/runtime/support.py
create mode 100644 python/tvm/script/ir_builder/relax/__init__.py
create mode 100644 python/tvm/script/ir_builder/relax/_ffi_api.py
create mode 100644 python/tvm/script/ir_builder/relax/distributed/__init__.py
create mode 100644 python/tvm/script/ir_builder/relax/distributed/_ffi_api.py
create mode 100644 python/tvm/script/ir_builder/relax/distributed/ir.py
create mode 100644 python/tvm/script/ir_builder/relax/frame.py
create mode 100644 python/tvm/script/ir_builder/relax/ir.py
create mode 100644 python/tvm/script/parser/relax/__init__.py
create mode 100644 python/tvm/script/parser/relax/dist.py
create mode 100644 python/tvm/script/parser/relax/entry.py
create mode 100644 python/tvm/script/parser/relax/parser.py
create mode 100644 python/tvm/script/relax.py
create mode 100644 python/tvm/script/tir.py
create mode 100644 python/tvm/topi/arm_cpu/dense_alter_op.py
create mode 100644 python/tvm/topi/arm_cpu/matmul.py
create mode 100644 python/tvm/topi/arm_cpu/pstate_attributes.py
create mode 100644 src/arith/scalable_expression.cc
create mode 100644 src/arith/scalable_expression.h
create mode 100644 src/contrib/msc/core/codegen/base_codegen.h
create mode 100644 src/contrib/msc/core/codegen/code_stack.cc
create mode 100644 src/contrib/msc/core/codegen/code_stack.h
create mode 100644 src/contrib/msc/core/codegen/codegen_json.cc
create mode 100644 src/contrib/msc/core/codegen/codegen_json.h
create mode 100644 src/contrib/msc/core/codegen/codegen_utils.cc
create mode 100644 src/contrib/msc/core/codegen/codegen_utils.h
create mode 100644 src/contrib/msc/core/codegen/cpp_codegen.h
create mode 100644 src/contrib/msc/core/codegen/py_codegen.h
create mode 100644 src/contrib/msc/core/ir/graph.cc
create mode 100644 src/contrib/msc/core/ir/graph.h
create mode 100644 src/contrib/msc/core/ir/graph_builder.cc
create mode 100644 src/contrib/msc/core/ir/graph_builder.h
create mode 100644 src/contrib/msc/core/ir/plugin.cc
create mode 100644 src/contrib/msc/core/ir/plugin.h
create mode 100644 src/contrib/msc/core/printer/cpp_printer.cc
create mode 100644 src/contrib/msc/core/printer/cpp_printer.h
create mode 100644 src/contrib/msc/core/printer/msc_base_printer.cc
create mode 100644 src/contrib/msc/core/printer/msc_base_printer.h
create mode 100644 src/contrib/msc/core/printer/msc_doc.cc
create mode 100644 src/contrib/msc/core/printer/msc_doc.h
create mode 100644 src/contrib/msc/core/printer/print_utils.cc
create mode 100644 src/contrib/msc/core/printer/print_utils.h
create mode 100644 src/contrib/msc/core/printer/prototxt_printer.cc
create mode 100644 src/contrib/msc/core/printer/prototxt_printer.h
create mode 100644 src/contrib/msc/core/printer/python_printer.cc
create mode 100644 src/contrib/msc/core/printer/python_printer.h
create mode 100644 src/contrib/msc/core/transform/bind_named_params.cc
create mode 100644 src/contrib/msc/core/transform/bind_shape.cc
create mode 100644 src/contrib/msc/core/transform/fuse_tuple.cc
create mode 100644 src/contrib/msc/core/transform/inline_params.cc
create mode 100644 src/contrib/msc/core/transform/layout_utils.cc
create mode 100644 src/contrib/msc/core/transform/layout_utils.h
create mode 100644 src/contrib/msc/core/transform/set_byoc_attrs.cc
create mode 100644 src/contrib/msc/core/transform/set_expr_layout.cc
create mode 100644 src/contrib/msc/core/transform/set_expr_name.cc
create mode 100644 src/contrib/msc/core/utils.cc
create mode 100644 src/contrib/msc/core/utils.h
create mode 100644 src/contrib/msc/framework/tensorflow/codegen.cc
create mode 100644 src/contrib/msc/framework/tensorflow/codegen.h
create mode 100644 src/contrib/msc/framework/tensorflow/codegen_utils.h
create mode 100644 src/contrib/msc/framework/tensorflow/config.h
create mode 100644 src/contrib/msc/framework/tensorflow/tf_v1_opcode.cc
create mode 100644 src/contrib/msc/framework/tensorflow/tf_v1_opcode.h
create mode 100644 src/contrib/msc/framework/tensorrt/codegen.cc
create mode 100644 src/contrib/msc/framework/tensorrt/codegen.h
create mode 100644 src/contrib/msc/framework/tensorrt/codegen_utils.h
create mode 100644 src/contrib/msc/framework/tensorrt/tensorrt_opcode.cc
create mode 100644 src/contrib/msc/framework/tensorrt/tensorrt_opcode.h
create mode 100644 src/contrib/msc/framework/tensorrt/transform_tensorrt.cc
create mode 100644 src/contrib/msc/framework/torch/codegen.cc
create mode 100644 src/contrib/msc/framework/torch/codegen.h
create mode 100644 src/contrib/msc/framework/torch/codegen_utils.h
create mode 100644 src/contrib/msc/framework/torch/config.h
create mode 100644 src/contrib/msc/framework/torch/torch_opcode.cc
create mode 100644 src/contrib/msc/framework/torch/torch_opcode.h
create mode 100644 src/contrib/msc/framework/tvm/codegen.cc
create mode 100644 src/contrib/msc/framework/tvm/codegen.h
create mode 100644 src/contrib/msc/framework/tvm/codegen_utils.h
create mode 100644 src/contrib/msc/framework/tvm/config.h
create mode 100644 src/contrib/msc/framework/tvm/relax_opcode.cc
create mode 100644 src/contrib/msc/framework/tvm/relax_opcode.h
create mode 100644 src/contrib/msc/plugin/base_codegen.h
create mode 100644 src/contrib/msc/plugin/codegen_utils.h
create mode 100644 src/contrib/msc/plugin/tensorrt_codegen.cc
create mode 100644 src/contrib/msc/plugin/tensorrt_codegen.h
create mode 100644 src/contrib/msc/plugin/torch_codegen.cc
create mode 100644 src/contrib/msc/plugin/torch_codegen.h
create mode 100644 src/contrib/msc/plugin/tvm_codegen.cc
create mode 100644 src/contrib/msc/plugin/tvm_codegen.h
create mode 100644 src/ir/apply_pass_to_function.cc
create mode 100644 src/ir/global_info.cc
create mode 100644 src/relax/analysis/analysis.cc
create mode 100644 src/relax/analysis/computable_at_compile_time.cc
create mode 100644 src/relax/analysis/detect_recursion.cc
create mode 100644 src/relax/analysis/layout_transformation.cc
create mode 100644 src/relax/analysis/shape_analysis.cc
create mode 100644 src/relax/analysis/struct_info_analysis.cc
create mode 100644 src/relax/analysis/tir_op_pattern_kind.cc
create mode 100644 src/relax/analysis/udchain.cc
create mode 100644 src/relax/analysis/var2value.cc
create mode 100644 src/relax/analysis/well_formed.cc
create mode 100644 src/relax/backend/contrib/codegen_json/codegen_json.h
create mode 100644 src/relax/backend/contrib/cublas/codegen.cc
create mode 100644 src/relax/backend/contrib/cudnn/codegen.cc
create mode 100644 src/relax/backend/contrib/cutlass/codegen.cc
create mode 100644 src/relax/backend/contrib/dnnl/codegen.cc
create mode 100644 src/relax/backend/contrib/tensorrt/codegen.cc
create mode 100644 src/relax/backend/contrib/utils.cc
create mode 100644 src/relax/backend/contrib/utils.h
create mode 100644 src/relax/backend/pattern_registry.cc
create mode 100644 src/relax/backend/pattern_registry.h
create mode 100644 src/relax/backend/task_extraction.cc
create mode 100644 src/relax/backend/vm/codegen_vm.cc
create mode 100644 src/relax/backend/vm/codegen_vm_tir.cc
create mode 100644 src/relax/backend/vm/exec_builder.cc
create mode 100644 src/relax/backend/vm/vm_builtin_lower.cc
create mode 100644 src/relax/backend/vm/vm_shape_lower.cc
create mode 100644 src/relax/distributed/axis_group_graph.cc
create mode 100644 src/relax/distributed/global_info.cc
create mode 100644 src/relax/distributed/struct_info.cc
create mode 100644 src/relax/distributed/transform/legalize_redistribute.cc
create mode 100644 src/relax/distributed/transform/lower_distir.cc
create mode 100644 src/relax/distributed/transform/lower_global_view_to_local_view.cc
create mode 100644 src/relax/distributed/transform/propagate_sharding.cc
create mode 100644 src/relax/distributed/transform/utils.cc
create mode 100644 src/relax/distributed/transform/utils.h
create mode 100644 src/relax/ir/binding_rewrite.cc
create mode 100644 src/relax/ir/block_builder.cc
create mode 100644 src/relax/ir/dataflow_matcher.cc
create mode 100644 src/relax/ir/dataflow_matcher_impl.h
create mode 100644 src/relax/ir/dataflow_pattern.cc
create mode 100644 src/relax/ir/dataflow_pattern_functor.cc
create mode 100644 src/relax/ir/emit_te.cc
create mode 100644 src/relax/ir/emit_te.h
create mode 100644 src/relax/ir/expr.cc
create mode 100644 src/relax/ir/expr_functor.cc
create mode 100644 src/relax/ir/py_expr_functor.cc
create mode 100644 src/relax/ir/struct_info.cc
create mode 100644 src/relax/ir/struct_info_functor.cc
create mode 100644 src/relax/ir/tir_pattern.cc
create mode 100644 src/relax/ir/transform.cc
create mode 100644 src/relax/ir/type.cc
create mode 100644 src/relax/op/ccl/ccl.cc
create mode 100644 src/relax/op/ccl/ccl.h
create mode 100644 src/relax/op/distributed/binary.cc
create mode 100644 src/relax/op/distributed/binary.h
create mode 100644 src/relax/op/distributed/ccl.cc
create mode 100644 src/relax/op/distributed/distributed.cc
create mode 100644 src/relax/op/distributed/distributed.h
create mode 100644 src/relax/op/distributed/linear_algebra.cc
create mode 100644 src/relax/op/distributed/linear_algebra.h
create mode 100644 src/relax/op/distributed/manipulate.cc
create mode 100644 src/relax/op/distributed/manipulate.h
create mode 100644 src/relax/op/distributed/nn.cc
create mode 100644 src/relax/op/distributed/nn.h
create mode 100644 src/relax/op/distributed/op.cc
create mode 100644 src/relax/op/distributed/statistical.cc
create mode 100644 src/relax/op/distributed/statistical.h
create mode 100644 src/relax/op/distributed/unary.cc
create mode 100644 src/relax/op/distributed/unary.h
create mode 100644 src/relax/op/distributed/utils.cc
create mode 100644 src/relax/op/distributed/utils.h
create mode 100644 src/relax/op/image/resize.cc
create mode 100644 src/relax/op/image/resize.h
create mode 100644 src/relax/op/memory/view.cc
create mode 100644 src/relax/op/memory/view.h
create mode 100644 src/relax/op/nn/attention.cc
create mode 100644 src/relax/op/nn/attention.h
create mode 100644 src/relax/op/nn/convolution.cc
create mode 100644 src/relax/op/nn/convolution.h
create mode 100644 src/relax/op/nn/nn.cc
create mode 100644 src/relax/op/nn/nn.h
create mode 100644 src/relax/op/nn/pooling.cc
create mode 100644 src/relax/op/nn/pooling.h
create mode 100644 src/relax/op/op.cc
create mode 100644 src/relax/op/op_common.cc
create mode 100644 src/relax/op/op_common.h
create mode 100644 src/relax/op/tensor/binary.cc
create mode 100644 src/relax/op/tensor/binary.h
create mode 100644 src/relax/op/tensor/create.cc
create mode 100644 src/relax/op/tensor/create.h
create mode 100644 src/relax/op/tensor/datatype.cc
create mode 100644 src/relax/op/tensor/datatype.h
create mode 100644 src/relax/op/tensor/grad.cc
create mode 100644 src/relax/op/tensor/grad.h
create mode 100644 src/relax/op/tensor/index.cc
create mode 100644 src/relax/op/tensor/index.h
create mode 100644 src/relax/op/tensor/inspect.cc
create mode 100644 src/relax/op/tensor/inspect.h
create mode 100644 src/relax/op/tensor/linear_algebra.cc
create mode 100644 src/relax/op/tensor/linear_algebra.h
create mode 100644 src/relax/op/tensor/manipulate.cc
create mode 100644 src/relax/op/tensor/manipulate.h
create mode 100644 src/relax/op/tensor/qdq.cc
create mode 100644 src/relax/op/tensor/qdq.h
create mode 100644 src/relax/op/tensor/search.cc
create mode 100644 src/relax/op/tensor/search.h
create mode 100644 src/relax/op/tensor/set.cc
create mode 100644 src/relax/op/tensor/set.h
create mode 100644 src/relax/op/tensor/sorting.cc
create mode 100644 src/relax/op/tensor/sorting.h
create mode 100644 src/relax/op/tensor/statistical.cc
create mode 100644 src/relax/op/tensor/statistical.h
create mode 100644 src/relax/op/tensor/ternary.cc
create mode 100644 src/relax/op/tensor/ternary.h
create mode 100644 src/relax/op/tensor/unary.cc
create mode 100644 src/relax/op/tensor/unary.h
create mode 100644 src/relax/testing/transform.cc
create mode 100644 src/relax/training/utils.cc
create mode 100644 src/relax/training/utils.h
create mode 100644 src/relax/transform/adjust_matmul_order.cc
create mode 100644 src/relax/transform/allocate_workspace.cc
create mode 100644 src/relax/transform/alter_op_impl.cc
create mode 100644 src/relax/transform/annotate_tir_op_pattern.cc
create mode 100644 src/relax/transform/attach_global_symbol.cc
create mode 100644 src/relax/transform/bind_params.cc
create mode 100644 src/relax/transform/bind_symbolic_vars.cc
create mode 100644 src/relax/transform/bundle_model_params.cc
create mode 100644 src/relax/transform/call_tir_rewrite.cc
create mode 100644 src/relax/transform/canonicalize_bindings.cc
create mode 100644 src/relax/transform/combine_parallel_matmul.cc
create mode 100644 src/relax/transform/compute_prim_value.cc
create mode 100644 src/relax/transform/convert_dataflow.cc
create mode 100644 src/relax/transform/convert_layout.cc
create mode 100644 src/relax/transform/dataflow_inplace.cc
create mode 100644 src/relax/transform/dead_code_elimination.cc
create mode 100644 src/relax/transform/decompose_ops.cc
create mode 100644 src/relax/transform/eliminate_common_subexpr.cc
create mode 100644 src/relax/transform/expand_matmul_of_sum.cc
create mode 100644 src/relax/transform/expand_tuple_arguments.cc
create mode 100644 src/relax/transform/few_shot_tuning.cc
create mode 100644 src/relax/transform/fold_constant.cc
create mode 100644 src/relax/transform/fuse_ops.cc
create mode 100644 src/relax/transform/fuse_tir.cc
create mode 100644 src/relax/transform/gradient.cc
create mode 100644 src/relax/transform/gradient_simplifier.cc
create mode 100644 src/relax/transform/gradient_simplifier.h
create mode 100644 src/relax/transform/infer_amp_utils.cc
create mode 100644 src/relax/transform/infer_amp_utils.h
create mode 100644 src/relax/transform/infer_layout_utils.cc
create mode 100644 src/relax/transform/infer_layout_utils.h
create mode 100644 src/relax/transform/inline_functions.cc
create mode 100644 src/relax/transform/kill_after_last_use.cc
create mode 100644 src/relax/transform/lambda_lift.cc
create mode 100644 src/relax/transform/lazy_transform_params.cc
create mode 100644 src/relax/transform/legalize_ops.cc
create mode 100644 src/relax/transform/lift_transform_params.cc
create mode 100644 src/relax/transform/lower_alloc_tensor.cc
create mode 100644 src/relax/transform/merge_composite_functions.cc
create mode 100644 src/relax/transform/meta_schedule.cc
create mode 100644 src/relax/transform/normalize.cc
create mode 100644 src/relax/transform/realize_vdevice.cc
create mode 100644 src/relax/transform/remove_purity_checking.cc
create mode 100644 src/relax/transform/remove_unused_outputs.cc
create mode 100644 src/relax/transform/remove_unused_parameters.cc
create mode 100644 src/relax/transform/reorder_permute_dims_after_concat.cc
create mode 100644 src/relax/transform/reorder_take_after_matmul.cc
create mode 100644 src/relax/transform/rewrite_cuda_graph.cc
create mode 100644 src/relax/transform/rewrite_dataflow_reshape.cc
create mode 100644 src/relax/transform/run_codegen.cc
create mode 100644 src/relax/transform/split_call_tir_by_pattern.cc
create mode 100644 src/relax/transform/static_plan_block_memory.cc
create mode 100644 src/relax/transform/to_mixed_precision.cc
create mode 100644 src/relax/transform/to_non_dataflow.cc
create mode 100644 src/relax/transform/topological_sort.cc
create mode 100644 src/relax/transform/tuning_api/database.cc
create mode 100644 src/relax/transform/tuning_api/primitives.cc
create mode 100644 src/relax/transform/update_param_struct_info.cc
create mode 100644 src/relax/transform/update_vdevice.cc
create mode 100644 src/relax/transform/utils.cc
create mode 100644 src/relax/transform/utils.h
create mode 100644 src/relax/utils.cc
create mode 100644 src/relay/backend/contrib/mrvl/codegen.cc
create mode 100644 src/relay/backend/contrib/mrvl/compiler_attr.cc
create mode 100644 src/runtime/contrib/cublas/cublas_json_runtime.cc
create mode 100644 src/runtime/contrib/cudnn/cudnn_json_runtime.cc
create mode 100644 src/runtime/contrib/cutlass/fp16_group_gemm.cu
create mode 100644 src/runtime/contrib/cutlass/fp8_group_gemm.cu
create mode 100644 src/runtime/contrib/cutlass/group_gemm_runner.cuh
create mode 100644 src/runtime/contrib/cutlass/weight_preprocess.cc
create mode 100644 src/runtime/contrib/mrvl/mrvl_base64.h
create mode 100644 src/runtime/contrib/mrvl/mrvl_runtime.cc
create mode 100644 src/runtime/contrib/mrvl/mrvl_sw_runtime_lib.cc
create mode 100644 src/runtime/contrib/mrvl/mrvl_sw_runtime_lib.h
create mode 100644 src/runtime/contrib/msc/tensorrt_runtime.cc
create mode 100644 src/runtime/contrib/mscclpp/allreduce.cu
create mode 100644 src/runtime/contrib/vllm/attention_kernels.cu
create mode 100644 src/runtime/contrib/vllm/attention_utils.cuh
create mode 100644 src/runtime/contrib/vllm/cache_alloc.cc
create mode 100644 src/runtime/contrib/vllm/cache_kernels.cu
create mode 100644 src/runtime/contrib/vllm/dtype_float16.h
create mode 100644 src/runtime/disco/bcast_session.cc
create mode 100644 src/runtime/disco/bcast_session.h
create mode 100644 src/runtime/disco/builtin.cc
create mode 100644 src/runtime/disco/cuda_ipc/cuda_ipc_memory.cc
create mode 100644 src/runtime/disco/cuda_ipc/custom_allreduce.cc
create mode 100644 src/runtime/disco/disco_worker.cc
create mode 100644 src/runtime/disco/disco_worker_thread.h
create mode 100644 src/runtime/disco/loader.cc
create mode 100644 src/runtime/disco/nccl/nccl.cc
create mode 100644 src/runtime/disco/nccl/nccl_context.h
create mode 100644 src/runtime/disco/process_session.cc
create mode 100644 src/runtime/disco/protocol.h
create mode 100644 src/runtime/disco/session.cc
create mode 100644 src/runtime/disco/threaded_session.cc
create mode 100644 src/runtime/disco/utils.h
create mode 100644 src/runtime/nvtx.cc
create mode 100644 src/runtime/regex.cc
create mode 100644 src/runtime/regex.h
create mode 100644 src/runtime/relax_vm/builtin.cc
create mode 100644 src/runtime/relax_vm/bytecode.cc
create mode 100644 src/runtime/relax_vm/cuda/cuda_graph_builtin.cc
create mode 100644 src/runtime/relax_vm/executable.cc
create mode 100644 src/runtime/relax_vm/hexagon/builtin.cc
create mode 100644 src/runtime/relax_vm/kv_state.cc
create mode 100644 src/runtime/relax_vm/kv_state.h
create mode 100644 src/runtime/relax_vm/lm_support.cc
create mode 100644 src/runtime/relax_vm/ndarray_cache_support.cc
create mode 100644 src/runtime/relax_vm/paged_kv_cache.cc
create mode 100644 src/runtime/relax_vm/rnn_state.cc
create mode 100644 src/runtime/relax_vm/vm.cc
create mode 100644 src/script/ir_builder/relax/distributed.cc
create mode 100644 src/script/ir_builder/relax/frame.cc
create mode 100644 src/script/ir_builder/relax/ir.cc
create mode 100644 src/script/ir_builder/relax/utils.h
create mode 100644 src/script/printer/ir/distributed.cc
create mode 100644 src/script/printer/relax/binding.cc
create mode 100644 src/script/printer/relax/call.cc
create mode 100644 src/script/printer/relax/distributed.cc
create mode 100644 src/script/printer/relax/expr.cc
create mode 100644 src/script/printer/relax/function.cc
create mode 100644 src/script/printer/relax/region.cc
create mode 100644 src/script/printer/relax/struct_info.cc
create mode 100644 src/script/printer/relax/tir.cc
create mode 100644 src/script/printer/relax/type.cc
create mode 100644 src/script/printer/relax/utils.h
create mode 100644 src/support/ordered_set.h
create mode 100644 src/support/process_id.h
create mode 100644 src/target/llvm/codegen_aarch64.cc
create mode 100644 src/target/source/literal/cuda_int8_t.h
create mode 100644 src/tir/analysis/is_pure_function.cc
create mode 100644 src/tir/op/runtime.cc
create mode 100644 src/tir/transforms/default_gpu_schedule.cc
create mode 100644 src/tir/transforms/force_narrow_index_to_i32.cc
delete mode 100644 src/tir/transforms/merge_dynamic_shared_memory_allocations.cc
create mode 100644 src/tir/transforms/merge_shared_memory_allocations.cc
create mode 100644 tests/cpp/nested_msg_test.cc
create mode 100644 tests/cpp/tir_scalable_datatype.cc
create mode 100644 tests/python/codegen/test_target_codegen_cuda_fp8.py
create mode 100644 tests/python/codegen/test_target_codegen_gpu_common.py
create mode 100644 tests/python/contrib/test_android/__init__.py
create mode 100644 tests/python/contrib/test_android/infrastructure.py
create mode 100644 tests/python/contrib/test_android/test_meta_schedule.py
create mode 100644 tests/python/contrib/test_hexagon/test_dma_builtin.py
create mode 100644 tests/python/contrib/test_hexagon/test_relax_2d_buffer_allocation.py
create mode 100644 tests/python/contrib/test_hexagon/test_relax_integration.py
create mode 100644 tests/python/contrib/test_mrvl/__init__.py
create mode 100644 tests/python/contrib/test_mrvl/infrastructure.py
create mode 100644 tests/python/contrib/test_mrvl/test_mrvl.py
create mode 100644 tests/python/contrib/test_msc/test_graph_build.py
create mode 100644 tests/python/contrib/test_msc/test_pipeline.py
create mode 100644 tests/python/contrib/test_msc/test_plugin.py
create mode 100644 tests/python/contrib/test_msc/test_runner.py
create mode 100644 tests/python/contrib/test_msc/test_tools.py
create mode 100644 tests/python/contrib/test_msc/test_transform.py
create mode 100644 tests/python/contrib/test_msc/test_translate_relax.py
create mode 100644 tests/python/contrib/test_msc/test_translate_relay.py
create mode 100644 tests/python/contrib/test_msc/test_translate_tensorflow.py
create mode 100644 tests/python/contrib/test_msc/test_translate_tensorrt.py
create mode 100644 tests/python/contrib/test_msc/test_translate_torch.py
create mode 100644 tests/python/disco/test_callback.py
create mode 100644 tests/python/disco/test_ccl.py
create mode 100644 tests/python/disco/test_custom_allreduce.py
create mode 100644 tests/python/disco/test_loader.py
create mode 100644 tests/python/disco/test_session.py
create mode 100644 tests/python/dlight/test_benchmark.py
create mode 100644 tests/python/dlight/test_gpu_fallback.py
create mode 100644 tests/python/dlight/test_gpu_gemv.py
create mode 100644 tests/python/dlight/test_gpu_general_reduction.py
create mode 100644 tests/python/dlight/test_gpu_low_batch_gemv.py
create mode 100644 tests/python/dlight/test_gpu_matmul.py
create mode 100644 tests/python/dlight/test_gpu_matmul_tensorize.py
create mode 100644 tests/python/dlight/test_gpu_reduction.py
create mode 100644 tests/python/dlight/test_gpu_rmsnorm.py
create mode 100644 tests/python/dlight/test_gpu_transpose.py
create mode 100644 tests/python/dlight/test_primitives.py
create mode 100644 tests/python/ir/test_dtype.py
create mode 100644 tests/python/ir/test_pass_instrument.py
create mode 100644 tests/python/relax/conftest.py
create mode 100644 tests/python/relax/distributed/test_distributed_dtensor_sinfo.py
create mode 100644 tests/python/relax/distributed/test_distributed_transform_legalize_redistribute.py
create mode 100644 tests/python/relax/distributed/test_distributed_transform_lower_distir.py
create mode 100644 tests/python/relax/distributed/test_distributed_transform_lower_global_to_local_view.py
create mode 100644 tests/python/relax/distributed/test_distributed_transform_propagate_sharding.py
create mode 100644 tests/python/relax/distributed/test_distributed_tvmscript_parser.py
create mode 100644 tests/python/relax/distributed/test_distributed_tvmscript_printer.py
create mode 100644 tests/python/relax/frontend_nn_extern_module.cc
create mode 100644 tests/python/relax/test_analysis.py
create mode 100644 tests/python/relax/test_analysis_computable_at_compile_time.py
create mode 100644 tests/python/relax/test_analysis_contains_impure_call.py
create mode 100644 tests/python/relax/test_analysis_detect_recursion.py
create mode 100644 tests/python/relax/test_analysis_estimate_memory_usage.py
create mode 100644 tests/python/relax/test_analysis_struct_info_analysis.py
create mode 100644 tests/python/relax/test_analysis_suggest_layout_transforms.py
create mode 100644 tests/python/relax/test_analysis_well_formed.py
create mode 100644 tests/python/relax/test_ast_printer.py
create mode 100644 tests/python/relax/test_backend_dispatch_sort_scan.py
create mode 100644 tests/python/relax/test_backend_transform_shape_lower.py
create mode 100644 tests/python/relax/test_bind_params.py
create mode 100644 tests/python/relax/test_bind_symbolic_vars.py
create mode 100644 tests/python/relax/test_binding_rewrite.py
create mode 100644 tests/python/relax/test_blockbuilder_core.py
create mode 100644 tests/python/relax/test_blockbuilder_emit_te.py
create mode 100644 tests/python/relax/test_codegen_coreml.py
create mode 100644 tests/python/relax/test_codegen_cublas.py
create mode 100644 tests/python/relax/test_codegen_cudnn.py
create mode 100644 tests/python/relax/test_codegen_cutlass.py
create mode 100644 tests/python/relax/test_codegen_dnnl.py
create mode 100644 tests/python/relax/test_codegen_tensorrt.py
create mode 100644 tests/python/relax/test_codegen_tir_cutlass.py
create mode 100644 tests/python/relax/test_contrib_vllm.py
create mode 100644 tests/python/relax/test_dataflow_inplace.py
create mode 100644 tests/python/relax/test_dataflow_pattern.py
create mode 100644 tests/python/relax/test_e2e_op_dynamic.py
create mode 100644 tests/python/relax/test_expr.py
create mode 100644 tests/python/relax/test_expr_args_converter.py
create mode 100644 tests/python/relax/test_expr_functor.py
create mode 100644 tests/python/relax/test_fast_math_transform.py
create mode 100644 tests/python/relax/test_frontend_common.py
create mode 100644 tests/python/relax/test_frontend_dynamo.py
create mode 100644 tests/python/relax/test_frontend_from_fx.py
create mode 100644 tests/python/relax/test_frontend_nn_debug.py
create mode 100644 tests/python/relax/test_frontend_nn_exporter.py
create mode 100644 tests/python/relax/test_frontend_nn_extern_module.py
create mode 100644 tests/python/relax/test_frontend_nn_jit.py
create mode 100644 tests/python/relax/test_frontend_nn_modules.py
create mode 100644 tests/python/relax/test_frontend_nn_mutator.py
create mode 100644 tests/python/relax/test_frontend_nn_op.py
create mode 100644 tests/python/relax/test_frontend_nn_packing.py
create mode 100644 tests/python/relax/test_frontend_nn_subroutines.py
create mode 100644 tests/python/relax/test_frontend_nn_tensor.py
create mode 100644 tests/python/relax/test_frontend_onnx.py
create mode 100644 tests/python/relax/test_frontend_stablehlo.py
create mode 100644 tests/python/relax/test_inline_functions.py
create mode 100644 tests/python/relax/test_json_compact.py
create mode 100644 tests/python/relax/test_kill_after_last_use.py
create mode 100644 tests/python/relax/test_lower_alloc_tensor.py
create mode 100644 tests/python/relax/test_meta_schedule_relax_integration.py
create mode 100644 tests/python/relax/test_op_binary.py
create mode 100644 tests/python/relax/test_op_ccl.py
create mode 100644 tests/python/relax/test_op_create.py
create mode 100644 tests/python/relax/test_op_datatype.py
create mode 100644 tests/python/relax/test_op_distributed.py
create mode 100644 tests/python/relax/test_op_grad.py
create mode 100644 tests/python/relax/test_op_gradient_numeric.py
create mode 100644 tests/python/relax/test_op_image.py
create mode 100644 tests/python/relax/test_op_index.py
create mode 100644 tests/python/relax/test_op_inspect.py
create mode 100644 tests/python/relax/test_op_linear_algebra.py
create mode 100644 tests/python/relax/test_op_manipulate.py
create mode 100644 tests/python/relax/test_op_misc.py
create mode 100644 tests/python/relax/test_op_nn.py
create mode 100644 tests/python/relax/test_op_nn_convolution.py
create mode 100644 tests/python/relax/test_op_nn_pooling.py
create mode 100644 tests/python/relax/test_op_qdq.py
create mode 100644 tests/python/relax/test_op_search.py
create mode 100644 tests/python/relax/test_op_set.py
create mode 100644 tests/python/relax/test_op_sort.py
create mode 100644 tests/python/relax/test_op_statistical.py
create mode 100644 tests/python/relax/test_op_take.py
create mode 100644 tests/python/relax/test_op_ternary.py
create mode 100644 tests/python/relax/test_op_unary.py
create mode 100644 tests/python/relax/test_op_view.py
create mode 100644 tests/python/relax/test_optimize_layout_transform.py
create mode 100644 tests/python/relax/test_pipeline.py
create mode 100644 tests/python/relax/test_relax_operators.py
create mode 100644 tests/python/relax/test_relay_translator.py
create mode 100644 tests/python/relax/test_remove_redundant_reshape.py
create mode 100644 tests/python/relax/test_runtime_builtin.py
create mode 100644 tests/python/relax/test_runtime_builtin_paged_attention_kv_cache_flashinfer.py
create mode 100644 tests/python/relax/test_runtime_builtin_paged_attention_kv_cache_tir.py
create mode 100644 tests/python/relax/test_runtime_builtin_rnn_state.py
create mode 100644 tests/python/relax/test_struct_info.py
create mode 100644 tests/python/relax/test_testing_nn.py
create mode 100644 tests/python/relax/test_training_append_loss.py
create mode 100644 tests/python/relax/test_training_loss.py
create mode 100644 tests/python/relax/test_training_optimizer.py
create mode 100644 tests/python/relax/test_training_optimizer_numeric.py
create mode 100644 tests/python/relax/test_training_setup_trainer.py
create mode 100644 tests/python/relax/test_training_trainer_numeric.py
create mode 100644 tests/python/relax/test_transform.py
create mode 100644 tests/python/relax/test_transform_adjust_matmul_order.py
create mode 100644 tests/python/relax/test_transform_allocate_workspace.py
create mode 100644 tests/python/relax/test_transform_alter_op_impl.py
create mode 100644 tests/python/relax/test_transform_annotate_tir_op_pattern.py
create mode 100644 tests/python/relax/test_transform_attach_global_symbol.py
create mode 100644 tests/python/relax/test_transform_bind_params.py
create mode 100644 tests/python/relax/test_transform_bind_symbolic_vars.py
create mode 100644 tests/python/relax/test_transform_bundle_model_params.py
create mode 100644 tests/python/relax/test_transform_canonicalize_bindings.py
create mode 100644 tests/python/relax/test_transform_codegen_pass.py
create mode 100644 tests/python/relax/test_transform_combine_parallel_matmul.py
create mode 100644 tests/python/relax/test_transform_compute_prim_value.py
create mode 100644 tests/python/relax/test_transform_convert_dataflow.py
create mode 100644 tests/python/relax/test_transform_convert_layout.py
create mode 100644 tests/python/relax/test_transform_cse.py
create mode 100644 tests/python/relax/test_transform_dead_code_elimination.py
create mode 100644 tests/python/relax/test_transform_decompose_ops.py
create mode 100644 tests/python/relax/test_transform_expand_matmul_of_sum.py
create mode 100644 tests/python/relax/test_transform_expand_tuple_args.py
create mode 100644 tests/python/relax/test_transform_few_shot_tuning.py
create mode 100644 tests/python/relax/test_transform_fold_constant.py
create mode 100644 tests/python/relax/test_transform_fuse_ops.py
create mode 100644 tests/python/relax/test_transform_fuse_ops_by_pattern.py
create mode 100644 tests/python/relax/test_transform_fuse_tir.py
create mode 100644 tests/python/relax/test_transform_gradient.py
create mode 100644 tests/python/relax/test_transform_gradient_checkpoint.py
create mode 100644 tests/python/relax/test_transform_gradient_numeric.py
create mode 100644 tests/python/relax/test_transform_gradient_te_register.py
create mode 100644 tests/python/relax/test_transform_inline_private_functions.py
create mode 100644 tests/python/relax/test_transform_ipc_allreduce_rewrite.py
create mode 100644 tests/python/relax/test_transform_lambda_lift.py
create mode 100644 tests/python/relax/test_transform_lazy_transform_params.py
create mode 100644 tests/python/relax/test_transform_legalize_ops.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_binary.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_ccl.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_create_datatype.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_distributed.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_grad.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_image.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_index_linear_algebra.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_manipulate.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_nn.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_qdq.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_search_statistical.py
create mode 100644 tests/python/relax/test_transform_legalize_ops_unary.py
create mode 100644 tests/python/relax/test_transform_lift_transform_params.py
create mode 100644 tests/python/relax/test_transform_lower_gpu_ipc_alloc_storage.py
create mode 100644 tests/python/relax/test_transform_merge_composite_functions.py
create mode 100644 tests/python/relax/test_transform_meta_schedule_apply_database.py
create mode 100644 tests/python/relax/test_transform_meta_schedule_tuning.py
create mode 100644 tests/python/relax/test_transform_normalize.py
create mode 100644 tests/python/relax/test_transform_normalize_global_var.py
create mode 100644 tests/python/relax/test_transform_operator_specific_normalization.py
create mode 100644 tests/python/relax/test_transform_realize_vdevice.py
create mode 100644 tests/python/relax/test_transform_remove_unused_outputs.py
create mode 100644 tests/python/relax/test_transform_remove_unused_parameters.py
create mode 100644 tests/python/relax/test_transform_reorder_permute_dims_after_concat.py
create mode 100644 tests/python/relax/test_transform_reorder_take_after_matmul.py
create mode 100644 tests/python/relax/test_transform_rewrite_cuda_graph.py
create mode 100644 tests/python/relax/test_transform_rewrite_dataflow_reshape.py
create mode 100644 tests/python/relax/test_transform_static_plan_block_memory.py
create mode 100644 tests/python/relax/test_transform_to_mixed_precision.py
create mode 100644 tests/python/relax/test_transform_topological_sort.py
create mode 100644 tests/python/relax/test_transform_update_param_struct_info.py
create mode 100644 tests/python/relax/test_transform_update_vdevice.py
create mode 100644 tests/python/relax/test_tuning_api.py
create mode 100644 tests/python/relax/test_tvmscript_ir_builder.py
create mode 100644 tests/python/relax/test_tvmscript_parser.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_arith_cmp.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_create.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_datatype.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_grad.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_image.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_index.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_linear_algebra.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_manipulate.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_nn.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_search.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_set.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_sort.py
create mode 100644 tests/python/relax/test_tvmscript_parser_op_statistical.py
create mode 100644 tests/python/relax/test_tvmscript_printer_relax.py
create mode 100644 tests/python/relax/test_utils.py
create mode 100644 tests/python/relax/test_vm_alloc_storage_with_scope.py
create mode 100644 tests/python/relax/test_vm_build.py
create mode 100644 tests/python/relax/test_vm_builtin.py
create mode 100644 tests/python/relax/test_vm_builtin_lower.py
create mode 100644 tests/python/relax/test_vm_callback_function.py
create mode 100644 tests/python/relax/test_vm_codegen_only.py
create mode 100644 tests/python/relax/test_vm_codegen_tir.py
create mode 100644 tests/python/relax/test_vm_cuda_graph.py
create mode 100644 tests/python/relax/test_vm_execbuilder.py
create mode 100644 tests/python/relax/test_vm_instrument.py
create mode 100644 tests/python/relax/test_vm_multi_device.py
create mode 100644 tests/python/relax/test_vm_profiler.py
create mode 100644 tests/python/relay/aot/aprofile_aem.mk
create mode 100644 tests/python/relay/aot/aprofile_extra_support_routines.c
create mode 100644 tests/python/relay/aot/test_aot_test_harness.py
create mode 100644 tests/python/relay/strategy/arm_cpu/scalable_utils.py
create mode 100644 tests/python/relay/strategy/arm_cpu/test_dense.py
delete mode 100644 tests/python/relay/strategy/arm_cpu/test_dense_dsp.py
create mode 100644 tests/python/relay/strategy/arm_cpu/test_matmul.py
create mode 100644 tests/python/runtime/test_runtime_device_api.py
create mode 100644 tests/python/runtime/test_runtime_nd_array.py
create mode 100644 tests/python/tir-analysis/test_tir_analysis_is_pure_function.py
create mode 100644 tests/python/tir-base/test_tir_scalable_datatype.py
create mode 100644 tests/python/tir-schedule/test_tir_schedule_partition.py
create mode 100644 tests/python/tir-transform/test_tir_transform_force_narrow_index_to_i32.py
create mode 100644 tests/python/tir-transform/test_tir_transform_merge_static_shared_memory_allocations.py
create mode 100644 tests/python/tir-transform/test_transform_default_gpu_schedule.py
delete mode 100644 tests/scripts/release/PRERELEASE_NOTES.md
create mode 100644 tests/scripts/release/test_release_package.sh
create mode 100755 tests/scripts/task_config_build_mrvl.sh
create mode 100644 tests/scripts/unity/README
create mode 100755 tests/scripts/unity/task_extra_lint.sh
create mode 100755 tests/scripts/unity/task_python_relax.sh
create mode 100755 tests/scripts/unity/task_python_relax_gpuonly.sh
delete mode 100644 vta/scripts/tune_resnet.py
delete mode 100644 vta/tutorials/autotvm/README.txt
delete mode 100644 vta/tutorials/autotvm/tune_alu_vta.py
delete mode 100644 vta/tutorials/autotvm/tune_relay_vta.py
delete mode 100644 vta/tutorials/frontend/deploy_classification.py
create mode 100644 web/apps/browser/rpc_plugin.html
delete mode 100644 web/rollup.config.js
create mode 100644 web/rollup.config.mjs
create mode 100644 web/run_jest.py
create mode 100644 web/src/artifact_cache.ts
create mode 100644 web/src/asyncify.ts
create mode 100644 web/src/tvmjs_runtime_wasi.d.ts
create mode 100644 web/tests/node/test_random_generator.js
create mode 100644 web/tests/node/test_relax_vm.js
create mode 100644 web/tests/python/relax_rpc_test.py