This is an automated email from the ASF dual-hosted git repository.
github-bot pushed a change to branch nightly
in repository https://gitbox.apache.org/repos/asf/tvm.git
from 170302bab2 [FFI][DOCS] Initial bringup of cpp docs (#18279)
add 6a38d926e2 [FFI][Bugfix] Fix bug of `ffi.cpp.load_inline` on Windows
(#18281)
add 349df2bc26 [FFI][REFACTOR] Cleanup namespace (#18280)
add bfd7e467bd [FFI] Relax default alignment and continguous requirement
(#18282)
add 0ae2dc17dc [Fix][Metal] Fix type for device array in Metal API (#18283)
add c852002334 [Relax] Add Relax to Python Function Converter (#18269)
add c3b168b8ea [FFI][REFACTOR] Introduce UnsafeInit and enhance ObjectRef
null safety (#18284)
add 010089eafe [Hotfix] Fix the conflicts about ffi-related updated names
(#18287)
add 36522b2e4c [FFI][Bugfix] Enable `load_inline` on macos (#18285)
add 3900556af9 [Metal] Fix MetalModuleCreate (#18290)
add c655f14e03 [3rdparty] Bump cutlass_fpA_intB_gemm to fix SM90 build
(#18291)
No new revisions were added by this update.
Summary of changes:
3rdparty/cutlass_fpA_intB_gemm | 2 +-
apps/hexagon_launcher/launcher_core.cc | 4 +-
apps/ios_rpc/tvmrpc/TVMRuntime.mm | 2 +-
docs/arch/pass_infra.rst | 32 +-
ffi/include/tvm/ffi/cast.h | 13 +-
ffi/include/tvm/ffi/container/array.h | 6 +-
ffi/include/tvm/ffi/container/map.h | 6 +-
ffi/include/tvm/ffi/container/shape.h | 17 +-
ffi/include/tvm/ffi/container/tensor.h | 26 +-
ffi/include/tvm/ffi/container/tuple.h | 14 +-
ffi/include/tvm/ffi/container/variant.h | 4 +-
ffi/include/tvm/ffi/dtype.h | 2 -
ffi/include/tvm/ffi/extra/module.h | 17 +-
ffi/include/tvm/ffi/function.h | 2 +-
ffi/include/tvm/ffi/function_details.h | 2 +-
ffi/include/tvm/ffi/memory.h | 2 -
ffi/include/tvm/ffi/object.h | 62 +-
ffi/include/tvm/ffi/optional.h | 19 +-
ffi/include/tvm/ffi/reflection/access_path.h | 4 +
ffi/include/tvm/ffi/reflection/registry.h | 10 +
ffi/include/tvm/ffi/rvalue_ref.h | 9 +-
ffi/include/tvm/ffi/string.h | 3 -
ffi/include/tvm/ffi/type_traits.h | 17 +-
ffi/python/tvm_ffi/_convert.py | 4 +-
ffi/python/tvm_ffi/cpp/load_inline.py | 65 +-
ffi/python/tvm_ffi/cython/function.pxi | 5 +-
ffi/python/tvm_ffi/cython/tensor.pxi | 43 +-
ffi/src/ffi/tensor.cc | 2 +-
ffi/tests/cpp/test_object.cc | 8 +
ffi/tests/cpp/testing_object.h | 10 +-
ffi/tests/python/test_load_inline.py | 4 -
include/tvm/arith/analyzer.h | 4 +-
include/tvm/arith/bound.h | 4 +-
include/tvm/arith/int_set.h | 48 +-
include/tvm/arith/int_solver.h | 37 +-
include/tvm/arith/iter_affine_map.h | 42 +-
include/tvm/arith/pattern.h | 4 +-
include/tvm/ir/analysis.h | 2 +-
include/tvm/ir/attrs.h | 34 +-
include/tvm/ir/diagnostic.h | 4 +-
include/tvm/ir/env_func.h | 12 +-
include/tvm/ir/expr.h | 22 +-
include/tvm/ir/function.h | 10 +-
include/tvm/ir/global_info.h | 2 +-
include/tvm/ir/global_var_supply.h | 6 +-
include/tvm/ir/instrument.h | 2 +-
include/tvm/ir/module.h | 46 +-
include/tvm/ir/name_supply.h | 13 +-
include/tvm/ir/op.h | 32 +-
include/tvm/ir/replace_global_vars.h | 4 +-
include/tvm/ir/source_map.h | 20 +-
include/tvm/ir/transform.h | 56 +-
include/tvm/ir/type.h | 12 +-
include/tvm/ir/type_functor.h | 2 +-
include/tvm/meta_schedule/arg_info.h | 4 +-
include/tvm/meta_schedule/builder.h | 27 +-
include/tvm/meta_schedule/cost_model.h | 34 +-
include/tvm/meta_schedule/database.h | 89 +-
include/tvm/meta_schedule/extracted_task.h | 8 +-
include/tvm/meta_schedule/feature_extractor.h | 14 +-
include/tvm/meta_schedule/measure_callback.h | 34 +-
include/tvm/meta_schedule/measure_candidate.h | 4 +-
include/tvm/meta_schedule/mutator.h | 20 +-
include/tvm/meta_schedule/postproc.h | 14 +-
include/tvm/meta_schedule/profiler.h | 8 +-
include/tvm/meta_schedule/runner.h | 28 +-
.../tvm/meta_schedule/schedule/cuda/thread_bind.h | 8 +-
include/tvm/meta_schedule/schedule_rule.h | 76 +-
include/tvm/meta_schedule/search_strategy.h | 37 +-
include/tvm/meta_schedule/space_generator.h | 51 +-
include/tvm/meta_schedule/task_scheduler.h | 71 +-
include/tvm/meta_schedule/tune_context.h | 26 +-
include/tvm/node/attr_registry_map.h | 4 +-
include/tvm/node/cast.h | 13 +-
include/tvm/node/reflection.h | 3 +-
include/tvm/node/repr_printer.h | 8 +-
include/tvm/node/script_printer.h | 24 +-
include/tvm/node/structural_equal.h | 4 +-
include/tvm/node/structural_hash.h | 4 +-
include/tvm/relax/analysis.h | 71 +-
include/tvm/relax/attrs/ccl.h | 2 +-
include/tvm/relax/attrs/image.h | 10 +-
include/tvm/relax/attrs/index.h | 4 +-
include/tvm/relax/attrs/linear_algebra.h | 2 +-
include/tvm/relax/attrs/manipulate.h | 26 +-
include/tvm/relax/attrs/nn.h | 138 +--
include/tvm/relax/attrs/op.h | 8 +-
include/tvm/relax/attrs/search.h | 2 +-
include/tvm/relax/attrs/sorting.h | 2 +-
include/tvm/relax/attrs/statistical.h | 4 +-
include/tvm/relax/binding_rewrite.h | 12 +-
include/tvm/relax/block_builder.h | 16 +-
include/tvm/relax/dataflow_matcher.h | 16 +-
include/tvm/relax/dataflow_pattern.h | 90 +-
include/tvm/relax/distributed/axis_group_graph.h | 15 +-
include/tvm/relax/distributed/global_info.h | 6 +-
include/tvm/relax/distributed/struct_info.h | 8 +-
include/tvm/relax/exec_builder.h | 2 +-
include/tvm/relax/expr.h | 104 +-
include/tvm/relax/expr_functor.h | 9 +-
include/tvm/relax/nested_msg.h | 69 +-
include/tvm/relax/op_attr_types.h | 4 +-
include/tvm/relax/struct_info.h | 40 +-
include/tvm/relax/tir_pattern.h | 10 +-
include/tvm/relax/transform.h | 87 +-
include/tvm/relax/utils.h | 10 +-
include/tvm/runtime/contrib/papi.h | 3 +-
include/tvm/runtime/disco/builtin.h | 8 +-
include/tvm/runtime/disco/disco_worker.h | 2 +-
include/tvm/runtime/disco/session.h | 5 +-
include/tvm/runtime/memory/memory_manager.h | 4 +-
include/tvm/runtime/module.h | 12 +-
include/tvm/runtime/object.h | 26 +-
include/tvm/runtime/profiling.h | 53 +-
include/tvm/runtime/tensor.h | 8 +-
include/tvm/runtime/vm/executable.h | 10 +-
include/tvm/runtime/vm/tensor_cache_support.h | 10 +-
include/tvm/runtime/vm/vm.h | 6 +-
include/tvm/script/ir_builder/base.h | 25 +-
include/tvm/script/ir_builder/ir/frame.h | 11 +-
include/tvm/script/ir_builder/ir/ir.h | 4 +-
include/tvm/script/ir_builder/relax/frame.h | 50 +-
include/tvm/script/ir_builder/relax/ir.h | 10 +-
include/tvm/script/ir_builder/tir/frame.h | 130 ++-
include/tvm/script/ir_builder/tir/ir.h | 118 ++-
include/tvm/script/printer/doc.h | 191 ++--
include/tvm/script/printer/ir_docsifier.h | 33 +-
include/tvm/script/printer/ir_docsifier_functor.h | 10 +-
include/tvm/target/tag.h | 26 +-
include/tvm/target/target.h | 46 +-
include/tvm/target/target_kind.h | 78 +-
include/tvm/target/virtual_device.h | 6 +-
include/tvm/te/operation.h | 114 +-
include/tvm/te/tensor.h | 23 +-
include/tvm/tir/analysis.h | 30 +-
include/tvm/tir/block_dependence_info.h | 2 +-
include/tvm/tir/block_scope.h | 19 +-
include/tvm/tir/buffer.h | 45 +-
include/tvm/tir/builtin.h | 2 +-
include/tvm/tir/data_layout.h | 24 +-
include/tvm/tir/data_type_rewriter.h | 7 +-
include/tvm/tir/expr.h | 62 +-
include/tvm/tir/function.h | 16 +-
include/tvm/tir/index_map.h | 33 +-
include/tvm/tir/op.h | 16 +-
include/tvm/tir/op_attr_types.h | 4 +-
include/tvm/tir/schedule/instruction.h | 31 +-
include/tvm/tir/schedule/schedule.h | 119 ++-
include/tvm/tir/schedule/state.h | 4 +-
include/tvm/tir/schedule/trace.h | 16 +-
include/tvm/tir/stmt.h | 125 +--
include/tvm/tir/stmt_functor.h | 32 +-
include/tvm/tir/transform.h | 12 +-
include/tvm/tir/var.h | 26 +-
include/tvm/topi/broadcast.h | 44 +-
include/tvm/topi/contrib/cublas.h | 4 +-
include/tvm/topi/contrib/rocblas.h | 4 +-
include/tvm/topi/detail/array_utils.h | 2 +-
include/tvm/topi/detail/broadcast.h | 20 +-
include/tvm/topi/detail/constant_utils.h | 6 +-
include/tvm/topi/detail/extern.h | 32 +-
include/tvm/topi/detail/fuse.h | 2 +-
include/tvm/topi/detail/pad_utils.h | 2 +-
include/tvm/topi/detail/ravel_unravel.h | 4 +-
include/tvm/topi/detail/strided_slice.h | 31 +-
include/tvm/topi/detail/tensor_utils.h | 6 +-
include/tvm/topi/einsum.h | 6 +-
include/tvm/topi/elemwise.h | 50 +-
include/tvm/topi/nn.h | 90 +-
include/tvm/topi/nn/bnn.h | 8 +-
include/tvm/topi/nn/dilate.h | 12 +-
include/tvm/topi/nn/flatten.h | 2 +-
include/tvm/topi/nn/group_norm.h | 17 +-
include/tvm/topi/nn/instance_norm.h | 10 +-
include/tvm/topi/nn/layer_norm.h | 10 +-
include/tvm/topi/nn/local_response_norm.h | 4 +-
include/tvm/topi/nn/pooling.h | 143 +--
include/tvm/topi/nn/rms_norm.h | 12 +-
include/tvm/topi/nn/softmax.h | 27 +-
include/tvm/topi/reduction.h | 110 +-
include/tvm/topi/transform.h | 424 ++++----
include/tvm/topi/utils.h | 8 +-
include/tvm/topi/vision/reorg.h | 2 +-
python/tvm/relax/relax_to_pyfunc_converter.py | 1106 ++++++++++++++++++++
python/tvm/runtime/_tensor.py | 10 +-
src/arith/analyzer.cc | 8 +-
src/arith/bound_deducer.cc | 13 +-
src/arith/canonical_simplify.cc | 36 +-
src/arith/const_fold.h | 40 +-
src/arith/const_int_bound.cc | 8 +-
src/arith/detect_common_subexpr.cc | 4 +-
src/arith/detect_linear_equation.cc | 20 +-
src/arith/domain_touched.cc | 16 +-
src/arith/int_constraints.cc | 73 +-
src/arith/int_set.cc | 128 +--
src/arith/ir_mutator_with_analyzer.cc | 20 +-
src/arith/ir_mutator_with_analyzer.h | 9 +-
src/arith/iter_affine_map.cc | 208 ++--
src/arith/modular_set.cc | 4 +-
src/arith/narrow_predicate_expression.cc | 22 +-
src/arith/narrow_predicate_expression.h | 2 +-
src/arith/pattern_match.h | 14 +-
src/arith/presburger_set.cc | 18 +-
src/arith/presburger_set.h | 18 +-
src/arith/rewrite_simplify.cc | 17 +-
src/arith/rewrite_simplify.h | 4 +-
src/arith/scalable_expression.cc | 4 +-
src/arith/scalable_expression.h | 4 +-
src/arith/solve_linear_equation.cc | 31 +-
src/arith/solve_linear_inequality.cc | 59 +-
src/arith/transitive_comparison_analyzer.cc | 2 +-
src/arith/unwrap_vector_expr.cc | 2 +-
src/contrib/msc/core/codegen/base_codegen.h | 41 +-
src/contrib/msc/core/codegen/code_stack.cc | 140 +--
src/contrib/msc/core/codegen/code_stack.h | 590 ++++++-----
src/contrib/msc/core/codegen/codegen_json.cc | 6 +-
src/contrib/msc/core/codegen/codegen_json.h | 10 +-
src/contrib/msc/core/codegen/codegen_utils.cc | 28 +-
src/contrib/msc/core/codegen/codegen_utils.h | 118 ++-
src/contrib/msc/core/codegen/cpp_codegen.h | 17 +-
src/contrib/msc/core/codegen/py_codegen.h | 9 +-
src/contrib/msc/core/ir/graph.cc | 401 +++----
src/contrib/msc/core/ir/graph.h | 209 ++--
src/contrib/msc/core/ir/graph_builder.cc | 171 +--
src/contrib/msc/core/ir/graph_builder.h | 89 +-
src/contrib/msc/core/ir/plugin.cc | 67 +-
src/contrib/msc/core/ir/plugin.h | 89 +-
src/contrib/msc/core/printer/cpp_printer.cc | 2 +-
src/contrib/msc/core/printer/cpp_printer.h | 2 +-
src/contrib/msc/core/printer/msc_base_printer.h | 4 +-
src/contrib/msc/core/printer/msc_doc.cc | 29 +-
src/contrib/msc/core/printer/msc_doc.h | 49 +-
src/contrib/msc/core/printer/print_utils.cc | 22 +-
src/contrib/msc/core/printer/print_utils.h | 58 +-
src/contrib/msc/core/printer/prototxt_printer.cc | 20 +-
src/contrib/msc/core/printer/prototxt_printer.h | 10 +-
src/contrib/msc/core/printer/python_printer.cc | 4 +-
src/contrib/msc/core/printer/python_printer.h | 4 +-
.../msc/core/transform/bind_named_params.cc | 33 +-
src/contrib/msc/core/transform/bind_shape.cc | 19 +-
src/contrib/msc/core/transform/fuse_tuple.cc | 44 +-
src/contrib/msc/core/transform/inline_params.cc | 31 +-
src/contrib/msc/core/transform/layout_utils.cc | 18 +-
src/contrib/msc/core/transform/layout_utils.h | 4 +-
src/contrib/msc/core/transform/rewrite_utils.cc | 10 +-
src/contrib/msc/core/transform/rewrite_utils.h | 8 +-
src/contrib/msc/core/transform/set_byoc_attrs.cc | 23 +-
src/contrib/msc/core/transform/set_expr_layout.cc | 268 ++---
src/contrib/msc/core/transform/set_expr_name.cc | 99 +-
src/contrib/msc/core/utils.cc | 136 +--
src/contrib/msc/core/utils.h | 111 +-
src/contrib/msc/framework/tensorflow/codegen.cc | 8 +-
src/contrib/msc/framework/tensorflow/codegen.h | 4 +-
.../msc/framework/tensorflow/tf_v1_opcode.cc | 49 +-
.../msc/framework/tensorflow/tf_v1_opcode.h | 15 +-
src/contrib/msc/framework/tensorrt/codegen.cc | 63 +-
src/contrib/msc/framework/tensorrt/codegen.h | 22 +-
src/contrib/msc/framework/tensorrt/codegen_utils.h | 12 +-
.../msc/framework/tensorrt/tensorrt_opcode.cc | 109 +-
.../msc/framework/tensorrt/tensorrt_opcode.h | 38 +-
.../msc/framework/tensorrt/transform_tensorrt.cc | 162 +--
src/contrib/msc/framework/torch/codegen.cc | 8 +-
src/contrib/msc/framework/torch/codegen.h | 4 +-
src/contrib/msc/framework/torch/codegen_utils.h | 4 +-
src/contrib/msc/framework/torch/torch_opcode.cc | 36 +-
src/contrib/msc/framework/torch/torch_opcode.h | 25 +-
src/contrib/msc/framework/tvm/codegen.cc | 16 +-
src/contrib/msc/framework/tvm/codegen.h | 6 +-
src/contrib/msc/framework/tvm/relax_opcode.cc | 42 +-
src/contrib/msc/framework/tvm/relax_opcode.h | 13 +-
src/contrib/msc/plugin/base_codegen.h | 84 +-
src/contrib/msc/plugin/tensorrt_codegen.cc | 58 +-
src/contrib/msc/plugin/tensorrt_codegen.h | 8 +-
src/contrib/msc/plugin/torch_codegen.cc | 54 +-
src/contrib/msc/plugin/torch_codegen.h | 14 +-
src/contrib/msc/plugin/tvm_codegen.cc | 74 +-
src/contrib/msc/plugin/tvm_codegen.h | 8 +-
src/ir/analysis.cc | 6 +-
src/ir/apply_pass_to_function.cc | 10 +-
src/ir/attrs.cc | 10 +-
src/ir/diagnostic.cc | 18 +-
src/ir/env_func.cc | 4 +-
src/ir/expr.cc | 18 +-
src/ir/function.cc | 30 +-
src/ir/global_info.cc | 4 +-
src/ir/global_var_supply.cc | 18 +-
src/ir/instrument.cc | 20 +-
src/ir/module.cc | 56 +-
src/ir/name_supply.cc | 21 +-
src/ir/op.cc | 38 +-
src/ir/replace_global_vars.cc | 18 +-
src/ir/source_map.cc | 38 +-
src/ir/transform.cc | 63 +-
src/ir/type.cc | 24 +-
src/ir/type_functor.cc | 14 +-
src/meta_schedule/arg_info.cc | 32 +-
src/meta_schedule/builder/builder.cc | 22 +-
src/meta_schedule/cost_model/cost_model.cc | 19 +-
src/meta_schedule/database/database.cc | 78 +-
src/meta_schedule/database/database_utils.cc | 18 +-
src/meta_schedule/database/json_database.cc | 34 +-
src/meta_schedule/database/memory_database.cc | 14 +-
.../database/ordered_union_database.cc | 16 +-
src/meta_schedule/database/schedule_fn_database.cc | 21 +-
src/meta_schedule/database/union_database.cc | 18 +-
src/meta_schedule/extracted_task.cc | 10 +-
.../feature_extractor/feature_extractor.cc | 6 +-
.../feature_extractor/per_store_feature.cc | 17 +-
.../measure_callback/add_to_database.cc | 12 +-
.../measure_callback/measure_callback.cc | 14 +-
.../measure_callback/remove_build_artifact.cc | 10 +-
.../measure_callback/update_cost_model.cc | 12 +-
src/meta_schedule/module_equality.cc | 12 +-
src/meta_schedule/module_equality.h | 2 +-
.../mutator/mutate_compute_location.cc | 14 +-
src/meta_schedule/mutator/mutate_parallel.cc | 30 +-
src/meta_schedule/mutator/mutate_thread_binding.cc | 18 +-
src/meta_schedule/mutator/mutate_tile_size.cc | 24 +-
src/meta_schedule/mutator/mutate_unroll.cc | 17 +-
src/meta_schedule/mutator/mutator.cc | 20 +-
.../postproc/disallow_async_strided_mem_copy.cc | 22 +-
.../postproc/disallow_dynamic_loop.cc | 4 +-
src/meta_schedule/postproc/postproc.cc | 26 +-
.../postproc/rewrite_cooperative_fetch.cc | 50 +-
src/meta_schedule/postproc/rewrite_layout.cc | 32 +-
.../postproc/rewrite_parallel_vectorize_unroll.cc | 25 +-
.../postproc/rewrite_reduction_block.cc | 23 +-
src/meta_schedule/postproc/rewrite_tensorize.cc | 22 +-
.../postproc/rewrite_unbound_block.cc | 16 +-
src/meta_schedule/postproc/verify_gpu_code.cc | 27 +-
src/meta_schedule/postproc/verify_vtcm_limit.cc | 4 +-
src/meta_schedule/profiler.cc | 18 +-
src/meta_schedule/runner/runner.cc | 24 +-
src/meta_schedule/schedule/cpu/winograd.cc | 20 +-
src/meta_schedule/schedule/cuda/thread_bind.cc | 30 +-
src/meta_schedule/schedule/cuda/winograd.cc | 40 +-
src/meta_schedule/schedule/generic/winograd.cc | 4 +-
src/meta_schedule/schedule_rule/add_rfactor.cc | 21 +-
.../schedule_rule/apply_custom_rule.cc | 21 +-
src/meta_schedule/schedule_rule/auto_bind.cc | 15 +-
src/meta_schedule/schedule_rule/auto_inline.cc | 35 +-
.../schedule_rule/cross_thread_reduction.cc | 52 +-
.../schedule_rule/multi_level_tiling.cc | 70 +-
.../schedule_rule/multi_level_tiling.h | 46 +-
.../multi_level_tiling_tensor_core.cc | 159 +--
.../multi_level_tiling_wide_vector.cc | 37 +-
.../multi_level_tiling_with_intrin.cc | 26 +-
.../schedule_rule/parallel_vectorize_unroll.cc | 14 +-
.../schedule_rule/random_compute_location.cc | 10 +-
src/meta_schedule/schedule_rule/schedule_rule.cc | 197 ++--
.../search_strategy/evolutionary_search.cc | 89 +-
src/meta_schedule/search_strategy/replay_func.cc | 37 +-
src/meta_schedule/search_strategy/replay_trace.cc | 47 +-
.../search_strategy/search_strategy.cc | 21 +-
.../space_generator/post_order_apply.cc | 24 +-
src/meta_schedule/space_generator/schedule_fn.cc | 16 +-
.../space_generator/space_generator.cc | 21 +-
.../space_generator/space_generator_union.cc | 22 +-
src/meta_schedule/task_scheduler/gradient_based.cc | 12 +-
src/meta_schedule/task_scheduler/round_robin.cc | 2 +-
src/meta_schedule/task_scheduler/task_scheduler.cc | 48 +-
src/meta_schedule/trace_apply.cc | 17 +-
src/meta_schedule/tune_context.cc | 25 +-
src/meta_schedule/utils.h | 70 +-
src/node/attr_registry.h | 20 +-
src/node/reflection.cc | 4 +-
src/node/script_printer.cc | 43 +-
src/node/structural_hash.cc | 8 +-
src/relax/analysis/analysis.cc | 48 +-
src/relax/analysis/collect_call_map.cc | 4 +-
src/relax/analysis/computable_at_compile_time.cc | 6 +-
src/relax/analysis/detect_recursion.cc | 10 +-
src/relax/analysis/graph_partitioner.cc | 10 +-
src/relax/analysis/graph_partitioner.h | 4 +-
src/relax/analysis/layout_transformation.cc | 62 +-
src/relax/analysis/shape_analysis.cc | 2 +-
src/relax/analysis/struct_info_analysis.cc | 157 +--
src/relax/analysis/tir_op_pattern_kind.cc | 39 +-
src/relax/analysis/udchain.cc | 18 +-
src/relax/analysis/var2value.cc | 23 +-
src/relax/analysis/well_formed.cc | 68 +-
src/relax/backend/contrib/clml/codegen.cc | 28 +-
src/relax/backend/contrib/codegen_c/codegen_c.h | 14 +-
.../backend/contrib/codegen_json/codegen_json.h | 30 +-
src/relax/backend/contrib/cublas/codegen.cc | 17 +-
src/relax/backend/contrib/cudnn/codegen.cc | 19 +-
src/relax/backend/contrib/cutlass/codegen.cc | 63 +-
src/relax/backend/contrib/dnnl/codegen.cc | 17 +-
src/relax/backend/contrib/hipblas/codegen.cc | 18 +-
src/relax/backend/contrib/nnapi/codegen.cc | 18 +-
src/relax/backend/contrib/tensorrt/codegen.cc | 26 +-
src/relax/backend/contrib/utils.cc | 8 +-
src/relax/backend/contrib/utils.h | 8 +-
src/relax/backend/pattern_registry.cc | 12 +-
src/relax/backend/pattern_registry.h | 8 +-
src/relax/backend/task_extraction.cc | 11 +-
src/relax/backend/vm/codegen_vm.cc | 48 +-
src/relax/backend/vm/codegen_vm_tir.cc | 101 +-
src/relax/backend/vm/exec_builder.cc | 19 +-
src/relax/backend/vm/lower_runtime_builtin.cc | 14 +-
src/relax/backend/vm/vm_shape_lower.cc | 57 +-
src/relax/distributed/axis_group_graph.cc | 35 +-
src/relax/distributed/global_info.cc | 10 +-
src/relax/distributed/struct_info.cc | 18 +-
.../distributed/transform/legalize_redistribute.cc | 2 +-
src/relax/distributed/transform/lower_distir.cc | 39 +-
.../transform/lower_global_view_to_local_view.cc | 67 +-
.../distributed/transform/propagate_sharding.cc | 71 +-
src/relax/distributed/transform/utils.cc | 6 +-
src/relax/distributed/transform/utils.h | 8 +-
src/relax/ir/binding_rewrite.cc | 18 +-
src/relax/ir/block_builder.cc | 119 +--
src/relax/ir/dataflow_block_rewriter.cc | 48 +-
src/relax/ir/dataflow_expr_rewriter.cc | 141 +--
src/relax/ir/dataflow_matcher.cc | 31 +-
src/relax/ir/dataflow_matcher.h | 10 +-
src/relax/ir/dataflow_pattern.cc | 156 +--
src/relax/ir/dataflow_rewriter.h | 53 +-
src/relax/ir/emit_te.cc | 6 +-
src/relax/ir/emit_te.h | 2 +-
src/relax/ir/expr.cc | 195 ++--
src/relax/ir/expr_functor.cc | 86 +-
src/relax/ir/py_expr_functor.cc | 38 +-
src/relax/ir/struct_info.cc | 65 +-
src/relax/ir/struct_info_functor.cc | 22 +-
src/relax/ir/tir_pattern.cc | 6 +-
src/relax/ir/transform.cc | 21 +-
src/relax/ir/type.cc | 10 +-
src/relax/op/ccl/ccl.cc | 12 +-
src/relax/op/ccl/ccl.h | 2 +-
src/relax/op/distributed/binary.h | 5 +-
src/relax/op/distributed/ccl.cc | 2 +-
src/relax/op/distributed/distributed.cc | 14 +-
src/relax/op/distributed/linear_algebra.cc | 15 +-
src/relax/op/distributed/manipulate.cc | 8 +-
src/relax/op/distributed/nn.cc | 3 +-
src/relax/op/distributed/statistical.cc | 5 +-
src/relax/op/distributed/unary.h | 5 +-
src/relax/op/distributed/utils.cc | 23 +-
src/relax/op/distributed/utils.h | 4 +-
src/relax/op/image/resize.cc | 27 +-
src/relax/op/image/resize.h | 7 +-
src/relax/op/memory/view.cc | 24 +-
src/relax/op/memory/view.h | 3 +-
src/relax/op/nn/attention.cc | 27 +-
src/relax/op/nn/attention.h | 5 +-
src/relax/op/nn/convolution.cc | 148 +--
src/relax/op/nn/convolution.h | 48 +-
src/relax/op/nn/nn.cc | 165 +--
src/relax/op/nn/nn.h | 12 +-
src/relax/op/nn/pooling.cc | 183 ++--
src/relax/op/nn/pooling.h | 16 +-
src/relax/op/op.cc | 105 +-
src/relax/op/op_common.cc | 38 +-
src/relax/op/op_common.h | 63 +-
src/relax/op/tensor/binary.cc | 20 +-
src/relax/op/tensor/create.cc | 41 +-
src/relax/op/tensor/create.h | 11 +-
src/relax/op/tensor/datatype.cc | 8 +-
src/relax/op/tensor/grad.cc | 32 +-
src/relax/op/tensor/grad.h | 22 +-
src/relax/op/tensor/index.cc | 56 +-
src/relax/op/tensor/index.h | 6 +-
src/relax/op/tensor/inspect.cc | 4 +-
src/relax/op/tensor/linear_algebra.cc | 32 +-
src/relax/op/tensor/linear_algebra.h | 4 +-
src/relax/op/tensor/manipulate.cc | 272 ++---
src/relax/op/tensor/manipulate.h | 30 +-
src/relax/op/tensor/qdq.cc | 12 +-
src/relax/op/tensor/sampling.cc | 2 +-
src/relax/op/tensor/search.cc | 25 +-
src/relax/op/tensor/search.h | 4 +-
src/relax/op/tensor/set.cc | 6 +-
src/relax/op/tensor/set.h | 2 +-
src/relax/op/tensor/sorting.cc | 12 +-
src/relax/op/tensor/sorting.h | 2 +-
src/relax/op/tensor/statistical.cc | 29 +-
src/relax/op/tensor/statistical.h | 26 +-
src/relax/op/tensor/ternary.cc | 10 +-
src/relax/training/utils.cc | 32 +-
src/relax/training/utils.h | 4 +-
src/relax/transform/adjust_matmul_order.cc | 12 +-
src/relax/transform/allocate_workspace.cc | 16 +-
src/relax/transform/alter_op_impl.cc | 128 +--
.../transform/attach_attr_layout_free_buffers.cc | 6 +-
src/relax/transform/attach_global_symbol.cc | 13 +-
src/relax/transform/bind_params.cc | 35 +-
src/relax/transform/bind_symbolic_vars.cc | 28 +-
src/relax/transform/bundle_model_params.cc | 18 +-
src/relax/transform/call_tir_rewrite.cc | 6 +-
src/relax/transform/canonicalize_bindings.cc | 58 +-
src/relax/transform/combine_parallel_matmul.cc | 26 +-
src/relax/transform/convert_dataflow.cc | 10 +-
src/relax/transform/convert_layout.cc | 60 +-
src/relax/transform/dataflow_inplace.cc | 106 +-
src/relax/transform/dead_code_elimination.cc | 5 +-
src/relax/transform/decompose_ops.cc | 16 +-
src/relax/transform/eliminate_common_subexpr.cc | 8 +-
src/relax/transform/expand_matmul_of_sum.cc | 4 +-
src/relax/transform/expand_tuple_arguments.cc | 12 +-
src/relax/transform/few_shot_tuning.cc | 25 +-
src/relax/transform/fold_constant.cc | 54 +-
src/relax/transform/fuse_ops.cc | 168 +--
src/relax/transform/fuse_tir.cc | 208 ++--
src/relax/transform/gradient.cc | 72 +-
src/relax/transform/gradient_simplifier.cc | 4 +-
src/relax/transform/infer_amp_utils.cc | 14 +-
src/relax/transform/infer_amp_utils.h | 8 +-
src/relax/transform/infer_layout_utils.cc | 7 +-
src/relax/transform/infer_layout_utils.h | 23 +-
src/relax/transform/inline_functions.cc | 21 +-
src/relax/transform/kill_after_last_use.cc | 13 +-
src/relax/transform/lambda_lift.cc | 102 +-
src/relax/transform/lazy_transform_params.cc | 35 +-
src/relax/transform/legalize_ops.cc | 15 +-
src/relax/transform/lift_transform_params.cc | 167 +--
src/relax/transform/lower_alloc_tensor.cc | 4 +-
src/relax/transform/merge_composite_functions.cc | 30 +-
src/relax/transform/meta_schedule.cc | 35 +-
src/relax/transform/normalize.cc | 26 +-
src/relax/transform/realize_vdevice.cc | 32 +-
src/relax/transform/remove_purity_checking.cc | 6 +-
src/relax/transform/remove_unused_outputs.cc | 18 +-
src/relax/transform/remove_unused_parameters.cc | 14 +-
.../transform/reorder_permute_dims_after_concat.cc | 21 +-
src/relax/transform/reorder_take_after_matmul.cc | 6 +-
src/relax/transform/replace_global_vars.cc | 12 +-
src/relax/transform/rewrite_cuda_graph.cc | 66 +-
src/relax/transform/rewrite_dataflow_reshape.cc | 12 +-
src/relax/transform/run_codegen.cc | 47 +-
src/relax/transform/split_call_tir_by_pattern.cc | 136 +--
.../transform/split_layout_rewrite_preproc.cc | 36 +-
src/relax/transform/static_plan_block_memory.cc | 61 +-
src/relax/transform/to_mixed_precision.cc | 49 +-
src/relax/transform/topological_sort.cc | 13 +-
src/relax/transform/update_param_struct_info.cc | 8 +-
src/relax/transform/update_vdevice.cc | 4 +-
src/relax/transform/utils.cc | 6 +-
src/relax/transform/utils.h | 43 +-
src/relax/utils.cc | 29 +-
src/runtime/const_loader_module.cc | 16 +-
src/runtime/contrib/arm_compute_lib/acl_runtime.cc | 10 +-
src/runtime/contrib/bnns/bnns_json_runtime.cc | 10 +-
src/runtime/contrib/clml/clml_runtime.cc | 18 +-
src/runtime/contrib/coreml/coreml_runtime.h | 2 +-
src/runtime/contrib/coreml/coreml_runtime.mm | 8 +-
src/runtime/contrib/cublas/cublas_json_runtime.cc | 12 +-
.../contrib/cudnn/cudnn_frontend/attention.h | 2 +-
src/runtime/contrib/cudnn/cudnn_json_runtime.cc | 10 +-
src/runtime/contrib/dnnl/dnnl_json_runtime.cc | 12 +-
src/runtime/contrib/edgetpu/edgetpu_runtime.cc | 2 +-
.../contrib/hipblas/hipblas_json_runtime.cc | 12 +-
src/runtime/contrib/json/json_runtime.h | 24 +-
src/runtime/contrib/mrvl/mrvl_hw_runtime.cc | 22 +-
src/runtime/contrib/mrvl/mrvl_runtime.cc | 18 +-
src/runtime/contrib/mrvl/mrvl_sw_runtime_lib.cc | 2 +-
src/runtime/contrib/msc/tensorrt_runtime.cc | 28 +-
src/runtime/contrib/nnapi/nnapi_runtime.cc | 12 +-
src/runtime/contrib/nvshmem/init.cc | 2 +-
src/runtime/contrib/nvshmem/memory_allocator.cc | 2 +-
src/runtime/contrib/papi/papi.cc | 28 +-
src/runtime/contrib/tensorrt/tensorrt_runtime.cc | 10 +-
src/runtime/contrib/tflite/tflite_runtime.cc | 4 +-
src/runtime/contrib/tflite/tflite_runtime.h | 2 +-
src/runtime/contrib/vllm/cache_alloc.cc | 6 +-
src/runtime/contrib/vllm/cache_kernels.cu | 2 +-
src/runtime/cuda/cuda_device_api.cc | 4 +-
src/runtime/cuda/cuda_module.cc | 12 +-
src/runtime/device_api.cc | 4 +-
src/runtime/disco/bcast_session.cc | 10 +-
src/runtime/disco/bcast_session.h | 2 +-
src/runtime/disco/builtin.cc | 16 +-
src/runtime/disco/cuda_ipc/cuda_ipc_memory.cc | 2 +-
src/runtime/disco/distributed/socket_session.cc | 14 +-
src/runtime/disco/loader.cc | 25 +-
src/runtime/disco/nccl/nccl.cc | 8 +-
src/runtime/disco/process_session.cc | 6 +-
src/runtime/disco/protocol.h | 8 +-
src/runtime/disco/threaded_session.cc | 2 +-
src/runtime/disco/utils.h | 2 +-
src/runtime/file_utils.cc | 16 +-
src/runtime/file_utils.h | 10 +-
src/runtime/hexagon/hexagon_buffer.cc | 6 +-
src/runtime/hexagon/hexagon_buffer.h | 6 +-
src/runtime/hexagon/hexagon_common.cc | 4 +-
src/runtime/hexagon/hexagon_device_api.cc | 6 +-
src/runtime/hexagon/hexagon_device_api.h | 2 +-
src/runtime/hexagon/hexagon_module.cc | 8 +-
src/runtime/hexagon/hexagon_module.h | 14 +-
src/runtime/hexagon/hexagon_thread_manager.cc | 4 +-
src/runtime/memory/memory_manager.cc | 14 +-
src/runtime/memory/naive_allocator.h | 2 +-
src/runtime/meta_data.h | 2 +-
src/runtime/metal/metal_device_api.mm | 2 +-
src/runtime/metal/metal_module.mm | 36 +-
src/runtime/module.cc | 2 +-
src/runtime/opencl/opencl_common.h | 26 +-
src/runtime/opencl/opencl_device_api.cc | 27 +-
src/runtime/opencl/opencl_module.cc | 12 +-
src/runtime/opencl/opencl_module_spirv.cc | 9 +-
src/runtime/profiling.cc | 159 +--
src/runtime/rocm/rocm_device_api.cc | 3 +-
src/runtime/rocm/rocm_module.cc | 10 +-
src/runtime/rpc/rpc_device_api.cc | 2 +-
src/runtime/rpc/rpc_endpoint.cc | 9 +-
src/runtime/rpc/rpc_module.cc | 26 +-
src/runtime/rpc/rpc_session.h | 3 +
src/runtime/rpc/rpc_socket_impl.cc | 2 +-
src/runtime/static_library.cc | 18 +-
src/runtime/static_library.h | 2 +-
src/runtime/tensor.cc | 5 +-
src/runtime/thread_pool.cc | 4 +-
src/runtime/vm/attn_backend.cc | 23 +-
src/runtime/vm/attn_backend.h | 14 +-
src/runtime/vm/attn_utils.h | 2 +-
src/runtime/vm/builtin.cc | 28 +-
src/runtime/vm/cuda/cuda_graph_builtin.cc | 10 +-
src/runtime/vm/executable.cc | 24 +-
src/runtime/vm/kv_state.cc | 4 +-
src/runtime/vm/kv_state.h | 11 +-
src/runtime/vm/lm_support.cc | 6 +-
src/runtime/vm/paged_kv_cache.cc | 58 +-
src/runtime/vm/rnn_state.cc | 42 +-
src/runtime/vm/tensor_cache_support.cc | 57 +-
src/runtime/vm/vm.cc | 46 +-
src/runtime/vulkan/vulkan_module.cc | 4 +-
src/runtime/vulkan/vulkan_wrapped_func.cc | 6 +-
src/runtime/vulkan/vulkan_wrapped_func.h | 6 +-
src/script/ir_builder/base.cc | 6 +-
src/script/ir_builder/ir/frame.cc | 2 +-
src/script/ir_builder/ir/ir.cc | 22 +-
src/script/ir_builder/ir/utils.h | 8 +-
src/script/ir_builder/relax/distributed.cc | 5 +-
src/script/ir_builder/relax/frame.cc | 26 +-
src/script/ir_builder/relax/ir.cc | 37 +-
src/script/ir_builder/relax/utils.h | 22 +-
src/script/ir_builder/tir/frame.cc | 14 +-
src/script/ir_builder/tir/ir.cc | 220 ++--
src/script/ir_builder/tir/utils.h | 25 +-
src/script/printer/doc.cc | 219 ++--
src/script/printer/doc_printer/base_doc_printer.cc | 2 +-
src/script/printer/doc_printer/base_doc_printer.h | 4 +-
.../printer/doc_printer/python_doc_printer.cc | 14 +-
src/script/printer/ir/distributed.cc | 2 +-
src/script/printer/ir/ir.cc | 2 +-
src/script/printer/ir/misc.cc | 16 +-
src/script/printer/ir/utils.h | 4 +-
src/script/printer/ir_docsifier.cc | 19 +-
src/script/printer/relax/binding.cc | 14 +-
src/script/printer/relax/call.cc | 77 +-
src/script/printer/relax/distributed.cc | 14 +-
src/script/printer/relax/expr.cc | 8 +-
src/script/printer/relax/function.cc | 22 +-
src/script/printer/relax/region.cc | 25 +-
src/script/printer/relax/struct_info.cc | 30 +-
src/script/printer/relax/tir.cc | 8 +-
src/script/printer/relax/type.cc | 6 +-
src/script/printer/relax/utils.h | 15 +-
src/script/printer/tir/block.cc | 22 +-
src/script/printer/tir/buffer.cc | 64 +-
src/script/printer/tir/expr.cc | 41 +-
src/script/printer/tir/for_loop.cc | 20 +-
src/script/printer/tir/function.cc | 19 +-
src/script/printer/tir/ir.cc | 4 +-
src/script/printer/tir/stmt.cc | 56 +-
src/script/printer/tir/utils.h | 18 +-
src/script/printer/utils.h | 30 +-
src/support/array.h | 54 +-
src/support/ffi_testing.cc | 42 +-
src/support/nd_int_set.h | 6 +-
src/target/build_common.h | 4 +-
src/target/intrin_rule.h | 2 +-
src/target/llvm/codegen_aarch64.cc | 2 +-
src/target/llvm/codegen_amdgpu.cc | 2 +-
src/target/llvm/codegen_arm.cc | 10 +-
src/target/llvm/codegen_cpu.cc | 24 +-
src/target/llvm/codegen_cpu.h | 18 +-
src/target/llvm/codegen_hexagon.cc | 38 +-
src/target/llvm/codegen_llvm.cc | 42 +-
src/target/llvm/codegen_llvm.h | 16 +-
src/target/llvm/codegen_nvptx.cc | 2 +-
src/target/llvm/intrin_rule_hexagon.cc | 6 +-
src/target/llvm/intrin_rule_llvm.cc | 2 +-
src/target/llvm/intrin_rule_llvm.h | 4 +-
src/target/llvm/intrin_rule_nvptx.cc | 2 +-
src/target/llvm/intrin_rule_rocm.cc | 2 +-
src/target/llvm/llvm_instance.cc | 45 +-
src/target/llvm/llvm_instance.h | 6 +-
src/target/llvm/llvm_module.cc | 62 +-
src/target/opt/build_cuda_on.cc | 4 +-
src/target/parsers/aprofile.cc | 47 +-
src/target/parsers/cpu.cc | 12 +-
src/target/parsers/mprofile.cc | 33 +-
src/target/source/codegen_c.cc | 21 +-
src/target/source/codegen_c.h | 15 +-
src/target/source/codegen_c_host.cc | 12 +-
src/target/source/codegen_c_host.h | 7 +-
src/target/source/codegen_cuda.cc | 7 +-
src/target/source/codegen_cuda.h | 4 +-
src/target/source/codegen_metal.cc | 9 +-
src/target/source/codegen_opencl.cc | 14 +-
src/target/source/codegen_source_base.h | 10 +-
src/target/source/codegen_webgpu.cc | 14 +-
src/target/source/intrin_rule_cuda.cc | 2 +-
src/target/source/intrin_rule_metal.cc | 2 +-
src/target/source/intrin_rule_opencl.cc | 3 +-
src/target/source/source_module.cc | 58 +-
src/target/spirv/intrin_rule_spirv.cc | 4 +-
src/target/spirv/spirv_support.cc | 5 +-
src/target/spirv/spirv_utils.cc | 2 +-
src/target/tag.cc | 133 +--
src/target/target.cc | 160 +--
src/target/target_kind.cc | 94 +-
src/target/virtual_device.cc | 5 +-
src/te/operation/compute_op.cc | 36 +-
src/te/operation/create_primfunc.cc | 151 +--
src/te/operation/create_primfunc.h | 12 +-
src/te/operation/extern_op.cc | 31 +-
src/te/operation/graph.cc | 12 +-
src/te/operation/graph.h | 6 +-
src/te/operation/placeholder_op.cc | 16 +-
src/te/operation/scan_op.cc | 33 +-
src/te/tensor.cc | 37 +-
src/tir/analysis/block_access_region_detector.cc | 48 +-
src/tir/analysis/buffer_access_lca_detector.cc | 14 +-
src/tir/analysis/calculate_allocated_memory.cc | 28 +-
src/tir/analysis/control_flow_graph.cc | 113 +-
src/tir/analysis/control_flow_graph.h | 39 +-
src/tir/analysis/deep_equal.cc | 6 +-
src/tir/analysis/estimate_flops.cc | 2 +-
src/tir/analysis/identify_memcpy.cc | 18 +-
src/tir/analysis/is_pure_function.cc | 2 +-
src/tir/analysis/oob_checker.cc | 6 +-
src/tir/analysis/stmt_finding.cc | 6 +-
src/tir/analysis/var_use_def_analysis.cc | 20 +-
src/tir/analysis/var_use_def_analysis.h | 6 +-
src/tir/analysis/verify_gpu_code.cc | 20 +-
src/tir/analysis/verify_memory.cc | 6 +-
src/tir/analysis/verify_ssa.cc | 2 +-
src/tir/analysis/verify_well_formed.cc | 4 +-
src/tir/ir/block_dependence_info.cc | 12 +-
src/tir/ir/block_scope.cc | 26 +-
src/tir/ir/buffer.cc | 81 +-
src/tir/ir/data_layout.cc | 42 +-
src/tir/ir/data_type_rewriter.cc | 108 +-
src/tir/ir/expr.cc | 147 +--
src/tir/ir/expr_functor.cc | 40 +-
src/tir/ir/function.cc | 24 +-
src/tir/ir/functor_common.h | 4 +-
src/tir/ir/index_map.cc | 116 +-
src/tir/ir/py_functor.cc | 10 +-
src/tir/ir/script/script_complete.cc | 15 +-
src/tir/ir/script/script_complete.h | 2 +-
src/tir/ir/specialize.cc | 39 +-
src/tir/ir/stmt.cc | 158 +--
src/tir/ir/stmt_functor.cc | 110 +-
src/tir/ir/tir_visitor_with_path.cc | 4 +-
src/tir/ir/tir_visitor_with_path.h | 6 +-
src/tir/ir/transform.cc | 7 +-
src/tir/op/builtin.cc | 8 +-
src/tir/op/op.cc | 12 +-
src/tir/schedule/analysis.h | 124 ++-
src/tir/schedule/analysis/analysis.cc | 304 +++---
src/tir/schedule/analysis/layout.cc | 45 +-
src/tir/schedule/analysis/reducer.cc | 88 +-
src/tir/schedule/analysis/verify.cc | 26 +-
src/tir/schedule/concrete_schedule.cc | 239 +++--
src/tir/schedule/concrete_schedule.h | 156 +--
src/tir/schedule/error.cc | 6 +-
src/tir/schedule/error.h | 14 +-
src/tir/schedule/instruction.cc | 36 +-
src/tir/schedule/instruction_traits.h | 124 +--
src/tir/schedule/ir_comparator.cc | 34 +-
src/tir/schedule/ir_comparator.h | 15 +-
src/tir/schedule/primitive.h | 88 +-
src/tir/schedule/primitive/annotate.cc | 38 +-
.../schedule/primitive/annotate_buffer_access.cc | 39 +-
src/tir/schedule/primitive/block_annotate.cc | 79 +-
src/tir/schedule/primitive/blockize_tensorize.cc | 218 ++--
src/tir/schedule/primitive/cache_index.cc | 67 +-
src/tir/schedule/primitive/cache_read_write.cc | 395 +++----
src/tir/schedule/primitive/compute_at.cc | 104 +-
src/tir/schedule/primitive/compute_inline.cc | 114 +-
src/tir/schedule/primitive/decompose_padding.cc | 63 +-
src/tir/schedule/primitive/for_kind.cc | 31 +-
src/tir/schedule/primitive/get_block_loop.cc | 50 +-
src/tir/schedule/primitive/hide_buffer_access.cc | 37 +-
.../schedule/primitive/layout_transformation.cc | 291 ++---
src/tir/schedule/primitive/loop_transformation.cc | 305 +++---
src/tir/schedule/primitive/pad_einsum.cc | 145 +--
src/tir/schedule/primitive/read_write_at.cc | 99 +-
src/tir/schedule/primitive/reduction.cc | 283 ++---
.../schedule/primitive/reorder_block_iter_var.cc | 33 +-
src/tir/schedule/primitive/rolling_buffer.cc | 80 +-
src/tir/schedule/primitive/sampling.cc | 57 +-
src/tir/schedule/schedule.cc | 18 +-
src/tir/schedule/state.cc | 118 ++-
src/tir/schedule/trace.cc | 159 +--
src/tir/schedule/traced_schedule.cc | 150 +--
src/tir/schedule/traced_schedule.h | 108 +-
src/tir/schedule/transform.cc | 174 +--
src/tir/schedule/transform.h | 45 +-
src/tir/schedule/utils.h | 54 +-
src/tir/transforms/annotate_device_regions.cc | 4 +-
src/tir/transforms/arg_binder.cc | 8 +-
src/tir/transforms/arg_binder.h | 6 +-
src/tir/transforms/bind_params.cc | 16 +-
src/tir/transforms/bind_target.cc | 20 +-
src/tir/transforms/bound_checker.cc | 29 +-
src/tir/transforms/common_subexpr_elim.cc | 18 +-
src/tir/transforms/common_subexpr_elim_tools.cc | 14 +-
src/tir/transforms/common_subexpr_elim_tools.h | 8 +-
src/tir/transforms/compact_buffer_region.cc | 41 +-
src/tir/transforms/convert_blocks_to_opaque.cc | 6 +-
src/tir/transforms/default_gpu_schedule.cc | 22 +-
src/tir/transforms/extract_constants.cc | 10 +-
src/tir/transforms/flatten_buffer.cc | 25 +-
src/tir/transforms/force_narrow_index_to_i32.cc | 2 +-
src/tir/transforms/hoist_expression.cc | 6 +-
src/tir/transforms/inject_double_buffer.cc | 4 +-
src/tir/transforms/inject_permuted_layout.cc | 9 +-
src/tir/transforms/inject_ptx_async_copy.cc | 4 +-
src/tir/transforms/inject_ptx_ldg32.cc | 4 +-
src/tir/transforms/inject_rolling_buffer.cc | 26 +-
src/tir/transforms/inject_software_pipeline.cc | 115 +-
src/tir/transforms/inject_virtual_thread.cc | 34 +-
src/tir/transforms/inline_private_functions.cc | 19 +-
src/tir/transforms/ir_utils.cc | 74 +-
src/tir/transforms/ir_utils.h | 28 +-
src/tir/transforms/lift_thread_binding.cc | 32 +-
src/tir/transforms/loop_partition.cc | 16 +-
src/tir/transforms/lower_async_dma.cc | 5 +-
src/tir/transforms/lower_cross_thread_reduction.cc | 162 +--
src/tir/transforms/lower_custom_datatypes.cc | 4 +-
src/tir/transforms/lower_device_kernel_launch.cc | 32 +-
src/tir/transforms/lower_init_block.cc | 2 +-
src/tir/transforms/lower_intrin.cc | 14 +-
src/tir/transforms/lower_match_buffer.cc | 22 +-
src/tir/transforms/lower_opaque_block.cc | 32 +-
src/tir/transforms/lower_thread_allreduce.cc | 42 +-
src/tir/transforms/lower_tvm_builtin.cc | 20 +-
src/tir/transforms/lower_vtcm_alloc.cc | 2 +-
src/tir/transforms/lower_warp_memory.cc | 6 +-
src/tir/transforms/make_packed_api.cc | 27 +-
src/tir/transforms/make_unpacked_api.cc | 14 +-
.../manifest_shared_memory_local_stage.cc | 35 +-
src/tir/transforms/memhammer_coalesce.cc | 35 +-
src/tir/transforms/memhammer_intermediate_stage.cc | 68 +-
src/tir/transforms/memhammer_lower_auto_copy.cc | 89 +-
src/tir/transforms/memhammer_rewrite_rule.h | 26 +-
src/tir/transforms/memhammer_tensorcore_rewrite.cc | 39 +-
.../transforms/merge_shared_memory_allocations.cc | 10 +-
src/tir/transforms/narrow_datatype.cc | 4 +-
.../plan_update_buffer_allocation_location.cc | 25 +-
src/tir/transforms/primfunc_utils.cc | 8 +-
src/tir/transforms/remap_thread_axis.cc | 6 +-
src/tir/transforms/remove_no_op.cc | 16 +-
.../remove_weight_layout_rewrite_block.cc | 31 +-
src/tir/transforms/renew_defs.cc | 46 +-
src/tir/transforms/replace_global_vars.cc | 8 +-
src/tir/transforms/simplify.cc | 20 +-
src/tir/transforms/split_host_device.cc | 10 +-
src/tir/transforms/storage_access.cc | 4 +-
src/tir/transforms/storage_access.h | 8 +-
src/tir/transforms/storage_rewrite.cc | 37 +-
src/tir/transforms/tensorcore_infer_fragment.cc | 4 +-
src/tir/transforms/thread_storage_sync.cc | 4 +-
src/tir/transforms/transform_mma_buffer_layout.cc | 8 +-
src/tir/transforms/unify_thread_binding.cc | 18 +-
src/tir/transforms/unroll_loop.cc | 8 +-
src/tir/transforms/unsupported_dtype_legalize.cc | 50 +-
src/tir/transforms/update_pointer_storage_scope.cc | 6 +-
src/tir/transforms/update_pointer_storage_scope.h | 2 +-
.../transforms/using_assume_to_reduce_branches.cc | 20 +-
src/tir/transforms/vectorize_loop.cc | 104 +-
src/topi/broadcast.cc | 3 +-
src/topi/einsum.cc | 70 +-
src/topi/elemwise.cc | 4 +-
src/topi/nn.cc | 50 +-
src/topi/reduction.cc | 2 +-
src/topi/transform.cc | 51 +-
src/topi/utils.cc | 6 +-
tests/cpp-runtime/hexagon/hexagon_buffer_tests.cc | 47 +-
.../hexagon/hexagon_device_api_tests.cc | 9 +-
.../cpp-runtime/hexagon/hexagon_user_dma_tests.cc | 5 +-
.../cpp-runtime/hexagon/hexagon_vtcm_pool_tests.cc | 17 +-
tests/cpp-runtime/opencl/opencl_compile_to_bin.cc | 2 +-
tests/cpp-runtime/opencl/texture_copy_test.cc | 4 +-
tests/cpp/data_type_rewriter_test.cc | 16 +-
tests/cpp/expr_test.cc | 2 +-
tests/cpp/ir_functor_test.cc | 12 +-
tests/cpp/nested_msg_test.cc | 8 +-
tests/cpp/object_protocol_test.cc | 7 +-
tests/cpp/target/parsers/aprofile_test.cc | 35 +-
tests/cpp/target/parsers/mprofile_test.cc | 20 +-
tests/cpp/target/virtual_device_test.cc | 16 +-
tests/cpp/target_test.cc | 139 +--
tests/python/relax/test_op_inspect.py | 2 +-
.../python/relax/test_relax_to_pyfunc_converter.py | 866 +++++++++++++++
web/emcc/tvmjs_support.cc | 8 +-
web/emcc/wasm_runtime.cc | 7 +-
web/emcc/webgpu_runtime.cc | 6 +-
902 files changed, 17743 insertions(+), 14573 deletions(-)
create mode 100644 python/tvm/relax/relax_to_pyfunc_converter.py
create mode 100644 tests/python/relax/test_relax_to_pyfunc_converter.py