This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch tvm-simplify-jenkins-pytest-sharding
in repository https://gitbox.apache.org/repos/asf/tvm.git
discard c73279e84b [CI] Simplify Jenkins pytest execution
omit 973ed0d3fe ci: inline automatic pytest parallelism
omit e74e2d1ebd ci: enable automatic pytest parallelism
add 45e1b8233a Refactor Tensor arithmetic dispatch away from tirx.generic
(#19943)
add adf8d6a463 [TIRx] Phase out duplicate Var type_annotation (#19944)
new 09f7de617a [CI] Simplify Jenkins pytest execution
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 (c73279e84b)
\
N -- N -- N refs/heads/tvm-simplify-jenkins-pytest-sharding
(09f7de617a)
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.
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
include/tvm/ir/base_expr.h | 16 +-
include/tvm/ir/type.h | 3 +
include/tvm/relax/nested_msg.h | 2 +-
include/tvm/te/operation.h | 24 +--
include/tvm/te/tensor.h | 4 +-
include/tvm/tirx/buffer.h | 6 +-
include/tvm/tirx/builtin.h | 2 +-
include/tvm/tirx/exec_scope.h | 6 +-
include/tvm/tirx/expr.h | 6 +-
include/tvm/tirx/function.h | 2 +-
include/tvm/tirx/index_map.h | 4 +-
include/tvm/tirx/op.h | 26 +--
include/tvm/tirx/predicate.h | 4 +-
include/tvm/tirx/script/builder/ir.h | 15 +-
include/tvm/tirx/stmt.h | 12 +-
include/tvm/tirx/stmt_functor.h | 67 ++++---
include/tvm/tirx/var.h | 103 +++++++++--
include/tvm/topi/broadcast.h | 42 ++---
include/tvm/topi/contrib/cublas.h | 6 +-
include/tvm/topi/detail/broadcast.h | 16 +-
include/tvm/topi/detail/extern.h | 31 ++--
include/tvm/topi/detail/strided_slice.h | 2 +-
include/tvm/topi/elemwise.h | 51 +++---
include/tvm/topi/nn.h | 55 +++---
include/tvm/topi/nn/bnn.h | 10 +-
include/tvm/topi/nn/dense.h | 6 +-
include/tvm/topi/nn/dilate.h | 2 +-
include/tvm/topi/nn/flatten.h | 2 +-
include/tvm/topi/nn/group_norm.h | 6 +-
include/tvm/topi/nn/instance_norm.h | 8 +-
include/tvm/topi/nn/layer_norm.h | 13 +-
include/tvm/topi/nn/local_response_norm.h | 6 +-
include/tvm/topi/nn/mapping.h | 10 +-
include/tvm/topi/nn/pooling.h | 42 +++--
include/tvm/topi/nn/rms_norm.h | 8 +-
include/tvm/topi/nn/softmax.h | 38 ++--
include/tvm/topi/reduction.h | 30 +--
include/tvm/topi/transform.h | 147 ++++++++-------
include/tvm/topi/vision/reorg.h | 2 +-
python/tvm/backend/cuda/op.py | 14 +-
.../backend/cuda/operator/intrinsics/cp_async.py | 4 +-
.../cuda/operator/tile_primitive/copy/reg.py | 2 +-
.../cuda/operator/tile_primitive/copy_async/tma.py | 4 +-
python/tvm/backend/metal/op.py | 8 +-
python/tvm/ir/expr.py | 17 +-
.../tvm/relax/frontend/nn/llm/_decode_kernels.py | 4 +-
python/tvm/relax/frontend/onnx/onnx_frontend.py | 3 +-
.../tvm/relax/transform/legalize_ops/inspect_op.py | 4 +-
python/tvm/relax/transform/legalize_ops/qdq.py | 4 +-
python/tvm/s_tir/schedule/schedule.py | 4 +-
python/tvm/s_tir/tensor_intrin/cuda.py | 8 +-
python/tvm/te/__init__.py | 2 -
.../_te_tensor_overload.py} | 54 +-----
python/tvm/te/tensor.py | 202 ++++++++++++++++++++-
python/tvm/tirx/__init__.py | 1 -
python/tvm/tirx/buffer.py | 10 +-
python/tvm/tirx/expr.py | 64 +++++--
python/tvm/tirx/generic.py | 145 ---------------
python/tvm/tirx/op.py | 125 +++++++++----
python/tvm/tirx/script/builder/external_kernel.py | 14 +-
python/tvm/tirx/script/builder/ir.py | 14 +-
python/tvm/tirx/script/builder/triton.py | 21 ++-
python/tvm/tirx/script/parser/parser.py | 20 +-
python/tvm/topi/__init__.py | 2 +-
python/tvm/topi/_te_tensor_overload.py | 64 +++++++
python/tvm/topi/generic_op_impl.py | 105 -----------
python/tvm/topi/gpu/scan.py | 56 +++---
python/tvm/topi/gpu/sort.py | 14 +-
python/tvm/topi/scan.py | 7 +-
src/arith/analyzer.cc | 4 +-
src/arith/canonical_simplify.cc | 4 +-
src/arith/detect_linear_equation.cc | 4 +-
src/arith/int_constraints.cc | 4 +-
src/arith/int_set.cc | 10 +-
src/arith/ir_mutator_with_analyzer.cc | 34 ++--
src/arith/ir_visitor_with_analyzer.cc | 8 +-
src/arith/iter_affine_map.cc | 6 +-
src/arith/pattern_match.h | 6 +-
src/arith/rewrite_simplify.cc | 13 +-
src/arith/solve_linear_equation.cc | 6 +-
src/arith/solve_linear_inequality.cc | 20 +-
src/arith/transitive_comparison_analyzer.cc | 9 +-
src/arith/unwrap_vector_expr.cc | 6 +-
src/backend/cuda/codegen/codegen_cuda.cc | 65 +++----
src/backend/cuda/codegen/codegen_cuda.h | 2 +-
src/backend/cuda/codegen/llvm/codegen_nvptx.cc | 12 +-
.../hexagon/codegen/llvm/codegen_hexagon.cc | 37 ++--
src/backend/metal/codegen/codegen_metal.cc | 30 ++-
src/backend/opencl/codegen/codegen_opencl.cc | 30 ++-
src/backend/rocm/codegen/llvm/codegen_amdgpu.cc | 7 +-
src/backend/trn/codegen/codegen_trn.cc | 5 +-
src/backend/vulkan/codegen/codegen_spirv.cc | 24 +--
src/backend/vulkan/codegen/codegen_spirv.h | 2 +-
src/backend/vulkan/codegen/ir_builder.cc | 157 +++++++++-------
src/backend/vulkan/codegen/ir_builder.h | 2 +-
src/backend/webgpu/codegen/codegen_webgpu.cc | 62 ++++---
src/backend/webgpu/codegen/codegen_webgpu.h | 3 +-
src/ir/module.cc | 2 +-
src/ir/type.cc | 13 +-
src/relax/analysis/layout_transformation.cc | 10 +-
src/relax/analysis/tir_op_pattern_kind.cc | 6 +-
.../backend/adreno/annotate_custom_storage.cc | 8 +-
.../backend/adreno/fold_vdevice_scope_change.cc | 2 +-
src/relax/backend/contrib/utils.cc | 2 +-
src/relax/backend/contrib/utils.h | 2 +-
src/relax/backend/vm/codegen_vm.cc | 10 +-
src/relax/backend/vm/codegen_vm_tir.cc | 127 +++++++------
src/relax/backend/vm/lower_runtime_builtin.cc | 24 +--
src/relax/backend/vm/vm_shape_lower.cc | 6 +-
src/relax/ir/binding_rewrite.cc | 33 +++-
src/relax/ir/block_builder.cc | 2 +-
src/relax/ir/expr.cc | 2 +-
src/relax/op/tensor/inspect.cc | 30 +--
src/relax/script/builder/ir.cc | 5 +-
src/relax/script/printer/tir.cc | 2 +-
src/relax/transform/adjust_matmul_order.cc | 8 +-
src/relax/transform/alter_op_impl.cc | 6 +-
.../transform/attach_attr_layout_free_buffers.cc | 2 +-
src/relax/transform/call_tir_rewrite.cc | 6 +-
src/relax/transform/canonicalize_bindings.cc | 4 +-
src/relax/transform/compute_prim_value.cc | 2 +-
src/relax/transform/convert_layout.cc | 6 +-
src/relax/transform/dataflow_inplace.cc | 14 +-
src/relax/transform/decompose_ops.cc | 10 +-
src/relax/transform/fuse_ops.cc | 25 ++-
src/relax/transform/fuse_tir.cc | 33 ++--
src/relax/transform/gradient.cc | 16 +-
src/relax/transform/gradient_simplifier.cc | 4 +-
src/relax/transform/lambda_lift.cc | 2 +-
src/relax/transform/lazy_transform_params.cc | 2 +-
src/relax/transform/legalize_ops.cc | 10 +-
src/relax/transform/lift_transform_params.cc | 9 +-
src/relax/transform/realize_vdevice.cc | 6 +-
src/relax/transform/remove_purity_checking.cc | 6 +-
src/relax/transform/remove_unused_parameters.cc | 2 +-
src/relax/transform/rewrite_cuda_graph.cc | 8 +-
src/relax/transform/rewrite_dataflow_reshape.cc | 2 +-
.../specialize_primfunc_based_on_callsite.cc | 6 +-
src/relax/transform/split_call_tir_by_pattern.cc | 8 +-
.../transform/split_layout_rewrite_preproc.cc | 4 +-
src/relax/transform/static_plan_block_memory.cc | 19 +-
src/s_tir/analysis/calculate_allocated_memory.cc | 2 +-
src/s_tir/analysis/estimate_flops.cc | 6 +-
.../analysis/sblock_access_region_detector.cc | 4 +-
src/s_tir/backend/adreno/inject_texture_alloc.cc | 12 +-
src/s_tir/backend/adreno/texture_flatten.cc | 12 +-
src/s_tir/data_layout.cc | 20 +-
.../feature_extractor/per_store_feature.cc | 9 +-
.../multi_level_tiling_tensor_core.cc | 19 +-
src/s_tir/schedule/analysis/analysis.cc | 11 +-
src/s_tir/schedule/analysis/layout.cc | 8 +-
src/s_tir/schedule/analysis/reducer.cc | 40 ++--
src/s_tir/schedule/concrete_schedule.h | 4 +-
src/s_tir/schedule/instruction.cc | 4 +-
src/s_tir/schedule/ir_comparator.cc | 36 +++-
src/s_tir/schedule/ir_comparator.h | 1 +
src/s_tir/schedule/primitive/blockize_tensorize.cc | 8 +-
src/s_tir/schedule/primitive/cache_index.cc | 11 +-
src/s_tir/schedule/primitive/cache_read_write.cc | 26 +--
src/s_tir/schedule/primitive/compute_at.cc | 7 +-
src/s_tir/schedule/primitive/compute_inline.cc | 6 +-
src/s_tir/schedule/primitive/decompose_padding.cc | 14 +-
src/s_tir/schedule/primitive/for_kind.cc | 8 +-
.../schedule/primitive/layout_transformation.cc | 50 ++---
.../schedule/primitive/loop_transformation.cc | 59 +++---
src/s_tir/schedule/primitive/pad_einsum.cc | 12 +-
src/s_tir/schedule/primitive/read_write_at.cc | 15 +-
src/s_tir/schedule/primitive/reduction.cc | 89 ++++++---
src/s_tir/schedule/primitive/rolling_buffer.cc | 4 +-
src/s_tir/schedule/trace.cc | 14 +-
src/s_tir/schedule/transform.cc | 16 +-
src/s_tir/schedule/utils.h | 2 +-
src/s_tir/transform/canonicalize_loop.cc | 2 +-
src/s_tir/transform/compact_buffer_region.cc | 6 +-
src/s_tir/transform/default_gpu_schedule.cc | 12 +-
src/s_tir/transform/hoist_expression.cc | 4 +-
src/s_tir/transform/inject_double_buffer.cc | 8 +-
src/s_tir/transform/inject_permuted_layout.cc | 17 +-
src/s_tir/transform/inject_ptx_async_copy.cc | 8 +-
src/s_tir/transform/inject_software_pipeline.cc | 32 ++--
src/s_tir/transform/inject_virtual_thread.cc | 38 ++--
src/s_tir/transform/lift_thread_binding.cc | 9 +-
src/s_tir/transform/loop_partition.cc | 38 ++--
src/s_tir/transform/lower_async_dma.cc | 21 +--
.../transform/lower_cross_thread_reduction.cc | 14 +-
src/s_tir/transform/lower_match_buffer.cc | 39 ++--
src/s_tir/transform/lower_opaque_block.cc | 7 +-
src/s_tir/transform/lower_thread_allreduce.cc | 5 +-
src/s_tir/transform/lower_vtcm_alloc.cc | 11 +-
.../manifest_shared_memory_local_stage.cc | 2 +-
src/s_tir/transform/memhammer_coalesce.cc | 37 ++--
.../transform/memhammer_intermediate_stage.cc | 8 +-
src/s_tir/transform/memhammer_lower_auto_copy.cc | 13 +-
.../transform/memhammer_tensorcore_rewrite.cc | 177 +++++++++---------
.../transform/merge_shared_memory_allocations.cc | 35 ++--
src/s_tir/transform/profile_instrumentation.cc | 8 +-
src/s_tir/transform/remove_store_undef.cc | 6 +-
.../remove_weight_layout_rewrite_block.cc | 4 +-
src/s_tir/transform/renew_defs.cc | 45 ++---
src/s_tir/transform/rewrite_unsafe_select.cc | 14 +-
src/s_tir/transform/storage_access.cc | 13 +-
src/s_tir/transform/unify_thread_binding.cc | 29 +--
src/target/build_common.h | 12 +-
src/target/intrin_rule.cc | 20 --
src/target/llvm/codegen_cpu.cc | 114 +++++++-----
src/target/llvm/codegen_cpu.h | 6 +-
src/target/llvm/codegen_llvm.cc | 116 ++++++------
src/target/llvm/codegen_llvm.h | 15 +-
src/target/llvm/intrin_rule_llvm.h | 4 +-
src/target/source/codegen_c.cc | 122 +++++++++----
src/target/source/codegen_c.h | 18 +-
src/target/source/codegen_c_host.cc | 8 +-
src/target/source/codegen_source_base.cc | 8 +-
src/target/source/codegen_source_base.h | 8 +-
src/target/source/source_module.cc | 2 +-
src/te/operation/compute_op.cc | 12 +-
src/te/operation/create_primfunc.cc | 40 ++--
src/te/operation/scan_op.cc | 4 +-
src/te/tensor.cc | 12 +-
src/tirx/analysis/deep_equal.cc | 22 ++-
src/tirx/analysis/verify_memory.cc | 6 +-
src/tirx/analysis/verify_ssa.cc | 6 +-
src/tirx/ir/buffer.cc | 35 ++--
src/tirx/ir/data_type_rewriter.cc | 54 +++---
src/tirx/ir/exec_scope.cc | 10 +-
src/tirx/ir/expr.cc | 30 ++-
src/tirx/ir/function.cc | 20 +-
src/tirx/ir/index_map.cc | 42 +++--
src/tirx/ir/predicate.cc | 7 +-
src/tirx/ir/specialize.cc | 47 +++--
src/tirx/ir/stmt.cc | 20 +-
src/tirx/ir/stmt_functor.cc | 60 +++---
src/tirx/ir/tir_visitor_with_path.cc | 5 +-
src/tirx/ir/tir_visitor_with_path.h | 15 +-
src/tirx/op/op.cc | 91 +++++++---
src/tirx/script/builder/ir.cc | 130 ++++++-------
src/tirx/script/printer/buffer.cc | 16 +-
src/tirx/script/printer/expr.cc | 82 ++++++---
src/tirx/script/printer/function.cc | 6 +-
src/tirx/script/printer/ir.cc | 8 +
src/tirx/script/printer/stmt.cc | 12 +-
src/tirx/script/printer/utils.h | 2 +
src/tirx/transform/bind_target.cc | 2 +-
src/tirx/transform/common_subexpr_elim.cc | 3 +-
src/tirx/transform/inline_private_functions.cc | 8 +-
src/tirx/transform/ir_utils.cc | 27 +--
src/tirx/transform/ir_utils.h | 32 ++--
src/tirx/transform/lower_intrin.cc | 74 ++++++--
src/tirx/transform/lower_tirx_dedup_tensormap.cc | 37 ++--
src/tirx/transform/lower_tirx_opaque.cc | 7 +-
src/tirx/transform/lower_tvm_builtin.cc | 146 ++++++++-------
src/tirx/transform/lower_warp_memory.cc | 12 +-
src/tirx/transform/make_packed_api.cc | 53 +++---
src/tirx/transform/remove_no_op.cc | 8 +
src/tirx/transform/split_host_device.cc | 43 +++--
src/tirx/transform/stmt_simplify.cc | 10 +-
src/tirx/transform/storage_rewrite.cc | 121 ++++++++----
src/tirx/transform/tile_primitive_dispatch.cc | 40 ++--
src/tirx/transform/tvm_ffi_binder.cc | 149 ++++++++++-----
src/tirx/transform/tvm_ffi_binder.h | 20 +-
src/tirx/transform/unsupported_dtype_legalize.cc | 74 +++++---
src/tirx/transform/update_pointer_storage_scope.cc | 2 +-
src/tirx/transform/vectorize_loop.cc | 137 +++++++++-----
src/topi/einsum.cc | 38 ++--
tests/cpp/expr_test.cc | 10 +-
tests/cpp/ir_functor_test.cc | 26 +--
tests/cpp/pattern_match_test.cc | 20 +-
tests/cpp/te_compute_test.cc | 11 +-
tests/cpp/tir_analysis_side_effect.cc | 8 +-
tests/cpp/topi_ewise_test.cc | 2 +-
tests/python/ir/test_ir_type.py | 4 +
tests/python/relax/test_analysis_well_formed.py | 2 +-
tests/python/relax/test_binding_rewrite.py | 7 +
...nsform_specialize_primfunc_based_on_callsite.py | 19 +-
.../test_s_tir_transform_inject_virtual_thread.py | 2 +-
.../test_s_tir_transform_lift_thread_binding.py | 4 +-
...s_tir_transform_lower_cross_thread_reduction.py | 38 ++--
...test_s_tir_transform_lower_thread_all_reduce.py | 24 +--
.../test_s_tir_transform_rewrite_unsafe_select.py | 18 ++
.../transform/test_s_tir_transform_thread_sync.py | 4 +-
tests/python/testing/test_tvm_testing_features.py | 41 -----
tests/python/tirx-base/test_tir_buffer.py | 9 +
tests/python/tirx-base/test_tir_nodes.py | 16 +-
tests/python/tirx-base/test_tir_op_types.py | 26 +++
.../test_tir_transform_bf16_legalize.py | 12 +-
.../test_tir_transform_lower_intrin.py | 28 +++
.../test_tir_transform_lower_tvm_builtin.py | 2 +-
.../test_tir_transform_make_packed_api.py | 24 +++
.../test_tir_transform_storage_rewrite.py | 8 +-
.../tvmscript/test_tvmscript_ir_builder_tir.py | 8 +-
.../python/tvmscript/test_tvmscript_parser_tir.py | 14 +-
.../python/tvmscript/test_tvmscript_printer_tir.py | 22 +++
tests/python/tvmscript/test_tvmscript_roundtrip.py | 73 +++++---
tests/python/tvmscript/test_tvmscript_type.py | 2 +-
294 files changed, 3953 insertions(+), 2921 deletions(-)
copy python/tvm/{ir/_tensor_expr_overload.py => te/_te_tensor_overload.py}
(64%)
delete mode 100644 python/tvm/tirx/generic.py
create mode 100644 python/tvm/topi/_te_tensor_overload.py
delete mode 100644 python/tvm/topi/generic_op_impl.py
