colin
/
onnx-mlir
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
onnx-mlir/src/CMakeLists.txt

136 lines
5.0 KiB
CMake
Raw Normal View History

flatten src directory structure
2019-12-23 13:13:52 +08:00
add_library(compiler
dialect/krnl/krnl_ops.cpp
dialect/krnl/krnl_ops.hpp
dialect/krnl/krnl_types.cpp
dialect/krnl/krnl_types.hpp
dialect/onnx/onnx_ops.cpp
dialect/onnx/onnx_ops.hpp
dialect/krnl/krnl_helper.cpp
dialect/krnl/krnl_helper.hpp
pass/shape_inference_interface.hpp
pass/onnx_combine.cpp
Rewrite ReduceL1, ReduceL2, ReduceLogSum, ReduceLogSumExp, ReduceSumSquare in the ONNX dialect (#38) * Rewrite ReduceSumSquare * Edit gen_doc.py * Revise the code * Do shape inference after canonicalization so that there is no need to implement shape inference of rewritten ops * Rewrite ReduceL2 * Add onnx_rewrite.cpp for all rewriting for ONNX ops * Rewrite ReduceL1, ReduceLogSum, ReduceLogSumExp * Edit comments * Change the use of -> to . * Checkout gen_doc.py from the master branch * Use emplace_back instead of push_back * Revise the code * Edit comments Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-01-31 19:00:39 +08:00
pass/onnx_rewrite.cpp
Add a pass to decompose ONNX operations (#9)
2020-03-04 23:53:59 +08:00
pass/onnx_decompose.cpp
Revert "Support attribute promotion." This reverts commit 955968b7504ef1fb226317b81ae644e27e4c64a4.
2020-03-17 17:41:59 +08:00
pass/passes.hpp)
flatten src directory structure
2019-12-23 13:13:52 +08:00
# Include root src directory.
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
target_include_directories(compiler PRIVATE ${ONNX_MLIR_SRC_ROOT})
flatten src directory structure
2019-12-23 13:13:52 +08:00
# Include tablegen generated header files.
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
target_include_directories(compiler PRIVATE ${ONNX_MLIR_BIN_ROOT})
flatten src directory structure
2019-12-23 13:13:52 +08:00
target_link_libraries(compiler
${CMAKE_THREAD_LIBS_INIT}
${CMAKE_DL_LIBS}
${MLIRLibs}
curses)
Add a pass to decompose ONNX operations (#9)
2020-03-04 23:53:59 +08:00
set(LLVM_TARGET_DEFINITIONS pass/onnx_decompose.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(onnx_decompose.inc -gen-rewriters)
Add a pass to decompose ONNX operations (#9)
2020-03-04 23:53:59 +08:00
add_public_tablegen_target(gen_onnx_decompose)
add_dependencies(compiler gen_onnx_decompose)
flatten src directory structure
2019-12-23 13:13:52 +08:00
set(LLVM_TARGET_DEFINITIONS pass/shape_inference_interface.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(shape_inference.hpp.inc -gen-op-interface-decls)
onnx_mlir_tablegen(shape_inference.cpp.inc -gen-op-interface-defs)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_public_tablegen_target(gen_shape_inference)
add_dependencies(compiler gen_shape_inference)
set(LLVM_TARGET_DEFINITIONS pass/onnx_combine.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(onnx_combine.inc -gen-rewriters)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_public_tablegen_target(gen_onnx_combine)
add_dependencies(compiler gen_onnx_combine)
Add result type inference to op definition (#87) * Add result type inference to op definition * Edit MLIR tests * Fix result type for Mul * Format comments * Return UnrankedTensorType as result type * Just for testing -split-input-file * Undo: Just for testing -split-input-file * Extract a function, get_operand_ins, that gets operand types; rewrite gen_attr_ins function * Generate custom builders * Call existing build methods * Add comments * Minor changes * Generate build methods with attributes * Add support of variadic type * Do not generate custom build methods for ops having only attributes Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-02-21 22:28:24 +08:00
set(LLVM_TARGET_DEFINITIONS pass/onnx_rewrite.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(onnx_rewrite.inc -gen-rewriters)
Add result type inference to op definition (#87) * Add result type inference to op definition * Edit MLIR tests * Fix result type for Mul * Format comments * Return UnrankedTensorType as result type * Just for testing -split-input-file * Undo: Just for testing -split-input-file * Extract a function, get_operand_ins, that gets operand types; rewrite gen_attr_ins function * Generate custom builders * Call existing build methods * Add comments * Minor changes * Generate build methods with attributes * Add support of variadic type * Do not generate custom build methods for ops having only attributes Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-02-21 22:28:24 +08:00
add_public_tablegen_target(gen_onnx_rewrite)
add_dependencies(compiler gen_onnx_rewrite)
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
add_subdirectory(interface)
flatten src directory structure
2019-12-23 13:13:52 +08:00
set(LLVM_TARGET_DEFINITIONS dialect/onnx/onnx.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(onnx.hpp.inc -gen-op-decls "-I${CMAKE_SOURCE_DIR}/compiler/pass")
onnx_mlir_tablegen(onnx.cpp.inc -gen-op-defs "-I${CMAKE_SOURCE_DIR}/compiler/pass")
Using Tablegen to Generate Op Documentation (#74) * Add dialect documentation. * Add a step in our CI to ensure documentation is up-to-date. * Add dialect documentation. * Fix config file mistake, using multi-line commands. * Fix a bug in DocCheck.
2020-02-11 03:18:54 +08:00
set(GEN_DOC_FILE ${CMAKE_BINARY_DIR}/docs/Dialects/onnx.md)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_public_tablegen_target(gen_onnx)
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
add_dependencies(gen_onnx gen_shape_inference)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_dependencies(compiler gen_onnx)
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
# TODO: onnx_mlir_gen_promotable_const_operands_op_interface is really a
# dependency of the onnx dialect library, which is currently part of `compiler`.
add_dependencies(compiler onnx_mlir_gen_promotable_const_operands_op_interface)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
add_onnx_mlir_dialect_doc(onnx dialect/onnx/onnx.td)
flatten src directory structure
2019-12-23 13:13:52 +08:00
set(LLVM_TARGET_DEFINITIONS dialect/krnl/krnl_ops.td)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
onnx_mlir_tablegen(krnl.hpp.inc -gen-op-decls)
onnx_mlir_tablegen(krnl.cpp.inc -gen-op-defs)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_public_tablegen_target(gen_krnl_ops)
add_dependencies(compiler gen_krnl_ops)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
add_library(onnx_mlir_onnx_decompose pass/onnx_decompose.cpp)
target_include_directories(onnx_mlir_onnx_decompose
PRIVATE ${ONNX_MLIR_SRC_ROOT} ${ONNX_MLIR_BIN_ROOT}
${ONNX_MLIR_SRC_ROOT})
target_link_libraries(onnx_mlir_onnx_decompose ${MLIRLibs})
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
add_dependencies(onnx_mlir_onnx_decompose gen_onnx)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
add_library(onnx_mlir_shape_inference pass/shape_inference_pass.cpp)
target_include_directories(onnx_mlir_shape_inference
PRIVATE ${ONNX_MLIR_SRC_ROOT} ${ONNX_MLIR_BIN_ROOT}
${ONNX_MLIR_SRC_ROOT})
target_link_libraries(onnx_mlir_shape_inference ${MLIRLibs})
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
add_dependencies(onnx_mlir_shape_inference gen_onnx)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
add_library(onnx_mlir_lower_frontend
[NFC] Change structure of conversion folder. (#96) * Change structure of conversion folder. * Fix comments. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-02-25 23:38:08 +08:00
conversion/onnx_to_krnl/onnx_to_krnl_common.cpp
conversion/onnx_to_krnl/onnx_to_krnl_common.hpp
conversion/onnx_to_krnl/math/elementwise.cpp
conversion/onnx_to_krnl/math/gemm.cpp
conversion/onnx_to_krnl/math/matmul.cpp
conversion/onnx_to_krnl/math/reduction.cpp
conversion/onnx_to_krnl/math/softmax.cpp
conversion/onnx_to_krnl/nn/conv.cpp
conversion/onnx_to_krnl/nn/normalization.cpp
Lower MaxPoolSingleOutOp to Krnl dialect (#1) * Lower MaxPoolSingleOutOp to Krnl dialect * Edit comments * Update changes according to the new folder structure * Add MLIR tests * Support ceil_mode * Merge the first two krnl loops into one krnl loop; remove attribute checks * Dynamically allocate memory for the result if the result has unknown dimensions Co-authored-by: Gheorghe-Teodor Bercea <gt.bercea@gmail.com>
2020-03-05 03:27:21 +08:00
conversion/onnx_to_krnl/nn/pooling.cpp
[NFC] Change structure of conversion folder. (#96) * Change structure of conversion folder. * Fix comments. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-02-25 23:38:08 +08:00
conversion/onnx_to_krnl/tensor/identity.cpp
conversion/onnx_to_krnl/tensor/reshape.cpp
Lower constant padding operation to KRNL dialect (#27)
2020-03-12 04:54:07 +08:00
conversion/onnx_to_krnl/tensor/padconstantvaluepad.cpp
[NFC] Change structure of conversion folder. (#96) * Change structure of conversion folder. * Fix comments. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-02-25 23:38:08 +08:00
conversion/onnx_to_krnl/tensor/transpose.cpp
conversion/onnx_to_krnl/tensor/unsqueeze.cpp
Lower ConstantOp (#28) * Lower ConstantOp * Refactor the code * Edit error messages * Check whether attribute is sparse or dense during shape inference
2020-03-12 22:58:42 +08:00
conversion/onnx_to_krnl/tensor/constant.cpp
[NFC] Change structure of conversion folder. (#96) * Change structure of conversion folder. * Fix comments. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-02-25 23:38:08 +08:00
conversion/onnx_to_krnl/convert_onnx_to_krnl.cpp)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
target_include_directories(onnx_mlir_lower_frontend
PRIVATE ${ONNX_MLIR_SRC_ROOT} ${ONNX_MLIR_BIN_ROOT}
${ONNX_MLIR_SRC_ROOT})
target_link_libraries(onnx_mlir_lower_frontend ${MLIRLibs})
add_dependencies(onnx_mlir_lower_frontend gen_krnl_ops)
flatten src directory structure
2019-12-23 13:13:52 +08:00
add_subdirectory(transform)
add_subdirectory(tool)
add_subdirectory(builder)
add_subdirectory(runtime)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
add_executable(onnx-mlir main.cpp)
[MLIR] Lowering of frontend dialect to KRNL dialect (#382) * Partial support for lowering operations to KRNL dialect. * Attempt to lower to KRNL IR. * Update file. * Add lowering. * Address comments. Fix alloc dynamic dimensions. Correctly link StandardOps. * Temporarily remove deallocation of locally allocated tensors.
2019-11-27 02:55:44 +08:00
Support attribute promotion. (#34) * Support attribute promotion. * Simplify op interface name. * 1. Add more comments to Attribute Promotion Pass. 2. Move Promotable Const Operand Interface to src/interface, and link against it. * Complete NFC change onnx -> onnx-mlir. * Move attribute_promotion pass to src/transform. * Nit: reword comment. * Support Attribute Promotion in gen_doc.py. * Add test. * Update ONNX doc. * Add negative test. * Rename onnxop.inc -> onnx_ops.td.inc. * Include onnx_ops.td.inc. * Nit: better comments. * Prettify CMake. * Remove original attribute_promotion code, improve comments. * Append '_op_interface' to op interface decl/defs. * Namespace cmake targets using onnx_mlir_ prefix. * Use updated header name. * Use new body file name. * Fix dependency. * Use new CMake target name. * Make attribute promotion self-contained by removing redundant constant operaions inside the pass execution. * Remove canonicalization pass. * Increase comments. * Use stricter checks. * Add one more test case. * Remove %arg1 as it's never used.
2020-03-19 15:03:37 +08:00
target_link_libraries(onnx-mlir
builder
${MLIRLibs}
onnx_mlir_transform
onnx_mlir_onnx_decompose
onnx_mlir_shape_inference
onnx_mlir_lower_frontend
onnx_mlir_attribute_promotion)
whole_archive_link_mlir(onnx-mlir
${MLIRWholeArchiveLibs})
Properly link with ZLIB. (#40)
2020-01-22 00:08:16 +08:00
find_package(ZLIB REQUIRED)
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
target_link_libraries(onnx-mlir ${ZLIB_LIBRARIES})
[MLIR] import attribute of onnx node (#383) * add attributes as NamedAttribute * support list value for attribute * use std::tie to avoid c++17 feature
2019-12-21 14:58:23 +08:00
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
target_include_directories(onnx-mlir PRIVATE ${CMAKE_SOURCE_DIR})
target_include_directories(onnx-mlir PRIVATE ${CMAKE_BINARY_DIR})
[MLIR] Krnl Dialect Definition (#357) * a complete, roud-trippable Krnl dialect operation definition * remove old dialect definition files, edit build files * register dialect * check in src for onnf_opt and dimension handler types * re-trigger jenkins * fix build * clarify operation semantics * add verifier for krnl.iterate * refactor to make things clear * do not hard code types * nit and add comments * fix rebase * update op implementation * fix merge * update kernel dialect definition * more comment on how to use the builder for krnl.iterate operation * ammend the comment * can parse krnl.iterate * can parse and print if bounds are not SSA values * address comments * better error handling * Update CMakeLists.txt * update comment * reflow comments
2019-11-12 10:31:56 +08:00
[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
install(TARGETS onnx-mlir DESTINATION bin)