onnx-mlir/test/numerical/TestConv.cpp

168 lines
6.2 KiB
C++
Raw Normal View History

#include <algorithm>
#include <cmath>
#include <iostream>
#include <random>
#include <rapidcheck.h>
#include <string>
#include <vector>
#include "mlir/IR/Module.h"
#include "llvm/Support/FileSystem.h"
#include "src/Dialect/ONNX/ONNXOps.hpp"
#include "src/MainUtils.hpp"
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
#include "src/Runtime/ExecutionSession.hpp"
#include "src/Runtime/OMTensorHelper.h"
#define SHARED_LIB_BASE string("./TestConv_main_graph")
using namespace std;
// Returns whether onnx-mlir compiled convolution is producing the same results
// as a naive implementation of convolution for a specific set of convolution
// parameters/configuration.
bool isOMConvTheSameAsNaiveImplFor(const int N, const int C, const int H,
const int W, const int kH, const int kW, const int pHBegin, const int pHEnd,
const int pWBegin, const int pWEnd) {
MLIRContext ctx;
Update llvm commit ID to 1d01fc1 (#292) * Fix for LLVM revision D85495 * Fix for LLVM revision DD86121 * Fix for LLVM revision D85622 (f9dc2b7) TODO: Change preloadDialectsInContext to false Memo for previous fixes: D86121 (250f43d), D85495 (575b22b) * clang-format * Update llvm commit ID of README and clone-mlir.sh * Updated llvm commit ID of README.md * Fix for passing backend tests * Removed the commented code * Empty commit for triggering rebuild * Test multi-stage travis build * Specify stage order. * Empty commit for triggering rebuild * Update prereq.s390x.Dockerfile Make it possible to execute s390x prereq docker multiple times. * Build prereq for each arch * Fix multi-arch prereq build. * timeout at 40m * Update .travis.yml * add ppc64le prereq builder * Run ppc docker prereq build multiple times * Do not test branch update unless it's mater. * Fix dockerfile. * Fix typo in travis.yml. * Fix ppc64 docker file * Update .travis.yml * turn off metacopy on ppc64le * Update .travis.yml * Turn off metacopy. * Turn off metacopy inside Dockerfile in ppc64. * No sudo in Docker. * Remove metacopy config from Dockerfile. * Change base image to be bionic. * Using newer linux distro for ppc64. * Turn off metacopy in before_install. * Fix sudo permission issue. * Run docker info. * Allow amd64 docker file to be built multiple times * Support building amd64 prereq. * Fix amd64 docker file typo. * fix ppc64le dockerfile typo. * timeout from 40m -> 30m * 40m->30m * 40m->30m * fix bug preventing incremental build. * fix bug preventing incremental build. * Bump CircleCI cache version. * Push to production prereq container repository and condition prereq docker rebuild on commit message. * Rebuild prereq docker. * Move default script to top-level. * Python not properly installed. * amd64 -> x86 * Rebuild prereq docker. * Rebuild prereq docker. * Rebuild prereq docker. * Restart all CI. * Disallow cache on Jenkins docker build. * Restart zJenkins. * Restart zJenkins. Co-authored-by: Haruki Imai <imaihal@jp.ibm.com> Co-authored-by: Alexandre Eichenberger <alexe@us.ibm.com>
2020-09-09 23:12:01 +08:00
registerDialects(ctx);
auto module = ModuleOp::create(UnknownLoc::get(&ctx));
OpBuilder builder(&ctx);
llvm::SmallVector<int64_t, 4> xShape = {N, C, H, W};
llvm::SmallVector<int64_t, 1> bShape = {C};
llvm::SmallVector<int64_t, 4> wShape = {C, C, kH, kW};
auto xType = RankedTensorType::get(xShape, builder.getF32Type());
auto wType = RankedTensorType::get(wShape, builder.getF32Type());
auto yType = UnrankedTensorType::get(builder.getF32Type());
llvm::SmallVector<Type, 2> inputsType{xType, wType};
llvm::SmallVector<Type, 1> outputsType{yType};
auto funcType = builder.getFunctionType(inputsType, outputsType);
Compiling Models with Large Constant Arrays (#146) * PoC works. * MNist works. * Clean up. * Fix test. * Make Linux work. * Use consistent symbol name. * Fix variable name. * Fix array addr access. * Bug fix. * Bug fix. * install before running e2e tests. * Fix build config. * Use sudo when installing. * Make embeddedDataLoader position independent. * Enable ResNet50. * Format code. * Format MainUtil. * Try not using sudo to install. * Supply runtime dir via environment variable. * Dump problematic operation. * Dump entire function. * Debug. * Dump input. * Dump constant op. * Debug. * Debug. * Debug. * Print to stderr. * take care of endianness. * Use endianness-aware execution session. * Fix ZLinux error. * Include warning when desired output endianness can't be deduced. * Remove debug code. * Remove debug code in shape inference. * Support binary-decoder for testing constants packing. * Support filename, move-to-file, elision-threshold configurations in constant packing pass for easy testing. * Add lit test, fix lit test type mismatch. * Add more consts packing tests. * Ensure intermediate files are properly cleaned up. * No need for constant elimination. * Link with threading libraries. * Remove debug code. * Format code. * More tests. * test nit. * Remove debug code. * Reduce hard-coded constants. * Use temporary and unique working directory for hosting model parameters. * Test if it works. * Try to find objcopy. * Rename symbols using objcopy. * Move sanitized name to linux section. * Use verbose mode for debugging. * Disambiguate pass constructor. * Fix symbol name. * Use Command API to build and execute commands. * Move linux to use Command API. * Fix reset args. * Execute redefine sym. * Format code. * Do not use verbose mode for CircleCI. * Remove debug code. * Prettify code, add comments. * getSegmentData -> getEmbeddedConstPool * vector -> std::vector. * Make sure we properly clean up intermediate files. * Fix test cases. * Add runtime directory. * Trigger rebuild. * [Merge with master] fix debug script. * Diable affine fusion pass for now. * Support generic fallback const packing mechanism. * Remove debug code. * Handle the case where objcopy is not available. * Fix Windows missing types. * Support int64. * Copy packed constant to a local directory for non-Linux/Mac platforms. * Nit: remove debug code, refactor const pack preprocessing out as a separate function. * Cannot make preprocessConstPack a standalone function because file removers are stack-allocated, and they are deallocated prematurely when function stack gets popped, deleteing intermediate files too early. * Don't require executable filename. * Import ONNX data types directly. * Fix LIT test. * Bug fix, use moved string value. * Remove redundant filenames. * Fix CMake script. * Embed endianness information as a symbol, and check during runtime. * More comments, update lit tests. * Fix lit test on BE machine. * Copyright notices.
2020-06-12 10:27:05 +08:00
string funcName = "main_graph";
llvm::SmallVector<NamedAttribute, 1> attrs;
auto funcOp =
builder.create<FuncOp>(UnknownLoc::get(&ctx), funcName, funcType, attrs);
auto entryBlock = funcOp.addEntryBlock();
builder.setInsertionPointToStart(entryBlock);
auto xVal = entryBlock->getArgument(0);
auto wVal = entryBlock->getArgument(1);
auto bVal =
builder.create<ConstantOp>(UnknownLoc::get(&ctx), builder.getUnitAttr())
.getResult();
auto dilations = builder.getI64ArrayAttr({1, 1});
auto kernel_shape = builder.getI64ArrayAttr({kH, kW});
auto pads = builder.getI64ArrayAttr({pHBegin, pWBegin, pHEnd, pWEnd});
auto strides = builder.getI64ArrayAttr({1, 1});
auto convOp = builder.create<ONNXConvOp>(UnknownLoc::get(&ctx),
/*Y=*/yType,
/*X=*/xVal, /*W=*/wVal, /*B=*/bVal,
/*auto_pad=*/builder.getStringAttr("NOTSET"),
/*dilations=*/dilations,
/*group=*/
IntegerAttr::get(builder.getIntegerType(64, /*isSigned=*/true),
APInt(64, 1, /*isSigned=*/true)),
/*kernel_shape=*/kernel_shape, /*pads=*/pads,
/*strides=*/strides);
// Use the convOp shape inference method to compute output shape, and unset
// the shape so that we don't leave IR in a inconsistent state.
convOp.inferShapes();
auto outputShape = convOp.getResult().getType().cast<ShapedType>().getShape();
auto NOut = outputShape[0];
auto COut = outputShape[1];
auto HOut = outputShape[2];
auto WOut = outputShape[3];
convOp.getResult().setType(yType);
llvm::SmallVector<Value, 1> results = {convOp.getResult()};
builder.create<ReturnOp>(UnknownLoc::get(&ctx), results);
module.push_back(funcOp);
// Emit the entry point operation which specifies the number of user
// inputs and outputs.
auto entryPoint = ONNXEntryPointOp::create(UnknownLoc::get(&ctx), funcOp,
/*numInputs=*/2,
/*numOutputs=*/1);
module.push_back(entryPoint);
OwningModuleRef moduleRef(module);
compileModule(moduleRef, ctx, SHARED_LIB_BASE, EmitLib);
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
onnx_mlir::ExecutionSession sess(SHARED_LIB_BASE + ".so", "run_main_graph");
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
std::vector<unique_ptr<OMTensor, decltype(&omTensorDestroy)>> inputs;
auto xOmt = unique_ptr<OMTensor, decltype(&omTensorDestroy)>(
omTensorCreateWithRandomData<float>({N, C, H, W}), omTensorDestroy);
inputs.emplace_back(move(xOmt));
auto wOmt = unique_ptr<OMTensor, decltype(&omTensorDestroy)>(
omTensorCreateWithRandomData<float>({C, C, kH, kW}), omTensorDestroy);
inputs.emplace_back(move(wOmt));
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
auto ref = omTensorCreateWithShape<float>({NOut, COut, HOut, WOut});
auto &img = inputs.at(0);
auto &filter = inputs.at(1);
for (int64_t n = 0; n < NOut; n++)
for (int64_t c = 0; c < COut; c++)
for (int64_t h = 0; h < HOut; h++)
for (int64_t w = 0; w < WOut; w++) {
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
omTensorGetElem<float>(ref, {n, c, h, w}) = 0;
for (int64_t ci = 0; ci < C; ci++)
for (int64_t kh = 0; kh < kH; kh++)
for (int64_t kw = 0; kw < kW; kw++)
if ((h + kh - pHBegin >= 0 && h + kh - pHBegin < H) &&
(w + kw - pWBegin >= 0 && w + kw - pWBegin < W))
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
omTensorGetElem<float>(ref, {n, c, h, w}) +=
omTensorGetElem<float>(img.get(),
{n, ci, h + kh - pHBegin, w + kw - pWBegin}) *
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
omTensorGetElem<float>(filter.get(), {c, ci, kh, kw});
}
auto outputs = sess.run(move(inputs));
auto &conv = outputs.at(0);
Cleanup rtmemref api (#238) * Detect llvm-project commit change in utils/clone-mlir.sh and rebuild llvm-project for zLinux Jenkins build bot * Cleanup RtMemRef API - use forward declaration to hide private data fields - RtMemRef.h: external user header, C/C++ - _RtMemRef.h: internal user header, C++ only - RtMemRef.hpp and RtMemRef.cpp: implementation header and file - add external APIs OrderedRtMemRefDict *ormrd_create(RtMemRef **rmrs, int n) RtMemRef **ormrd_getRmrs(OrderedRtMemRefDict *ormrd) int ormrd_getNumOfRmrs(OrderedRtMemRefDict *ormrd) for creating and querying OrderedRtMemRefDict with RtMemRef arrays - data buffer installed by rmr_setData() will be managed by user - unique_ptr<RtMemRef> must use custom deleter <RtMemRef,decltype(&rmr_destroy)> * See if I have write access. * Remove test CMake code. * Use new API. * Format code. * Format code & rename variables for readability. * Remove used API spec. * Rename OrderedRtMemRefDict -> RtMemRefList, _dataMalloc -> _owningData. * OrderedRtMemRefDict -> RtMemRefList * Update KrnlToLLVM.cpp * Trigger Jenkins * Restart Jenkins * OrderedRtMemRefDict -> RtRmrRefList * More OrderedRtMemRefDict -> RtMemRefList. * Format jni wrapper. * Rename API functions to maintain stylistic consistency. * Bug fix. * Bug fix. * Format code. * Fix RtMemRefUtils. * Format code. * Using llvm function naming scheme. * Rename runtime api file name to project name (onnx-mlir) as per convention. * Include the new runtime header file. * Reflect api header file name change in build script. * Bug fix. * Remove C++ code. * Revert "Remove C++ code." This reverts commit b217dfabae99e42db30721600cb5507866d4dc98. * Clarify memory management responsibility. * Add constructor to specify name & data ownership. * Include stdbool. * Remove dictionary semantics from RtMemRefList * Bug fix. * Format code. * Format code. * Use macro to define database of metadata. * Prevent formatter from acting on metadata decl. * Nit. * Restore backend unit tests. * Use spaces instead of tabs for better formatting. * Use explicit template instantiation. * Update RtMemRef struct doc. * Make runtime compilable both in c and c++ mode. Build two versions of the runtime library, one c version as the user-facing c runtime, and one c++ version as the one used inside this project. * Bug fix, avoid stack allocation for output rmr list. * Change _dyn_entry_point_main_graph -> run_main_graph for better memorability. * Write a complete introductory tutorial on c99 Runtime and a test for it. * Add onnx installation as dependency. * Use target_include_directory to avoid installation. * Format code. * Fix cmake target_include_directories. * Address compiler warning. * First pass of RtMemRef->OMTensor. * Second pass of RtMemRef -> OMTensor. * nit, omtList -> omTensorList. * omt -> omTensor for clarity. * Rename OnnxMlirInternal.h -> OnnxMlirRuntime.hpp because there's no internal/external API, only C/C++ API. * Format code. * Restructure Runtime source code and move header -> /include and test -> /test/unit. * Bugfix. * Format code. * Add unit test for OMTensor ctor. * Update JNI CMake include directory. * Bugfix. * No need to re-declare ONNX types. * Disable runtime doc test on non-x86 platforms. * Rename OMTensor fields to be more sensible. * Fix data type mismatch. * size_t -> int64_t, prefer fixed width integers. * Use consistent header guard style. * Further tweak OMTensor API. * Bugfix. * Bugfix. * Format code. * Add doxygen config file. * Tweak OMTensor API. * Tweak API doc, hide OMTensorList implementation. * Format code. * Add new documentation item for Runtime API. * Hide internal use only API declarations, move their comments to their implementations. * Clarify ownership semantics in relevant API documentations. * Fix PyRuntime. * Remove alignment concerns from public API and include explaination of alignment issue in struct OMTensor definition. * Print out unsupported numpy dtype. * Use preferred way of type comparison in pybind11. * Debug s390x issue. * Remove debug code. * Clarify semantics of strides/shape setter/getter, use \brief to include short description of API function. * Improve documentation. * Single out unpolished C++ API declarations. * Clarify OMTensorList API. * Bugfix. * Bugfix. * Assert after malloc. * Handle malloc failures. * Nit. * Tweak legal notices. * Format code. * Remove doxygen generated files. * Tweak legal notice format. * Upgrade Cython Numpy installation depends on Cython. Co-authored-by: Tian Jin <tjingrant@gmail.com>
2020-10-10 22:32:09 +08:00
return omTensorAreTwoOmtsClose<float>(conv.get(), ref);
}
int main(int argc, char *argv[]) {
setExecPath(argv[0], (void *)main);
llvm::FileRemover remover(SHARED_LIB_BASE + ".so");
// RapidCheck test case generation.
rc::check("convolution implementation correctness", []() {
const auto N = *rc::gen::inRange(1, 10);
const auto C = *rc::gen::inRange(1, 20);
const auto H = *rc::gen::inRange(5, 20);
const auto W = *rc::gen::inRange(5, 20);
const auto kH = *rc::gen::inRange(1, 15);
const auto kW = *rc::gen::inRange(1, 15);
// We don't want an entire window of padding.
const auto pHBegin = *rc::gen::inRange(0, kH - 1);
const auto pHEnd = *rc::gen::inRange(0, kH - 1);
const auto pWBegin = *rc::gen::inRange(0, kW - 1);
const auto pWEnd = *rc::gen::inRange(0, kW - 1);
// Make sure we have at least 1 output per dimension.
RC_PRE((H >= kH) && (W > kW));
RC_ASSERT(isOMConvTheSameAsNaiveImplFor(
N, C, H, W, kH, kW, pHBegin, pHEnd, pWBegin, pWEnd));
});
// Exhaustive test case generation.
for (int pHBegin = 0; pHBegin < 3; pHBegin++)
for (int pHEnd = 0; pHEnd < 3; pHEnd++)
for (int pWBegin = 0; pWBegin < 3; pWBegin++)
for (int pWEnd = 0; pWEnd < 3; pWEnd++)
assert(isOMConvTheSameAsNaiveImplFor(
2, 4, 5, 5, 3, 3, pHBegin, pHEnd, pWBegin, pWEnd));
return 0;
}