Merge branch 'master' of https://github.com/microsoft/onnxruntime int…

…o optrf

.gitignore

@@ -14,6 +14,7 @@ cmake-build-debug
 gen
 *~
 .vs
+.DS_Store
 TestResults/
 .idea/
 onnxruntime.egg-info

cgmanifests/cgmanifest.json

@@ -343,7 +343,7 @@
          "component": {
             "type": "git",
             "git": {
-               "commitHash": "f97fd8664fa824e5d081d776cf4a811dbf09bca4",
+               "commitHash": "bc8ce45b35ade9b8f08d88d4d9fdfd12b4f6f310",
                "repositoryUrl": "https://github.com/pypa/manylinux"
             },
             "comments": "For building our CI build docker image"
@@ -363,20 +363,20 @@
          "component": {
             "type": "git",
             "git": {
-               "commitHash": "439c93d51f45c50541fc755b597725168ecd939a",
+               "commitHash": "9cf6752276e6fcfd0c23fdb064ad27f448aaaf75",
                "repositoryUrl": "https://github.com/python/cpython"
             },
-            "comments": "Python 3.9.0rc1"
+            "comments": "Python 3.9.0"
          }
       },
       {
          "component": {
             "type": "git",
             "git": {
-               "commitHash": "580fbb018fd0844806119614d752b41fc69660f9",
+               "commitHash": "db455296be5f792b8c12b7cd7f3962b52e4f44ee",
                "repositoryUrl": "https://github.com/python/cpython"
             },
-            "comments": "Python 3.8.5"
+            "comments": "Python 3.8.6"
          }
       },
       {
@@ -403,10 +403,10 @@
          "component": {
             "type": "git",
             "git": {
-               "commitHash": "e5f6aba872e66bfd86eb592214696a519cded197",
+               "commitHash": "426b022776672fdf3d71ddd98d89af341c88080f",
                "repositoryUrl": "https://github.com/python/cpython"
             },
-            "comments": "Python 3.5.9"
+            "comments": "Python 3.5.10"
          }
       },
       {

cgmanifests/submodules/cgmanifest.json

@@ -1,6 +1,50 @@
 {
   "Version": 1,
   "Registrations": [
+    {
+      "Component": {
+        "Type": "other",
+        "other": {
+          "Name": "automake",
+          "Version": "1.16.2",
+          "DownloadUrl": "http://ftp.gnu.org/gnu/automake/automake-1.16.2.tar.gz"
+        },
+        "comments": "manylinux dependency"
+      }
+    },
+    {
+      "Component": {
+        "Type": "other",
+        "other": {
+          "Name": "libtool",
+          "Version": "2.4.6",
+          "DownloadUrl": "http://ftp.gnu.org/gnu/libtool/libtool-2.4.6.tar.gz"
+        },
+        "comments": "manylinux dependency"
+      }
+    },
+    {
+      "Component": {
+        "Type": "other",
+        "other": {
+          "Name": "sqlite",
+          "Version": "3330000",
+          "DownloadUrl": "https://www.sqlite.org/2020/sqlite-autoconf-3330000.tar.gz"
+        },
+        "comments": "manylinux dependency"
+      }
+    },
+    {
+      "Component": {
+        "Type": "other",
+        "other": {
+          "Name": "git",
+          "Version": "2.29.1",
+          "DownloadUrl": "https://www.kernel.org/pub/software/scm/git/git-2.29.1.tar.gz"
+        },
+        "comments": "manylinux dependency"
+      }
+    },
     {
       "component": {
         "type": "git",

cgmanifests/submodules/generate_submodule_cgmanifest.py

@@ -4,10 +4,54 @@
 import os
 import subprocess
 import sys
+import re
 
 SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__))
 REPO_DIR = os.path.normpath(os.path.join(SCRIPT_DIR, "..", ".."))
 
+package_name = None
+package_filename = None
+package_url = None
+
+registrations = []
+
+with open(os.path.join(REPO_DIR,'tools','ci_build','github','linux','docker','manylinux2014_build_scripts','build_env.sh'), "r") as f:
+  for line in f:
+    if not line.strip():      
+      package_name = None
+      package_filename = None
+      package_url = None
+    if package_filename is None:
+      m = re.match("(.+?)_ROOT=(.*)$",line)
+      if m != None:
+        package_name = m.group(1)
+        package_filename = m.group(2)
+      else:
+        m = re.match("(.+?)_AUTOCONF_VERSION=(.*)$",line)
+        if m != None:
+          package_name = m.group(1)
+          package_filename = m.group(2)        
+    elif package_url is None:
+      m = re.match("(.+?)_DOWNLOAD_URL=(.+)$",line)
+      if m != None:
+        package_url = m.group(2) + "/" + package_filename + ".tar.gz"
+        registration = {
+          "Component": {
+            "Type": "other",
+            "other": {
+                "Name": package_name.lower(),
+                "Version": package_filename.split("-")[-1],
+                "DownloadUrl": package_url,
+            },
+            "comments": "manylinux dependency"
+          }
+        }
+        registrations.append(registration)
+        package_name = None
+        package_filename = None
+        package_url = None
+
+
 proc = subprocess.run(
     ["git", "submodule", "foreach", "--quiet", "--recursive", "{} {} $toplevel/$sm_path".format(
         sys.executable, os.path.join(SCRIPT_DIR, "print_submodule_info.py"))],
@@ -17,7 +61,7 @@
     stderr=subprocess.PIPE,
     universal_newlines=True)
 
-registrations = []
+
 submodule_lines = proc.stdout.splitlines()
 for submodule_line in submodule_lines:
     (absolute_path, url, commit) = submodule_line.split(" ")

cmake/external/pybind11.cmake

@@ -8,7 +8,7 @@ if(NOT TARGET pybind11::module)
 
   set(pybind11_INCLUDE_DIRS ${CMAKE_CURRENT_BINARY_DIR}/pybind11/src/pybind11/include)
   set(pybind11_URL https://github.com/pybind/pybind11.git)
-  set(pybind11_TAG v2.4.0)
+  set(pybind11_TAG v2.6.1)
 
   ExternalProject_Add(pybind11
         PREFIX pybind11
@@ -21,4 +21,4 @@ if(NOT TARGET pybind11::module)
   set(pybind11_dep pybind11)
 else()
   set(pybind11_lib pybind11::module)
-endif()
+endif()

cmake/onnxruntime_mlas.cmake

@@ -124,6 +124,15 @@ if(MSVC)
     )
   endif()
 else()
+  if (CMAKE_OSX_ARCHITECTURES STREQUAL "arm64")
+    set(ARM64 TRUE)
+  elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "arm")
+    set(ARM TRUE)
+  elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "x86_64")
+    set(X86_64 TRUE)
+  elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "i386")
+    set(X86 TRUE)
+  endif()
   if (CMAKE_SYSTEM_NAME STREQUAL "Android")
     if (CMAKE_ANDROID_ARCH_ABI STREQUAL "armeabi-v7a")
       set(ARM TRUE)
@@ -136,15 +145,6 @@ else()
     endif()
   elseif(CMAKE_SYSTEM_NAME STREQUAL "iOS" OR CMAKE_SYSTEM_NAME STREQUAL "iOSCross")
     set(IOS TRUE)
-    if (CMAKE_OSX_ARCHITECTURES STREQUAL "arm64")
-      set(ARM64 TRUE)
-    elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "arm")
-      set(ARM TRUE)
-    elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "x86_64")
-      set(X86_64 TRUE)
-    elseif (CMAKE_OSX_ARCHITECTURES STREQUAL "i386")
-      set(X86 TRUE)
-    endif()
   else()
     execute_process(
       COMMAND ${CMAKE_C_COMPILER} -dumpmachine

csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests/runtest-docker-gpu.sh

@@ -27,6 +27,6 @@ docker run --gpus all --rm \
         -e "PackageName=$PackageName" \
         -e "RunTestCsharp=$RunTestCsharp" \
         -e "RunTestNative=$RunTestNative" \
-        onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntimecentosgpubuild:ch4 \
+        onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntimegpubuild:ch4 \
         /bin/bash /onnxruntime_src/csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests/runtest.sh \
         /home/onnxruntimedev/$NUGET_REPO_DIRNAME /onnxruntime_src /home/onnxruntimedev $CurrentOnnxRuntimeVersion

csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests/runtest-docker.sh

@@ -35,6 +35,6 @@ docker run --rm \
         -e "DisableMlOps=$DISABLEMLOPS" \
         -e "RunTestCsharp=$RunTestCsharp" \
         -e "RunTestNative=$RunTestNative" \
-        onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntimecentoscpubuild:ch3 \
+        onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntimecentoscpubuild:chr \
         /bin/bash /onnxruntime_src/csharp/test/Microsoft.ML.OnnxRuntime.EndToEndTests/runtest.sh \
         /home/onnxruntimedev/$NUGET_REPO_DIRNAME /onnxruntime_src /home/onnxruntimedev $CurrentOnnxRuntimeVersion

include/onnxruntime/core/session/onnxruntime_cxx_api.h

@@ -349,7 +349,7 @@ struct Session : Base<OrtSession> {
   // Run that will allocate the output values
   std::vector<Value> Run(const RunOptions& run_options, const char* const* input_names, const Value* input_values, size_t input_count,
                          const char* const* output_names, size_t output_count);
-  // Run for when there is a list of prealloated outputs
+  // Run for when there is a list of preallocated outputs
   void Run(const RunOptions& run_options, const char* const* input_names, const Value* input_values, size_t input_count,
            const char* const* output_names, Value* output_values, size_t output_count);
 

onnxruntime/core/providers/nnapi/nnapi_builtin/builders/helper.cc

@@ -1,6 +1,5 @@
-//
-// Created by daquexian on 8/3/18.
-//
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
 
 #include <iostream>
 #include <string>

onnxruntime/core/providers/nnapi/nnapi_builtin/builders/helper.h

@@ -1,6 +1,6 @@
-//
-// Created by daquexian on 5/21/18.
-//
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
 #pragma once
 
 #include <string>

onnxruntime/core/providers/nnapi/nnapi_builtin/builders/op_builder.cc

@@ -1767,6 +1767,10 @@ Status ConcatOpBuilder::AddToModelBuilderImpl(ModelBuilder& model_builder, const
 class SqueezeOpBuilder : public BaseOpBuilder {
  public:
   void AddInitializersToSkip(ModelBuilder& model_builder, const Node& node) const override;
+  static Status AddSqueezeOp(ModelBuilder& model_builder,
+                             const std::string& node_name,
+                             const std::string& input, const std::string& output,
+                             vector<int32_t> axes) ORT_MUST_USE_RESULT;
 
  private:
   Status AddToModelBuilderImpl(ModelBuilder& model_builder, const Node& node) const override ORT_MUST_USE_RESULT;
@@ -1779,6 +1783,49 @@ void SqueezeOpBuilder::AddInitializersToSkip(ModelBuilder& model_builder, const
   }
 }
 
+/* static */ Status SqueezeOpBuilder::AddSqueezeOp(ModelBuilder& model_builder,
+                                                   const std::string& node_name,
+                                                   const std::string& input, const std::string& output,
+                                                   vector<int32_t> axes) {
+  auto& shaper(model_builder.GetShaper());
+  const auto& operand_indices(model_builder.GetOperandIndices());
+  const auto& operand_types(model_builder.GetOperandTypes());
+
+  const auto& input_shape(shaper[input]);
+  auto input_dims = input_shape.size();
+  for (auto& axis : axes) {
+    axis = static_cast<int32_t>(HandleNegativeAxis(axis, input_dims));
+  }
+
+  // Despite the spec of ANEURALNETWORKS_SQUEEZE at
+  // https://developer.android.com/ndk/reference/group/neural-networks
+  // states, that the axes (input 1 of ANEURALNETWORKS_SQUEEZE) is optional.
+  //
+  // The actual code of NNAPI requires the axes to be provided
+  // https://android.googlesource.com/platform/frameworks/ml/+/master/nn/common/operations/Squeeze.cpp#31
+  if (axes.empty()) {  // Squeeze all
+    for (size_t i = 0; i < input_dims; i++) {
+      if (input_shape[i] == 1)
+        axes.push_back(i);
+    }
+  }
+
+  const auto axes_name = model_builder.GetUniqueName(node_name + input + "_axes");
+  Shape axes_dimen = {static_cast<uint32_t>(axes.size())};
+  const OperandType axes_operand_type(Type::TENSOR_INT32, axes_dimen);
+  ORT_RETURN_IF_ERROR(model_builder.AddOperandFromPersistMemoryBuffer(axes_name, axes.data(), axes_operand_type));
+
+  std::vector<uint32_t> input_indices;
+  input_indices.push_back(operand_indices.at(input));      // input
+  input_indices.push_back(operand_indices.at(axes_name));  // axes
+
+  ORT_RETURN_IF_ERROR(shaper.Squeeze(input, axes, output));
+  const OperandType output_operand_type(operand_types.at(input).type, shaper[output]);
+  ORT_RETURN_IF_ERROR(model_builder.AddOperation(ANEURALNETWORKS_SQUEEZE, input_indices,
+                                                 {output}, {output_operand_type}, {false}));
+  return Status::OK();
+}
+
 /* static */ vector<int32_t> SqueezeOpBuilder::GetAxes(ModelBuilder& model_builder, const Node& node) {
   vector<int32_t> axes;
   // Squeeze opset 13 use input as axes
@@ -1804,47 +1851,13 @@ void SqueezeOpBuilder::AddInitializersToSkip(ModelBuilder& model_builder, const
 }
 
 Status SqueezeOpBuilder::AddToModelBuilderImpl(ModelBuilder& model_builder, const Node& node) const {
-  auto& shaper(model_builder.GetShaper());
-  const auto& operand_indices(model_builder.GetOperandIndices());
-  const auto& operand_types(model_builder.GetOperandTypes());
-
   auto input = node.InputDefs()[0]->Name();
   if (model_builder.IsOperandNHWC(input)) {
     // We want to transpose nhwc operand back to nchw before squeeze
     ORT_RETURN_IF_ERROR(GetNCHWInput(model_builder, node, 0, input));
   }
 
-  NodeAttrHelper helper(node);
-  vector<int32_t> axes = GetAxes(model_builder, node);
-  const auto& input_shape(shaper[input]);
-  auto input_dims = input_shape.size();
-  for (auto& axis : axes) {
-    axis = static_cast<int32_t>(HandleNegativeAxis(axis, input_dims));
-  }
-
-  if (axes.empty()) {  // Squeeze all
-    for (size_t i = 0; i < input_dims; i++) {
-      if (input_shape[i] == 1)
-        axes.push_back(i);
-    }
-  }
-
-  const auto axes_name = model_builder.GetUniqueName(node.Name() + input + "_axes");
-  Shape axes_dimen = {static_cast<uint32_t>(axes.size())};
-  shaper.AddShape(axes_name, axes_dimen);
-  const OperandType axes_operand_type(Type::TENSOR_INT32, axes_dimen);
-  ORT_RETURN_IF_ERROR(model_builder.AddOperandFromPersistMemoryBuffer(axes_name, axes.data(), axes_operand_type));
-
-  std::vector<uint32_t> input_indices;
-  input_indices.push_back(operand_indices.at(input));      // input
-  input_indices.push_back(operand_indices.at(axes_name));  // axes
-
-  const auto& output = node.OutputDefs()[0]->Name();
-  ORT_RETURN_IF_ERROR(shaper.Squeeze(input, axes, output));
-  const OperandType output_operand_type(operand_types.at(input).type, shaper[output]);
-  ORT_RETURN_IF_ERROR(model_builder.AddOperation(ANEURALNETWORKS_SQUEEZE, input_indices,
-                                                 {output}, {output_operand_type}, {false}));
-  return Status::OK();
+  return AddSqueezeOp(model_builder, node.Name(), input, node.OutputDefs()[0]->Name(), GetAxes(model_builder, node));
 }
 
 #pragma endregion