Upgrade oneDNN to v2.3.3 and package oneDNN Graph API together (#63748)

Summary:
This PR upgrades oneDNN to [v2.3.3](https://github.com/oneapi-src/oneDNN/releases/tag/v2.3.3) and includes [Graph API preview release](https://github.com/oneapi-src/oneDNN/releases/tag/graph-v0.2) in one package.

- oneDNN will be located at `pytorch/third_party/ideep/mkl-dnn/third_party/oneDNN`
- The version of oneDNN will be [v2.3.3](https://github.com/oneapi-src/oneDNN/releases/tag/v2.3.3)
  The main changes on CPU:

  - v2.3
    - Extended primitive cache to improve primitive descriptor creation performance.
    - Improved primitive cache performance in multithreaded configurations.
    - Introduced initial optimizations for bfloat16 compute functionality for future Intel Xeon Scalable processor (code name Sapphire Rapids).
    - Improved performance of binary primitive and binary post-op for cases with broadcast and mixed source and destination formats.
    - Improved performance of reduction primitive
    - Improved performance of depthwise convolution primitive with NHWC activations for training cases
  - v2.3.1
    -  Improved int8 GEMM performance for processors with Intel AVX2 and Intel DL Boost support
    - Fixed integer overflow for inner product implementation on CPUs
    - Fixed out of bounds access in GEMM implementation for Intel SSE 4.1
  - v2.3.2
    - Fixed performance regression in fp32 inner product primitive for processors with Intel AVX512 support
  - v2.3.3
    - Reverted check for memory descriptor stride validity for unit dimensions
    - Fixed memory leak in CPU GEMM implementation

  More changes can be found in https://github.com/oneapi-src/oneDNN/releases.
- The Graph API provides flexible API for aggressive fusion, and the preview2 supports fusion for FP32 inference.  See the [Graph API release branch](https://github.com/oneapi-src/oneDNN/tree/dev-graph-preview2) and [spec](https://spec.oneapi.io/onednn-graph/latest/introduction.html) for more details. A separate PR will be submitted to integrate the oneDNN Graph API to Torchscript graph.

Pull Request resolved: #63748

Reviewed By: albanD

Differential Revision: D32153889

Pulled By: malfet

fbshipit-source-id: 536071168ffe312d452f75d54f34c336ca3778c1

cmake/Modules/FindMKLDNN.cmake

@@ -18,7 +18,7 @@ SET(MKLDNN_LIBRARIES)
 SET(MKLDNN_INCLUDE_DIR)
 
 SET(IDEEP_ROOT "${PROJECT_SOURCE_DIR}/third_party/ideep")
-SET(MKLDNN_ROOT "${IDEEP_ROOT}/mkl-dnn")
+SET(MKLDNN_ROOT "${IDEEP_ROOT}/mkl-dnn/third_party/oneDNN")
 
 FIND_PACKAGE(BLAS)
 FIND_PATH(IDEEP_INCLUDE_DIR ideep.hpp PATHS ${IDEEP_ROOT} PATH_SUFFIXES include)

third_party/mkl-dnn.BUILD

@@ -5,53 +5,54 @@ _DNNL_RUNTIME_OMP = {
     "#cmakedefine DNNL_CPU_THREADING_RUNTIME DNNL_RUNTIME_${DNNL_CPU_THREADING_RUNTIME}": "#define DNNL_CPU_THREADING_RUNTIME DNNL_RUNTIME_OMP",
     "#cmakedefine DNNL_CPU_RUNTIME DNNL_RUNTIME_${DNNL_CPU_RUNTIME}": "#define DNNL_CPU_RUNTIME DNNL_RUNTIME_OMP",
     "#cmakedefine DNNL_GPU_RUNTIME DNNL_RUNTIME_${DNNL_GPU_RUNTIME}": "#define DNNL_GPU_RUNTIME DNNL_RUNTIME_NONE",
+    "#cmakedefine DNNL_USE_RT_OBJECTS_IN_PRIMITIVE_CACHE": "/* undef DNNL_USE_RT_OBJECTS_IN_PRIMITIVE_CACHE */",
     "#cmakedefine DNNL_WITH_SYCL": "/* #undef DNNL_WITH_SYCL */",
     "#cmakedefine DNNL_WITH_LEVEL_ZERO": "/* #undef DNNL_WITH_LEVEL_ZERO */",
     "#cmakedefine DNNL_SYCL_CUDA": "/* #undef DNNL_SYCL_CUDA */",
 }
 
 template_rule(
-    name = "include_dnnl_version",
-    src = "include/oneapi/dnnl/dnnl_version.h.in",
-    out = "include/oneapi/dnnl/dnnl_version.h",
+    name = "third_party/oneDNN/include_dnnl_version",
+    src = "third_party/oneDNN/include/oneapi/dnnl/dnnl_version.h.in",
+    out = "third_party/oneDNN/include/oneapi/dnnl/dnnl_version.h",
     substitutions = {
         "@DNNL_VERSION_MAJOR@": "2",
-        "@DNNL_VERSION_MINOR@": "2",
+        "@DNNL_VERSION_MINOR@": "3",
         "@DNNL_VERSION_PATCH@": "3",
-        "@DNNL_VERSION_HASH@": "7336ca9f055cf1bfa13efb658fe15dc9b41f0740",
+        "@DNNL_VERSION_HASH@": "f40443c413429c29570acd6cf5e3d1343cf647b4",
     },
 )
 
 template_rule(
-    name = "include_dnnl_config",
-    src = "include/oneapi/dnnl/dnnl_config.h.in",
-    out = "include/oneapi/dnnl/dnnl_config.h",
+    name = "third_party/oneDNN/include_dnnl_config",
+    src = "third_party/oneDNN/include/oneapi/dnnl/dnnl_config.h.in",
+    out = "third_party/oneDNN/include/oneapi/dnnl/dnnl_config.h",
     substitutions = _DNNL_RUNTIME_OMP,
 )
 
 cc_library(
     name = "mkl-dnn",
     srcs = glob([
-        "src/common/*.cpp",
-        "src/cpu/**/*.cpp",
+        "third_party/oneDNN/src/common/*.cpp",
+        "third_party/oneDNN/src/cpu/**/*.cpp",
     ], exclude=[
-        "src/cpu/aarch64/**/*.cpp",
+        "third_party/oneDNN/src/cpu/aarch64/**/*.cpp",
     ]),
     hdrs = glob([
-        "include/oneapi/dnnl/*.h",
-        "include/oneapi/dnnl/*.hpp",
-        "include/*.h",
-        "include/*.hpp",
-        "src/cpu/**/*.hpp",
-        "src/cpu/**/*.h",
-        "src/common/*.hpp",
-        "src/common/ittnotify/jitprofiling.h",
+        "third_party/oneDNN/include/oneapi/dnnl/*.h",
+        "third_party/oneDNN/include/oneapi/dnnl/*.hpp",
+        "third_party/oneDNN/include/*.h",
+        "third_party/oneDNN/include/*.hpp",
+        "third_party/oneDNN/src/cpu/**/*.hpp",
+        "third_party/oneDNN/src/cpu/**/*.h",
+        "third_party/oneDNN/src/common/*.hpp",
+        "third_party/oneDNN/src/common/ittnotify/jitprofiling.h",
     ], exclude=[
-        "src/cpu/aarch64/**/*.hpp",
-        "src/cpu/aarch64/**/*.h",
+        "third_party/oneDNN/src/cpu/aarch64/**/*.hpp",
+        "third_party/oneDNN/src/cpu/aarch64/**/*.h",
     ]) + [
-        "include/oneapi/dnnl/dnnl_config.h",
-        "include/oneapi/dnnl/dnnl_version.h",
+        "third_party/oneDNN/include/oneapi/dnnl/dnnl_config.h",
+        "third_party/oneDNN/include/oneapi/dnnl/dnnl_version.h",
     ],
     copts = [
         "-DUSE_AVX",
@@ -69,13 +70,13 @@ cc_library(
         "//conditions:default": ["-DDNNL_CPU_RUNTIME=2"],
     }),
     includes = [
-        "include/",
-        "include/oneapi/",
-        "include/oneapi/dnnl/",
-        "src/",
-        "src/common/",
-        "src/cpu/",
-        "src/cpu/x64/xbyak/",
+        "third_party/oneDNN/include/",
+        "third_party/oneDNN/include/oneapi/",
+        "third_party/oneDNN/include/oneapi/dnnl/",
+        "third_party/oneDNN/src/",
+        "third_party/oneDNN/src/common/",
+        "third_party/oneDNN/src/cpu/",
+        "third_party/oneDNN/src/cpu/x64/xbyak/",
     ],
     visibility = ["//visibility:public"],
     linkopts = [