Merge branch 'main' into new-delete-aarch64

Author: josephhackman

.github/CODEOWNERS

@@ -17,17 +17,9 @@
 # Default
 * @oneapi-src/onednn-arch @intel-innersource/dnn-arch
 
-# Codeowners
-/.github/CODEOWNERS @oneapi-src/onednn-maintain
-/MAINTAINERS.md @oneapi-src/onednn-maintain
-
 # Github automation
 /.github/ @oneapi-src/onednn-devops
 
-# Documentation
-*.md  @oneapi-src/onednn-doc @oneapi-src/onednn-arch @intel-innersource/dnn-doc @intel-innersource/dnn-arch
-/doc/ @oneapi-src/onednn-doc @oneapi-src/onednn-arch @intel-innersource/dnn-doc @intel-innersource/dnn-arch
-
 # CPU Engine
 /src/cpu/aarch64/ @oneapi-src/onednn-cpu-aarch64 @intel-innersource/dnn-arch
 /src/cpu/x64/ @oneapi-src/onednn-cpu-x64 @intel-innersource/dnn-cpu
@@ -52,3 +44,15 @@
 # Graph compiler
 /src/graph/backend/graph_compiler/ @intel-innersource/dnn-compiler
 /tests/gtests/graph/unit/backend/graph_compiler/ @intel-innersource/dnn-compiler
+
+# Documentation
+*.md  @oneapi-src/onednn-doc @oneapi-src/onednn-arch @intel-innersource/dnn-doc @intel-innersource/dnn-arch
+/doc/ @oneapi-src/onednn-doc @oneapi-src/onednn-arch @intel-innersource/dnn-doc @intel-innersource/dnn-arch
+
+# Governance and process
+/.github/CODEOWNERS @oneapi-src/onednn-maintain
+/SECURITY.md @oneapi-src/onednn-maintain
+/MAINTAINERS.md @oneapi-src/onednn-maintain
+/CONTRIBUTING.md @oneapi-src/onednn-maintain
+/CODING_STANDARDS.md @oneapi-src/onednn-maintain
+/CODE_OF_CONDUCT.md @oneapi-src/onednn-maintain

.github/dependabot.yml

@@ -20,5 +20,6 @@ updates:
     directory: "/"
     schedule:
       interval: "weekly"
+    labels: [ "devops" ]
     commit-message:
       prefix: "github: workflows"

.github/labels.yml

@@ -0,0 +1,85 @@
+#===============================================================================
+# Copyright 2024 Intel Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#===============================================================================
+
+# This is configuration file for Labeler workflow. See documenation for syntax
+# reference: https://github.com/marketplace/actions/labeler
+
+# Labels based on area of responsibility
+# Process definition and code owners
+governance:
+- changed-files:
+  - any-glob-to-any-file:
+    - '.github/CODEOWNERS'
+    - 'SECURITY.md'
+    - 'MAINTAINERS.md'
+    - 'CONTRIBUTING.md'
+    - 'CODING_STANDARDS.md'
+    - 'CODE_OF_CONDUCT.md'
+
+# Github automation
+devops:
+- all:
+  - changed-files:
+    - any-glob-to-any-file: '.github/**'
+    - all-globs-to-all-files: '!.github/CODEOWNERS'
+
+# Documentation
+documentation:
+- changed-files:
+  - any-glob-to-any-file: ['**/*.md', 'doc/**']
+
+# CPU Engine
+platform:cpu-aarch64:
+- changed-files:
+  - any-glob-to-any-file: 'src/cpu/aarch64/**'
+
+platform:cpu-ppc64:
+- changed-files:
+  - any-glob-to-any-file: 'src/cpu/ppc64/**'
+
+platform:cpu-rv64:
+- changed-files:
+  - any-glob-to-any-file: 'src/cpu/rv64/**'
+
+platform:cpu-s390x:
+- changed-files:
+  - any-glob-to-any-file: 'src/cpu/s390x/**'
+
+platform:cpu-x64:
+- changed-files:
+  - any-glob-to-any-file: ['src/cpu/x64/**', 'src/cpu/rnn/**']
+
+# GPU Engine
+platform:gpu-amd:
+- changed-files:
+  - any-glob-to-any-file: 'src/gpu/amd/**'
+
+platform:gpu-intel:
+- changed-files:
+  - any-glob-to-any-file: 'src/gpu/intel/**'
+
+platform:gpu-nvidia:
+- changed-files:
+  - any-glob-to-any-file: 'src/gpu/nvidia/**'
+
+platform:gpu-generic:
+- changed-files:
+  - any-glob-to-any-file: 'src/gpu/generic/**'
+
+# Labels based on target branch
+# Backport
+backport:
+- base-branch: 'rls-*'

.github/workflows/labeler.yml

@@ -0,0 +1,34 @@
+#===============================================================================
+# Copyright 2024 Intel Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#===============================================================================
+
+name: Labeler
+on: [pull_request_target]
+
+# Declare default permissions as read only.
+permissions: read-all
+
+jobs:
+  label:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: write
+
+    steps:
+      - uses: actions/labeler@v5.0.0
+        with:
+          sync-labels: true
+          configuration-path: '.github/labels.yml'

README.binary.in

@@ -39,8 +39,9 @@ The library is optimized for the following CPUs:
 * Intel Xeon Scalable processor (formerly Skylake, Cascade Lake, Cooper
   Lake, Ice Lake, Sapphire Rapids, and Emerald Rapids)
 * Intel Xeon CPU Max Series (formerly Sapphire Rapids HBM)
-* future Intel Xeon Scalable processors (code name Sierra Forest
-and Granite Rapids)
+* Intel Core Ultra processors (formerly Meteor Lake, Arrow Lake,
+  and Lunar Lake)
+* Intel Xeon 6 processors (formerly Sierra Forest and Granite Rapids)
 
 oneDNN detects the instruction set architecture (ISA) at runtime and uses
 just-in-time (JIT) code generation to deploy the code optimized
@@ -52,11 +53,13 @@ for more details.
 
 The library is optimized for the following GPUs:
 * Intel Graphics for 11th-14th Generation Intel Core Processors
-* Intel Graphics for Intel Core Ultra Processors (formerly Meteor Lake)
 * Intel Iris Xe MAX Graphics (formerly DG1)
-* Intel Arc(TM) Graphics (formerly Alchemist)
+* Intel Arc(TM) graphics (formerly Alchemist)
 * Intel Data Center GPU Flex Series (formerly Arctic Sound)
 * Intel Data Center GPU Max Series (formerly Ponte Vecchio)
+* Intel Graphics and Intel Arc graphics for Intel Core Ultra processors
+ (formerly Meteor Lake, Arrow Lake and Lunar Lake)
+* future Intel Arc graphics (code name Battlemage)
 
 Support
 -------

README.md

@@ -84,14 +84,14 @@ The library is optimized for the following CPUs:
 * Intel 64/AMD64 architecture
   * Intel Atom(R) processor (at least Intel SSE4.1 support is required)
   * Intel Core(TM) processor (at least Intel SSE4.1 support is required)
-  * Intel Core Ultra processors (formerly Meteor Lake)
   * Intel Xeon(R) processor E3, E5, and E7 family (formerly Sandy Bridge,
     Ivy Bridge, Haswell, and Broadwell)
   * Intel Xeon Scalable processor (formerly Skylake, Cascade Lake, Cooper
     Lake, Ice Lake, Sapphire Rapids, and Emerald Rapids)
   * Intel Xeon CPU Max Series (formerly Sapphire Rapids HBM)
-  * future Intel Xeon Scalable processors (code name Sierra Forest
-and Granite Rapids)
+  * Intel Core Ultra processors (formerly Meteor Lake, Arrow Lake,
+    and Lunar Lake)
+  * Intel Xeon 6 processors (formerly Sierra Forest and Granite Rapids)
 * AArch64 architecture
   * Arm Neoverse(TM) N1 and V1 processors
 
@@ -111,11 +111,13 @@ require the use of run-time controls to enable them. See
 
 The library is optimized for the following GPUs:
 * Intel Graphics for 11th-14th Generation Intel Core Processors
-* Intel Graphics for Intel Core Ultra processors (formerly Meteor Lake)
 * Intel Iris Xe MAX Graphics (formerly DG1)
 * Intel Arc(TM) graphics (formerly Alchemist)
 * Intel Data Center GPU Flex Series (formerly Arctic Sound)
 * Intel Data Center GPU Max Series (formerly Ponte Vecchio)
+* Intel Graphics and Intel Arc graphics for Intel Core Ultra processors
+ (formerly Meteor Lake, Arrow Lake and Lunar Lake)
+* future Intel Arc graphics (code name Battlemage)
 
 [CPU dispatcher control]: https://oneapi-src.github.io/oneDNN/dev_guide_cpu_dispatcher_control.html
 [Linking Guide]: https://oneapi-src.github.io/oneDNN/dev_guide_link.html
@@ -171,7 +173,7 @@ On a CPU based on Arm AArch64 architecture, oneDNN CPU engine can be built with
 machine learning applications and provides AArch64 optimized implementations
 of core functions. This functionality currently requires that ACL is downloaded
 and built separately. See [Build from Source] section of the Developer Guide for
-details. oneDNN only supports Compute Library versions 24.07 or later.
+details. oneDNN only supports Compute Library versions 24.08 or later.
 
 [Arm Compute Library (ACL)]: https://github.com/arm-software/ComputeLibrary
 
@@ -201,6 +203,9 @@ is enabled:
     version of MIOpen)
     * [rocBLAS] version 2.45.0 or later (optional if AMD ROCm includes
     the required version of rocBLAS)
+* SYCL runtime with a generic GPU support requires
+    * oneAPI DPC++/C++ Compiler that supports the target GPU. Refer to the
+    [generic GPU vendor readme](src/gpu/generic/sycl/README.md) for more information.
 
 > **WARNING**
 >
@@ -374,9 +379,12 @@ discussions.
 
 # Contributing
 
-We welcome community contributions to oneDNN. If you have an idea on how
-to improve the library:
+We welcome community contributions to oneDNN. You can find the oneDNN release
+schedule and work already in progress towards future milestones in Github's
+[Milestones] section. If you are looking for a specific task to start,
+consider selecting from issues that are marked with the [help wanted] label.
 
+If you have an idea on how to improve the library:
 * For changes impacting the public API or library overall, such as adding new
 primitives or changes to the architecture, submit an [RFC pull request].
 * Ensure that the changes are consistent with the [code contribution guidelines]
@@ -397,6 +405,8 @@ contributors are expected to adhere to the
 [code contribution guidelines]: CONTRIBUTING.md#code-contribution-guidelines
 [coding standards]: CONTRIBUTING.md#coding-standards
 [pull request]: https://github.com/oneapi-src/oneDNN/pulls
+[Milestones]: https://github.com/oneapi-src/oneDNN/milestones
+[help wanted]: https://github.com/oneapi-src/oneDNN/issues?q=is%3Aissue+is%3Aopen+label%3A%22help+wanted%22
 
 # License
 

cmake/ACL.cmake

@@ -31,7 +31,7 @@ endif()
 
 find_package(ACL REQUIRED)
 
-set(ACL_MINIMUM_VERSION "24.07")
+set(ACL_MINIMUM_VERSION "24.08")
 
 if(ACL_FOUND)
     file(GLOB_RECURSE ACL_VERSION_FILE ${ACL_INCLUDE_DIR}/*/arm_compute_version.embed)

cmake/SYCL.cmake

@@ -74,16 +74,19 @@ macro(adjust_headers_priority targets)
     endif()
 endmacro()
 
-if(DNNL_SYCL_CUDA)
+macro(suppress_warnings_for_nvidia_target)
     # XXX: Suppress warning coming from SYCL headers:
     #   error: use of function template name with no prior declaration in
     #   function call with eplicit template arguments is a C++20 extension
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-c++20-extensions")
+    append(CMAKE_CXX_FLAGS "-Wno-c++20-extensions")
 
     # Suppress LLVM warning about not supporting latest cuda. It's safe enough
     # as long as no new cuda features are used in SYCL kernels.
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-unknown-cuda-version")
+    append(CMAKE_CXX_FLAGS "-Wno-unknown-cuda-version")
+endmacro()
 
+if(DNNL_SYCL_CUDA)
+    suppress_warnings_for_nvidia_target()
     find_package(cuBLAS REQUIRED)
     find_package(cuDNN REQUIRED)
 
@@ -102,6 +105,12 @@ elseif(DNNL_SYCL_HIP)
 
     list(APPEND EXTRA_SHARED_LIBS HIP::HIP rocBLAS::rocBLAS MIOpen::MIOpen)
     message(STATUS "DPC++ support is enabled (HIP)")
+elseif(DNNL_SYCL_GENERIC)
+    CHECK_CXX_COMPILER_FLAG("-fsycl -fsycl-targets=nvptx64-nvidia-cuda" NVIDIA_TARGET_SUPPORTED)
+
+    if(NVIDIA_TARGET_SUPPORTED)
+        suppress_warnings_for_nvidia_target()
+    endif()
 else()
     find_library(OPENCL_LIBRARY OpenCL PATHS ENV LIBRARY_PATH ENV LIB NO_DEFAULT_PATH)
     if(OPENCL_LIBRARY)

cmake/options.cmake

@@ -298,7 +298,7 @@ if(NOT DNNL_GPU_RUNTIME STREQUAL "NONE" AND DNNL_GPU_VENDOR STREQUAL "NONE")
     set(DNNL_GPU_VENDOR "INTEL")
 endif()
 
-if(NOT "${DNNL_GPU_VENDOR}" MATCHES "^(NONE|INTEL|NVIDIA|AMD)$")
+if(NOT "${DNNL_GPU_VENDOR}" MATCHES "^(NONE|INTEL|NVIDIA|AMD|GENERIC)$")
     message(FATAL_ERROR "Unsupported GPU vendor: ${DNNL_GPU_VENDOR}")
 endif()
 
@@ -322,12 +322,16 @@ if(DNNL_GPU_RUNTIME STREQUAL "DPCPP" OR DNNL_GPU_RUNTIME STREQUAL "SYCL")
     set(DNNL_GPU_SYCL true)
     set(DNNL_SYCL_CUDA OFF)
     set(DNNL_SYCL_HIP OFF)
+    set(DNNL_SYCL_GENERIC OFF)
     if(DNNL_GPU_VENDOR STREQUAL "NVIDIA")
         set(DNNL_SYCL_CUDA ON)
     endif()
     if(DNNL_GPU_VENDOR STREQUAL "AMD")
         set(DNNL_SYCL_HIP ON)
     endif()
+    if(DNNL_GPU_VENDOR STREQUAL "GENERIC")
+        set(DNNL_SYCL_GENERIC ON)
+    endif()
 else()
     set(DNNL_GPU_SYCL false)
 endif()