Update DeepSpeed copyright license to Apache 2.0 (#3111)

Co-authored-by: Jeff Rasley <jerasley@microsoft.com>

.pre-commit-config.yaml

@@ -48,7 +48,7 @@ repos:
         name: check-license
         entry: ./scripts/check-license.py
         language: script
-        files: \.(py|cc|cu|h|cuh|hip)$
+        files: \.(py|c|cpp|cu|cc|h|hpp|cuh|hip|tr)$
 
 -   repo: https://github.com/codespell-project/codespell
     rev: v2.1.0

LICENSE

@@ -1,21 +1,13 @@
-    MIT License
+Copyright 2023 Microsoft Corporation
 
-    Copyright (c) Microsoft 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
 
-    Permission is hereby granted, free of charge, to any person obtaining a copy
-    of this software and associated documentation files (the "Software"), to deal
-    in the Software without restriction, including without limitation the rights
-    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-    copies of the Software, and to permit persons to whom the Software is
-    furnished to do so, subject to the following conditions:
+    http://www.apache.org/licenses/LICENSE-2.0
 
-    The above copyright notice and this permission notice shall be included in all
-    copies or substantial portions of the Software.
-
-    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-    SOFTWARE
+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.

accelerator/__init__.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 from .abstract_accelerator import DeepSpeedAccelerator
 from .real_accelerator import get_accelerator, set_accelerator

accelerator/abstract_accelerator.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import abc
 from abc import ABC

accelerator/cuda_accelerator.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import os
 import pkgutil

accelerator/real_accelerator.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 try:
     from accelerator.abstract_accelerator import DeepSpeedAccelerator as dsa1

benchmarks/__init__.py

@@ -1 +1,5 @@
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 '''Copyright The Microsoft DeepSpeed Team'''

benchmarks/communication/__init__.py

@@ -1 +1,5 @@
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 '''Copyright The Microsoft DeepSpeed Team'''

benchmarks/communication/all_gather.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import sys, os, time

benchmarks/communication/all_reduce.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import sys, os, time

benchmarks/communication/all_to_all.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import sys, os, time

benchmarks/communication/broadcast.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import sys, os, time

benchmarks/communication/constants.py

@@ -1,4 +1,8 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
 from deepspeed.accelerator import get_accelerator
 
 DEFAULT_WARMUPS = 5

benchmarks/communication/pt2pt.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import sys, os, time

benchmarks/communication/run_all.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import sys, os
 

benchmarks/communication/utils.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import os, sys

benchmarks/inference/bert-bench.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import torch
 import time

benchmarks/inference/collect_results.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import os
 import re

benchmarks/inference/gpt-bench.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 import os
 import torch

csrc/adagrad/cpu_adagrad.cpp

@@ -1,3 +1,8 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include "cpu_adagrad.h"
 #include <torch/extension.h>
 #include <iostream>

csrc/adam/cpu_adam.cpp

@@ -1,3 +1,8 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include "cpu_adam.h"
 #include <torch/extension.h>
 #include <cassert>

csrc/adam/fused_adam_frontend.cpp

@@ -1,3 +1,8 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include <torch/extension.h>
 
 void multi_tensor_adam_cuda(int chunk_size,

csrc/adam/multi_tensor_adam.cu

@@ -1,6 +1,11 @@
-/* Copyright 2020 The Microsoft DeepSpeed Team
-   Copyright NVIDIA/apex
-   This file is adapted from fused adam in NVIDIA/apex, commit a109f85
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
+/*
+Copyright NVIDIA/apex
+This file is adapted from fused adam in NVIDIA/apex, commit a109f85
 */
 
 #include <ATen/ATen.h>

csrc/adam/multi_tensor_apply.cuh

@@ -1,6 +1,11 @@
-/* Copyright 2020 The Microsoft DeepSpeed Team
-   Copyright NVIDIA/apex
-   This file is adapted from fused adam in NVIDIA/apex, commit a109f85
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
+/*
+Copyright NVIDIA/apex
+This file is adapted from fused adam in NVIDIA/apex, commit a109f85
 */
 
 #include <ATen/ATen.h>

csrc/aio/common/deepspeed_aio_common.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/common/deepspeed_aio_common.h

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/common/deepspeed_aio_types.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/common/deepspeed_aio_types.h

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/common/deepspeed_aio_utils.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/common/deepspeed_aio_utils.h

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_lib/deepspeed_aio_thread.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_lib/deepspeed_aio_thread.h

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_lib/deepspeed_pin_tensor.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2023 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for managing CPU tensors occupying page-locked memory.
 */
 

csrc/aio/py_lib/deepspeed_pin_tensor.h

@@ -1,11 +1,14 @@
-/*
-Copyright 2023 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for managing CPU tensors occupying page-locked memory.
 TODO: Implement a full-featured manager that
- 1. Avoid page-locked memory leaks
- 2. Minimize page-locked memory usage by reducing internal fragmentation
+1. Avoid page-locked memory leaks
+2. Minimize page-locked memory usage by reducing internal fragmentation
+Functionality for managing CPU tensors occupying page-locked memory.
 */
 
 #include <map>

csrc/aio/py_lib/deepspeed_py_aio.cpp

@@ -1,3 +1,7 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 /*
 Copyright 2020 The Microsoft DeepSpeed Team

csrc/aio/py_lib/deepspeed_py_aio.h

@@ -1,3 +1,7 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 /*
 Copyright 2020 The Microsoft DeepSpeed Team

csrc/aio/py_lib/deepspeed_py_aio_handle.cpp

@@ -1,3 +1,7 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 /*
 Copyright 2020 The Microsoft DeepSpeed Team

csrc/aio/py_lib/deepspeed_py_aio_handle.h

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_lib/deepspeed_py_copy.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_lib/deepspeed_py_copy.h

@@ -1,3 +1,7 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 /*
 Copyright 2020 The Microsoft DeepSpeed Team

csrc/aio/py_lib/py_ds_aio.cpp

@@ -1,7 +1,9 @@
-/*
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Functionality for swapping optimizer tensors to/from (NVMe) storage devices.
 */
 

csrc/aio/py_test/aio_bench_generate_param.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2021 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 import os

csrc/aio/py_test/aio_bench_perf_sweep.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2021 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 import os

csrc/aio/py_test/ds_aio_basic.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 

csrc/aio/py_test/ds_aio_handle.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 

csrc/aio/py_test/parse_aio_stats.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 

csrc/aio/py_test/perf_sweep_utils.py

@@ -1,4 +1,7 @@
-'''Copyright The Microsoft DeepSpeed Team'''
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
 
 SCRIPT_PREFIX = '_aio_bench'
 WRITE_OP_DESC = 'write'

csrc/aio/py_test/test_ds_aio.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 

csrc/aio/py_test/test_ds_aio_utils.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2020 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 

csrc/aio/py_test/validate_async_io.py

@@ -1,7 +1,8 @@
-"""
-Copyright 2021 The Microsoft DeepSpeed Team
-Licensed under the MIT license.
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
 
+# DeepSpeed Team
+"""
 Functionality of swapping optimizer tensors to/from (NVMe) storage devices.
 """
 from deepspeed.ops.op_builder import AsyncIOBuilder

csrc/common/custom_cuda_kernel.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "custom_cuda_layers.h"
 

csrc/includes/StopWatch.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 #ifdef _WIN32

csrc/includes/Timer.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #ifndef __TIMER_H__
 #define __TIMER_H__

csrc/includes/compat.h

@@ -1,6 +1,11 @@
-/* Copyright 2020 The Microsoft DeepSpeed Team
-   Copyright NVIDIA/apex
-   This file is adapted from fused adam in NVIDIA/apex, commit a109f85
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
+/*
+Copyright NVIDIA/apex
+This file is adapted from fused adam in NVIDIA/apex, commit a109f85
 */
 
 #ifndef TORCH_CHECK

csrc/includes/context.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/conversion_utils.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/cpu_adagrad.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/cpu_adam.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/cublas_wrappers.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/custom_cuda_layers.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/dequantization_utils.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "conversion_utils.h"
 #include "ds_kernel_utils.h"

csrc/includes/dropout.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/ds_kernel_utils.h

@@ -1,6 +1,9 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
 
+// DeepSpeed Team
+
+/*
 Centralized header file for preprocessor macros and constants
 used throughout the codebase.
 */

csrc/includes/ds_transformer_cuda.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/feed_forward.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #ifndef __FEEDFORWARD_H__
 #define __FEEDFORWARD_H__

csrc/includes/gelu.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/gemm_test.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/general_kernels.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <cuda.h>
 #include <cuda_fp16.h>

csrc/includes/memory_access_utils.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/normalize_layer.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/quantization.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/quantization_utils.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <cassert>
 #include "conversion_utils.h"

csrc/includes/quantizer.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/reduction_utils.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/simd.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/softmax.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/strided_batch_gemm.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/includes/type_shim.h

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 /* Taken from NVIDIA/apex commit 855808f3fc268e9715d613f3c2e56469d8c986d8 */
 #include <ATen/ATen.h>

csrc/lamb/fused_lamb_cuda.cpp

@@ -1,4 +1,8 @@
-/* Copyright 2019 The Microsoft DeepSpeed Team */
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include <torch/extension.h>
 
 // CUDA forward declaration

csrc/lamb/fused_lamb_cuda_kernel.cu

@@ -1,4 +1,8 @@
-/* Copyright 2019 The Microsoft DeepSpeed Team */
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include <cuda.h>
 #include <cuda_runtime.h>
 #include <stdio.h>

csrc/quantization/dequantize.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "dequantization_utils.h"
 #include "memory_access_utils.h"

csrc/quantization/fake_quantizer.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <math.h>
 #include "custom_cuda_layers.h"

csrc/quantization/pt_binding.cpp

@@ -1,3 +1,8 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include <ATen/cuda/CUDAContext.h>
 #include <torch/extension.h>
 #include <cassert>

csrc/quantization/quantize.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "ds_kernel_utils.h"
 #include "memory_access_utils.h"

csrc/random_ltd/gather_scatter.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "custom_cuda_layers.h"
 #include "memory_access_utils.h"

csrc/random_ltd/pt_binding.cpp

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <torch/extension.h>
 #include <vector>

csrc/random_ltd/slice_attn_masks.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "custom_cuda_layers.h"
 #include "memory_access_utils.h"

csrc/random_ltd/token_sort.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <cassert>
 #include "custom_cuda_layers.h"

csrc/sparse_attention/utils.cpp

@@ -1,5 +1,12 @@
-// DeepSpeed note, code taken & adapted from commit 9aa94789f13ada713af36cfd8cca2fc9a7f6b79a
-// https://github.com/ptillet/torch-blocksparse/blob/master/csrc/utils.cpp
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
+/*
+DeepSpeed note, code taken & adapted from commit 9aa94789f13ada713af36cfd8cca2fc9a7f6b79a
+ https:github.com/ptillet/torch-blocksparse/blob/master/csrc/utils.cpp
+*/
 
 #include <torch/extension.h>
 #include <string>

csrc/spatial/csrc/opt_bias_add.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <cassert>
 #include "memory_access_utils.h"

csrc/spatial/csrc/pt_binding.cpp

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include <c10/cuda/CUDAStream.h>
 #include <torch/extension.h>

csrc/spatial/includes/spatial_cuda_layers.h

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #pragma once
 

csrc/transformer/cublas_wrappers.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "cublas_wrappers.h"
 

csrc/transformer/dropout_kernels.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "custom_cuda_layers.h"
 

csrc/transformer/ds_transformer_cuda.cpp

@@ -1,3 +1,8 @@
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
+
 #include <torch/extension.h>
 
 #include <cublas_v2.h>

csrc/transformer/gelu_kernels.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "custom_cuda_layers.h"
 

csrc/transformer/general_kernels.cu

@@ -1,6 +1,7 @@
-/*
-Copyright The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "general_kernels.h"
 

csrc/transformer/inference/csrc/apply_rotary_pos_emb.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "inference_cuda_layers.h"
 

csrc/transformer/inference/csrc/dequantize.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "inference_cuda_layers.h"
 

csrc/transformer/inference/csrc/gelu.cu

@@ -1,6 +1,7 @@
-/*
-Copyright 2022 The Microsoft DeepSpeed Team
-*/
+// Copyright (c) Microsoft Corporation.
+// SPDX-License-Identifier: Apache-2.0
+
+// DeepSpeed Team
 
 #include "conversion_utils.h"
 #include "inference_cuda_layers.h"