chore: import upstream snapshot with attribution

2026-07-13 12:40:42 +08:00
commit e25996e7db
15472 changed files with 3536181 additions and 0 deletions
@@ -0,0 +1,103 @@
+// Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+//
+// 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.
+
+#include "paddle/phi/kernels/reduce_sum_kernel.h"
+
+#include "paddle/phi/backends/all_context.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/reduce_kernel_impl.h"
+
+namespace phi {
+
+template <typename T, typename Context>
+void SumKernel(const Context& dev_ctx,
+               const DenseTensor& x,
+               const IntArray& dims,
+               DataType out_dtype,
+               bool keep_dim,
+               DenseTensor* out) {
+  bool reduce_all = recompute_reduce_all(x, dims);
+  SumRawKernel<T, Context>(
+      dev_ctx, x, dims, keep_dim, reduce_all, out_dtype, out);
+}
+
+}  // namespace phi
+
+PD_REGISTER_KERNEL(sum,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::SumKernel,
+                   bool,
+                   float,
+                   double,
+                   phi::float16,
+                   phi::bfloat16,
+                   int16_t,
+                   int,
+                   int64_t,
+                   uint8_t,
+                   int8_t,
+                   phi::complex64,
+                   phi::complex128) {
+  kernel->OutputAt(0).SetDataType(phi::DataType::UNDEFINED);
+}
+
+#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
+PD_REGISTER_KERNEL(sum,
+                   GPU,
+                   ALL_LAYOUT,
+                   phi::SumKernel,
+                   bool,
+                   float,
+                   double,
+                   phi::float16,
+                   phi::bfloat16,
+                   int16_t,
+                   int,
+                   int64_t,
+                   uint8_t,
+                   int8_t,
+                   phi::complex64,
+                   phi::complex128) {
+  kernel->OutputAt(0).SetDataType(phi::DataType::UNDEFINED);
+}
+#endif
+
+#if defined(PADDLE_WITH_XPU_KP) && !defined(PADDLE_WITH_XPU)
+PD_REGISTER_KERNEL(sum, KPS, ALL_LAYOUT, phi::SumKernel, float) {
+  kernel->OutputAt(0).SetDataType(phi::DataType::UNDEFINED);
+}
+#endif
+
+#if defined(PADDLE_WITH_DNNL)
+PD_REGISTER_KERNEL(sum, OneDNN, ONEDNN, phi::SumKernel, float, phi::bfloat16) {
+  kernel->check_if_onednn_kernel_support_ = phi::ReduceCheckIfOneDNNSupport;
+}
+#endif
+
+#if defined(PADDLE_WITH_XPU)
+PD_REGISTER_KERNEL(sum,
+                   XPU,
+                   ALL_LAYOUT,
+                   phi::SumKernel,
+                   float,
+                   phi::float16,
+                   phi::bfloat16,
+                   int8_t,
+                   int,
+                   int64_t,
+                   bool) {
+  kernel->OutputAt(0).SetDataType(phi::DataType::UNDEFINED);
+}
+#endif