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 +1,67 @@
+// Copyright (c) 2022 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/api/ext/dispatch.h"
+#include "paddle/phi/backends/cpu/cpu_context.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/cpu/elementwise.h"
+#include "paddle/phi/kernels/impl/elementwise_kernel_impl.h"
+
+namespace phi {
+
+template <typename T, typename Context>
+void DivideKernel(const Context& dev_ctx,
+                  const DenseTensor& x,
+                  const DenseTensor& y,
+                  DenseTensor* out) {
+  if (x.numel() == 0 || y.numel() == 0) {
+    dev_ctx.template Alloc<T>(out);
+    return;
+  }
+  dev_ctx.template Alloc<T>(out);
+  if (x.dims() == y.dims() && std::is_floating_point<T>::value) {
+    SameDimsElementwiseCompute<SameDimsDivideFunctor<CPUContext, T>>()(
+        dev_ctx, x, y, out);
+  } else {
+    auto x_dims = x.dims();
+    auto y_dims = y.dims();
+    if (x_dims.size() >= y_dims.size()) {  // NOLINT
+      funcs::ElementwiseCompute<funcs::DivideFunctor<T>, T>(
+          dev_ctx, x, y, funcs::DivideFunctor<T>(), out, -1);
+    } else {
+      funcs::ElementwiseCompute<funcs::InverseDivideFunctor<T>, T>(
+          dev_ctx, x, y, funcs::InverseDivideFunctor<T>(), out, -1);
+    }
+  }
+}
+
+}  // namespace phi
+
+// NOTE(chenweihang): using bfloat16 will cause redefine with xpu bfloat16
+// using bfloat16 = ::phi::bfloat16;
+
+PD_REGISTER_KERNEL(divide,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::DivideKernel,
+                   float,
+                   double,
+                   int8_t,
+                   uint8_t,
+                   int16_t,
+                   int,
+                   int64_t,
+                   bool,
+                   phi::complex64,
+                   phi::complex128) {}