Remove references to exec_aten::RuntimeContext

extension/llm/custom_ops/op_sdpa_aot.cpp

@@ -33,7 +33,7 @@ Tensor& sdpa_with_kv_cache_out_no_context(
     // @lint-ignore CLANGTIDY facebook-hte-ParameterMightThrowOnCopy
     const optional<double> scale,
     Tensor& output) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::native::sdpa_with_kv_cache_out(
       context,
       q_projected,

extension/llm/custom_ops/op_sdpa_test.cpp

@@ -28,7 +28,7 @@ exec_aten::Tensor op_scaled_dot_product_attention(
     bool is_causal,
     exec_aten::optional<double> scale,
     exec_aten::Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::native::flash_attention_kernel_out(
       context, query, key, value, attn_mask, dropout_p, is_causal, scale, out);
 }

extension/llm/custom_ops/op_sdpa_with_kv_cache_test.cpp

@@ -31,7 +31,7 @@ exec_aten::Tensor op_sdpa_with_kv_cache(
     bool is_causal,
     exec_aten::optional<double> scale,
     exec_aten::Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::native::sdpa_with_kv_cache_out(
       context,
       query,

kernels/aten/cpu/op__to_dim_order_copy.cpp

@@ -115,7 +115,7 @@ Tensor& _to_dim_order_copy_out(
     bool non_blocking,
     OptionalArrayRef<int64_t> dim_order,
     Tensor& out) {
-  exec_aten::RuntimeContext ctx{};
+  executorch::runtime::KernelRuntimeContext ctx{};
   return _to_dim_order_copy_out(ctx, self, non_blocking, dim_order, out);
 }
 

kernels/optimized/cpu/op_gelu.cpp

@@ -38,7 +38,7 @@ namespace {
  */
 template <typename CTYPE>
 void gelu(
-    exec_aten::RuntimeContext& context,
+    executorch::runtime::KernelRuntimeContext& context,
     const Tensor& input,
     string_view approximate,
     Tensor& output) {

kernels/portable/cpu/op__to_dim_order_copy.cpp

@@ -118,7 +118,7 @@ Tensor& _to_dim_order_copy_out(
     bool non_blocking,
     OptionalArrayRef<int64_t> dim_order,
     Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return _to_dim_order_copy_out(context, self, non_blocking, dim_order, out);
 }
 

kernels/portable/test/op_gelu_test.cpp

@@ -25,7 +25,7 @@ using torch::executor::testing::TensorFactory;
 // executorch/kernels/test/op_gelu_test.cpp instead.
 
 Tensor& op_gelu_out(const Tensor& self, string_view approximate, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::native::gelu_out(context, self, approximate, out);
 }
 

kernels/quantized/test/op_add_test.cpp

@@ -21,10 +21,10 @@
 using namespace ::testing;
 using exec_aten::ArrayRef;
 using exec_aten::optional;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::native::add_out;
 using torch::executor::native::dequantize_per_tensor_out;
 using torch::executor::native::quantize_per_tensor_out;
@@ -193,7 +193,7 @@ TEST(OpQuantizeAddTest, ConsitencyWithReferencePattern) {
 
   optional<ScalarType> out_dtype = optional<ScalarType>();
 
-  RuntimeContext context{};
+  KernelRuntimeContext context{};
   // q -> qadd -> dq
   // 3.5 / 0.5 + 1 = 8
   quantize_per_tensor_out(

kernels/quantized/test/op_embedding4b_test.cpp

@@ -19,9 +19,9 @@
 using namespace ::testing;
 using exec_aten::ArrayRef;
 using exec_aten::optional;
-using exec_aten::RuntimeContext;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::native::quantized_embedding_4bit_out;
 
 using torch::executor::testing::TensorFactory;
@@ -62,7 +62,7 @@ TEST(OpQuantizedEmbedding4bTest, TestGroupWiseQuantizedEmbedding) {
   EXPECT_TENSOR_EQ(out, expected);
 
   out = tf.zeros({3, 4});
-  auto context = RuntimeContext();
+  auto context = KernelRuntimeContext();
   torch::executor::native::quantized_embedding_4bit_out(
       context,
       qweight,

kernels/quantized/test/op_embedding_test.cpp

@@ -21,10 +21,10 @@
 using namespace ::testing;
 using exec_aten::ArrayRef;
 using exec_aten::optional;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::native::dequantize_per_tensor_out;
 using torch::executor::native::embedding_out;
 using torch::executor::native::quantize_per_tensor_out;
@@ -120,7 +120,7 @@ TEST(OpQuantizedEmbeddingTest, ConsitencyWithReferencePattern) {
 
   TensorFactory<ScalarType::Byte> tfo;
   Tensor qweight = tfo.zeros({3, 1});
-  RuntimeContext context{};
+  KernelRuntimeContext context{};
   // 3.5 / 0.5 + 1 = 8
   // 5.5 / 0.5 + 1 = 12
   // 1 / 0.5 + 1 = 3

kernels/quantized/test/op_mixed_linear_test.cpp

@@ -18,9 +18,9 @@
 
 using namespace ::testing;
 using exec_aten::optional;
-using exec_aten::RuntimeContext;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::native::quantized_mixed_linear_out;
 using torch::executor::testing::TensorFactory;
 
@@ -57,7 +57,7 @@ void test_dtype() {
       /*sizes=*/{1, 2},
       /*data=*/{2.3, 3.6});
 
-  RuntimeContext ctx{};
+  KernelRuntimeContext ctx{};
 
   quantized_mixed_linear_out(
       ctx, input, weight, weight_scales, opt_weight_zp, opt_dtype_out, out);
@@ -112,7 +112,7 @@ void test_dtype_partials() {
       {(1.0 * 5 + 1.5 * 3) * 0.2 + 2.0 * 1 * 1,
        (1.0 * 4 + 1.5 * 2) * 0.4 + 2.0 * 1 * 0.5});
 
-  RuntimeContext ctx{};
+  KernelRuntimeContext ctx{};
 
   quantized_mixed_linear_out(
       ctx, input, weight, weight_scales, opt_weight_zp, opt_dtype_out, out);

kernels/quantized/test/op_mixed_mm_test.cpp

@@ -18,9 +18,9 @@
 
 using namespace ::testing;
 using exec_aten::optional;
-using exec_aten::RuntimeContext;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::native::quantized_mixed_mm_out;
 using torch::executor::testing::TensorFactory;
 
@@ -55,7 +55,7 @@ void test_dtype() {
       /*sizes=*/{1, 2},
       /*data=*/{3.8, 3.0});
 
-  RuntimeContext ctx{};
+  KernelRuntimeContext ctx{};
 
   quantized_mixed_mm_out(ctx, input, weight, weight_scales, opt_weight_zp, out);
 

kernels/test/TestUtil.h

@@ -116,6 +116,6 @@ class OperatorTest : public ::testing::Test {
   }
 
  protected:
-  exec_aten::RuntimeContext context_;
+  executorch::runtime::KernelRuntimeContext context_;
   bool expect_failure_;
 };

kernels/test/custom_kernel_example/op_relu.cpp

@@ -18,8 +18,8 @@ namespace native {
 
 using Tensor = exec_aten::Tensor;
 using ScalarType = exec_aten::ScalarType;
-using exec_aten::RuntimeContext;
 using executor::Error;
+using executorch::runtime::KernelRuntimeContext;
 
 namespace {
 
@@ -61,7 +61,8 @@ void relu(const Tensor& input, Tensor& output) {
  *
  * relu.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)
  */
-Tensor& my_relu_out(RuntimeContext& context, const Tensor& input, Tensor& out) {
+Tensor&
+my_relu_out(KernelRuntimeContext& context, const Tensor& input, Tensor& out) {
   (void)context;
   resize(out, input.sizes());
   ET_KERNEL_CHECK(

kernels/test/op_atan2_test.cpp

@@ -23,7 +23,7 @@ using torch::executor::testing::SupportedFeatures;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_atan2_out(const Tensor& self, const Tensor& other, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::atan2_outf(context, self, other, out);
 }
 

kernels/test/op_cdist_forward_test.cpp

@@ -28,7 +28,7 @@ Tensor& op_cdist_forward_out(
     double p,
     optional<int64_t> compute_mode,
     Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::_cdist_forward_outf(
       context, x1, x2, p, compute_mode, out);
 }

kernels/test/op_clamp_test.cpp

@@ -260,7 +260,7 @@ class OpClampTensorOutTest : public OperatorTest {
       const optional<Tensor>& min,
       const optional<Tensor>& max,
       Tensor& out) {
-    exec_aten::RuntimeContext context{};
+    executorch::runtime::KernelRuntimeContext context{};
     return torch::executor::aten::clamp_outf(context, self, min, max, out);
   }
 };

kernels/test/op_diagonal_copy_test.cpp

@@ -27,7 +27,7 @@ Tensor& op_diagonal_copy_out(
     int64_t dim1,
     int64_t dim2,
     Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::diagonal_copy_outf(
       context, input, offset, dim1, dim2, out);
 }

kernels/test/op_expm1_test.cpp

@@ -23,7 +23,7 @@ using torch::executor::testing::SupportedFeatures;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_expm1_out(const Tensor& a, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::expm1_outf(context, a, out);
 }
 

kernels/test/op_flip_test.cpp

@@ -22,7 +22,7 @@ using exec_aten::Tensor;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_flip_out(const Tensor& input, IntArrayRef dims, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::flip_outf(context, input, dims, out);
 }
 

kernels/test/op_ge_test.cpp

@@ -15,10 +15,10 @@
 #include <gtest/gtest.h>
 
 using namespace ::testing;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::testing::TensorFactory;
 
 class OpGeTensorOutTest : public OperatorTest {

kernels/test/op_gt_test.cpp

@@ -15,10 +15,10 @@
 #include <gtest/gtest.h>
 
 using namespace ::testing;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::testing::TensorFactory;
 
 class OpGtScalarOutTest : public OperatorTest {

kernels/test/op_le_test.cpp

@@ -15,10 +15,10 @@
 #include <gtest/gtest.h>
 
 using namespace ::testing;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::testing::TensorFactory;
 
 class OpLeScalarOutTest : public OperatorTest {

kernels/test/op_log10_test.cpp

@@ -23,7 +23,7 @@ using torch::executor::testing::SupportedFeatures;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_log10_out(const Tensor& a, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::log10_outf(context, a, out);
 }
 

kernels/test/op_log1p_test.cpp

@@ -23,7 +23,7 @@ using torch::executor::testing::SupportedFeatures;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_log1p_out(const Tensor& a, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::log1p_outf(context, a, out);
 }
 

kernels/test/op_log2_test.cpp

@@ -23,7 +23,7 @@ using torch::executor::testing::SupportedFeatures;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_log2_out(const Tensor& a, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::log2_outf(context, a, out);
 }
 

kernels/test/op_lt_test.cpp

@@ -15,10 +15,10 @@
 #include <gtest/gtest.h>
 
 using namespace ::testing;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::testing::TensorFactory;
 
 class OpLtScalarOutTest : public OperatorTest {

kernels/test/op_maximum_test.cpp

@@ -21,7 +21,7 @@ using exec_aten::Tensor;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_maximum_out(const Tensor& self, const Tensor& other, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::maximum_outf(context, self, other, out);
 }
 

kernels/test/op_native_batch_norm_test.cpp

@@ -61,7 +61,7 @@ class OpNativeBatchNormLegitOutTest : public OperatorTest {
       exec_aten::Tensor& out0,
       exec_aten::Tensor& out1,
       exec_aten::Tensor& out2) {
-    exec_aten::RuntimeContext context{};
+    executorch::runtime::KernelRuntimeContext context{};
     return torch::executor::aten::_native_batch_norm_legit_outf(
         context,
         input,

kernels/test/op_native_group_norm_test.cpp

@@ -32,7 +32,7 @@ ::std::tuple<Tensor&, Tensor&, Tensor&> op_native_group_norm_out(
     Tensor& out0,
     Tensor& out1,
     Tensor& out2) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::native_group_norm_outf(
       context, input, weight, bias, N, C, HxW, group, eps, out0, out1, out2);
 }

kernels/test/op_ne_test.cpp

@@ -15,10 +15,10 @@
 #include <gtest/gtest.h>
 
 using namespace ::testing;
-using exec_aten::RuntimeContext;
 using exec_aten::Scalar;
 using exec_aten::ScalarType;
 using exec_aten::Tensor;
+using executorch::runtime::KernelRuntimeContext;
 using torch::executor::testing::TensorFactory;
 
 class OpNeTest : public OperatorTest {
@@ -34,7 +34,7 @@ class OpNeTest : public OperatorTest {
     Tensor a = tf_input.make(/*sizes=*/{2, 2}, /*data=*/{2, 3, 2, 4});
     Tensor b = tf_input.make({2, 2}, {2, 2, 2, 2});
     Tensor out = tf_bool.zeros({2, 2});
-    RuntimeContext context{};
+    KernelRuntimeContext context{};
 
     torch::executor::aten::ne_outf(context, a, b, out);
     EXPECT_TENSOR_EQ(out, tf_bool.make({2, 2}, {false, true, false, true}));

kernels/test/op_pdist_forward_test.cpp

@@ -22,7 +22,7 @@ using exec_aten::Tensor;
 using torch::executor::testing::TensorFactory;
 
 Tensor& op_pdist_forward_out(const Tensor& input, double p, Tensor& out) {
-  exec_aten::RuntimeContext context{};
+  executorch::runtime::KernelRuntimeContext context{};
   return torch::executor::aten::_pdist_forward_outf(context, input, p, out);
 }