Ordered the header file inclusion. Used allocate_temp instead of malloc in mean operator

Author: [email protected]

backends/cadence/fusion_g3/operators/op_cat.cpp

@@ -7,11 +7,14 @@
  */
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
+
+#include <cstring>
+
+#include <xa_nnlib_kernels_api.h>
+
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/util/copy_ops_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <xa_nnlib_kernels_api.h>
-#include <cstring>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_dequantize.cpp

@@ -8,14 +8,15 @@
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
 
+#include <algorithm>
+#include <cinttypes>
+#include <cmath>
+
 #include <xa_nnlib_kernels_api.h>
 
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/util/reduce_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <algorithm>
-#include <cinttypes>
-#include <cmath>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_div.cpp

@@ -8,6 +8,8 @@
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
 
+#include <cmath>
+
 #include <xa_nnlib_kernels_api.h>
 
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
@@ -16,7 +18,6 @@
 #include <executorch/kernels/portable/cpu/util/math_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
 #include <executorch/runtime/platform/assert.h>
-#include <cmath>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_exp.cpp

@@ -7,11 +7,14 @@
  */
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
+
+#include <cmath>
+
+#include <xa_nnlib_kernels_api.h>
+
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/pattern/pattern.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <xa_nnlib_kernels_api.h>
-#include <cmath>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_mean.cpp

@@ -146,8 +146,10 @@ Tensor& mean_dim_out(
       scratch_size *= inp_shape[i];
     }
 
-    void* __restrict__ p_scratch_in =
-        (void* __restrict__)malloc(scratch_size * sizeof(float));
+    executorch::runtime::Result<void*> temp_mem =
+        ctx.allocate_temp(scratch_size * sizeof(float));
+
+    void* __restrict__ p_scratch_in = (void* __restrict__)(temp_mem.get());
 
     XT_KERNEL_CHECK(
         ctx,

backends/cadence/fusion_g3/operators/op_native_layer_norm.cpp

@@ -7,13 +7,16 @@
  */
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
+
+#include <cmath>
+#include <tuple>
+
+#include <xa_nnlib_kernels_api.h>
+
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/util/normalization_ops_util.h>
 #include <executorch/kernels/portable/cpu/vec_ops.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <xa_nnlib_kernels_api.h>
-#include <cmath>
-#include <tuple>
 
 using Tensor = ::executorch::aten::Tensor;
 using ScalarType = ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_permute_copy.cpp

@@ -11,7 +11,6 @@
 #include <xa_nnlib_kernels_api.h>
 
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
-
 #include <executorch/kernels/portable/cpu/util/copy_ops_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
 

backends/cadence/fusion_g3/operators/op_quantize.cpp

@@ -6,12 +6,12 @@
  * LICENSE file in the root directory of this source tree.
  */
 
+#include <executorch/backends/cadence/fusion_g3/operators/operators.h>
+
 #include <algorithm>
 #include <cinttypes>
 #include <cmath>
 
-#include <executorch/backends/cadence/fusion_g3/operators/operators.h>
-
 #include <xa_nnlib_kernels_api.h>
 
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>

backends/cadence/fusion_g3/operators/op_slice_copy.cpp

@@ -8,12 +8,13 @@
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
 
+#include <cstring>
+
 #include <xa_nnlib_kernels_api.h>
 
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/util/slice_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <cstring>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;

backends/cadence/fusion_g3/operators/op_softmax.cpp

@@ -7,13 +7,16 @@
  */
 
 #include <executorch/backends/cadence/fusion_g3/operators/operators.h>
+
+#include <cmath>
+
+#include <xa_nnlib_kernels_api.h>
+
 #include <executorch/backends/cadence/fusion_g3/operators/xt_macros.h>
 #include <executorch/kernels/portable/cpu/util/activation_ops_util.h>
 #include <executorch/kernels/portable/cpu/util/functional_util.h>
 #include <executorch/kernels/portable/cpu/util/reduce_util.h>
 #include <executorch/runtime/kernel/kernel_includes.h>
-#include <xa_nnlib_kernels_api.h>
-#include <cmath>
 
 using ::executorch::aten::Scalar;
 using ::executorch::aten::ScalarType;