Apply clang-format on repository

Code is formatted as per a revised clang format configuration
file(not part of this delivery). Version 14.0.6 is used.

Exclusion List:
- files with .cl extension
- files that are not strictly C/C++ (e.g. Android.bp, Sconscript ...)
And the following directories
- compute_kernel_writer/validation/
- tests/
- include/
- src/core/NEON/kernels/convolution/
- src/core/NEON/kernels/arm_gemm/
- src/core/NEON/kernels/arm_conv/
- data/

There will be a follow up for formatting of .cl files and the
files under tests/ and compute_kernel_writer/validation/.

Signed-off-by: Felix Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
Change-Id: Ib7eb1fcf4e7537b9feaefcfc15098a804a3fde0a
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10391
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>

5 files changed, 401 insertions, 180 deletions

diff --git a/src/cpu/kernels/elementwise_binary/generic/sve/fp16.cpp b/src/cpu/kernels/elementwise_binary/generic/sve/fp16.cpp
index 85224351df..600c7f1c05 100644
--- a/src/cpu/kernels/elementwise_binary/generic/sve/fp16.cpp
+++ b/src/cpu/kernels/elementwise_binary/generic/sve/fp16.cpp
@@ -25,6 +25,7 @@
 #if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
 
 #include "arm_compute/core/Helpers.h"
+
 #include "src/cpu/kernels/elementwise_binary/generic/sve/impl.h"
 namespace arm_compute
 {
@@ -36,14 +37,38 @@ void sve_fp16_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor
     return elementwise_arithmetic_op<float16_t>(in1, in2, out, op, window);
 }
 
-template void sve_fp16_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1,
+                                                                             const ITensor *in2,
+                                                                             ITensor       *out,
+                                                                             const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1,
+                                                                      const ITensor *in2,
+                                                                      ITensor       *out,
+                                                                      const Window  &window);
+template void sve_fp16_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1,
+                                                                      const ITensor *in2,
+                                                                      ITensor       *out,
+                                                                      const Window  &window);
 
 template <ComparisonOperation op>
 void sve_fp16_comparison_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
@@ -51,14 +76,32 @@ void sve_fp16_comparison_elementwise_binary(const ITensor *in1, const ITensor *i
     return elementwise_comparison_op<float16_t>(in1, in2, out, op, window);
 }
 
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1,
+                                                                                 const ITensor *in2,
+                                                                                 ITensor       *out,
+                                                                                 const Window  &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1,
+                                                                                    const ITensor *in2,
+                                                                                    ITensor       *out,
+                                                                                    const Window  &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1,
+                                                                                   const ITensor *in2,
+                                                                                   ITensor       *out,
+                                                                                   const Window  &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1,
+                                                                                        const ITensor *in2,
+                                                                                        ITensor       *out,
+                                                                                        const Window  &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1,
+                                                                                const ITensor *in2,
+                                                                                ITensor       *out,
+                                                                                const Window  &window);
+template void sve_fp16_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1,
+                                                                                     const ITensor *in2,
+                                                                                     ITensor       *out,
+                                                                                     const Window  &window);
 
 } // namespace cpu
 } // namespace arm_compute
 
-#endif /* defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS) */
\ No newline at end of file
+#endif /* defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS) */
diff --git a/src/cpu/kernels/elementwise_binary/generic/sve/fp32.cpp b/src/cpu/kernels/elementwise_binary/generic/sve/fp32.cpp
index 2b479f76f1..832a966883 100644
--- a/src/cpu/kernels/elementwise_binary/generic/sve/fp32.cpp
+++ b/src/cpu/kernels/elementwise_binary/generic/sve/fp32.cpp
@@ -23,6 +23,7 @@
  */
 
 #include "arm_compute/core/Helpers.h"
+
 #include "src/cpu/kernels/elementwise_binary/generic/sve/impl.h"
 namespace arm_compute
 {
@@ -34,26 +35,68 @@ void sve_fp32_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor
     return elementwise_arithmetic_op<float32_t>(in1, in2, out, op, window);
 }
 
-template void sve_fp32_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1,
+                                                                    const ITensor *in2,
+                                                                    ITensor       *out,
+                                                                    const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1,
+                                                                             const ITensor *in2,
+                                                                             ITensor       *out,
+                                                                             const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1,
+                                                                      const ITensor *in2,
+                                                                      ITensor       *out,
+                                                                      const Window  &window);
+template void sve_fp32_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1,
+                                                                      const ITensor *in2,
+                                                                      ITensor       *out,
+                                                                      const Window  &window);
 
 template <ComparisonOperation op>
 void sve_fp32_comparison_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
 {
     return elementwise_comparison_op<float>(in1, in2, out, op, window);
 }
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1,
+                                                                                 const ITensor *in2,
+                                                                                 ITensor       *out,
+                                                                                 const Window  &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1,
+                                                                                    const ITensor *in2,
+                                                                                    ITensor       *out,
+                                                                                    const Window  &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1,
+                                                                                   const ITensor *in2,
+                                                                                   ITensor       *out,
+                                                                                   const Window  &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1,
+                                                                                        const ITensor *in2,
+                                                                                        ITensor       *out,
+                                                                                        const Window  &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1,
+                                                                                const ITensor *in2,
+                                                                                ITensor       *out,
+                                                                                const Window  &window);
+template void sve_fp32_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1,
+                                                                                     const ITensor *in2,
+                                                                                     ITensor       *out,
+                                                                                     const Window  &window);
 
 } // namespace cpu
 } // namespace arm_compute
diff --git a/src/cpu/kernels/elementwise_binary/generic/sve/impl.cpp b/src/cpu/kernels/elementwise_binary/generic/sve/impl.cpp
index c0515f2abc..fa48407e9b 100644
--- a/src/cpu/kernels/elementwise_binary/generic/sve/impl.cpp
+++ b/src/cpu/kernels/elementwise_binary/generic/sve/impl.cpp
@@ -23,7 +23,9 @@
  */
 
 #include "src/cpu/kernels/elementwise_binary/generic/sve/impl.h"
+
 #include "src/core/NEON/SVEMath.h"
+
 #include <arm_sve.h>
 
 namespace arm_compute
@@ -33,7 +35,8 @@ namespace cpu
 using namespace arm_compute::wrapper;
 
 template <typename ScalarType>
-void elementwise_arithmetic_op(const ITensor *in1, const ITensor *in2, ITensor *out, ArithmeticOperation op, const Window &window)
+void elementwise_arithmetic_op(
+    const ITensor *in1, const ITensor *in2, ITensor *out, ArithmeticOperation op, const Window &window)
 {
     using VectorType = typename sve_vector<ScalarType>::type;
 
@@ -51,7 +54,7 @@ void elementwise_arithmetic_op(const ITensor *in1, const ITensor *in2, ITensor *
     const auto window_end_x          = static_cast<int>(window.x().end());
     const bool is_broadcast_across_x = in1->info()->tensor_shape().x() != in2->info()->tensor_shape().x();
 
-    if(is_broadcast_across_x)
+    if (is_broadcast_across_x)
     {
         const bool     is_broadcast_input_2 = input2_win.x().step() == 0;
         Window         broadcast_win        = is_broadcast_input_2 ? input2_win : input1_win;
@@ -66,37 +69,40 @@ void elementwise_arithmetic_op(const ITensor *in1, const ITensor *in2, ITensor *
         Iterator non_broadcast_input(non_broadcast_tensor, non_broadcast_win);
         Iterator output(out, win);
 
-        execute_window_loop(win, [&](const Coordinates &)
-        {
-            auto             output_ptr              = reinterpret_cast<ScalarType *>(output.ptr());
-            const auto       non_broadcast_input_ptr = reinterpret_cast<const ScalarType *>(non_broadcast_input.ptr());
-            const ScalarType broadcast_value         = *reinterpret_cast<const ScalarType *>(broadcast_input.ptr());
-            const auto       broadcast_vector        = svdup_n(broadcast_value);
-
-            int x = window_start_x;
-
-            svbool_t pg = svwhilelt<ScalarType>(x, window_end_x);
-            do
+        execute_window_loop(
+            win,
+            [&](const Coordinates &)
             {
-                const auto non_broadcast_vector = svld1(pg, non_broadcast_input_ptr + x);
-                VectorType res{};
+                auto       output_ptr              = reinterpret_cast<ScalarType *>(output.ptr());
+                const auto non_broadcast_input_ptr = reinterpret_cast<const ScalarType *>(non_broadcast_input.ptr());
+                const ScalarType broadcast_value   = *reinterpret_cast<const ScalarType *>(broadcast_input.ptr());
+                const auto       broadcast_vector  = svdup_n(broadcast_value);
 
-                if(is_broadcast_input_2)
-                {
-                    res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(pg, non_broadcast_vector, broadcast_vector, op);
-                }
-                else
+                int x = window_start_x;
+
+                svbool_t pg = svwhilelt<ScalarType>(x, window_end_x);
+                do
                 {
-                    res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(pg, broadcast_vector, non_broadcast_vector, op);
-                }
-                svst1(pg, output_ptr + x, res);
-
-                x += svcnt<ScalarType>();
-                pg = svwhilelt<ScalarType>(x, window_end_x);
-            }
-            while(svptest_any(all_true_pg, pg));
-        },
-        broadcast_input, non_broadcast_input, output);
+                    const auto non_broadcast_vector = svld1(pg, non_broadcast_input_ptr + x);
+                    VectorType res{};
+
+                    if (is_broadcast_input_2)
+                    {
+                        res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(pg, non_broadcast_vector,
+                                                                                               broadcast_vector, op);
+                    }
+                    else
+                    {
+                        res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(
+                            pg, broadcast_vector, non_broadcast_vector, op);
+                    }
+                    svst1(pg, output_ptr + x, res);
+
+                    x += svcnt<ScalarType>();
+                    pg = svwhilelt<ScalarType>(x, window_end_x);
+                } while (svptest_any(all_true_pg, pg));
+            },
+            broadcast_input, non_broadcast_input, output);
     }
     else
     {
@@ -108,39 +114,46 @@ void elementwise_arithmetic_op(const ITensor *in1, const ITensor *in2, ITensor *
         Iterator input2(in2, input2_win);
         Iterator output(out, win);
 
-        execute_window_loop(win, [&](const Coordinates &)
-        {
-            auto       output_ptr = reinterpret_cast<ScalarType *>(output.ptr());
-            const auto input1_ptr = reinterpret_cast<const ScalarType *>(input1.ptr());
-            const auto input2_ptr = reinterpret_cast<const ScalarType *>(input2.ptr());
+        execute_window_loop(
+            win,
+            [&](const Coordinates &)
+            {
+                auto       output_ptr = reinterpret_cast<ScalarType *>(output.ptr());
+                const auto input1_ptr = reinterpret_cast<const ScalarType *>(input1.ptr());
+                const auto input2_ptr = reinterpret_cast<const ScalarType *>(input2.ptr());
 
-            int x = window_start_x;
+                int x = window_start_x;
 
-            svbool_t pg = svwhilelt<ScalarType>(x, window_end_x);
-            do
-            {
-                const auto in1 = svld1(pg, input1_ptr + x);
-                const auto in2 = svld1(pg, input2_ptr + x);
-                const auto res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(pg, in1, in2, op);
-                svst1(pg, output_ptr + x, res);
-
-                x += svcnt<ScalarType>();
-                pg = svwhilelt<ScalarType>(x, window_end_x);
-            }
-            while(svptest_any(all_true_pg, pg));
-        },
-        input1, input2, output);
+                svbool_t pg = svwhilelt<ScalarType>(x, window_end_x);
+                do
+                {
+                    const auto in1 = svld1(pg, input1_ptr + x);
+                    const auto in2 = svld1(pg, input2_ptr + x);
+                    const auto res = elementwise_arithmetic_op<typename sve_vector<ScalarType>::type>(pg, in1, in2, op);
+                    svst1(pg, output_ptr + x, res);
+
+                    x += svcnt<ScalarType>();
+                    pg = svwhilelt<ScalarType>(x, window_end_x);
+                } while (svptest_any(all_true_pg, pg));
+            },
+            input1, input2, output);
     }
 }
-template void elementwise_arithmetic_op<float32_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
-template void elementwise_arithmetic_op<float16_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
-template void elementwise_arithmetic_op<int16_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
-template void elementwise_arithmetic_op<int32_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
+template void elementwise_arithmetic_op<float32_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
+template void elementwise_arithmetic_op<float16_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
+template void elementwise_arithmetic_op<int16_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
+template void elementwise_arithmetic_op<int32_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ArithmeticOperation op, const Window &window);
 
 template <typename InputScalarType, typename OutputScalarType>
-void elementwise_comparison_op(const ITensor *in1, const ITensor *in2, ITensor *out, ComparisonOperation op, const Window &window)
+void elementwise_comparison_op(
+    const ITensor *in1, const ITensor *in2, ITensor *out, ComparisonOperation op, const Window &window)
 {
-    static_assert(sizeof(InputScalarType) >= sizeof(OutputScalarType), "input data type's width should be equal to or greater than output data type's width");
+    static_assert(sizeof(InputScalarType) >= sizeof(OutputScalarType),
+                  "input data type's width should be equal to or greater than output data type's width");
 
     using OutputVectorType = typename sve_vector<OutputScalarType>::type;
     const auto all_true_pg = svptrue<InputScalarType>();
@@ -157,7 +170,7 @@ void elementwise_comparison_op(const ITensor *in1, const ITensor *in2, ITensor *
     const auto window_end_x          = static_cast<int>(window.x().end());
     const bool is_broadcast_across_x = in1->info()->tensor_shape().x() != in2->info()->tensor_shape().x();
 
-    if(is_broadcast_across_x)
+    if (is_broadcast_across_x)
     {
         const bool     is_broadcast_input_2 = input2_win.x().step() == 0;
         Window         broadcast_win        = is_broadcast_input_2 ? input2_win : input1_win;
@@ -172,37 +185,44 @@ void elementwise_comparison_op(const ITensor *in1, const ITensor *in2, ITensor *
         Iterator non_broadcast_input(non_broadcast_tensor, non_broadcast_win);
         Iterator output(out, win);
 
-        execute_window_loop(win, [&](const Coordinates &)
-        {
-            auto                  output_ptr              = reinterpret_cast<OutputScalarType *>(output.ptr());
-            const auto            non_broadcast_input_ptr = reinterpret_cast<const InputScalarType *>(non_broadcast_input.ptr());
-            const InputScalarType broadcast_value         = *reinterpret_cast<const InputScalarType *>(broadcast_input.ptr());
-            const auto            broadcast_vector        = svdup_n(broadcast_value);
+        execute_window_loop(
+            win,
+            [&](const Coordinates &)
+            {
+                auto       output_ptr = reinterpret_cast<OutputScalarType *>(output.ptr());
+                const auto non_broadcast_input_ptr =
+                    reinterpret_cast<const InputScalarType *>(non_broadcast_input.ptr());
+                const InputScalarType broadcast_value =
+                    *reinterpret_cast<const InputScalarType *>(broadcast_input.ptr());
+                const auto broadcast_vector = svdup_n(broadcast_value);
 
-            int x = window_start_x;
+                int x = window_start_x;
 
-            svbool_t pg = svwhilelt<InputScalarType>(x, window_end_x);
-            do
-            {
-                const auto       non_broadcast_vector = svld1(pg, non_broadcast_input_ptr + x);
-                const svbool_t   output_pg            = narrow_to_byte_predicate<sizeof(InputScalarType)>(pg);
-                OutputVectorType res{};
-                if(is_broadcast_input_2)
-                {
-                    res = elementwise_comparison_op<typename sve_vector<InputScalarType>::type, typename sve_vector<OutputScalarType>::type>(pg, non_broadcast_vector, broadcast_vector, op);
-                }
-                else
+                svbool_t pg = svwhilelt<InputScalarType>(x, window_end_x);
+                do
                 {
-                    res = elementwise_comparison_op<typename sve_vector<InputScalarType>::type, typename sve_vector<OutputScalarType>::type>(pg, broadcast_vector, non_broadcast_vector, op);
-                }
-                svst1(output_pg, output_ptr + x, res);
-
-                x += svcnt<InputScalarType>();
-                pg = svwhilelt<InputScalarType>(x, window_end_x);
-            }
-            while(svptest_any(all_true_pg, pg));
-        },
-        broadcast_input, non_broadcast_input, output);
+                    const auto       non_broadcast_vector = svld1(pg, non_broadcast_input_ptr + x);
+                    const svbool_t   output_pg            = narrow_to_byte_predicate<sizeof(InputScalarType)>(pg);
+                    OutputVectorType res{};
+                    if (is_broadcast_input_2)
+                    {
+                        res = elementwise_comparison_op<typename sve_vector<InputScalarType>::type,
+                                                        typename sve_vector<OutputScalarType>::type>(
+                            pg, non_broadcast_vector, broadcast_vector, op);
+                    }
+                    else
+                    {
+                        res = elementwise_comparison_op<typename sve_vector<InputScalarType>::type,
+                                                        typename sve_vector<OutputScalarType>::type>(
+                            pg, broadcast_vector, non_broadcast_vector, op);
+                    }
+                    svst1(output_pg, output_ptr + x, res);
+
+                    x += svcnt<InputScalarType>();
+                    pg = svwhilelt<InputScalarType>(x, window_end_x);
+                } while (svptest_any(all_true_pg, pg));
+            },
+            broadcast_input, non_broadcast_input, output);
     }
     else
     {
@@ -214,37 +234,45 @@ void elementwise_comparison_op(const ITensor *in1, const ITensor *in2, ITensor *
         Iterator input2(in2, input2_win);
         Iterator output(out, win);
 
-        execute_window_loop(win, [&](const Coordinates &)
-        {
-            auto       output_ptr = reinterpret_cast<OutputScalarType *>(output.ptr());
-            const auto input1_ptr = reinterpret_cast<const InputScalarType *>(input1.ptr());
-            const auto input2_ptr = reinterpret_cast<const InputScalarType *>(input2.ptr());
+        execute_window_loop(
+            win,
+            [&](const Coordinates &)
+            {
+                auto       output_ptr = reinterpret_cast<OutputScalarType *>(output.ptr());
+                const auto input1_ptr = reinterpret_cast<const InputScalarType *>(input1.ptr());
+                const auto input2_ptr = reinterpret_cast<const InputScalarType *>(input2.ptr());
 
-            int x = window_start_x;
+                int x = window_start_x;
 
-            svbool_t pg = svwhilelt<InputScalarType>(x, window_end_x);
-            do
-            {
-                const auto     in1       = svld1(pg, input1_ptr + x);
-                const auto     in2       = svld1(pg, input2_ptr + x);
-                const auto     res       = elementwise_comparison_op<typename sve_vector<InputScalarType>::type, typename sve_vector<OutputScalarType>::type>(pg, in1, in2, op);
-                const svbool_t output_pg = narrow_to_byte_predicate<sizeof(InputScalarType)>(pg);
-                svst1(output_pg, output_ptr + x, res);
-
-                x += svcnt<InputScalarType>();
-                pg = svwhilelt<InputScalarType>(x, window_end_x);
-            }
-            while(svptest_any(all_true_pg, pg));
-        },
-        input1, input2, output);
+                svbool_t pg = svwhilelt<InputScalarType>(x, window_end_x);
+                do
+                {
+                    const auto in1 = svld1(pg, input1_ptr + x);
+                    const auto in2 = svld1(pg, input2_ptr + x);
+                    const auto res =
+                        elementwise_comparison_op<typename sve_vector<InputScalarType>::type,
+                                                  typename sve_vector<OutputScalarType>::type>(pg, in1, in2, op);
+                    const svbool_t output_pg = narrow_to_byte_predicate<sizeof(InputScalarType)>(pg);
+                    svst1(output_pg, output_ptr + x, res);
+
+                    x += svcnt<InputScalarType>();
+                    pg = svwhilelt<InputScalarType>(x, window_end_x);
+                } while (svptest_any(all_true_pg, pg));
+            },
+            input1, input2, output);
     }
 }
 
-template void elementwise_comparison_op<float32_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
-template void elementwise_comparison_op<float16_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
-template void elementwise_comparison_op<uint8_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
-template void elementwise_comparison_op<int16_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
-template void elementwise_comparison_op<int32_t>(const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
+template void elementwise_comparison_op<float32_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
+template void elementwise_comparison_op<float16_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
+template void elementwise_comparison_op<uint8_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
+template void elementwise_comparison_op<int16_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
+template void elementwise_comparison_op<int32_t>(
+    const ITensor *in1, const ITensor *in2, ITensor *out, const ComparisonOperation op, const Window &window);
 
 template <>
 svint32_t elementwise_pow<svint32_t>(svbool_t &pg, const svint32_t &a, const svint32_t &b)
diff --git a/src/cpu/kernels/elementwise_binary/generic/sve/impl.h b/src/cpu/kernels/elementwise_binary/generic/sve/impl.h
index 860c50a1e0..4c61b9f315 100644
--- a/src/cpu/kernels/elementwise_binary/generic/sve/impl.h
+++ b/src/cpu/kernels/elementwise_binary/generic/sve/impl.h
@@ -25,6 +25,7 @@
 #define SRC_CORE_SVE_KERNELS_ELEMENTWISE_LIST_H
 
 #include "arm_compute/core/Helpers.h"
+
 #include "src/core/NEON/wrapper/intrinsics/intrinsics.h"
 #include "src/core/NEON/wrapper/svtraits.h"
 
@@ -51,7 +52,7 @@ svbool_t narrow_to_byte_predicate(svbool_t pg)
 {
     const auto all_false = svpfalse();
 
-    switch(bytewidth)
+    switch (bytewidth)
     {
         case 8:
             pg = svuzp1_b32(pg, all_false);
@@ -74,7 +75,7 @@ VectorType elementwise_arithmetic_op(svbool_t &pg, const VectorType &a, const Ve
     using ScalarType = typename wrapper::sve_scalar<VectorType>::type;
     VectorType res{};
 
-    switch(op)
+    switch (op)
     {
         case ArithmeticOperation::MAX:
             res = svmax_z(pg, a, b);
@@ -114,11 +115,12 @@ VectorType elementwise_arithmetic_op(svbool_t &pg, const VectorType &a, const Ve
 }
 
 template <typename InputVectorType, typename OutputVectorType>
-OutputVectorType elementwise_comparison_op(svbool_t &pg, const InputVectorType &a, const InputVectorType &b, ComparisonOperation op)
+OutputVectorType
+elementwise_comparison_op(svbool_t &pg, const InputVectorType &a, const InputVectorType &b, ComparisonOperation op)
 {
     svbool_t selection_vector{};
 
-    switch(op)
+    switch (op)
     {
         case ComparisonOperation::Equal:
             selection_vector = svcmpeq(pg, a, b);
@@ -154,10 +156,12 @@ OutputVectorType elementwise_comparison_op(svbool_t &pg, const InputVectorType &
 }
 
 template <typename ScalarType>
-void elementwise_arithmetic_op(const ITensor *in1, const ITensor *in2, ITensor *out, ArithmeticOperation op, const Window &window);
+void elementwise_arithmetic_op(
+    const ITensor *in1, const ITensor *in2, ITensor *out, ArithmeticOperation op, const Window &window);
 
 template <typename ScalarType, typename OutputScalarType = uint8_t>
-void elementwise_comparison_op(const ITensor *in1, const ITensor *in2, ITensor *out, ComparisonOperation op, const Window &window);
+void elementwise_comparison_op(
+    const ITensor *in1, const ITensor *in2, ITensor *out, ComparisonOperation op, const Window &window);
 } // namespace cpu
 } // namespace arm_compute
 #endif /* SRC_CORE_SVE_KERNELS_ELEMENTWISE_LIST_H */
diff --git a/src/cpu/kernels/elementwise_binary/generic/sve/integer.cpp b/src/cpu/kernels/elementwise_binary/generic/sve/integer.cpp
index c313fc6e04..f7714ff7e9 100644
--- a/src/cpu/kernels/elementwise_binary/generic/sve/integer.cpp
+++ b/src/cpu/kernels/elementwise_binary/generic/sve/integer.cpp
@@ -23,6 +23,7 @@
  */
 
 #include "arm_compute/core/Helpers.h"
+
 #include "src/cpu/kernels/elementwise_binary/generic/sve/impl.h"
 namespace arm_compute
 {
@@ -33,64 +34,166 @@ void sve_s32_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor
 {
     return elementwise_arithmetic_op<int32_t>(in1, in2, out, op, window);
 }
-template void sve_s32_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1,
+                                                                            const ITensor *in2,
+                                                                            ITensor       *out,
+                                                                            const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1,
+                                                                     const ITensor *in2,
+                                                                     ITensor       *out,
+                                                                     const Window  &window);
+template void sve_s32_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1,
+                                                                     const ITensor *in2,
+                                                                     ITensor       *out,
+                                                                     const Window  &window);
 
 template <ArithmeticOperation op>
 void sve_s16_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
 {
     return elementwise_arithmetic_op<int16_t>(in1, in2, out, op, window);
 }
-template void sve_s16_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::ADD>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::SUB>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::DIV>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::MIN>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::MAX>(const ITensor *in1,
+                                                                   const ITensor *in2,
+                                                                   ITensor       *out,
+                                                                   const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::SQUARED_DIFF>(const ITensor *in1,
+                                                                            const ITensor *in2,
+                                                                            ITensor       *out,
+                                                                            const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::POWER>(const ITensor *in1,
+                                                                     const ITensor *in2,
+                                                                     ITensor       *out,
+                                                                     const Window  &window);
+template void sve_s16_elementwise_binary<ArithmeticOperation::PRELU>(const ITensor *in1,
+                                                                     const ITensor *in2,
+                                                                     ITensor       *out,
+                                                                     const Window  &window);
 
 template <ComparisonOperation op>
 void sve_u8_comparison_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
 {
     return elementwise_comparison_op<uint8_t>(in1, in2, out, op, window);
 }
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_u8_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1,
+                                                                               const ITensor *in2,
+                                                                               ITensor       *out,
+                                                                               const Window  &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1,
+                                                                                  const ITensor *in2,
+                                                                                  ITensor       *out,
+                                                                                  const Window  &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1,
+                                                                                 const ITensor *in2,
+                                                                                 ITensor       *out,
+                                                                                 const Window  &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1,
+                                                                                      const ITensor *in2,
+                                                                                      ITensor       *out,
+                                                                                      const Window  &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1,
+                                                                              const ITensor *in2,
+                                                                              ITensor       *out,
+                                                                              const Window  &window);
+template void sve_u8_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1,
+                                                                                   const ITensor *in2,
+                                                                                   ITensor       *out,
+                                                                                   const Window  &window);
 
 template <ComparisonOperation op>
 void sve_s16_comparison_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
 {
     return elementwise_comparison_op<int16_t>(in1, in2, out, op, window);
 }
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s16_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1,
+                                                                                const ITensor *in2,
+                                                                                ITensor       *out,
+                                                                                const Window  &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1,
+                                                                                   const ITensor *in2,
+                                                                                   ITensor       *out,
+                                                                                   const Window  &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1,
+                                                                                  const ITensor *in2,
+                                                                                  ITensor       *out,
+                                                                                  const Window  &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1,
+                                                                                       const ITensor *in2,
+                                                                                       ITensor       *out,
+                                                                                       const Window  &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1,
+                                                                               const ITensor *in2,
+                                                                               ITensor       *out,
+                                                                               const Window  &window);
+template void sve_s16_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1,
+                                                                                    const ITensor *in2,
+                                                                                    ITensor       *out,
+                                                                                    const Window  &window);
 
 template <ComparisonOperation op>
 void sve_s32_comparison_elementwise_binary(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window)
 {
     return elementwise_comparison_op<int32_t>(in1, in2, out, op, window);
 }
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
-template void sve_s32_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Equal>(const ITensor *in1,
+                                                                                const ITensor *in2,
+                                                                                ITensor       *out,
+                                                                                const Window  &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::NotEqual>(const ITensor *in1,
+                                                                                   const ITensor *in2,
+                                                                                   ITensor       *out,
+                                                                                   const Window  &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Greater>(const ITensor *in1,
+                                                                                  const ITensor *in2,
+                                                                                  ITensor       *out,
+                                                                                  const Window  &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::GreaterEqual>(const ITensor *in1,
+                                                                                       const ITensor *in2,
+                                                                                       ITensor       *out,
+                                                                                       const Window  &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::Less>(const ITensor *in1,
+                                                                               const ITensor *in2,
+                                                                               ITensor       *out,
+                                                                               const Window  &window);
+template void sve_s32_comparison_elementwise_binary<ComparisonOperation::LessEqual>(const ITensor *in1,
+                                                                                    const ITensor *in2,
+                                                                                    ITensor       *out,
+                                                                                    const Window  &window);
 
 } // namespace cpu
 } // namespace arm_compute