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>

1 files changed, 30 insertions, 12 deletions

diff --git a/src/runtime/heuristics/matmul_native/ClMatMulNativeHelpers.cpp b/src/runtime/heuristics/matmul_native/ClMatMulNativeHelpers.cpp
index 1e06e84d4d..89cad30214 100644
--- a/src/runtime/heuristics/matmul_native/ClMatMulNativeHelpers.cpp
+++ b/src/runtime/heuristics/matmul_native/ClMatMulNativeHelpers.cpp
@@ -26,6 +26,7 @@
 #include "arm_compute/core/KernelDescriptors.h"
 #include "arm_compute/core/TensorInfo.h"
 #include "arm_compute/core/TensorShape.h"
+
 #include "src/gpu/cl/kernels/ClMatMulNativeKernel.h"
 
 #include <limits>
@@ -37,22 +38,32 @@ namespace cl_matmul
 {
 MatMulKernelInfo select_info(const MatMulKernelInfo &info0,
                              const MatMulKernelInfo &info1,
-                             unsigned int m, unsigned int n, unsigned int k, unsigned int b, DataType data_type, bool rhs_lock_padding)
+                             unsigned int            m,
+                             unsigned int            n,
+                             unsigned int            k,
+                             unsigned int            b,
+                             DataType                data_type,
+                             bool                    rhs_lock_padding)
 {
-    ARM_COMPUTE_ERROR_ON_MSG(info1.export_rhs_to_cl_image == true, "The fallback MatMul configuration cannot have export_to_cl_image = true");
-    ARM_COMPUTE_ERROR_ON_MSG(info0.adj_lhs != info1.adj_lhs, "The MatMul configurations must have the same adj_lhs value");
-    ARM_COMPUTE_ERROR_ON_MSG(info0.adj_rhs != info1.adj_rhs, "The MatMul configurations must have the same adj_rhs value");
+    ARM_COMPUTE_ERROR_ON_MSG(info1.export_rhs_to_cl_image == true,
+                             "The fallback MatMul configuration cannot have export_to_cl_image = true");
+    ARM_COMPUTE_ERROR_ON_MSG(info0.adj_lhs != info1.adj_lhs,
+                             "The MatMul configurations must have the same adj_lhs value");
+    ARM_COMPUTE_ERROR_ON_MSG(info0.adj_rhs != info1.adj_rhs,
+                             "The MatMul configurations must have the same adj_rhs value");
 
     const bool adj_lhs = info0.adj_lhs;
     const bool adj_rhs = info0.adj_rhs;
 
-    TensorInfo lhs_info = !adj_lhs ? TensorInfo(TensorShape(k, m, b), 1, data_type) : TensorInfo(TensorShape(m, k, b), 1, data_type);
-    TensorInfo rhs_info = !adj_rhs ? TensorInfo(TensorShape(n, k, b), 1, data_type) : TensorInfo(TensorShape(k, n, b), 1, data_type);
+    TensorInfo lhs_info =
+        !adj_lhs ? TensorInfo(TensorShape(k, m, b), 1, data_type) : TensorInfo(TensorShape(m, k, b), 1, data_type);
+    TensorInfo rhs_info =
+        !adj_rhs ? TensorInfo(TensorShape(n, k, b), 1, data_type) : TensorInfo(TensorShape(k, n, b), 1, data_type);
     TensorInfo dst_info;
 
-    if(rhs_lock_padding == false)
+    if (rhs_lock_padding == false)
     {
-        if(bool(opencl::kernels::ClMatMulNativeKernel::validate(&lhs_info, &rhs_info, nullptr, &dst_info, info0)))
+        if (bool(opencl::kernels::ClMatMulNativeKernel::validate(&lhs_info, &rhs_info, nullptr, &dst_info, info0)))
         {
             return info0;
         }
@@ -67,7 +78,13 @@ MatMulKernelInfo select_info(const MatMulKernelInfo &info0,
     }
 }
 
-MatMulKernelInfo find_info(const MatMulNativeConfigsMatrix &configs, bool adj_lhs, bool adj_rhs, unsigned int m, unsigned int n, unsigned int k, unsigned int b)
+MatMulKernelInfo find_info(const MatMulNativeConfigsMatrix &configs,
+                           bool                             adj_lhs,
+                           bool                             adj_rhs,
+                           unsigned int                     m,
+                           unsigned int                     n,
+                           unsigned int                     k,
+                           unsigned int                     b)
 {
     size_t min_acc = std::numeric_limits<size_t>::max();
     size_t min_idx = 0;
@@ -76,12 +93,13 @@ MatMulKernelInfo find_info(const MatMulNativeConfigsMatrix &configs, bool adj_lh
     const size_t num_rows = configs.size();
     const size_t num_cols = configs[0].size();
 
-    ARM_COMPUTE_ERROR_ON_MSG(num_cols != 8U, "The entry should have 8 integer values representing: M, N, K, B, M0, N0. K0, IMG_RHS");
+    ARM_COMPUTE_ERROR_ON_MSG(num_cols != 8U,
+                             "The entry should have 8 integer values representing: M, N, K, B, M0, N0. K0, IMG_RHS");
     ARM_COMPUTE_UNUSED(num_cols);
 
     // Find nearest GeMM workload
     // Note: the workload does not depend on the K dimension
-    for(size_t y = 0; y < num_rows; ++y)
+    for (size_t y = 0; y < num_rows; ++y)
     {
         size_t mc0 = static_cast<size_t>(configs[y][0]);
         size_t nc0 = static_cast<size_t>(configs[y][1]);
@@ -94,7 +112,7 @@ MatMulKernelInfo find_info(const MatMulNativeConfigsMatrix &configs, bool adj_lh
         acc += (k - kc0) * (k - kc0);
         acc += (b - bc0) * (b - bc0);
         acc = std::sqrt(acc);
-        if(acc < min_acc)
+        if (acc < min_acc)
         {
             min_acc = acc;
             min_idx = y;