Formatting

src/include/migraphx/type_name.hpp

@@ -18,7 +18,7 @@ const std::string& get_type_name()
         name = typeid(PrivateMigraphTypeNameProbe).name();
         name = name.substr(7);
 #else
-        const char parameter_name[] = "PrivateMigraphTypeNameProbe =";// NOLINT
+        const char parameter_name[] = "PrivateMigraphTypeNameProbe ="; // NOLINT
 
         name = __PRETTY_FUNCTION__;
 

src/onnx/mnist.cpp

@@ -36,8 +36,8 @@ read_mnist_images(const std::string& full_path, int& number_of_images, int& imag
     if(file.is_open())
     {
         int magic_number = 0;
-        int n_rows = 0;
-        int n_cols = 0;
+        int n_rows       = 0;
+        int n_cols       = 0;
 
         file.read(reinterpret_cast<char*>(&magic_number), sizeof(magic_number));
         magic_number = reverse_int(magic_number);

src/targets/gpu/abs.cpp

@@ -19,7 +19,7 @@ argument miopen_abs::compute(context& ctx,
                              const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

src/targets/gpu/batchnorm.cpp

@@ -23,7 +23,7 @@ argument miopen_batch_norm_inference::compute(context& ctx,
     auto bn_desc = make_tensor(args[3].get_shape());
 
     float alpha = 1.0;
-    float beta = 0.0f;
+    float beta  = 0.0f;
 
     miopenBatchNormalizationForwardInference(ctx.get_stream().get_miopen(),
                                              miopenBatchNormMode_t(op.bn_mode),

src/targets/gpu/convolution.cpp

@@ -22,7 +22,7 @@ argument miopen_convolution::compute(context& ctx,
     auto y_desc = make_tensor(output_shape);
 
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     miopenConvolutionForward(ctx.get_stream().get_miopen(),
                              &alpha,
                              x_desc.get(),

src/targets/gpu/elu.cpp

@@ -19,7 +19,7 @@ argument miopen_elu::compute(context& ctx,
                              const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

src/targets/gpu/fuse_ops.cpp

@@ -266,7 +266,7 @@ struct miopen_conv_bias
     {
         auto fargs  = make_fused_args();
         float alpha = 1;
-        float beta = 0;
+        float beta  = 0;
         miopenSetOpArgsConvForward(fargs.get(), conv, &alpha, &beta, args[1].implicit());
         miopenSetOpArgsBiasForward(fargs.get(), bias, &alpha, &beta, args[3].implicit());
         return f.execute(ctx, fargs, args[0], args[4]);
@@ -310,7 +310,7 @@ struct miopen_conv_bias_relu
     {
         auto fargs  = make_fused_args();
         float alpha = 1;
-        float beta = 0;
+        float beta  = 0;
         miopenSetOpArgsConvForward(fargs.get(), conv, &alpha, &beta, args[1].implicit());
         miopenSetOpArgsBiasForward(fargs.get(), bias, &alpha, &beta, args[3].implicit());
         miopenSetOpArgsActivForward(fargs.get(), relu, &alpha, &beta, 0, 0, 0);

src/targets/gpu/leaky_relu.cpp

@@ -19,7 +19,7 @@ argument miopen_leaky_relu::compute(context& ctx,
                                     const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

src/targets/gpu/pooling.cpp

@@ -21,7 +21,7 @@ argument miopen_pooling::compute(context& ctx,
     auto y_desc = make_tensor(output_shape);
 
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
 
     miopenPoolingForward(ctx.get_stream().get_miopen(),
                          pd.get(),

src/targets/gpu/relu.cpp

@@ -19,7 +19,7 @@ argument miopen_relu::compute(context& ctx,
                               const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

src/targets/gpu/sigmoid.cpp

@@ -19,7 +19,7 @@ argument miopen_sigmoid::compute(context& ctx,
                                  const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

src/targets/gpu/softmax.cpp

@@ -19,7 +19,7 @@ argument miopen_softmax::compute(context& ctx,
                                  const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenSoftmaxForward(ctx.get_stream().get_miopen(),

src/targets/gpu/tanh.cpp

@@ -19,7 +19,7 @@ argument miopen_tanh::compute(context& ctx,
                               const std::vector<argument>& args) const
 {
     float alpha = 1;
-    float beta = 0;
+    float beta  = 0;
     auto x_desc = make_tensor(args[0].get_shape());
     auto y_desc = make_tensor(output_shape);
     miopenActivationForward(ctx.get_stream().get_miopen(),

test/cpu_ops_test.cpp

@@ -333,15 +333,15 @@ TEST_CASE(im2col_3x3_with_padding_test)
 TEST_CASE(batch_norm_inference_test)
 {
     migraphx::program p;
-    const size_t width = 2;
-    const size_t height = 2;
-    const size_t channels = 4;
-    const size_t batches = 2;
-    const float x_val = 8.0;
-    const float mean_val = 2.0;
+    const size_t width       = 2;
+    const size_t height      = 2;
+    const size_t channels    = 4;
+    const size_t batches     = 2;
+    const float x_val        = 8.0;
+    const float mean_val     = 2.0;
     const float variance_val = 4.0;
-    const float scale_val = 2.0f;
-    const float bias_val   = 1.0f;
+    const float scale_val    = 2.0f;
+    const float bias_val     = 1.0f;
     const float output_val = scale_val * (x_val - mean_val) / (std::sqrt(variance_val)) + bias_val;
 
     migraphx::shape s{migraphx::shape::float_type, {batches, channels, height, width}};

test/matcher.cpp

@@ -385,7 +385,10 @@ struct match_find_sum
     migraphx::instruction_ref ins;
     auto matcher() const { return match::name("sum"); }
 
-    void apply(migraphx::program&, const match::matcher_result& r) const { EXPECT(bool{r.result == ins}); }
+    void apply(migraphx::program&, const match::matcher_result& r) const
+    {
+        EXPECT(bool{r.result == ins});
+    }
 };
 
 struct match_find_literal