rename device convolution file and function name

host/driver_offline/include/device_convolution_backward_weight_implicit_gemm_v4r4r3_xdlops_nchw_kcyx_nkhw.hpp → host/driver_offline/include/device_convolution_backward_weight_implicit_gemm_v4r4r2_xdlops_atomic_nchw_kcyx_nkhw.hpp

@@ -14,7 +14,7 @@ template <typename TInWei,
           typename ConvDilations,
           typename InLeftPads,
           typename InRightPads>
-void device_convolution_backward_weight_implicit_gemm_v4r4r3_xdlops_nchw_kcyx_nkhw(
+void device_convolution_backward_weight_implicit_gemm_v4r4r2_xdlops_atomic_nchw_kcyx_nkhw(
     const InLengths& in_n_c_hi_wi_lengths,
     const WeiLengths& wei_k_c_y_x_lengths,
     const OutLengths& out_n_k_ho_wo_lengths,

host/driver_offline/include/driver_gemm_xdlops_v2r4.hpp

@@ -128,9 +128,7 @@ __host__ float driver_gemm_xdlops_v2r4(const FloatAB* p_a_grid,
     // const auto kbatch = GridwiseGemm::CalculateKBatch(c_m_n_grid_desc, b_k0_n_k1_grid_desc);
     const auto a_b_k0_m_k1_grid_desc = GridwiseGemm::MakeABK0MK1GridDescriptor(a_k0_m_k1_grid_desc);
     const auto b_b_k0_n_k1_grid_desc = GridwiseGemm::MakeBBK0NK1GridDescriptor(b_k0_n_k1_grid_desc);
-    {
-        // std::cout << "k batch number is: " << kbatch << std::endl;
-    }
+
     if(!GridwiseGemm::CheckValidity(a_k0_m_k1_grid_desc, b_k0_n_k1_grid_desc, c_m_n_grid_desc))
     {
         throw std::runtime_error(
@@ -150,7 +148,7 @@ __host__ float driver_gemm_xdlops_v2r4(const FloatAB* p_a_grid,
 
     const index_t grid_size = GridwiseGemm::CalculateGridSize(c_m_n_grid_desc);
     {
-        std::cout << "gridSize : " << grid_size << grid_size << std::endl;
+        std::cout << "gridSize : " << grid_size << std::endl;
     }
     const auto kernel = kernel_gemm_xdlops_v2r4<GridwiseGemm,
                                                 FloatAB,

host/driver_offline/src/conv_wrw_driver_offline.cpp

@@ -13,16 +13,16 @@
 #include "host_conv_bwd_weight.hpp"
 #include "device_tensor.hpp"
 #include "device_convolution_backward_weight_implicit_gemm_v4r4r2_xdlops_nchw_kcyx_nkhw.hpp"
-#include "device_convolution_backward_weight_implicit_gemm_v4r4r3_xdlops_nchw_kcyx_nkhw.hpp"
+#include "device_convolution_backward_weight_implicit_gemm_v4r4r2_xdlops_atomic_nchw_kcyx_nkhw.hpp"
 
 #define USE_DYNAMIC_MODE 1
 #define USE_CONV_WRW_V4R4R2_XDL_NCHW 1
-#define USE_CONV_WRW_V4R4R3_XDL_NCHW 1
+#define USE_CONV_WRW_V4R4R2_XDL_ATOMIC_NCHW 1
 
 enum ConvBackwardWeightAlgo
 {
     V4R4R2XDLNCHW,
-    V4R4R3XDLNCHW,
+    V4R4R2XDLATOMICNCHW,
 };
 
 int main(int argc, char* argv[])
@@ -262,8 +262,8 @@ int main(int argc, char* argv[])
     }
 #endif
 
-#if USE_CONV_WRW_V4R4R3_XDL_NCHW
-    if(algo == ConvBackwardWeightAlgo::V4R4R3XDLNCHW)
+#if USE_CONV_WRW_V4R4R2_XDL_ATOMIC_NCHW
+    if(algo == ConvBackwardWeightAlgo::V4R4R2XDLATOMICNCHW)
     {
         if(layout != ConvTensorLayout::NCHW)
         {
@@ -272,20 +272,20 @@ int main(int argc, char* argv[])
 
         const auto tmp = f_make_for_device_nchw();
 
-        device_convolution_backward_weight_implicit_gemm_v4r4r3_xdlops_nchw_kcyx_nkhw<in_data_t,
-                                                                                      acc_data_t,
-                                                                                      out_data_t>(
-            tmp[I0],
-            tmp[I1],
-            tmp[I2],
-            tmp[I3],
-            tmp[I4],
-            tmp[I5],
-            tmp[I6],
-            in,
-            wei_device,
-            out,
-            nrepeat);
+        device_convolution_backward_weight_implicit_gemm_v4r4r2_xdlops_atomic_nchw_kcyx_nkhw<
+            in_data_t,
+            acc_data_t,
+            out_data_t>(tmp[I0],
+                        tmp[I1],
+                        tmp[I2],
+                        tmp[I3],
+                        tmp[I4],
+                        tmp[I5],
+                        tmp[I6],
+                        in,
+                        wei_device,
+                        out,
+                        nrepeat);
     }
 #endif