fix

driver/device_convolution_implicit_gemm_v4_nchw_kcyx_nkhw.hpp

@@ -88,7 +88,7 @@ void device_convolution_implicit_gemm_v4_nchw_kcyx_nkhw(InDesc,
 #endif
 
     constexpr index_t GridSize =
-        ((B + BPerBlock - 1) / BPerBlock) * ((K + KPerBlock - 1) / KPerBlock);
+        ((B + BPerBlock - 1) / BPerBlock) * ((K + KPerBlock - 1) / KPerBlock) / (Strides{}.Get(I1) * Strides{}.Get(I0));
 
     printf("%s: BlockSize %u, GridSize %u \n", __func__, BlockSize, GridSize);
 

driver/driver.hip.cpp

@@ -16,6 +16,17 @@
 #include "device_convolution_implicit_gemm_v3_nchw_cyxk_nkhw.hpp"
 #include "device_convolution_implicit_gemm_v4_nchw_kcyx_nkhw.hpp"
 
+struct GeneratorTensor_0
+{
+    template <class... Is>
+    double operator()(Is... is)
+    {
+        return 0;
+    }
+};
+
+
+
 struct GeneratorTensor_1
 {
     template <class... Is>
@@ -196,12 +207,14 @@ void host_direct_convolution_back(Tensor<TOut>& in_nchw,
         {
             for(int y = 0; y < wei_kcyx.mDesc.GetLengths()[2]; ++y)
             {
-                int ho = (hi - y * dilation_h + h_pad_low) / stride_h;
+                int ho_ = (hi - y * dilation_h + h_pad_low);
+                int ho = ho_ / stride_h; 
                 for(int x = 0; x < wei_kcyx.mDesc.GetLengths()[3]; ++x)
                 {
-                    int wo = (wi - x * dilation_w + w_pad_low) / stride_w;
-                    if(ho >= 0 && hi < out_nkhw.mDesc.GetLengths()[2] && wo >= 0 &&
-                       wo < out_nkhw.mDesc.GetLengths()[3] && ho % stride_h == 0 && wo % stride_w == 0)
+                    int wo_ = (wi - x * dilation_w + w_pad_low);
+                    int wo = wo_ / stride_w;
+                    if(ho >= 0 && ho < out_nkhw.mDesc.GetLengths()[2] && wo >= 0 &&
+                       wo < out_nkhw.mDesc.GetLengths()[3] && ho_ % stride_h == 0 && wo_ % stride_w == 0)
                     {
                         v += double(out_nkhw(n, k, ho, wo)) * double(wei_kcyx(k, c, y, x));
                     }
@@ -489,12 +502,12 @@ int main(int argc, char* argv[])
     constexpr index_t WDilation = 1;
 
     constexpr index_t Direction = 2; //1: Forward; 2:Backward
-#if 1
+#if 0
     constexpr index_t N  = 8;
     constexpr index_t C  = 128;
-    constexpr index_t HI = 16;
-    constexpr index_t WI = 16;
-    constexpr index_t K  = 128;
+    constexpr index_t HI = 2;
+    constexpr index_t WI = 32;
+    constexpr index_t K  = 16;
     constexpr index_t Y  = 1;
     constexpr index_t X  = 1;
 
@@ -706,9 +719,9 @@ int main(int argc, char* argv[])
         in_nchw.GenerateTensorValue(GeneratorTensor_3{}, num_thread);
         wei_kcyx.GenerateTensorValue(GeneratorTensor_1{}, num_thread);
 #elif 1
-        in_nchw.GenerateTensorValue(GeneratorTensor_2{-5, 5}, num_thread);
-        out_nkhw.GenerateTensorValue(GeneratorTensor_2{-5, 5}, num_thread);
-        wei_kcyx.GenerateTensorValue(GeneratorTensor_2{-5, 5}, num_thread);
+        in_nchw.GenerateTensorValue(GeneratorTensor_0{}, num_thread);
+        out_nkhw.GenerateTensorValue(GeneratorTensor_1{}, num_thread);
+        wei_kcyx.GenerateTensorValue(GeneratorTensor_1{}, num_thread);
 #elif 0
         in_nchw.GenerateTensorValue(GeneratorTensor_2{1, 5}, num_thread);
 
@@ -785,10 +798,10 @@ int main(int argc, char* argv[])
         }
 
 #if 0
-        LogRange(std::cout << "in_nchw : ", in_nchw.mData, ",") << std::endl;
+        //LogRange(std::cout << "in_nchw : ", in_nchw.mData, ",") << std::endl;
         LogRange(std::cout << "wei_kcyx: ", wei_kcyx.mData, ",") << std::endl;
-        LogRange(std::cout << "out_nkhw_host  : ", out_nkhw_host.mData, ",") << std::endl;
-        LogRange(std::cout << "out_nkhw_device: ", out_nkhw_device.mData, ",") << std::endl;
+        LogRange(std::cout << "in_nchw_host  : ", in_nchw.mData, ",") << std::endl;
+        LogRange(std::cout << "in_nchw_device: ", in_nchw_device.mData, ",") << std::endl;
 #endif
     }
 }