Align conv2d behavior to flash attention's

Author: daniandtheweb

examples/cli/main.cpp

@@ -237,7 +237,9 @@ void print_usage(int argc, const char* argv[]) {
     printf("                                     Might lower quality, since it implies converting k and v to f16.\n");
     printf("                                     This might crash if it is not supported by the backend.\n");
     printf("  --diffusion-conv-direct            use Conv2D direct in the diffusion model");
+    printf("                                     This might crash if it is not supported by the backend.\n");
     printf("  --vae-conv-direct                  use Conv2D direct in the vae model (should improve the performance)");
+    printf("                                     This might crash if it is not supported by the backend.\n");
     printf("  --control-net-cpu                  keep controlnet in cpu (for low vram)\n");
     printf("  --canny                            apply canny preprocessor (edge detection)\n");
     printf("  --color                            colors the logging tags according to level\n");

ggml_extend.hpp

@@ -1518,15 +1518,8 @@ class Conv2d : public UnaryBlock {
         if (bias) {
             b = params["bias"];
         }
-        #if defined(SD_USE_OPENCL)
-            direct = true
-        #endif
         if (direct) {
-            #if defined(SD_USE_CUDA) || defined(SD_USE_SYCL) || defined(SD_USE_METAL)
-                return ggml_nn_conv_2d(ctx, x, w, b, stride.second, stride.first, padding.second, padding.first, dilation.second, dilation.first);
-            #else
-                return ggml_nn_conv_2d_direct(ctx, x, w, b, stride.second, stride.first, padding.second, padding.first, dilation.second, dilation.first);
-            #endif
+            return ggml_nn_conv_2d_direct(ctx, x, w, b, stride.second, stride.first, padding.second, padding.first, dilation.second, dilation.first);
         } else {
             return ggml_nn_conv_2d(ctx, x, w, b, stride.second, stride.first, padding.second, padding.first, dilation.second, dilation.first);
         }