llama : print devices used on model load

Author: slaren

ggml/src/ggml-backend.cpp

@@ -555,7 +555,6 @@ struct ggml_backend_registry {
         register_backend(ggml_backend_cuda_reg());
 #endif
 #ifdef GGML_USE_METAL
-
         register_backend(ggml_backend_metal_reg());
 #endif
 #ifdef GGML_USE_VULKAN
@@ -565,7 +564,7 @@ struct ggml_backend_registry {
         register_backend(ggml_backend_blas_reg());
 #endif
 
-        // TODO: sycl, vulkan, kompute, cann
+        // TODO: sycl, kompute, cann
 
         register_backend(ggml_backend_cpu_reg());
     }

src/llama.cpp

@@ -19100,8 +19100,13 @@ struct llama_model * llama_load_model_from_file(
 
             case GGML_BACKEND_DEVICE_TYPE_GPU:
             case GGML_BACKEND_DEVICE_TYPE_GPU_FULL:
+            {
+                size_t free, total; // NOLINT
+                ggml_backend_dev_memory(dev, &free, &total);
+                LLAMA_LOG_INFO("%s: using device %s (%s) - %zu MiB free\n", __func__, ggml_backend_dev_name(dev), ggml_backend_dev_description(dev), free/1024/1024);
                 model->devices.push_back(dev);
                 break;
+            }
         }
     }