Function main

samples/render_graph/rg-triangle/triangle.cpp:134–252

}

int main(int argc, char* argv[])
{
    if (SDL_Init(SDL_INIT_EVERYTHING) != 0) return -1;
    sdl_window = SDL_CreateWindow((const char*)gCGPUBackendNames[backend],
        SDL_WINDOWPOS_CENTERED, SDL_WINDOWPOS_CENTERED,
        BACK_BUFFER_WIDTH, BACK_BUFFER_HEIGHT,
        SDL_WINDOW_VULKAN | SDL_WINDOW_RESIZABLE);
        SDL_VERSION(&wmInfo.version);
    SDL_GetWindowWMInfo(sdl_window, &wmInfo);
    create_api_objects();
    create_render_pipeline();
    // initialize
    namespace render_graph = skr::render_graph;
    auto graph = render_graph::RenderGraph::create(
    [=](render_graph::RenderGraphBuilder& builder) {
        builder.with_device(device)
        .with_gfx_queue(gfx_queue);
    });
    // loop
    bool quit = false;
    while (!quit)
    {
        SDL_Event event;
        while (SDL_PollEvent(&event))
        {
            if (SDL_GetWindowID(sdl_window) == event.window.windowID)
            {
                if (!SDLEventHandler(&event, sdl_window))
                {
                    quit = true;
                }
            }
            if (event.type == SDL_WINDOWEVENT)
            {
                Uint8 window_event = event.window.event;
                if (window_event == SDL_WINDOWEVENT_SIZE_CHANGED)
                {
                    cgpu_wait_queue_idle(gfx_queue);
                    cgpu_free_swapchain(swapchain);
                    int width = 0, height = 0;
                    SDL_GetWindowSize(sdl_window, &width, &height);
                    CGPUSwapChainDescriptor chain_desc = {};
                    chain_desc.present_queues = &gfx_queue;
                    chain_desc.present_queues_count = 1;
                    chain_desc.width = width;
                    chain_desc.height = height;
                    chain_desc.surface = surface;
                    chain_desc.image_count = 3;
                    chain_desc.format = CGPU_FORMAT_R8G8B8A8_UNORM;
                    chain_desc.enable_vsync = true;
                    swapchain = cgpu_create_swapchain(device, &chain_desc);
                }
            }
        }
        // acquire frame
        cgpu_wait_fences(&present_fence, 1);
        CGPUAcquireNextDescriptor acquire_desc = {};
        acquire_desc.fence = present_fence;