Refactor LVGL code into kernel module (#472)

* **New Features**
  * Added a standalone LVGL module and enabled LVGL support in the simulator for richer local UI testing.

* **Refactor**
  * HAL and LVGL split into distinct modules; startup and device attach/detach flows reorganized for clearer lifecycle management.
  * Public APIs tightened with clearer nullability/documentation.

* **Bug Fixes**
  * More consistent LVGL start/stop and device attach/detach behavior for improved stability.
This commit is contained in:
Ken Van Hoeylandt
2026-02-01 22:57:45 +01:00
committed by GitHub
parent 3fe1dc0312
commit 9f721e6655
135 changed files with 1553 additions and 667 deletions
+3 -2
View File
@@ -27,9 +27,10 @@ void run(const Configuration& config, Module* platformModule, Module* deviceModu
/**
* While technically nullable, this instance is always set if tt_init() succeeds.
* Could return nullptr if init was not called.
* @return the Configuration instance that was passed on to tt_init() if init is successful
*/
const Configuration* _Nullable getConfiguration();
const Configuration* getConfiguration();
/** Provides access to the dispatcher that runs on the main task.
* @warning This dispatcher is used for WiFi and might block for some time during WiFi connection.
@@ -40,7 +41,7 @@ Dispatcher& getMainDispatcher();
namespace hal {
/** While technically this configuration is nullable, it's never null after initHeadless() is called. */
const Configuration* _Nullable getConfiguration();
const Configuration* getConfiguration();
} // namespace hal