[core] Thread GLObjectStore through to Holder objects

This eliminates the reliance on ThreadContext to provide GLObjectStore, and statically enforces that GL cleanup functions happen only when GLObjectStore::performCleanup is called.

With the elimination of the Map thread, this becomes important because there may be multiple GLObjectStore's per-thread, and Map will need to ensure that the correct context is active when calling GLObjectStore::performCleanup.

1 files changed, 2 insertions, 2 deletions

diff --git a/src/mbgl/sprite/sprite_atlas.hpp b/src/mbgl/sprite/sprite_atlas.hpp
index ff9cb44e4d..ba592c5acb 100644
--- a/src/mbgl/sprite/sprite_atlas.hpp
+++ b/src/mbgl/sprite/sprite_atlas.hpp
@@ -52,14 +52,14 @@ public:
     optional<SpriteAtlasPosition> getPosition(const std::string& name, bool repeating = false);
 
     // Binds the atlas texture to the GPU, and uploads data if it is out of date.
-    void bind(bool linear = false);
+    void bind(bool linear, gl::GLObjectStore&);
 
     // Updates sprites in the atlas texture that may have changed in the source SpriteStore object.
     void updateDirty();
 
     // Uploads the texture to the GPU to be available when we need it. This is a lazy operation;
     // the texture is only bound when the data is out of date (=dirty).
-    void upload();
+    void upload(gl::GLObjectStore&);
 
     inline dimension getWidth() const { return width; }
     inline dimension getHeight() const { return height; }