GLES2 2d Batch rendering

2d rendering is currently bottlenecked by drawing rects one at a time, limiting OpenGL efficiency. This PR batches rects and renders in fewer drawcalls, resulting in significant performance improvements. This also speeds up text rendering.

The code dynamically chooses between a vertex format with and without color, depending on the input data for a frame, in order to optimize throughput and maximize batch size.

drivers/gles2/rasterizer_array.h

@@ -0,0 +1,170 @@
+#pragma once
+
+/*************************************************************************/
+/*  rasterizer_array.h                                                   */
+/*************************************************************************/
+/*                       This file is part of:                           */
+/*                           GODOT ENGINE                                */
+/*                      https://godotengine.org                          */
+/*************************************************************************/
+/* Copyright (c) 2007-2020 Juan Linietsky, Ariel Manzur.                 */
+/* Copyright (c) 2014-2020 Godot Engine contributors (cf. AUTHORS.md).   */
+/*                                                                       */
+/* Permission is hereby granted, free of charge, to any person obtaining */
+/* a copy of this software and associated documentation files (the       */
+/* "Software"), to deal in the Software without restriction, including   */
+/* without limitation the rights to use, copy, modify, merge, publish,   */
+/* distribute, sublicense, and/or sell copies of the Software, and to    */
+/* permit persons to whom the Software is furnished to do so, subject to */
+/* the following conditions:                                             */
+/*                                                                       */
+/* The above copyright notice and this permission notice shall be        */
+/* included in all copies or substantial portions of the Software.       */
+/*                                                                       */
+/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,       */
+/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF    */
+/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/
+/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY  */
+/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,  */
+/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE     */
+/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.                */
+/*************************************************************************/
+
+/**
+ * Fast single-threaded growable array for POD types.
+ * For use in render drivers, not for general use.
+*/
+
+#include "core/os/memory.h"
+#include "core/vector.h"
+
+#include <string.h>
+
+template <class T>
+class RasterizerArray {
+public:
+	RasterizerArray() {
+		_list = 0;
+		_size = 0;
+		_max_size = 0;
+	}
+	~RasterizerArray() { free(); }
+
+	T &operator[](unsigned int ui) { return _list[ui]; }
+	const T &operator[](unsigned int ui) const { return _list[ui]; }
+
+	void free() {
+		if (_list) {
+			memdelete_arr(_list);
+			_list = 0;
+		}
+		_size = 0;
+		_max_size = 0;
+	}
+
+	void create(int p_size) {
+		free();
+		_list = memnew_arr(T, p_size);
+		_size = 0;
+		_max_size = p_size;
+	}
+
+	void reset() { _size = 0; }
+
+	// none of that inefficient pass by value stuff here, thanks
+	T *request() {
+		if (_size < _max_size) {
+			return &_list[_size++];
+		}
+		return 0;
+	}
+
+	// four verts at a time
+	T *request_four() {
+		int old_size = _size;
+		_size += 4;
+
+		if (_size <= _max_size) {
+			return &_list[old_size];
+		}
+
+		// revert
+		_size = old_size;
+		return 0;
+	}
+
+	int size() const { return _size; }
+	int max_size() const { return _max_size; }
+	const T *get_data() const { return _list; }
+
+	bool copy_from(const RasterizerArray<T> &o) {
+		// no resizing done here, it should be done manually
+		if (o.size() > _max_size)
+			return false;
+
+		// pod types only please!
+		memcpy(_list, o.get_data(), o.size() * sizeof(T));
+		_size = o.size();
+		return true;
+	}
+
+	// if you want this to be cheap, call reset before grow,
+	// to ensure there is no data to copy
+	void grow() {
+		unsigned int new_max_size = _max_size * 2;
+		T *new_list = memnew_arr(T, new_max_size);
+
+		// copy .. pod types only
+		memcpy(new_list, _list, _size * sizeof(T));
+
+		unsigned int new_size = size();
+		free();
+		_list = new_list;
+		_size = new_size;
+		_max_size = new_max_size;
+	}
+
+private:
+	T *_list;
+	int _size;
+	int _max_size;
+};
+
+template <class T>
+class RasterizerArray_non_pod {
+public:
+	RasterizerArray_non_pod() {
+		_size = 0;
+	}
+
+	const T &operator[](unsigned int ui) const { return _list[ui]; }
+
+	void create(int p_size) {
+		_list.resize(p_size);
+		_size = 0;
+	}
+	void reset() { _size = 0; }
+
+	void push_back(const T &val) {
+		while (true) {
+			if (_size < max_size()) {
+				_list.set(_size, val);
+				_size++;
+				return;
+			}
+
+			grow();
+		}
+	}
+
+	int size() const { return _size; }
+	int max_size() const { return _list.size(); }
+
+private:
+	void grow() {
+		_list.resize(_list.size() * 2);
+	}
+
+	Vector<T> _list;
+	int _size;
+};